Proceedings of the Annual Meeting of the Cognitive Science Society

Variation in second language acquisition is evident from earliest stages. This study examined effects of learning tasks (retrieval practice, comprehension, verbal repetition) on comprehension of Turkish as a new language. Undergraduates (N = 156) engaged with Turkish spoken dialogues in a computer-assisted language learning session via Zoom, with learning tasks manipulated between-subjects. Participants completed pre/posttests assessing comprehension of Turkish number and case marking, a vocabulary test, and open-response questions gauging explicit awareness. The retrieval-practice group showed highest performance overall, after controlling for significant effects of nonverbal ability and pretest. For comprehension of number/case marking, the comprehension group performed comparably to the retrieval-practice group. For vocabulary comprehension, the verbal-repetition group performed comparably to the retrieval-practice group. Differential performance associated with learning tasks indicates benefits of testing and production and aligns with transfer-appropriate processing. As predicted by the noticing hypothesis, explicit awareness of number and case marking correlated with comprehension accuracy.

From the logical point of view, abduction is a procedure in which something that lacks classical explanatory epistemic virtue can be accepted because it has virtue of another kind: the GW-Model contends that abduction presents an ignorancepreserving or (ignorance-mitigating) character. From this perspective abductive reasoning is a response to an ignoranceproblem. Is abduction really ignorance-preserving? To better answer this question I will take advantage of my eco-cognitive model (EC-model) of abduction. It will be illustrated, also thanks to cognitive and epistemological considerations, that through abduction, knowledge can be enhanced, even when abduction is not considered an inference to the best explanation in the classical sense of the expression, that is an inference necessarily characterized by an empirical evaluation phase.

Along with using geometric information to define spatial terms such as “in”, Coventry and Garrod (2004) have proposed the use of dynamic-kinematic (DK) routines which relate to how two objects interact (kinematic) over time (dynamic). For the spatial description, “The penny is in the bowl,” the penny is contained by a bounding box around the bowl as well as the DK location-control routine that if the bowl moves, so will the penny. In two experiments a speeded prime/ probe picturesentence verification task was used to gather evidence for the existence of DK routines. The first experiment found evidence for priming location-control for the preposition “in”. The second experiment examined location-control routines for both “in” and “on” using the same prepositions for both location-control and embedded spatial relationships. A significant response time benefit for priming location-control routines was found. These experiments provided evidence for priming location-control routines independent of semantic priming.

Despite an extensive literature on the brain substrates of response inhibition, it still remains to be determined which is the best functional comparison to tease apart neural activity underlying this cognitive process. Here we aimed to shed light on this issue be recording event-related potentials while participants performed a modified stop-signal task that allowed us to compare the following conditions: successful versus unsuccessful response inhibitions, successful response inhibitions versus successful response executions, and easy versus difficult response inhibitions. Electrophysiological activity related to response inhibition was best isolated by comparing easy and difficult inhibitions. This activity was observed at fronto-central scalp electrodes between 260 and 300 milliseconds after stop-stimulus presentation. Notably, the stop-signal reaction time (an estimate of the time required to inhibit the motor response) fell within this window interval.

How much information does a large-scale cortical network process when it’s conscious and/or unconscious? Can the complexity of such networks be quantified and be coupled to brain function and consciousness? Recently, measures of network complexity such as integrated information have been proposed. However, we show that these approaches are computationally intractable for realistic brain networks. We propose alternative quantifications that allow precise computations for large-scale networks including their stochastic dynamics, plasticity and perturbations. Even for stable stationary dynamics our measure shows that the processed information of a realistic network sharply rises at the edge of criticality. In particular, we demonstrate that the specific topology of the human brain generates greater informational complexity compared to randomly rewired networks. We analyze to what extent these results and their associated measures are specific to levels of consciousness or simply a hallmark of how neuronal systems process information.

Cognitive representations of musical structure have long been of interest to psychologists and musicians. This project addresses comprehension of and long-term memory for musical form, a debated topic in music cognition. We use three methods: a corpus of Billboard magazine’s top 10 songs for each of the last 20 years; a probabilistic grammar derived from this corpus; and an experiment testing predictions of the grammar. Two statistical analyses of the corpus are presented here, dealing with its zero- and first-order Markov properties. These provide a probablistic grammar of form in popular songs. We have tested this grammar by prompted recall of listeners’ memory for popular songs they claim to know well. Recalls average over 70% correct; errors in these recalls most often correspond to low-frequency 2-tuples in our networks. Our results show that listeners learn statistical regularities of form in popular music, much as they learn melodic and harmonic structure.

Mathematical problem solving requires cognitive flexibility. Bilingual children show advantages in cognitive flexibility as compared to monolingual children. Whether this bilingual cognitive advantage extends to mathematical skills is the focus of the present study. To measure children’s use of cognitive flexibility, worksheets containing 60 arithmetic problems of different operations were administered to first- through fifth-grade monolingual and bilingual children; children were given 60 seconds to complete as many grade-appropriate problems as possible. Performance on the math worksheet was analyzed as the number of problems completed and solved correctly. Results indicate that bilingualism affects arithmetic problem solving in third- through fifth-graders (problems completed: F(1,24)=9.20, p<.01; accuracy: F(1,24)=3.30, p=.08) but not first- and second-graders (problems completed: F(1,38)=0.38, p=.54; accuracy: F(1,38)=0.83, p=.37). Findings from this study thus suggest that bilingual cognitive flexibility extends to mathematical problem solving, and that this cognitive flexibility develops over the elementary school years in bilingual children.

All languages have ways of expressing location in time, but they differ widely in their grammatical tense systems. At the same time, there are tense systems that recur across unrelated languages. What explains this wide but constrained variation? Taking a functionalist perspective, we propose that tense systems are shaped by the need to support efficient communication–a need that has recently been shown to explain cross-language semantic variation in other domains. We test this proposal computationally against the tense systems of 64 languages. We find that most languages in the sample support near-optimally efficient communication, but with some interesting and potentially illuminating exceptions. We conclude that efficient communication may play an important role in explaining why tense systems vary across languages in the ways they do.

A number of frameworks for capturing insight phenomena have been proposed, but there are no executable models of knowledge-lean insight problem-solving. Here, an ACT-R model is presented for the nine-dot problem, which implements the Criterion for Satisfactory Progress theory for this problem. The model has two main components: a mechanism for searching for possible moves in the problem representation, and a mechanism for expanding the search to discover new moves not immediately available in the initial problem representation. The model accounts for key phenomena including impasse, fixation and the ‘aha’ moment, as well as predicting the relative difficulty of different problem variants.

The blue circle on smartphone displays is an everyday visualization of uncertainty; with the circle size indicating uncertainty of one’s location. Like error bars on graphs, it is a discrete visualization of a graded probability function. Two experiments examined the effectiveness of different visualizations of location estimates varying whether and how uncertainty was visualized (uniform blue circle showing confidence interval, faded circle showing graded probability, or both). Given a known location and visualizations of the estimates of two “smartphones” of that location, participants judged which smartphone showed the better location estimation. Participants reported using two primary heuristics (1) choosing the blue circle that was closest to the known location (distance) and (2) choosing the smaller circle (size). Visualizing graded probability with faded circles biased participants towards the distance heuristic. Visualizing confidence intervals with uniform circles biased participants towards the size heuristic (and using uncertainty information) and produced more accurate judgments

Society depends on knowledge workers (KWs) to identify, characterize and propose solutions to the many significant challenges it faces. KWs contend with ever changing information technology (IT) and bemoan ”information overload.” They commonly consult literature (e.g., Allen, 2001) and use productivity software that, regrettably, fail to leverage key findings in cognitive science. Can cognitive science help KWs process information and learn with technology? Yes, provided we directly address their problems. We present the Cognitive Productivity Research Project (Beaudoin, 2014) which is: characterizing information processing (IP) challenges KWs face (e.g., cognitive illusions, missing concepts and learning strategies); exploring gaps in cognitive science, including under-explored concepts (e.g., meta-effectiveness, monitors) and phenomena (e.g., KWs’ self-regulated learning when using IT tools to draw on source material); marshaling an IP architecture and principles to address these issues; and proposing practical IP strategies for KWs that emphasize meta-documentation and productive practice.

It has been shown that the eyes anticipate the target of the next manual object interaction. Meanwhile, manual interactions anticipate object features for grasp adjustments (first order) and the most convenient end-state in anticipation of subsequent tasks (second order). Moreover, grasping kinematics for the same object can vary depending on the final goal (third order planning). In an eye-tracking experiment we show that these factors can be measured already in eye fixations prior to grasping objects with different orientations (upright vs. inverted) and for different tasks (drink vs. hand over). Fixation measures show significant effects of object, task, and orientation and significant interactions. These results show for the first time end-state comfort effects in the eyes and suggest a tighter coupling of oculo-motor and motor programming than assumed so far. The insights suggest that even more intricate derivations of manipulation intentions can be derived from eye gaze data.

Fictive Motion is a type of figurative language used to express static visual scenes in terms of motion, for example, ”The road runs along the river” or ”The scar runs down his back”. Previous research suggests that we mentally simulate fictive motion (Matlock 2004, Matlock & Bergmann, in press), but little is known about the use of fictive motion in real discourse. Our study is the first to look at discourse data, analyzing videos taken from the TV News Archive containing fictive motion utterances. Our results show that the conceptual structure of the trajector (road, scar) influences both gestures produced with the utterance and linguistic properties of the utterance. Our study not only shows how fictive motion is used in speech, but also provides more insight on the mental processes involved in understanding and producing fictive motion, and more generally, figurative language.

Individuals with elevated symptoms of depression exhibit deficits in decision-making. Depressed individuals show decreased sensitivity to rewards, but increased sensitivity to punishments. This may be critical to understanding depressionrelated decision-making deficits, yet the computational nature of these effects is poorly understood. Participants (N=161) completed a decision-making task wherein they chose between two options on each of 150 trials. Rewards for both options were drawn from skewed-normal distributions with mean reward values of 0 points. For one option the reward distribution was positively skewed—more frequently giving losses than gains. For the other option the reward distribution was negatively skewed—more frequently giving gains than losses. Preference for the negatively-skewed option increased linearly as a function of the degree of depressive symptoms. Modeling analyses indicate that depressive symptoms are associated with less effective processing of reward magnitude and greater reliance on reward valence (gains vs. losses) in decision-making.

The ability to create and understand novel communicative signals is exemplary of people’s creative and inferential abilities. For example, when traveling and unable to speak the local language, we can make ourselves understood by creating novel gestures. This ability is a form of abductive inference, and requires people to generate novel hypotheses about possible meanings of signals (abduction proper). We propose that novel hypotheses may be generated from scratch by re-conceptualizing perceptual and conceptual representations through analogical augmentation. We plan to use robotics methodology to assess the plausibility of this model. By enhancing a robot with analogical augmentation we aim to enable it to generate novel gestures based on analogies. This lays the groundwork for more natural human-robot interaction. Furthermore, by studying the robot’s gestures and to what extent people can understand them, we gain better understanding of the abduction-based computational processes underlying communication.

The Mental Rotations Test (Vandenberg & Kuse, 1978) consistently produces large gender differences favoring males (Voyer, Voyer, & Bryden, 1995). This test requires participants to select two of four answer choices that are rotations of a probe stimulus. The incorrect choices (i.e., foils) are either mirror reflections of the probe or structurally different. Two experiments investigated the hypothesis that males notice structural differences more than females and a strategy of capitalizing on structural differences, accounts for the gender difference. Trials with structurally different foils showed higher accuracy and faster reaction times for both males and females. A significant male advantage was found for both foil trial types; however, an interaction between trial type and gender was not present. Moreover, males and females did not differ in reaction time. Thus, no evidence was found to suggest that strategy differences account for the large gender difference in mental rotation tasks.

Success in calculus undoubtedly requires the ability to coordinate multiple representations (CMR; i.e., coordinate among graphs, equations, tables representing the same function). This presentation will describe a study in which calculus and pre-calculus high school students are presented with CMR activities involving graphs and equations, first in their standard format, and then in an enhanced format designed to facilitate coordination between the order/sign of the function and the shape/direction of the graph. Students will also be tested on their visuospatial working memory, conceptual knowledge of calculus, spatial skills, and their knowledge of strategies for completing CMR problems. We will investigate whether individual or group-level differences in these background measures and their class placement lead to different levels of responsiveness to the enhanced presentation format. Both success vs. failure on the CMR problems and the particular strategies students use to solve the problems will be evaluated.

Psychological essentialism is the belief that members of a category share deep, underlying commonalities. Previous evidence suggests that essentialism is stronger for masculine than feminine properties and more evident in monolingual than bilingual children. Here, we investigated essentialism of gender by monolingual and bilingual adults, focusing on two distinct dimensions of essentialism: naturalness (regarding categories as natural kinds) and entitativity (regarding categories as homogeneous groups). Participants indicated their agreement with statements assessing beliefs about men and women on these two dimensions. The results replicated previous work showing that men are essentialized more than women, but revealed that this effect may be specific to the entitativity dimension. We also found that, compared to monolinguals, bilinguals were less likely to essentialize gender in terms of naturalness. Our findings converge with previous research highlighting the bidimensionality of essentialism and suggest that early effects of language experience on essentialist beliefs may persist into adulthood.

Mathematical competence requires pattern recognition for appropriate application of concepts and procedures. Research demonstrates that interventions targeting perceptual learning (PL) improve math performance. Little research, however, has directly investigated encoding changes, measured psychophysically, that may accompany PL gains in real-world tasks. We sought evidence of lasting encoding changes for mathematical objects for participants who used an Algebraic Transformations PLM and compared them to a Control group who did a different task, one that did not target PL of equation structure, but provided equal exposure to the same equations as in the PLM. All participants completed a speeded same/different psychophysical task comparing equations at pretest and delayed posttest. The PLM group improved on the psychophysical task more than controls, demonstrating that PLM structure, not mere exposure to equations, caused the information encoding gains. Perceptual learning interventions can accelerate expertise in complex domains, and these learning gains produce detectable, durable encoding changes.

Difficulty in interpreting and using spatial information from You-Are-Here maps (YAH maps specify a viewer’s location and orientation) stems from a misalignment between one’s physical orientation and the map’s orientation. The current research investigates the relationship between the participant’s physical location and orientation in accurately placing the location and orientation of another individual onto a map – essentially creating a YAH map for another individual. The other individual’s location and orientation was conveyed to the participant using a photograph. The photographs were of highly familiar building facades from around campus, which were oriented towards the cardinal directions. This research reveals how the participant’s location and orientation interacts with the photographer’s location and orientation, given the properties of the environment. Task performance was related to individual difference factors, such as self-assessed sense-of-direction, gender, familiarity with photographed locations, and the spatial reference frame used by the participant.

We examine the effects of age and education on the production of variable grammatical morphology among speakers of Yucatec Maya. Our investigation focuses on the use of optional plural morphology when describing pictures of one, two and seven entities performing an action in Yucatec Maya. Our sample (N=69) compares children and adults ranging in age from 5 to 48 years old (mean=19.6, SD=12.1) with levels of education (Spanish-based) ranging from no formal education to some college. Since the use of plural morphology is not optional in Spanish, this is a particularly interesting test of the effects of education in Spanish on sentence production in Yucatec Maya. Mixed effects logit models reveal that set size is a significant predictor of the use of plural morphology and numeral mention. Education, but not age, predicts the use of plural morphology, though only on nouns. We propose a language-internal and language contact-based explanation.

The social psychologists Fritz Heider and Marianne Simmel have shown in the 1950s that movies about very simple object interactions are typically interpreted in a very social manner, quickly perceiving motivations and emotions. We have developed a tool to generate Heider-Simmel-like videos. In contrast to the Heider-Simmel Interactive Theater project by Andrew Gordon and colleagues, though, our tool enables the generation of story-lines. The user is offered a manifold of individual object behaviors, such as approaching, avoiding, or circling around another object, and a manifold of perceptual events, such as the detection of another object, the touch of another object, etc. Behavior-event complexes can be chosen and concatenated by an intuitive user interface. The resulting story lines are then played-out by the tool in different scenarios, yielding different videos about socially similar interactions. The tool may be very useful for conducting future social- and object-interaction-related cognitive science projects.

Several studies have found gender differences in Iowa Gambling Task (IGT) performance in which males typically outperform females. However, the precise mechanism that underlies this effect remains unclear, and prior modeling efforts have been unable to pinpoint specific gender differences in behavioral aspects of the IGT. Our results replicated the behavioral gender difference finding and showed that females select the disadvantageous Deck B more than males. We fit the data with versions of the Expectancy Valence and Prospect Valence Learning models that included a parameter to account for participants’ perseverative tendencies. The addition of this parameter to the models led to a substantial improvement in the fit to the data. An examination of the best fitting parameter differences suggests that females give greater weight to recent events than males, which may lead females to discount the large, infrequent losses given by Deck B more and select that option more frequently

Landmarks can be helpful guides when completing spatial tasks, such as giving directions. Past research has shown that the salience of landmarks can influence their use, with salience based on perceptual features, or on the spatial relation between the landmark and a target location. In two experiments, students were asked to give directions to locations on campus to other students, alumni, or visitors. In Experiment 1, speakers’ ratings of the imaginability and frequency of use for 20 buildings along the directed paths impacted whether they were included in their directions to other students. In Experiment 2, these features did not impact speakers’ directions to alumni or visitors, suggesting a different prioritization of salience. These results suggest that the experience of the speakers and the identity of the receivers play a role in which landmarks speakers choose to include in their directions.

Spatial configurations allow for many different kinds of spatial relations between objects. Previous cross-linguistic work in this domain relies on a valuable but restricted stimulus set, the Topological Relations Picture Series (TRPS), which has two major limitations: (1) it covers a small subset of the spatial semantic domain, focusing on the IN/ON area, and (2) it covers that subset in an unsystematic way. We propose to create a large stimulus set of spatial relations that covers the space of possible relations in a more comprehensive way and includes the TRPS as a subset. The extended set will be systematically generated from a large family of spatial features describing relations between figure and ground objects, such as contact, support, attachment-by-spiking, and others that have been previously proposed. All stimuli will be rendered in 3D and released to the public to aid basic research in spatial language and cognition.

Exploration is an important component of learning. Previous research has focused on how information in a particular situation affects exploratory play. An important question is the degree to which past exploration and learning experiences affect exploration in novel contexts. In particular, children often get information about the difficulty of a completed task (“that was hard!”), but we do not know whether such information affects how children approach new tasks. We present preschoolers with information about an initial task, explaining that it was hard or easy (or no information). Preschoolers are then invited to explore a novel toy where no information about the difficulty of the toy is given. We find that preschoolers explore and discover more features following information that the initial task was difficult, than following easy or neutral descriptions. We suggest that children are extending expectations about previous events to novel ones.

Prior research shows that shared book reading promotes preschoolers´ language and literacy skills. However, little is known about the potential role of books´ features –e.g., wordless picture books vs. books with text– in children and teachers´ spontaneous language production. In this study, we transcribed verbal interactions of thirteen Colombian teachers reading to groups of preschooler students (aged 43 to 55 months) during reading sessions in Spanish using wordless picture books (condition 1) and prototypical storybook with text (condition 2). Books were matched for page length, genre and theme. Using Computerized Language Analysis (CLAN), we found important differences in children and teachers´ spontaneous language production. Specifically, paired t-test comparisons revealed that in the wordless-picture-book condition: (a) children produced significantly more word tokens, word types, utterances and questions (all p´s < .05), and (b) teachers produced significantly more word tokens, questions and levels of instructional support (all p´s <.05).

The present study investigated the differences between e-reading and paper-reading in their breadth and depth. Our results showed that (1) breadth and depth of reading were both greater in e-reading than in paper-reading; (2) possession of a tablet tended to facilitate breadth of e-reading; (3) breadth of e-reading was greater than breadth of paper-reading for news, magazines, and others, but not for novels; (4) depth of e-reading was greater than depth of paper-reading for novels, but the reverse was true for news and magazines; (5) people tended to read research articles, books and magazines on paper, but news and others on digital devices; (6) people tended to read longer on paper than on digital devices, but the percentage of contents they could remember was no different between e-reading and paper-reading. We conclude that modern readers have become accustomed to e-reading and can do it more efficiently than paper-reading.

Chinese Lexicon Project (Sze et al., 2014) summarized lexical decision response data of 2,500 Chinese characters. The original analysis has showed that the newest character frequency norm accounts the most variance of reaction times. The variance of these response data are analyzed in terms of the character frequency, strokes, and structures in use of the norms from Taiwan. First of all, simplified characters ranked as high frequency have greater performance on reaction times, but these characters ranked as lower frequency in traditional scripts have overestimated response points. Secondly many simplified characters are transformed from complex to simple, and strokes are substantial discrepancy between Chinese scripts. Finally, stuructre of character takes a large proportion of variance in the response data. The covariance of structure and character frequency also shows a significant trend. Our current work reveals some critical thinkings on using mega-data as the approach to study Chinese character processing.

Previous work using the Implicit Association Test (IAT) has revealed an implicit association between academic disciplines and gender (Nosek et al., 2002). Namely, participants appear to have an implicit association between men and science, and women and the humanities. The purpose of this research is to examine whether the former implicit association is rooted in an association between men and mathematics or math-heavy fields. This might explain the fact of relatively low representation by women in math-heavy sciences, and relatively high representation by women in areas of science that do not require advanced mathematical training. One hypothesis is that this difference in representation can be at least partially explained by an implicit association between math or math-intensive fields and men. We hypothesize that subjects will exhibit an implicit association between women and non-math intensive sciences, and men and math-intensive sciences. This research is conducted using mouse-tracking on Amazon Mechanical Turk.

Past research often found that older adults searched less in terms of browsing and generating keywords; few studies examined the processes and underlying mechanism that caused the age-related reduction on search. In the current study, 20 younger and older adults performed ill-defined search tasks with a search box we implemented. In addition to the age differences in the quantities of search, results showed that there were qualitative age differences in allocating resources to exploration and exploitation across tasks varying in difficulties. Older adults were found to do more exploitation within one information cluster as defined by the keywords than younger adults. There was also age difference in the ways to reformulate keywords, that older doing more total changes in keywords and younger doing more partial changes in terms of narrowing or broadening the search. The links between search processes and the age differences in cognitive profiles were also discussed.

Embodied cognition is a theory that emphasizes the importance of sensorimotor experiences for cognition. Therefore, the present study focused on how the facial expression manipulation influences the property judgments toward Chinese emotional words. 41 college students were divided into “biting the pen with smile” group and control group to rate the same 26 Chinese emotional words chosen from a Chinese Emotions Corpora (Cho, Chen, and Cheng, 2013). After having the instructed expression, subjects evaluated several semantic dimensions of emotional words immediately. The findings show that the “biting the pen with smile” group has higher rating values for dimension ‘valence’, ‘frequency’ and ‘continuance’ for the ‘disgust’ words, and the dimension ‘valence’ for the ‘angry’ words. The study found that positive facial expression indeed influenced the semantic properties of negative words, not the positive emotional words. The results are useful for investigating how word meaning is built in children and clinical applications.

We studied the role of action in a conceptual combination task by varying whether word stimuli could be physically grasped and arranged (words displayed individually on tangible cubes) or only touched and pointed to (words printed on a poster paper). Middle-school aged participants combined nouns from different taxonomic categories then described creative meanings. Descriptions contained more between-category relations (e.g., “shaped like” and “looks like” analogies) in the poster condition than when combining words using cubes. Conversely, participants produced more within-category descriptions (e.g., taxonomic declarations “it’s a X”, and metaphorically blended categories) when interacting with cubes than with a poster. These results suggest embodied explanations, and are consist with developmental studies that find categorization is differentially organized by shape and taxonomy. We propose that hardcopy and traditional keyboard-display computers which afford pointing and touching may engage categorization differently than tangible computers based on physical objects which afford grasping and arranging.

In an election year, political messaging can become feisty or even violent. In the 2012 U.S. presidential election, Americans were inundated with statements that metaphorically referred to violence, such as “Romney slams Obama” and”Romney slaughtered Obama.” Such expressions grab our attention and resonate with our understanding of actual physical violence. Despite their frequent use in election discourse, little is known about how such messages affect voters. Here we report the results of a novel experiment that examines how varying degree of violence in these metaphors influences inferences people make about politicians and election outcomes. Our results indicate that participants perceive candidates differently depending on degree of violence in descriptions of their performance in presidential debates. The results are informative and valuable because they shed new light on how framing works in election messages, especially how varying source domain information can lead to notable differences in reasoning.

Sampling biases are often assumed to arise from the type of information that learners sample (Fiedler, 2008), the possibility of negative payoffs (Denrell, 2001), or the prevalence of small samples (Kareev et al., 2002). Here, we show that even in a natural sampling situation (repeated Bernoulli trials), in which a learner’s only decision is when to stop sampling, different sampling goals can have an impact on sample composition and on inferences drawn from them. Specifically, we find that learners sampling with a binary goal (”more heads/tails?”) versus a distributional goal (”how many heads?”) end up with samples that differ not only in size but also content. Binary sampling leads to more samples with extreme distributions (many more heads or tails) compared to distributional sampling. In this project, we explore the impact of those sampling goals on subsequent decision-making on the basis of those samples.

We investigated human learning and generalization of three novel category structures based on eight exemplars in a continuous (9x9) stimulus space. Each category requires attention to both dimensions, but they differ in their organization. Critically, all three category types are matched on within- and between-category exemplar distances. The first category structure conforms to a condensation or information-integration type of problem with two classes separable by a diagonal bound. The other category structures cannot be solved with a linear decision boundary. We found that learners trained on the diagonal bound structure showed significantly better learning and generalization performance. In computational simulations, we found that an exemplar model (ALCOVE) could not account for the observed pattern. We posit that ALCOVE is constrained by the matched distances to learn these category structures at the same speed. Another similarity-based model with different basic design principles (DIVA) provided a good account of the behavioral data.

How do people reason about complex bodies of legal evidence? The story model of juror decision-making posits that people construct stories to determine guilt. But the story model does not model how evidence items relate to elements within the story, how the credibility and reliability of the evidence (e.g., witness testimony) is assessed, or how this affects story evaluation. Recent empirical work suggests that people reason using qualitative causal networks. In two studies mock jurors read a real legal case and judged the probability of the defendant’s guilt, the credibility of the victim and of key witnesses. Study 1 showed that an inconsistent testimony decreased the victim’s credibility and defendant’s guilt, also increasing the defendant’s credibility. Study 2 replicated this finding with a different population. These findings suggest that people draw inferences about credibility and reliability of evidence that filter into their network of beliefs about the crime.

This study looks into the conception that drawing or copying a face that is vertically inverted will improve the accuracy of the drawing by preventing holistic interference. We had participants draw parameterized face profiles (both upright and inverted) that were sampled from face space (see Davidenko, 2007). In each trial, participants were shown a face on the left side of the screen and asked to copy it on the right side. We then recorded the location of 66 landmark points on each face drawing, allowing us to compute a distance metric between each drawing and its corresponding original face. This distance metric served as a measure of accuracy, with higher distances corresponding to greater errors. Contrary to common belief, people’s drawings were significantly more accurate for upright versus inverted faces (t(15) = 4.9; p=0.0002). Our results suggest that holistic processing improves, rather than impairs, the accuracy of face drawing.

A common form of embodied social engagement involves acting as an ensemble, with all participants aware of each other and mutually engaged. In this naturalistic comparative study, we term this kind of triadic socio-cognitive activity ‘fluid collaboration’. Fluid collaboration occurs when participants establish the pace of ensemble activity mutually, with their moves flexibly adjusted and responsive to one another and to the demands of their shared endeavors. The moves of a fluidly collaborating ensemble depend on each other and might be described as harmonious. This kind of ensemble behavior appears to be the activity of “one organism with many limbs”. Utilizing a micro-analytic qualitative method, our finding of cultural differences in fluid collaboration relates to previous ethnographic and comparative work involving patterns of collaborative activity in Indigenous and Indigenous heritage communities in the Americas.

Controlling for alternative causes is essential for learning the strength of any one cause on an effect. Several processes have been proposed for how people control for alternative causes, including probabilistic contrasts within focal sets and associative processes. We investigated another mechanism called the informative transitions heuristic; people selectively attend to temporally adjacent observations (informative transitions; IT) in which the state of the target cause changes but the alternative causes remain the same. Within ITs, whether the effect also changes in the same direction, does not change, or changes in the opposite direction implies that the target cause has a positive, neutral, or negative influence on the effect. Participants judged the strength of the relationship between two drugs and a side effect in a trial-by-trial learning task. Causes with more positive as opposed to neutral ITs were judged to have stronger causal relations, consistent with the IT heuristic.

Insomnia affects about 33% of Americans according to Harvey & Tang (2003) who called for new cognitive treatments. We will report preliminary results from a test of (a) the Somnolent Mentation theory (SMT) of sleep onset (SO) and (b) a new cognitive treatment for insomnia, the cognitive shuffle (CS), derived from the SMT (Beaudoin, 2013, 2014). According to SMT, incoherent mentation characteristic of SO is not merely a side-effect of the SO period but promotes it, meaning it is somnolent. The SMT identifies several types of insomnolent mentation, which involve sense making (e.g., problem solving). SMT postulates counter-insomnolent mentation, thought patterns that interfere with insomnolent mentation. The CS is predicted to be both somnolent and counter-insomnolent (super-somnolent). Participants either engage in constructive worry Carney & Waters (2006) or in the CS using SomnoTest an iOS app developed by CogSci Apps Corp. (led by Beaudoin) based on mySleepButton R .

Knowledge of natural language quantifiers (like all, some, many) correlates with number acquisition (i.e., counting abilities). At the same time, number acquisition and approximate number skills are closely linked. These findings raise the question whether quantifier comprehension is exclusively related to exact number skills like counting or whether the relationship also extends to approximate number skills. To find out, we tested 3- to 6-year-old German speaking children on a quantifier comprehension task and on two counting tasks (‘how-many task’, ‘give-a-number task’). Additionally, we assessed children’s approximate number skills (ANS acuity) in a non-symbolic number comparison task. Quantifier comprehension was found to correlate with exact number skills even when age was controlled for. However, no significant correlation between quantifier comprehension and ANS acuity was obtained. Our findings support the notion of two distinct number systems: Only exact number skills appear to benefit from quantifier comprehension, whereas approximate number skills do not.

When nervous, unprepared, or less knowledgeable about a subject area, speakers may become increasingly worried that they are revealing telling acoustic cues about their anxiety level (e.g., hesitation before speaking or vocal jitter) to their audience. The current study is a two-part (production/perception) experiment that sought to evaluate when 1) speakers produce telling acoustics and 2) listeners become sensitive to vocal confidence cues produced by the speaker. The results indicated that when a speaker produced discriminating (un)confidence cues (e.g., rising intonation and delayed speech onset), the listener was significantly better able to predict the speaker’s confidence level. Interestingly, the speaker was significantly more likely to produce discriminating acoustic cues when more social pressure was applied, suggesting that speakers may intentionally communicate information about confidence. This indicates that confidence cues may be produced for the benefit of the listener and not to the detriment of the speaker.

Humans readily deploy external strategies in an attempt to offload cognitive work, a process commonly referred to as cognitive offloading. For example, individuals will often rotate their head in an attempt to normalize a rotated display (i.e., external normalization). Previous work has emphasized how various manipulations affect the overall use of the behavior to better understand the underlying decision processes. This approach, however, has overlooked the potential utility in investigating how these manipulations affect the time to initiate the use of strategy. We manipulated task difficulty with upright and rotated displays and measured initiation time and the overall use of external normalization. Analyses demonstrated that when individuals rotate, rotations take the same amount of time to initiate across tasks, whereas overall frequencies of rotations varied as a function of task. This dissociation suggests that the time to initiate the strategy and external strategy selection are at least partly independent

In the classic horizontal vertical illusion (HVI), vertical lines appear 5-6% longer than horizontal lines. However, in outdoor scenes vertical poles of several meters appear as much as 25% longer than frontal ground extents. This large-scale HVI is consistent with angular scale expansion theory (Durgin & Li, 2011). It is known that the classic HVI is yoked to the reference frame of the eye itself, such that the illusion reverses when the observer is on his or her side. In a series of experiments conducted both in real outdoor spaces and in immersive virtual environments we examined how the large-scale HVI was affected by reorienting the observer, and found that the large scale HVI was reduced, but not reversed. The amount of reduction was quantitatively consistent with a retinotopic contribution of the classical HVI (6%). Most of the large-scale HVI is world-centric.

In the study of verbal memory, a critical question is the extent to which recognition is influenced by the prior contexts in which items have appeared (‘context noise’), as opposed to competition from other items present within the immediate task context (‘item noise’). In a standard recognition task, subjects study a list of words, and at test, discriminate between studied items (targets) and novel items (foils). To disentangle the contributions of context and item noise, we systematically manipulated both the contexts in which critical items had been encountered prior to study, and the composition of the recognition list, varying semantic similarity among items. Our results suggest independent contributions of each factor, with word frequency and temporal lag as important mediating variables. These findings can be interpreted within both associative learning and memory paradigms

The capability of estimating emotions is an important feature needed for intelligent systems to interact with humans. . In this paper, we propose a computational model to calculate a user’s desirability as one of the most important emotions, especially in Intelligent Tutoring Systems (ITS). The main purpose of this research is to find a relationship between personality and desirability in virtual learning environments. The proposed model can determine the desirability of mastery motivational oriented students considering their personality, which is determined using MBTI. Based on the OCC emotion model, the goals and events are considered which can affect desirability. Then a cognitive map is developed between the personality dimensions and the goals and events. The proposed model has been implemented and evaluated in a simulated virtual learning environment and the results show that the proposed model can formulate the relationship between personality and desirability with high precision.

Building upon the finding that people can assess the logicality of conjunctive probability statements intuitively (Vallee-Tourangeau & Faure-Bloom, 2015), we examine whether increasing the salience of people’s implicit knowledge about the logic of sets can impact the influence of logical considerations on probability judgments. We compared the rate of heuristic responding in a control group with that observed in two experimental groups where participants were either explicitly or implicitly primed to reflect on the logic of sets inclusions prior to completing a conjunction probability task. Explicit priming involved asking them to rate a series of statements such as ‘I am more likely to meet a bank teller than a bank teller who is also a feminist.’ Implicit priming involved experiencing a series of events with the co-occurring presence and absence of a target characteristic (e.g., Feminist) and an alternative characteristic (e.g., Bank Teller) using a computer-based dynamic learning task

Emoticons in informal text communication are common worldwide. They have the potential to reveal emotion and social functions, analogous to facial expression and body gestures in face-to-face verbal communication. Our findings from a corpus study of online text communication by a group of scientists, some of whom were bilingual and others monolingual, suggested that patterns of emoticon use depend on a variety of factors, including emoticon valence and language of texting (Aragon et al., 2014). In the present study we bring these effects into the laboratory by examining the interrelation of emoticons and words in lexical decision (LD) experiments with sequential (SOA 200 ms) but spatially superimposed emoticon-word pairs. Monolingual speakers showed a reliable interaction of emoticon valence with lexicality but interactions with word valence were unreliable. Results will be compared to those from comparable word-word pairs.

In this study compared single stopping rules models to the Cast-Net stopping rule model. The Cast-Net model assumes that several stopping rules can be used simultaneously to determine the stopping point to stop information search and to proceed to making a final decision. We analyzed whether the Cast-net model would pay the price for being more complex when compare to single stopping rule models (critical difference, fixed-sample size and runs). The models were compared under different decision making conditions (time pressure and validity of recommendations). The model fitting procedure was conducted on the full data stopping-value distributions, by simultaneously fitting the correct and incorrect responses. Across variety of experimental conditions, the general results supported the validity of the Cast-Net model. These results challenge many decision making models that utilize only one type of a stopping rule, and may provide a new direction in the exploration of cognitive computational models.

The capacity to use language to form new representations and to revise existing knowledge is a crucial aspect of human cognition. Here we examined whether infants can use language to adjust their existing representation of a recently encoded scene. Using an eye-tracking paradigm, we asked if 16-month-old infants (N = 26; mean age: 16;03, range: 14;15-17;15) use new linguistic information about an occluded event to inform their expectation about what the visual world should look like. We compared looking time to outcomes that matched this information to those that did not. Infants looked significantly longer when the outcome did not match the input, suggesting that they generated an expectation of the outcome based on language alone. This effect was unrelated to infants’ vocabulary size. Thus, using language to form expectations about the visual world is present at an early developmental stage, even when language skills are rudimentary.

This research develops three text analytic techniques to improve survey questionnaires. The first is open-ended response mining. Narrative responses on a survey are mined for themes then used to develop new questions. Closed-ended responses identify subgroups who agree/disagree with the question. Then open-ended responses examined for systematic differences which suggest new constructs that distinguish the groups. The second is used during question development. Agree/disagree questions are examined for similarity in language using latent semantic analysis. The matrix of similarity coefficients is used to make scale assembly and predicted item performance decisions in advance of field test data. The third involves replacing zero with LSA-derived coefficients as baseline comparisons for correlation coefficients to identify interesting relationships between rating questions. Semantic similarity of question stems suggests a degree of relationship between questions. This, rather than zero, is the appropriate expected value of a correlation between two items.

Policy decisions on political, economic, legal, and health issues are often made by groups that rarely exceed 40 members and are typically much smaller. Given that wisdom is usually attributed to large crowds, should committees be larger? Using computer simulations and mathematical analyses we show that group accuracy, averaged across the range of difficulty that would be encountered in real-world tasks, is often maximized for moderate-sized groups. The result holds whenever the accuracy in easy tasks is above chance more than the accuracy in difficult tasks is below chance and the easy tasks are encountered more often. Note that for the result to hold, it is not necessary to assume selective sampling of group members according to their individual accuracy. A smaller committee that would produce more accurate decisions in our model can simply be selected randomly out of a larger group of experts.

The use of video for instructional purposes has exploded online, but little is known about how people interact with video differently than other instructional materials and how much they learn from those lessons. Kintsch’s (1994) model of text comprehension provides a productive model for studying how students build a semantic representation of materials that unfold over time but have an overarching conceptual structure. The current study seeks to replicate an earlier finding from the text comprehension literature (Mannes & Kintsch, 1987) that providing students with an informational outline before reading text can improve recall of the text and/or transfer of ideas in the texts to new contexts, depending on the structural relationship of the outline to the text itself (consistent or inconsistent structure). This study was replicated using instructional videos rather than texts. Results will be discussed with respect to Mannes and Kintsch’s original study.

Previous literature has shown that spatial anxiety relates to navigation abilities (Hund & Gill, 2014). How spatial anxiety effects the spatial-temporal perception of one’s environment is not well known. The present student aimed to examine how spatial anxiety related to the memory of distances and time to landmarks in the surrounding area. Participants completed a battery of navigation questionnaires and reported how far (both in distance and time) different known landmarks in the surrounding area were. Data show a trend suggesting that females overestimated distances whereas males were more accurate in estimates to the five furthest landmarks. Spatial anxiety did not predict distance estimates; however, mobility within the surrounding area was marginally predictive of distance estimates for females. These findings suggest that spatial anxiety does not predict the remembered distances and time estimates to landmarks, but that mobility may be a more important predictive factor in remembered distances to landmarks

The way participants adapt their search to the specifics of different types of analogies is not fully understood. We compared the effects of two types of semantic distractors. The first were related to C by a semantic relation which had nothing to do with the semantic relation used in the A:B pairs, whereas the second, the so-called ”double distractors”, were not only related to C but also had a semantic relation similar to the one linking A to B. We used eye-tracking measurements in addition to reaction time and performance indices. We found that performance decreased, and that the solution set was less explored visually with the double distractors than with the former. This suggests that the analysis of the A:B pair activates a set of relations which can prime irrelevant relations which compete with other relations while searching for the relevant dimensions.

In causal inference, people place greatest weight on cases where a hypothesized cause and its outcome are simultaneously present, potentially reflecting a positive test or confirmatory strategy. We hypothesized that individuals may display more confirmation seeking when an outcome has few, versus many, causal alternatives and that this relation may vary with actively open-minded thinking (AOT) or need for cognition (NFC). Subjects learned about implausible or plausible causes of outcomes that had many or few causal alternatives (e.g., stress vs. colon cancer). On each of 16 trials, subjects received frequency data and made a causal judgment, after which they completed the AOT and NFC scales. As hypothesized, subjects weighted confirming data more heavily with fewer vs. many causal alternatives, but this relationship only held for plausible causes. AOT interacted with causal alternatives: With few alternatives, AOT was unrelated to data-weighting. However, with many alternatives, data-weighting increased with increases in AOT.

Digital devices are becoming ubiquitous fixtures in classrooms nationwide. Despite this, the costs and benefits of digitization are understudied. For example, Mangen et al. (2013) found advantages in reading with physical books compared to digital readers. The current study extends these findings to physical and digital versions of spatial puzzles. Participants completing a series of physical tangram puzzles were both faster and more accurate than those completing digital versions of identical puzzles on tablet computers. Those in the physical condition were also faster and less error-prone on a subsequent arithmetic test. These results suggest that the current trend of increased digitization in education may have far-reaching and unexpected implications that could compromise learning. Follow-up studies aim to identify the cognitive mechanisms that cause these differences. These findings can be used to develop a set of best practices for incorporating digital teaching tools in the classroom.

Many experiments within the memory literature show evidence for context-dependent effects in memory (Gieselman & Bjork, 1980; Godden & Baddeley, 1980; Marian & Neisser, 2000; Smith, 1986; Smith & Vela, 2001). There have also been a number of failed replications of this effect (Farnsworth, 1930; Reed, 1931; Smith, Glenberg, & Bjork, 1978; Godden & Baddeley, 1980; Jacoby, 1983). If the context-dependent effect exists, there are many unanswered questions about it, such as the mechanism behind the phenomenon and at what level of detail context is encoded. In order to explore these questions, we need a reliable way of observing context-dependence that holds across experiments. Here we attempt to find a paradigm that can be reliably used to elicit context-dependence, so that we can begin answering some of these questions.

Persistent virtue requires commitment to values not easily operationalized in the metrics of cognitive science. Caregivers living in L’Arche communities commit years and decades of their lives to compassion and social justice in caring for unrelated developmentally disabled adults. We examine generosity and fairness among 48 L’Arche caregivers using economic behavioral tasks and life and identity narrative interviews. Analysis of forced dictator economic behavioral tasks found heterogeneity in generosity and in approaches to fairness utilizing economic equity and efficiency. Latent semantic analysis of transcribed interviews showed a significant difference between probes measuring just, brave, caring, and interpersonal generosity characteristics for all participants. The integrative analysis of economic behavioral tasks and semantic analyses of life and identity narrative interviews illuminates challenges in using laboratory methods to examine qualities of exemplary virtue and demonstrates the fruitfulness of examining exemplars of caring for insight into moral cognition.

Empirical evidence suggests infants can make inferences about uncertain future events using probabilistic data (Denison & Xu, 2010; Teglas et al., 2007). However, it is unclear if infants are using information about proportions to make these decisions. Infants were presented with population jars that contained large distributions of objects, and were tasked with deciding which of the two jars was more likely to yield a desirable object on a single draw. In Experiment 1, most infants chose the jar that contained only preferred objects over a jar that contained a 3:1 ratio of preferred objects. Most infants also made the correct choice when presented with the reverse ratios (choosing a 1:3 over a 0:1 ratio). In Experiment 2, infants correctly chose the jar that contained a 4:1 ratio of preferred objects over a jar that contained a 1.5:1 ratio. However, infants performed at chance when presented with the reverse ratios.

Although many information system are employing autonomous agents for the purpose of user-friendly interface, cognitive mechanisms of their effects are not clear yet. In this study, we compared three types of UI, an embodied agent system equipped with a direct anthropomorphization robot, an only voice agent, and without agent system condition, of a microwave oven. Thirty-six older (65 years or above) and 36 younger adults (undergraduate students) participated in the usability testing experiment with one of the three agent conditions. Analysis of interaction between a participant and the oven, through participants’ utterance and the personal-space data, showed large differences between two age groups; younger adults entertained the interaction both the voice and the embodiment agent conditions, while older adults evaluated higher only with the embodiment agent condition. Differences in mental models of older- and younger adults, and reasons of those aging effects will be discussed.

This study examined how customers’ preference and purchase intention change depending on presentation type of information. Products’ information was manipulated on scale and order dimension. Participants made a decision on three situations choosing music(mp3) download plan, cell phone data plan, drinks voucher plan for a coffee shop as well. All stimuli were appeared through the computer monitor in laboratory. Preference and purchase intention were measured by 7 likert scale on two kinds of plans presented in two different scales. Findings show that consumers had more preference and purchase intention when the product’s information was displayed on an expanded scale than on a contracted scale. Unfortunately the displaying order didn’t have any significant impact on preference of products. However, effect of an order on preference appeared differently depending on the product type. In case of the music plan, price-first condition was most preferred type of information.

Female performance on tests of spatial ability may be hindered by the presence of stereotype threat. We examined sex differences in performance on two perspective taking tests when these tests were framed as measuring either spatial or social (empathy) abilities. In the spatial condition, the tasks were framed as spatial and participants were reminded of the male advantage on some spatial tasks. The social condition included modified versions of the tasks to include avatars of human figures, and framed the tasks as social tasks with a female advantage. Results showed a gender difference in favor of males in the spatial condition, but not in the social condition. Framing did not affect male performance. However, females in the social condition outperformed females in the spatial condition. These results suggest that females may underperform on spatial tests in part because of negative performance expectations rather than their actual spatial abilities.

Studies of mother-infant interaction indicate that sensitivity to interactional timing begins developing around 3–4 months, but there is currently no evidence bearing on when children start to understand conversational timing rules; how to transition from one speaker to the next. We showed twelve- and thirty-month-old children videos of conversation featuring puppets using typical (200ms inter-turn silence) and atypical (1200ms silence and 3+ syllables vocal overlap) turn-timing. We assessed children’s timing preferences by then showing them the two puppets (typical and atypical) for a sustained period and then by asking them to choose one puppet to hold. Preliminary results suggest that, overall, children were more likely to discriminate between typical and atypical timing for vocal overlap than long silent gaps. This maps nicely onto findings about children’s spontaneous turn-taking: they learn to minimize overlaps before gaps.

Previous studies have shown that underlining sentences while reading is an effective comprehension strategy that many people use spontaneously. We examined which components of this underlining strategy could facilitate comprehension. The effects on comprehension of lined traces and hand movement were examined independently. Eighty-two undergraduates were assigned to one of the four conditions: both traces and movement, movement and no trace, traces and no movement, and neither trace nor movement. After reading the expository text as instructed for ten minutes, participants were instructed to solve a Sudoku puzzle as a distracter task for three minutes. They were then asked to summarize, title, and generate three keywords for the text in ten minutes. The results showed that hand movement facilitated appropriate summarization and that lined traces enhanced appropriate titling. We interpreted the results in terms of cognitive load theory and external memory aid.

What cognitive processes underlie scientific explanation? Although scientific reasoning is often careful and methodical, we hypothesize that it is also influenced by an intuitive explanatory process: namely, an inherence heuristic (Cimpian & Salomon, 2014, BBS). The central claim of the inherence heuristic proposal is that, when people construct explanations, they oversample inherent facts about the entities whose behavior they are attempting to explain. We investigated the influence of this heuristic process on explanations for novel and historical scientific phenomena in chemistry, biology, and physics. Participants were provided with short vignettes describing unexpected outcomes of experiments and were asked to explain these outcomes. As predicted, explanations were couched primarily in terms of inherent features of the entities involved. Importantly, this was so even though such features were not mentioned in the vignettes but extrinsic factors were (e.g., high altitude, unusual location). These findings elucidate the psychological processes that underlie scientific explanation.

Previous studies found that humor and empathy are associated with interpersonal relationships. Particularly, Hampes (2001) reported that humor styles and empathy had a positive correlation in adults. The purpose of the study is to explore the link between empathy and humor use in teenagers and to investigate if gender differences exist as well. 115 adolescents between 11-12 years old participated the study and filled out The Empathy Quotient and Taiwanese Adolescent Humor Instruments. We found that empathy and the sense of humor had a significant correlation both in boys and girls. However, the scenarios of humor using and the purposes of humor using only had a positive correlation with empathy in girls. The findings offer a supplementary evidence for the developmental link between empathy and humor. The implications of the study shed light on the developmental discrepancies between genders in adolescents.

Uncertainty represented in visualizations is often ignored or misunderstood by the non-expert user. The National Hurricane Center displays hurricane forecasts using a track forecast cone, depicting the expected track of the storm and the uncertainty in the forecast. Our goal was to test whether different graphical displays of a hurricane forecast containing uncertainty would influence a decision about storm characteristics. Participants viewed one of five different visualization types. Three varied the currently used forecast cone, one presented a track with no uncertainty, and one presented an ensemble of multiple possible hurricane tracks. Results show that individuals make different decisions using uncertainty visualizations with different visual properties, demonstrating that basic visual properties must be considered in visualization design and communication.

The cross-modal literature posits a weak-to-strong continuum of synesthesia. One extreme views cross-modal associations as idiosyncratic and unique to synesthetes. The other extreme suggests that cross-modal associations follow a general pattern across individuals, and are mediated by emotional associations. We tested these views by examining differences between music-color synesthetes and non-synesthetes in their consistency of color associations and memory for music. We find that music-color associations follow the same general pattern across these groups. A two-dimensional mapping is found to mode (major/minor) and tempo. Slow-minor music (thought to convey sadness) is associated with blue, fast-minor with red (anger), fast-major with yellow (happiness), and slow-major with green (calmness). Both groups are consistent in their associations over time, and synesthesia has no effect on memory. We conclude that music-color synesthesia may be an extension of normal psychological processes that govern cross-modal associations, with individuals aligning music and color based on emotional congruence.

The purpose of the present study was to examine the relationship between art experience and artistic inspiration. Focusing on attitudes and behaviors in appreciation and creation as mediating variables, it was hypothesized that (a) the method of appreciation with comparison between one’s creations and others’ creations is the best predictor of artistic inspiration, and (b) art experiences might affect artistic inspiration, meditated by competence in artistic creation and the method of art appreciation. A total of 185 Japanese undergraduate and graduate students completed the research questionnaire. Data was analyzed using multiple linear regression for the first hypothesis and structural equation modeling for the second hypothesis. The two hypotheses were supported. The findings suggest that people with more extensive art experience develop competence in artistic creation and consider their own creations when appreciating others’ artwork. In addition, they experience artistic inspiration more frequently and intensely

It has been argued that linguistic color categories, despite their number can vary between languages, are themselves universal and based on physiologically conditioned distinctions. Recently it has been demonstrated that Russians, whose language has separate terms for light and dark blue, discriminate faster between objects of corresponding hues than Englishmen. In our study we tested if language also conditions the physiologically more sound discrimination between cold and warm hues. We compare Russians to speakers of Komi language, where green and yellow make up the same category and there is only one category for blue. Russians outperformed Komi in discriminating between light and dark blue objects as well as yellow and green objects. However Komi were faster in discriminating yellow and green objects then light and dark blue objects. Therefore, language influences discrimination between warm and cold hues, but this impact is weaker than in the case of two cold hues.

Relational retrieval—retrieval that is based on common relational structure, such as an underlying principle or pattern, is typically rare. Previously, we found that providing relational labels at encoding and/or test can improve relational retrieval (Jamrozik & Gentner, 2013). In the current work, we tested the specificity of the labeling effect by comparing the effects of relational labels (e.g., inoculation) with domain labels (e.g., psychology). Because people are naturally likely to attend to domain information, we predicted that domain labels would have a smaller effect on domain retrieval. Using a cued-recall paradigm, we varied the presence of relational and domain labels at encoding and test. Relational labels increased relational retrieval, but domain labels had no effect on domain retrieval. These results suggest that relational labels have a strong effect on retrieval (relative to other kinds of labels) since they increase people’s attention to information that is not naturally salient.

This paper presents an exploration of the ontological shift from musical materials (i.e. melody, harmony, rhythm, texture, timbre, register) to activities in music performance analysis. The “dogmas” extend Herbert H. Clark’s conceptual framework for the study of joint activity in language use to explore music performance in the WAM tradition. A systematic analysis of London Symphony Orchestra masterclasses examines the basic mechanisms of music making in four main areas: representation, audience, interaction, and tacit knowledge. This exploration leads to a broader account of cognition and creativity in music performance, one that bridges inner and outer processes of awareness around domains of coordination in joint activities. In this view, material conceptualizations are viewed as targets of focal awareness rather than the basis for cognition in music making. This account, grounded in a rich third-person phenomenological analysis of instructional materials, paves the way for a “meaningful analytics” of musical practice.

This project explores how the ability to ask informative questions changes during development. We hypothesized an intrinsic link between the ability to update beliefs given evidence and the ability to ask informative questions. To study the developmental trajectory of this behavior, five to ten-year-old children played an iPad game asking them to identify a hidden bug. Learners could either ask about individual bugs, or make a series of feature queries (e.g., “Does the hidden bug have antenna?”) that more efficiently narrow the hypothesis space. The iPad display either assisted children with updating their beliefs or required them to update themselves. We analyze the relationship between belief updating and information seeking behavior as a function of age, along with how their strategies for acquiring information change. The broader context of the work is to better understand how to structure informal science exhibits in ways that are developmentally appropriate.

Compound words are cross-linguistic morphological phenomena that occur in all languages. Compound words are widely accepted to be stored in the lexicon but their constituents need to be accessed during both language learning and production processes. In this study, the use of corpora was investigated for how to differentiate single-stem words from single-word compounds and then how to segment compound words when no phonological information is available. Stems and morphs discovered in manual segmentations of the METU-Sabancı Turkish Treebank and the CHILDES were employed in the compound word recognition task and the results were compared. The METU Turkish Corpus (with about 2 million words) and a webcorpus (with about 490 million of Turkish words) were utilized in the segmentation task. The results emphasize that the lexicon can be morpheme-based; and lexical frequencies are effective heuristics in compound word recognition and segmentation

In this paper, NARS (Non-Axiomatic Reasoning System) is discussed in the context of Cognitive Science. NARS is an artificial general intelligence system designed under the assumption that the system usually has insufficient knowledge and resources with respect to the problems to be solved, and must adapt to its environment. Since the human mind was evolved under the same restriction, this normative model shows many human-like properties. In this paper, NARS is used to reinterpret several well-known results in cognitive science, such as Wason’s selection task, the Linda problem, and U-shaped learning. These results cannot be explained by traditional normative models, while now they can be handled by NARS in a unified way.

A central idea in social psychology is that people can construe other people in terms of two types of mental representations: social categories (e.g. male) and attributes (e.g. intelligent). It is assumed that assigning a person to a social category (i.e. social categorization) is one of the most important causes of stereotyping. However, no theory has yet successfully explicated the properties that distinguish social categories from attributes and how those distinct properties may cause stereotyping. We show that an interpretation of social categories as mental representations that strongly exclude other mental representations (i.e. ’excluders’) can explain how social categories may cause stereotyping. In addition, we present computer simulations that implement the assumed principles in a connectionist model where social categories are interpreted as nodes with strong inhibitory links. We argue that our model solves fundamental ambiguities in social categorization theories and unifies these theories with connectionist models of person perception.

This paper presents a novel computational model of jazz improvisation based on n-gram language models. Recent functional neuroimaging studies suggest that the brain processes structural elements of improvised music and conversational language in a similar manner. We hypothesized that if musi- cal improvisation and language share a common cognitive and neurological foundation, then statistical techniques for modeling one domain should be capable of successfully modeling the other domain. Accordingly, we demonstrate that n-grams (an archetypal language model) can successfully model jazz improvisation when trained on a large corpus of expert-level jazz saxophone solos. Furthermore, we propose perplexity as a novel method of evaluation of jazz improvisation models.

Acquiring novel speech categories is necessary in spoken language learning. The dual-learning systems (DLS) approach posits that two competitive systems underlie the category learning process: an explicit hypothesis-testing system, and an implicit procedural system. DLS assumes that the explicit system dominates early and control is passed to the implicit system when optimal. Evidence from our work, including the finding that minimally informative feedback enhances speech learning relative to fully informative feedback, supports the claim that the implicit system optimally mediates speech learning in adulthood. Experiment 1 replicates this finding. Experiment 2 tests the DLS prediction that explicit processing dominates early by comparing performance across two conditions. The optimal condition includes full feedback early and minimal feedback later. The suboptimal condition includes minimal feedback early and full feedback later. In both conditions, real-time computational modeling individualized when feedback transitions occurred. As predicted from DLS, learning was superior in the optimal condition.

It is well-know that the action-sentence compatibility effect (ACE) is activated by mental simulation triggered by visual or linguistic stimuli (Glenberg and Kaschak 2002 and many others). However, no study has compared the extent of ACE activated by literal and figurative understanding of honorific Japanese verbs. Two experiments, consisting of 59 Japanese university students was conducted using two regular and two honorific Japanese verbs in sentences where participants’ literal or metaphorical interpretation of the verbs was measured by their vertical or horizontal hand movement. The findings from this research found metaphorical, not literal, meaning activated a stronger degree of ACE in regards to Japanese sentences. In addition, social norms of Japanese society partly played a crucial role regarding ACE. 2921

How might neuroaesthetics move beyond beauty and begin to study meaning? While neuroaesthetics is broadly concerned with the brain’s role in processing art, it has typically focused on perceptual preferences concerning the question of beauty. To cut deeper ontologically the field might consider exploring other basic kinds of meaningful categorical judgments people routinely make about artifacts, including those related to an object’s purpose and status as art. Providing a point of entry for an empirical approach, a methodology is described where participants judge objects (chairs) for higher-order qualities beyond beauty, including functionality, and art-objecthood. Results suggest that artifacts can be used for probing deeper empirical questions concerning the neural basis for aesthetic judgments and object processing. In this manner, we can begin to understand the meaning of art with respect to both its form and function.

Automated assessment and immediate feedback are staple features of modern e-learning systems. In the case of programming exercises, most systems only provide binary (correct/incorrect) feedback, which is often inadequate for students struggling with the material, as they may need expert guidance in order to successfully overcome obstacles to understanding. We propose a Case-Based Reasoning (CBR) approach to improve the quality of feedback on programming exercises. CBR is a machine learning technique that solves problems based on previous experiences (cases). Every time the instructor provides feedback to a student on a particular exercise, the information is stored in a database as a past case. When students experience similar problems in the future, knowledge contained in past cases is used to guide the students to a solution. While the system will provide detailed feedback automatically, this feedback will have been previously crafted by human instructors, leveraging their pedagogical expertise.

Policy capturing is a decision analysis technique normally using linear statistical models to infer the basis of expert judgments. The purpose of the present work was to test if the C4.5 decision tree (DT) algorithm (a non-linear machine learning method) is more effective at capturing individual’s decision policies than the standard linear technique. Human classification behavior was measured in a simulated naval air-defence task to compare decision tree models and linear logistic regression models in terms of their descriptive and predictive accuracy. Results show that C4.5 was superior in terms of goodness-of-fit and cross-validation performance. Decision tree complexity was significantly correlated to individuals’ response times. The classification rules derived from each individual were actually more reliable than their human counterparts – replicating a classic finding in policy capturing. We conclude that C4.5 is a useful policy capturing tool in the context of a complex dynamic task environment.

Research suggests that guided methods of instruction, such as worked examples, reduce the cognitive load placed on learners, which allows them to learn new information more efficiently and effectively. The current study examined the effect of worked examples on transfer of statistical reasoning, as compared to traditional study techniques. Students from an introductory college-level psychology course learned information related to basic statistics and hypothesis testing from a computerized instructional program. The experimental group completed a computerized program which contained worked examples and practice and feedback. The control group consisted of students who went through a computer program through which they read excerpts from a textbook used in Queens College statistical reasoning classes. The same topics were covered in both computerized programs. On posttests, the experimental group performed significantly better than the control group. This provides support for computerized worked examples as effective instruction on the college level.

Music-to-color associations show emotionally-mediated cross-modal correspondences (Palmer et al., 2013): people choose colors as going best with music when their emotional content matches (e.g., happy-looking colors go best with happysounding music). What musical/acoustic features underlie such correspondences? And are music-to-texture correspondences also evident? Experiments using highly-controlled melodies that varied in tonality (major/minor), note-rate (fast/medium/slow), and register (high/low) revealed systematic correspondences between musical/acoustic and colorimetric dimensions: faster, major, higher-pitched melodies were associated with more saturated, lighter, yellower colors, whereas slower, minor, lowerpitched melodies were associated with more muted, darker, bluer colors. Further experiments revealed emotion-mediated associations from music to texture, although agitated/calm and angry/not-angry emotions were stronger with textures, whereas happy/sad emotions were stronger with colors. Systematic associations were also evident between visual/spatial features of texture (e.g., Sharp/Smooth, Curved/Straight) and musical dimensions (e.g., note-rate and piano/cello timbre).

While conversationally common, sarcasm presents identification challenges in writing. For example, ”Oh wow!” may be expressed both sincerely and ironically. Research suggests that interpretation of ambiguous sarcasm may be context dependent, though precisely how visual and auditory cues contribute to comprehension remains unclear. To explore this, we recorded an actor performing 45 phrases sarcastically and sincerely. In two calibration studies, we selected 16 ambiguous phrases that were identifiable with all cues available. In the main study, participants classified these phrases as sarcastic or sincere in three conditions: audio-video (N=32), audio-only (N=29), or video-only (N=26). Performance was high (91%) with both cues available. Intriguingly, the drops in performance with audio only (82%, p<.001) and video only (80%, p<.0001) were small and not significantly different from one another (p>.5) suggesting auditory and visual cues contribute similarly to sarcasm comprehension. We replicated these results in a separate within-subjects experiment (N=77).

The semantic richness of a word is multidimensional, and includes dimensions such as semantic neighborhood density, imageability, number of features, and valence. While certain dimensions (e.g., imageability) have been examined in the memory domain, the bulk of semantic richness research has been confined to visual word recognition tasks. Therefore, it is unclear if other dimensions influence memory and in what manner. Our aim was to extend previous works by investigating the relative contributions of these dimensions in memory using the megastudy approach. This approach allows the language to define the stimuli, rather than have the experimenter select stimuli based on a limited set of criteria. 120 participants studied 532 words and they had to either recall or recognize these words. We found that although semantically richer words were generally more memorable, this did not generalize to all dimensions. The implications of these findings will be discussed.

Previous research showed improved working memory through cognitive training. However, comparatively little is known about whether long-term memory can be improved by training. Several cognitive processes, including working memory and executive function, subsume long-term memory functioning. Thus, if the cognitive processes that subserve longterm memory could be improved via training, we predicted that it should lead to broad improvements in long-term memory. The current study examined whether training using three different cognitive training conditions would improve performance in tasks measuring executive functions and long-term memory. Participants were assigned one of three possible training conditions (working memory training, mindfulness training, and retrieval practice) over 20 days. Results suggested that retrieval practice and mindfulness led to changes that improve participants’ ability to store and retrieve new episodic information formed while reading text passages. Other benefits include improvements in working memory and executive functions. Limitations and directions for future research will be discussed.

In many languages, tone (i.e., pitch patterns) is part of the phonological system; two words with the same sequences of segments can differ only in tone. Tone does not distinguish word meanings in English, so English-learning infants can ignore tone when learning words, but do they? We examined the encoding of tonal detail in word learning by monolingual English-learning 14- and 17-month-olds. Infants were habituated to a novel word with a Mandarin tone (/k´a/) paired with a novel object. Test trials alternated between the same pairing (Same), and the same object paired with the word with a different tone (/k`a/, Switch). Longer looks to the unfamiliar mapping indicate infants noticed the switch and attended to tone contrasts. Overall, neither age group discriminated the tone contrast; however, infants with larger vocabularies looked longer to the novel mapping (r=.32, p=.007), suggesting a common underlying mechanism between general word learning and tone sensitivity.

Within the fast-and-frugal heuristics framework several strategies have been proposed to describe how people infer unknown criteria from knowledge stored in memory. An open question is how people select between the set of available strategies. We build upon previous work that maps environmental structures into mental representations to carve out for each strategy a cognitive niche, or area of applicability. Based on patterns of occurrences and co-occurrences of objects and facts in the internet, we predict the probability and latency of retrieval of factual knowledge about these objects. This allows us to simulate the applicability of different knowledge-based strategies as a function of the distribution of decision relevant information in the environment. We conclude that the problem of strategy selection might be restricted when the pattern of information occurrence in the environment and the resulting accessibility of knowledge about the decision objects in memory are accounted for.

This research investigated how pseudoscientific, paranormal, and conspiracy beliefs relate to each other. Preliminary research indicates that holding one type of unsubstantiated belief predicts holding other types of belief (Lobato et al., 2014). We administered a survey (n=420) asking about belief in specific pseudoscientific, paranormal, and conspiracy claims. We also examined cognitive predispositions towards analytical and intuitive thinking, open-minded thinking, and ontological knowledge. Pseudoscientific beliefs were predicted by beliefs about paranormal and conspiracy claims; paranormal beliefs were predicted by beliefs about pseudoscientific and conspiracy claims; and conspiracy beliefs were predicted by beliefs about pseudoscientific and paranormal claims. Other individual difference variables were minimally predictive of each kind of belief. However, individuals predisposed towards intuitive thinking and who made ontological confusions were more likely to endorse paranormal and conspiracy claims. These results partially replicate Lobato et al. (2014), but provide a more nuanced description of the characteristics of believers and skeptics.

Emoticons are graphic signs, such as :-)(smiley face), that often accompany textual computer-mediated communication (Dresner & Herring, 2010). Emoticons are thought to be used to convey emotion (Derks et al, 2007; Wolf, 2000). Specifically, it is believed that emoticons supplement electronic communication with non-verbal cues, such as those often seen in face-to-face communication (Lo, 2008). We collected text messages from participants who also rated the messages across a number of emotional dimensions. We asked a separate group to rate the same messages on the same dimensions. By comparing when a particular emoticon was used, the emotional qualities of the message as rated by both the sender and reviewer, we aim to investigate which emoticon is likely to be included in text messages according to emotional intent, and whether this emotional intent will likely be understood.

This study aimed to identify which ERP are specifically related to response inhibition. Electrophysiological activity was recorded from 30 subjects, and was submitted to a temporospatial principal component analysis to detect and quantify the main components associated with response inhibition. A modified go/nogo composed of three types of stimuli (frequentGo, infrequentGo, and infrequentNogo) was used to dissociate activity related to response inhibition from that related to novelty processing. InfrequentGo and infrequentNogo trials differed in the type of response (execution vs. inhibition), but not in their frequency of appearance. Neither the anterior nor the posterior N2 displayed larger amplitudes for infrequentNogo than infrequentGo trials. By contrast, both the anterior and the posterior P3 showed larger amplitudes for infrequentNogo than for infrequentGo trials. Present results suggest that P3 plays a key role in motor response inhibition. These findings substantiate and extend the current evidence and previous findings from our group.

The visual system, like the brain more broadly, relies heavily on categorical representations. It is easier to spot a visual difference that crosses a category boundary, e.g., between blue and green, or between vertical and oblique. Here we show that topological relations between objects are similarly categorical. When asked to detect changes between object arrangements, participants were better at detecting those changes that crossed hypothesized category boundaries, such as ’overlapping’, or ’touching’, compared to equally-sized changes that did not. These effects were magnified at increased memory load, presumably because this categorization forms a more efficient code. This finding, consistent with previous computational modeling work, suggests that categorical relations are critical for remembering and comparing complex images.

Students’ beliefs about a subject influence their comprehension and learning of that subject (Ornek et. al., 2008). Many students consider science as a difficult subject to learn. Therefore, this study explored a new way in helping elementary children understand abstract science concepts using embodiment, or physically moving their own bodies. Students engaged in activities that helped them learn about abstract science concepts by physically performing tasks related to these science concepts. The purpose of this study was to examine the importance and role of embodiment in students’ understanding and motivation in elementary science learning. The results provide evidence to suggest that embodiment has remarkable potential to enhance both children’s understanding and motivation in abstract scientific concepts through the use embodiment.

Past research has shown that people use temporal information to detect and discriminate between different causal relationships and that timing-based causal inferences are modulated by explicit information and domain-appropriate expectations. Many of these past results suggest that learners make inferences about hidden causes from timing information, but there have been no systematic studies of the ways in which subtle changes in temporal information can shape inferences about the presence and nature of hidden causes. We present new results showing that people make nuanced causal inferences when faced with streams of events, using temporal information to infer the presence of simple generative relationships, independent and common hidden causes, and causal cycles. Interpreted in a Bayesian framework, these results shed light on the cues and tacit temporal expectations that people use to make efficient use of temporal information.

This paper presents an agent-based argumentation framework. Different from probabilistic, fuzzy and weighted approaches, this framework considers the strength of arguments and attacks from two aspects: the inner structure of arguments and the beliefs of agents. A key concept in this framework is the notion of satisfiability degree, which is used to define the intrinsic strength of attacks and the extrinsic strength of arguments. These two kinds of strengths are combined into the degree of attack/support. Then, new semantics of this framework are defined and the relation with Dung’s approach is discussed. 2938

This study investigated how knowledge of causality representations affects the reading in Japanese. In Experiment 1, 24 participants read events presented in cause-to-effect or effect-to-cause order without causal conjunctions. The latter received longer reading times and more regressions than the former even when probability and predictability were balanced. Experiment 2 examined whether this reading order effect was derived from a default reasoning asymmetry (i.e., reasoning from cause to effect is favored over that from effect to cause). We utilized epistemic constructions to explicitly identify the reasoning direction. Self-paced reading results from 24 participants showed that readers were more efficient when reading cause-event as the evidence, effect-event as the conclusion vis-a-vis the counterpart, confirming the reasoning asymmetry. However, reading order effect remained robust, presumably reflecting that causal reasoning is temporally embodied.

Present study examines the effect of incidental haptic sensations on acceptance of utilitarian judgment—seeking greater happiness in exchange for a few victims—under personal and impersonal moral dilemmas. Recently, Nakamura et al. (2014) indicated that physical coldness reduced empathic concern and facilitated utilitarian judgment in personal moral dilemma. It is also shown that tactile sensations such as softness and hardness affect our social judgment and empathic feeling to others. In this experiment, participants palmed either a soft cotton cushion or a hard iron tube while making moral judgments. Results showed that the hardness did not solely facilitate utilitarian judgment, however sex-tactile sensations interaction was modestly significant in personal dilemma. Specifically, males seemed to be more utilitarian while palming the iron tube. Results also showed that males were more utilitarian when they felt the sacrificed individuals were socially distant. Furthermore, contrary to expectation, high empathic feelings strengthened utilitarian judgment.

Neural precursors of voluntary actions appear before subjects report having decided on their behavior, leading some to dismiss a causal role for consciousness in decision-making. But the voluntary actions studied are typically arbitrary – bearing no purpose, meaning or consequence. We used EEG to directly compare deliberate and arbitrary decisions in a donation-preference task. Two NPOs appeared on the left/right of the screen, and subjects pressed the left/right button with the corresponding hand. In the deliberate condition, subjects’ choices led to monetary donations to the NPOs. In the arbitrary condition, both NPOs received donations irrespective of the choice. Early left/right ERP differences appeared 1s before the action only for arbitrary decisions. Following our earlier work, we interpreted these ERPs as reflecting random bias activity disjoint from decision-making processes. Our findings challenge previous studies, suggesting that early predictability of voluntary action does not generalize from arbitrary to more-interesting deliberate decisions.

This study discusses differences in problem solving performance among different domains presumably caused by the same instructional intervention. Discpline-based education research (DBER) acknowledges similarities in learners’ cognitive architecture that allow interventions to transfer among domains, but it also argues that each domain has characteristics that might affect how interventions impact learning. The present study uses an instructional design technique that had previously improved learners’ problem solving performance in programming: subgoal labeled procedural instructions and worked examples. The present study explores the effect of this technique for solving problems in statistics and chemistry. The problem solving procedures in the three domains have different characteristics. Similarly, each of the three experiments has a different pattern of results for problem solving performance. This study concludes that subgoal labeled worked examples seem to be equally effective across the different domains. Subgoal labeled procedural instructions, however, seem to be most effective for more complex procedures.

We examined the role of text in learning to replace science misconceptions. Undergraduates’ beliefs about where a coin falls when dropped by someone walking were assessed. A common misconception is that a coin will fall straight down, but its forward motion actually continues before it hits the ground. 135 students who expressed this misconception read one of three passages about the issue. The passages differed in whether the misconception was explicitly stated, only implied, or not mentioned at all. Past research shows that calling a misconception to the foreground helps people overcome the misconception (Broughton & Sinatra, 2010). We found a significant difference across conditions, with 86% of those who saw the explicit misconception, 72% of those who saw the implicit reference to the misconception, and 59% of those who saw no reference to the misconception correcting their mistake (2(2, N = 135) = 8.22, p = .016).

The linguistic input a child receives in the first years of life is foundational for cognitive and language development. In a corpus analysis, the vocabulary in picture books was richer and more extensive than that found in child-directed and even adult-directed speech. The grammar and complexity of these communication media, measured by reading grade level, indicated that picture books averaged two grades higher than child-directed speech and one grade higher than adult-directed speech. These differences between written and spoken language can be more adequately described by formal versus informal genres rather than their oral or written media. Given that the child will read words and grammar not experienced in speech, these results question the feasibility of the popular view that a child’s reading task is simply to “decode” the written language into spoken language. A framework of acquiring literacy informally before schooling begins is described and explained.

Recently, the user interface (UI) has become more important because devices have become complex and give us excessive information. In this study, we investigated “what users learn” through information search training. We tested the hypothesis that graphical UI users would acquire perceptual knowledge (the knowledge of the visual features of the display) rather than structural knowledge (the knowledge of how information is categorized). The results of the experiment showed that participants acquired little structural knowledge. Further, the computer simulation results indicated that structural knowledge was inadequate to explain the search patterns observed in the experiment. The model that had perceptual knowledge was better able to explain the results of the experiment.

A key problem for models of joint action is to explain how co-ordination is established and sustained. Existing accounts emphasize the importance of interaction, demonstrating how collaborative feedback leads to more systematized, stable, and partner-specific referring conventions. However, in addition to conventionalizing referring expressions, recent work demonstrates how interlocutors also rapidly establish procedural conventions for resolving sequential and temporal co-ordination problems in the interaction. It is unclear, however, whether interlocutors associate these procedural conventions with specific conversational partners. To address this question, we report a collaborative, 3-participant, computer-mediated task which presents participants with the recurrent co-ordination problem of ordering their actions and utterances into a single sequence. Artificially generated clarification requests are inserted into the dialogue, that appear, to each participant, as if they originate from either of the 2 other participants. We argue that participants’ responses to these clarifications provide evidence of interlocutors associating procedural conventions with specific partners.

We used a novel experimental paradigm to investigate cognitive principles underpinning human communication. Through a computer game simulating different virtual scenes, pairs of participants sent and interpreted non-linguistic, minimal signals to achieve common goals. Participants’ signalling and interpretative actions demonstrated both flexibility and sensitivity to variations in the context of the shared visual scene: the same signal in one context could ‘flip’ its meaning in a new context. Such ‘invertible’ signals in the lab have their counterparts in patterns of real-world natural language use—from the phenomenon of enantiosemy (words/phrases that contain their ‘opposite’ meaning) to the pragmatics of satire and irony. But the emergence of such signals in our experiment challenges both correlational associative and recursive-mentalizing (‘mind-reading’) accounts of human communication and language. Instead, we point to a pragmatics-central perspective in which what is vital is our capacity for joint inference and coordination.

We conducted an experiment using two Japanese typographies and sixteen words, having positive/negative emotional valence, to investigate the effects of harmonization between a word’s meaning and typography for harmonized and ill-harmonized conditions. Four different words - two positive and two negative - were simultaneously presented at the four corners of a display. Participants’ eye movements were recorded as they search a word on the display after hearing it. We analyzed the gaze duration in each region of the display (target/distractors). In the ill-harmonized condition the target was found earlier compared with the harmonized condition. However the words were finished reading at same time in both condition. The results imply that the target and distractors were processed automatically because of the increase in words’ perceptual fluency due to harmonization, thereby resulting in the delayed finding of the target.

Reading to young, pre-literate children is associated with better language and reading outcomes, but the underlying mechanisms are poorly understood. The goal of this work is to better understand the potential mechanisms. We hypothesized that vocabulary diversity and sentence complexity might vary between picture books and child-directed speech, and we wanted to quantify those potential differences. We built a corpus consisting of the text of 100 picture books that caregivers might read to pre-literate children. We compared the distributions of vocabulary and certain complex sentences of that corpus to child-directed speech from the CHILDES corpus. We found that picture books contained a higher number of unique word types for a given number of tokens, and contained a higher proportion of complex sentences. The mechanisms by which shared book reading may contribute to improved language outcomes is by exposing children to words and sentence structures that they would not encounter otherwise.

Assuming that photographs accumulated in a personal computer reflect life history of a user, a model of one’s autobiographical memory could be constructed. Such a model will be useful to support memory problems caused by such as aging. Based on this idea, we constructed an image recommender system including an ACT-R model. We build the model using the first author’s private photo library, consisting 3,202 photos. We run a simulation manipulating the activation noise of declarative chunks. As a result, we found strong influence of the noise on memory retrieval. When the noise level was low, the model retrieved a few memory items that occurred recently. On the other hand, when the noise level was high, the retrieval process was like random walk over a memory network. Repeated recalls of old photos occurred. The result suggests a condition of an ACT-R model enabling a travel into a distant past.

Consciousness’ role in high-level semantic integration is still unclear. Here, we presented masked pairs of images, which could be unrelated (e.g., a broken plate and an eagle), associatively related (e.g., a broken plate and a fork) or abstractly related (e.g., a broken plate and a fighting couple). Low-level features of the pairs were controlled for. In each trial, a masked pair (prime) was followed by a second pair (target) of a similar or different type. When the prime pair was visible, equal priming effects were found for both associatively related and abstractly related pairs. Yet when primes were rendered invisible, only associatively related pairs affected target processing. Our findings go beyond previous ones by demonstrating that two simultaneously presented distinct objects can be unconsciously integrated. Critically however, they suggest a crucial role for consciousness in processing semantic relations that transcend those of simple categorical associations

A key challenge faced by children in vocabulary acquisition is learning which of the many possible meanings is appropriate for a word. The word generalization problem refers to how children associate a word such as dog with a meaning at the appropriate category level in the taxonomy of objects, such as Dalmatians, dogs, or animals. We present extensions to a cross-situational learner that enable the first computational study of word generalization integrated within a word learning model. The model simulates child patterns of word generalization due to the interaction of type and token frequencies in the input data, an influence often observed in usage-based approaches to underlie people’s generalization of linguistic categories.

UK bookies (bookmakers) herd geographically in less-affluent areas. The present work shows that UK bookies also herd with the special bets that they advertise to consumers, both in their shop window advertising and on TV adverts as shown to millions of viewers. I report an observational study of betting adverts over the 2014 soccer World Cup. Bet types vary in complexity, with complex types having the highest expected losses. Bookies herded on a common strategy of advertising special bets on two levels: by almost exclusively advertising complex bet types with high expected losses, and by advertising representative events within a given complex bet type. This evidence is most consistent with bookies’ advertising targeting a representativeness heuristic amongst bettors. Bookies may know how to nudge bettors toward larger losses

In some tasks, children show improved performance when allowed to make choices about task features. We conducted two experiments to examine the nature of this effect on 4-year-olds’ cognitive flexibility. Children completed one of two conditions of the dimensional change card sort (DCCS), wherein children are asked to sort items by one dimension (e.g., shape), and then to switch and sort by another (e.g., color). In the standard condition, children were instructed to switch and sort by the second dimension after 6 pre-switch trials. In the choice condition, children were additionally allowed to make a choice before the rule switch (e.g., to touch either a sun or a moon icon). Children’s post-switch performance was significantly higher in the choice condition than in the instructed condition, indicating that giving children choice can aid their cognitive flexibility, even if this choice is not task relevant.

To facilitate collaborative tasks that include different but related subordinate tasks, it is important to consider the consistency and coordination between the objectives of the subordinate tasks (local objectives) and the purposes of the entire task (global purposes). In this study, we propose a method that alternately propagates local objectives and global purposes during a collaboration through interactions using a two-layer model. We investigated the effects of the proposed method on human stress levels. We conducted an experiment in which we used two types of agents to evaluate the effect. Questionnaires confirm that the proposed method significantly improved impressions of consistency and naturalness. The results of our electrocardiogram analyses confirm that the participants’ stress levels increased throughout the task when they interacted with the traditional agent. The analyses show the possibility that physiological indices can use to evaluate the collaborative task performance from the viewpoint of human stress.

Considerable studies indicate that structuring concepts within a diagram enhances learning (e.g. concept mapping), as opposed to learning by just reading the information. We examine whether the differences in the structure of information (hierarchical or linear string) affect learning based on the formation and drawing of a diagram. In our experiment, participants learned a family tree consisting of 6 members (Hierarchical) or the order of 6 participants in a relay race (Linear). While learning, half of the participants in each condition produced a drawing of the family tree or flow diagram of relay order (Drawing). In contrast, the other half read presented diagrams and wrote down the names (Reading). The results revealed that while participants in Hierarchical- Drawing performed better on the post-test than those in Hierarchical-Reading, there was no difference between the performance of participants in Linear-Drawing and Linear-Reading. That suggests the structural factor would affect the learning.

Previous studies indicated that facilitating effects of implicit hints on insight problem solving are not universal. To clarify the mechanisms of this variability, the relationship between the use of hints and individual differences in personality traits were investigated. Participants engaged in a Remote Associates Test in which solution words were subliminally presented in one third of the trials. During the test, participant’s pulse rate was measured as an indicator of arousal. After the test, participants completed the Big Five personality scale (TIPI-J). The participants’ “extroversion” and “openness” were positively correlated with the effect of hints during low pulse rate, whereas they were negatively correlated during high pulse rate. These results suggest that solver during low arousal could utilize the cues, and their search through the problem space may become broader. During high arousal, however, their focus attention may become narrower, and extrinsic cues may not be associated with the problem.

I examined the effect of the working memory capacity (WMC) of older adult participants on tasks using touch interfaces, by using an extreme-groups design. Older participants (N = 100) completed a single tapping task and verbal, numerical, and spatial WMC tasks. To test whether the response time in the single tapping task differed as a result of the WMC, I performed a 2 x 2 Analyses of Variance with WMC (high, n = 25 /low, n = 25) as the between-subjects factors and the tapping interface (a touch pen, a finger, or a computer mouse) as within-subject factors. This indicated a significant interaction between the WMC and the tapping interface. The results suggested that the response time of participants with high WMC was shorter than the response time of participants with a low WMC, when using a touch pen and a computer mouse interface

Gigerenzer and colleagues have proposed the ‘adaptive toolbox of heuristics’ as an account of resource-bounded human decision-making. According to these authors, evolution has endowed such toolboxes with ‘ecological rationality’, defined as the ability to make good quality decisions in their specific environments. Here we explore to what extent the mechanisms of evolution alone are sufficient to explain the emergence of ecologically rational toolboxes. It is not clear how evolution can lead to ecologically rational toolboxes within the space of possible toolboxes. That is, even if one assumes a very simple environment (e.g., 10 cues and 50 decisions), the number of possible toolboxes (10ˆ72) is still astronomical. By using artificial evolution simulations we investigated the evolvability of ecologically rational toolboxes. We present preliminary results showing that evolution can produce toolboxes of heuristics that are “good enough” to survive, but those toolboxes are not ecologically rational.

We investigated (1) how focal brain-injured patients describe causal events (causal verb like “push” and the instrument of the action like “the stick”) in speech and co-speech gestures and (2) whether gestures compensate for their impaired verbalization. 16 left hemisphere damaged (LHD), 16 right hemisphere damaged (RHD) and 14 controls were asked to describe causal events (22 video clips). The correct use of causal action components in speech and iconic gestures referring to these actions were coded. Results indicated that LHD patients were less accurate in using both components in speech compared to RHD and controls. There was no difference in the number of iconic gestures among groups. Yet, LHD patients were more likely to omit or misuse both components in speech and in gesture than RHD and controls. Particularly, damage to the left inferior and middle frontal gyrus resulted in problems in both modalities, suggesting conceptual deficits of causality.

There are mixed findings with respect to individual or gender differences in virtual Morris water maze tasks, which may be attributed to variations in the scale of the space, the cues provided, and differences in spatial navigation experience and abilities. We explore the question of scale and context by presenting participants with either a large (146 m) or small (36 m) outdoor virtual Morris maze, along with a measure of lifetime mobility and mental rotation skills. Results of this study suggest that, for the small-scale environment, males and females performed similarly when asked to navigate with only close visual cues. However, males outperformed females when only far cues were visible. In the large-scale environment, males outperformed females in both cue conditions. Additionally, mental rotation abilities predicted better navigation performance with close cues only. Finally, we found that highly mobile females and males perform equally well when navigating with close cues.

The present study was designed to explore how learning magnitude knowledge and learning procedural knowledge, with respect to both whole numbers and fractions, might be causally related. Neither magnitude knowledge nor procedural knowledge is necessary or sufficient for learning the other, and yet, correlations between the two are ubiquitous (e.g., Siegler & Pyke, 2013). Using correlational data (Structural Equation Models) and accuracy data (Knowledge Space Theory), potential causal models to describe the data were tested. Structural equation models did not differentiate between learning magnitude knowledge helping to learn procedural knowledge or vice versa. However, knowledge space model testing models of accuracy data provides support for the notion that learning procedural knowledge helps learning magnitude knowledge, and evidence against the reverse notion that learning magnitude knowledge helps learning procedural knowledge. Key Words: Magnitude; Procedure; Structural Equation Models; Knowledge Space Theory

Finger gnosis and magnitude comparison were examined as predictors of adult numeracy. Previous findings were extended by (1) controlling for domain-general comparison processes (using a luminance judgment task), (2) controlling for visuo-spatial memory, and (3) examining the robustness of the relations across different numeracy tests, including exact and approximate calculations. Control variables were entered in the first step of a multiple regression, with finger gnosis and magnitude comparison entered as a second step. Finger gnosis and symbolic magnitude comparison predicted unique variance in adults’ calculation fluency, computational estimation, and Woodcock Johnson calculation scores. The control variables, luminance comparison and visuo-spatial memory, did not account for significant variance in the numeracy outcomes, nor did non-symbolic magnitude comparison. These findings suggest that (1) the relation between finger gnosis and numeracy does not reflect visuo-spatial memory and (2) the relation between magnitude comparison and numeracy reflects number representations, rather than domain general processes.

Crossmodal music-to-color correspondences are mediated by emotion for classical music (Palmer et al., 2013) and diverse other genres. People tend to choose colors as going best with music when both have similar emotional associations (e.g., happy-looking colors go best with happy-sounding music). Lower level musical stimuli, including single-line melodies, two-note intervals, and instrumental timbres show analogous emotional effects, as do music-to-texture associations (Peterson et al., VSS-2014). Other crossmodal correspondences without emotional mediation (e.g., size/pitch associations) affect classification performance, modulate motion perception, and influence multisensory integration/perception (Spence, 2011). Do emotionally mediated crossmodal correspondences also produce such effects? We find that people are slower and/or less accurate at classifying the emotionality of stimuli (e.g., saturated yellow as happy) when simultaneously presented auditory stimuli are emotionally incongruent (e.g., the sad sound of a clarinet or minor chord) than when it is congruent (e.g., the happy sound of a piano or major chord).

Linearly organized structures in language are supposed to be easy, while hierarchical information is difficult to process. Traditional accounts attribute the difficulty of processing hierarchical sentences (the dog the man walks, barks) to the long positional distances between dependencies (Gibson, 1998). Alternately, linear structures (the man walks the dog that barks) are easier to process. In a sentence comprehension study, structure (i.e., positional distance between dependencies) was manipulated (hierarchical versus linear), and congruency between the semantic and the positional dependencies, being either congruent as in the dog the man walks, barks, neutral, as in the dog the man sees, walks, or incongruent as in the man the dog walks, barks (barks being syntactically dependent on man, but semantically on dog). The data show that structure did not, whilst semantic-syntactic congruency did strongly affect comprehension, suggesting a striking new perspective on the cognitive versus formal complexity of human language.

As children gain knowledge about the world, the organization of their conceptual knowledge becomes increasingly complex, as reflected by the successive emergence of sensitivity to different types of similarity over the course of development. At the start, when judging the similarity of two objects, infants rely on perceptual similarity. By age 3, children often make these judgments based on object co-occurrence, or thematic similarity. Finally, after ages 5-6, children reliably start to prefer taxonomic similarity. Though these phenomena have been well-studied, they are often explained by reference to separate mechanisms that are stipulated to come on-line at specific ages. We present a PDP model that learns from the structure of its environment and exhibits transitions in the relative salience of perceptual, thematic, and taxonomic similarity, as observed empirically, without any changes in its underlying learning mechanism or training environment.

How is the outcome of the categorization process represented? This question has gone virtually unaddressed. A notable exception is Barsalou (2012) which proposes that categorization results in a type–token predication, whereby the type representation (e.g. DOG) is predicated of the token representation of the categorized individual (e.g. Fido). Another is Jackendoff (1983) which proposes that categorization results in a token representation being related to the type representation via a two-place IS-AN-INSTANCE-OF function. Despite important differences, both proposals assume that type and token representations are extrinsically related to one another. This contrasts with recent research (Prasada & Dillingham, 2009; Prasada, 2013) which makes use of instance-of-kind representations in which type and token are intrinsically related. This poster identifies theoretical and empirical implications of the two approaches for representing the output of categorization, and argues that these favor the instance-of-kind representations.

Infants undergo a series of dynamic changes during the first year of their life, starting at nearly complete dependence upon others for all functions, culminating with the ability to self-transport and rudimentarily converse around the age of 12 months. Previous research indicates an interaction between walking and expressive and receptive language development. Given that phonology underlies expressive language production, in this study we are exploring potential relationships between locomotion and phonological development by examining phonetic inventories of age-matched peers who are walking or crawling. We are transcribing canonical utterances from high-volubility samples taken from daylong home audio recordings of 18 English-learning infants. Various phonetic features of the infants’ productions are compared across locomotor groups, in search of an interaction between phonological and locomotor development. This work is informing our understanding of the mechanisms through which locomotor and language development are interrelated.

Although simulations can be useful tools to train dynamic decision making (DDM) skills, studies show that mere practice with simulated environments leads to limited improvements in performance. Simulated environments often show little or no transparency about the underlying structure. Making information about the system and the consequences of decisions available to users has been found to enhance learning. We tested a glass-box approach using highly interactive feedback tools to support implicit learning in a 3-hour DDM training session. Ninety participants were assigned to either the control (no training) or implicit learning condition. While performance on the training scenario improved over time, learning took place mostly in the beginning of the training session, and final performance remained far from optimal. Performance in the training scenario was positively correlated to performance in a test scenario. However, implicit learning did not improve performance on the test scenario compared to the control group.

The present study addresses (1) whether Chinese classifier-noun integration is syntactic or semantic in nature, and (2) whether the Anterior Negativity in brainwaves is a separable component indexing automatic morphosyntactic processing (Hagoort, 2003) or instead results from overlapping N400 and P600 components (Tanner, 2014). In Chinese, classifiers (e.g., a sheet of) must be used whenever any noun is quantified or specified and must be congruent with noun meaning. Thirty-three Mandarin speakers read 120 sets of sentences that manipulated classifier-noun congruency (There is a machine-like-classifier/sheet-likeclassifier computer on the table) and classifier presence (a machine-like-classifier computer vs. *a computer). A larger N400 component in the incongruent condition suggests that classifier-noun integration is primarily semantic. In the classifier-absent condition, a P600 was observed during the first half of the experiment but that diminished during the second half and an apparent Anterior Negativity emerged, suggesting that readers changed their processing strategy over time.

This research project has two parts. First, I argue that empathy is not necessary and not sufficient for morality. I use autistic individuals and individuals who suffer from psychopathy as my primary examples to show that empathy is not a significant source of morality. I review the literature in moral psychology and emotion research to show that emotions other than empathy, primarily disgust, are responsible for morality. Second, I provide initial findings from my meta-analysis to show which emotions are most prominent in moral judgment and moral behaviour.

Researchers argue that dissatisfaction with a misconception is a prerequisite for adopting an alternative conception and that having clear feedback aids learning. The present study investigated the importance of ambiguity (having response options that support both the misconception and target learning category), falsification, and category induction opportunities when overriding a prior conception in favor of a new conception. The results suggest that ambiguity and direct falsification opportunities may aid in learning more than having both direct falsification and induction opportunities, which may be better than ambiguity and providing induction opportunities without direct falsification. Ambiguity may improve learning when coupled with falsification opportunities. Implications are discussed.

A central goal of typological research is to characterize linguistic features in terms of both their functional role and their fit to social and cognitive systems. One longstanding puzzle concerns why certain languages employ grammatical gender, which assigns nouns to distinct classes and marks neighboring words for agreement. While historically noun classification has been viewed as a useless ornament with little apparent rhyme or reason, there is an accumulating body of evidence that native speakers use determiners to guide lexical access. Here, we compare the communicative function of gender marking in German (a deterministic system) to that of prenominal adjective use in English (a probabilistic one), finding that despite their differences, both systems efficiently smooth information over discourse, making upcoming nouns more equally predictable in context. We hypothesize that evolutionary pressures may favor one system over another on account of how easy they are for children and adults to acquire.

Much of our interpersonal communication conforms to Gricean-style norms governing the truthfulness, informativeness and relevance of the information exchanged. But we also experience untruthful, uninformative, and misleading communication when these norms are violated. How do people draw upon this experience when attempting to conceal the truth? We introduce a computational model which predicts how people should best conceal the truth when required to reveal information to another (and lying is not an option). We argue that when placed in such situations, people will take into account the other’s expectations of whether Gricean norms apply. This notion is incorporated in our model, which we test with an experiment that manipulates people’s assumptions in this regard. Results show that revealing informative but misleading information is an acceptable strategy when the other expects cooperation; otherwise, being uninformative is overwhelmingly preferred. We analyse how our model and alternatives account for these results.

The ability of animals to learn complex tasks from reward is still not fully understood. While models of such reinforcement learning exist (e.g., Sutton & Barto 1990), it is unclear how these models might reflect the functioning of biological neural networks. One recent model has shown how networks of spiking neurons can model reinforcement learning in animals (Chorley & Seth 2011), however, it and the existing models of animal behavior do not fully account for learning phenomena such as extinction, spontaneous recovery, and gradual extinction (Gershman et al. 2013). Interestingly, the model’s failures in extinction and spontaneous recovery are the same as those of rodents with hippocampal lesions (e.g., Kimble & Kimble 1970), suggesting that adding a spiking model of the hippocampus will better model these phenomena. We present data on the original model’s performance in learning, extinction, and spontaneous recovery, and we explore modifications to better capture these phenomena.

The Emory-Tibet Science Initiative (ETSI), a two-way exchange betweenWestern science and Tibetan Buddhism, is a partnership between Emory University, the Dalai Lama, and the Library of TibetanWorks and Archives in Dharamsala, India. ETSI is a comprehensive 6-year science curriculum being implemented at Tibetan Buddhist monasteries in India, representing the most significant change in 600 years for the Tibetan Buddhist monastic curriculum. This two-way exchange between science and Buddhism offers potential for mutual enrichment leading to new discoveries. Yet a cross-cultural challenge exists between science faculty and monastic students in teaching and learning science, as both traditions can hold quite different understandings of fundamental concepts, including sentience, awareness, attention, and perception. Using cultural consensus modeling, we estimate Tibetan Buddhist concepts of core cognitive science constructs and compare them with Western scientific definitions. Results can enhance cross-cultural science education by supporting faculty in understanding students’ cultural concepts, and vice versa.

Standing balance control relies on multisensory feedback, but little is known about the influence of periodically varying sounds on this process. The level of sensorimotor activation has been shown to be highly correlated with the concept of musical groove. We presented musical stimuli with high and low groove ratings to participants (N=40) as center of pressure (CoP) was recorded using a force platform. We found an effect of groove on radial sway in both non-musician and musician groups, with the high groove condition accompanied by the least amount of sway. Further analysis revealed a stronger correlation between musical events and postural sway deviations in the high groove condition when compared with the low groove condition, providing support for auditory-motor entrainment in postural sway. Our results show that periodicity in music can reduce sway variability in standing balance, possibly due to involuntary motor entrainment.

Recently, both psychology and machine learning have explored the ability of learners to ask questions. However, much of this work has focused on a single type of question: a “label query”. When making a label query, the learner selects an unfamiliar (unlabeled) item and requests a label for it (e.g., ”What is this?”). We hypothesized that people often prefer much richer types of questions (e.g., feature queries: “Is this feature relevant?”, demonstration queries: “Can I see an example of a ladybug?”, etc.). To study this behavior, we had people play a simple game where they generated natural language questions to determine a hidden configuration of objects. We compute the normative value of these rich questions as measured by modelbased analyses (e.g., information gain). A second experiment evaluates the ability of human observers to judge this value when the demands of question generation are removed.

Previous studies on movement control modulated target size or distance in order to investigate the mechanism underlying control strategies. While these elucidated the contribution of targets in the control, how trajectory itself influences movement strategy has received little attention. Here, using event-related fMRI, we examined neural processes of trajectorybased movement control as well as those of target-based control; we manipulated the focus of movement control by varying the size of a target (target-based control) or the window through which the movement had to pass (trajectory-based control). The brain activation maps showed that the increase in task difficulty with target-based control was associated with greater activation at right parietal and ventrolateral occipital regions, while that with trajectory-based control was correlated with more extensive activation at medial frontal and ventromedial occipital regions. These differential brain activities indicate that the neural mechanisms for target selection and trajectory control are distinct during visually-guided movements.

To investigate accuracy and awareness of human perceptions of image manipulation, we examined 80 participants’ verbal reports of image manipulation in 14 manipulated and unmanipulated images, and compared these responses to data recordings of their eye movements. We found that in aggregate participants achieved only 56.0% success in correctly identifying whether images have been manipulated or not in verbal responses. Examination of participant eyegaze recordings shows that in some cases the participants’ eyegaze fixates on the manipulated elements of images although they fail to report this verbally, indicating that they may perceive the manipulations at a non-conscious level.

Dual-process theories (DPT) of cognition posit that performance differences in reasoning stem from an interplay between heuristics-based processing (i.e., System 1) and more controlled, rule-based processing (System 2). Emerging evidence suggests that solving classic base-rate problems via Bayesian inference depends on adequately inhibiting the prepotent representations elicited by System 1 (De Neys, 2014). We propose that DPTs may benefit probabilistic models of reasoning by providing a framework on which to map individual difference predictions (e.g., how inhibitory capacity, prior knowledge, and motivation influence adherence to probabilistic rules). We present a dual-process computational model that implements various normative (i.e., Bayesian) and non-normative rules, which in turn are probabilistically fired based on a functional relationship between relative (de)activations of each system and variability in agents’ inhibitory capacity and motivation. Simulation results map onto behavioral data and replicate a variety of base-rate performance patterns, including base-rate neglect

When asked to combine two pieces of evidence, one diagnostic and one non-diagnostic, people show a dilution effect: the addition of non-diagnostic evidence dilutes the overall strength of the evidence. This non-normative effect has been found in a variety of tasks and has been taken as evidence that people inappropriately combine information. We investigated the dilution effect using simple perceptual stimuli but unlike in previous work we asked participants to judge likelihoods ratios, allowing us to assess not just ordinal relationship between judgments but also whether each individual judgment was accurate. We found the dilution effect, but surprisingly it was not due to inaccurate combination of diagnostic and non-diagnostic information. People were accurate at judging diagnostic evidence combined with non-diagnostic evidence, but overestimated the strength of diagnostic evidence alone. We explain this within-participants dilution effect as the result of inference about missing features rather than incorrect combination of information.

Participants pressed computer keys to keep a moving dot stimulus within a rectangle, either alone or with a partner they could neither see nor hear. Pressing the A-key or L-key caused the dot to move right or left, respectively, for as long as the key was pressed. Switching between the A and L keys (i.e., turning) proved challenging: concurrently pressing both keys made the stimulus move upward, while pressing neither key made it move downward. Individuals performed better than dyads because they turned the dot near the edge of the rectangle and let it coast back and forth within the rectangle. Dyads turned the dot in the middle of the rectangle because they pressed their buttons as quickly as possible. These findings support the assertion that pairs require external feedback regarding the other’s actions during tasks necessitating anticipatory actions (Knoblich & Jordan, 2003; Van Der Wel, Knoblich, & Sebanz, 2011).

Little is known about how the experience of being bilingual and speaking two languages leads to advantages in cognitive control and flexibility (e.g., Bialystok & Martin, 2004). This study investigated productive vocabulary and knowledge of translation equivalents (TEs) as possible mechanisms underlying the bilingual advantage in cognitive flexibility and control. Spanish-English bilingual two-year-olds performed the Reverse Categorization Task (RCT; Carlson et al., 2004), which requires cognitive flexibility and control. Each child’s caregiver also completed the MCDI for English and Spanish to obtain a measure of each child’s productive vocabulary and knowledge of TEs. Correlation analyses showed that performance on the RCT was significantly correlated with productive knowledge of TEs but not cognates. These findings suggest that the experience of producing different words with the same meaning in two languages, as well as choosing between those words, may be an early mechanism underlying the bilingual advantage in cognitive control.

Prior research suggests that discounting rates– how much interest a person requests for waiting a period of time before collecting benefits– are influenced by perceptions of time. Other research, however, suggests people understand time via spatial representations. Thus, the current research examined whether underlying spatial representations of time influence these rates. Interest rate preferences were assessed twice over a semester from forty students in either art or cognition courses after drawing a picture with perspective or no perspective. Results revealed that drawing pictures without perspective led to higher average interest rates than drawing pictures with perspective. Additionally, there was an interaction of session and course; cognitive students’ rates increased substantially over time, while art students (i.e., students with practiced spatial representations) did not show this effect. These results are a preliminary step in suggesting that spatial priming can affect temporal representations, which in turn change discounting rates.

This study investigated the effect of empathy in text comprehension. After 89 university students read a document which described how to write an educational practical report, they took a comprehension test and responded on the following scales: parallel empathy, reactive empathy, subjective comprehension, and attitude. In the framework of dual-process theory, parallel empathy depends on system 1 while reactive empathy is controlled by system 2. As a result, the mean comprehension test score in the condition in which students read the document describing only procedure and some cautions was higher, but the mean reactive empathy score was lower than that in the condition in which students read the document including empathic episodes of the author with illustrations, in addition to the procedure and cautions. An analysis by structural equation modeling with previous data suggested that adding empathic episode disturbed text comprehension, but enhancing empathy promoted subjective comprehension and attitude change.

We examine whether the familiarity of thematic music affects the degree to which concepts associated with the music are activated after listening to the music and subsequently affect generation task performance. In two experiments, participants listened to one of two excerpts of war-themed music varying in familiarity either before or after completing Amabile’s (1985) American Haiku task. Haikus were examined to determine the degree to which concepts associated with the music were affected by music familiarity. Experiment 1 demonstrated that associated music concepts for both familiar and unfamiliar music were included in the haiku at equal rates when the music was listened to prior to writing the haiku. Experiment 2 demonstrated that listening to moderately familiar rather than unfamiliar music before the haiku task resulted in more music associates being included. Explanations of how familiarity and other factors affect incorporation of war-themed music concepts into the haiku will be discussed.

Highly predictive sentential contexts can facilitate the generation of expectancies for low-level physical form-based properties of upcoming linguistic input. When contextual information (such as words in a sentence) constrains upcoming input to a specific syntactic category, words with category-typical physical form-based properties are processed more quickly than category atypical words. We aim to determine whether expectancies for grammatical category can be induced in experimental paradigms where words are presented in isolation. We demonstrate that properties of the stimuli to which participants respond can facilitate, through experience with the task, the generation of category-based expectancies. When all words were nouns, participants were more accurate on lexical decision and category judgments when targets possessed category-typical form-based features. When words from multiple categories were present, however, the typicality effect disappeared, suggesting higher-level expectancies can be induced without sentential context and modulate the effects of lexical- and form-based properties of words. 2988

In parallel with the development of the theory of distributed cognition, the study of distributed computing has progressed rapidly. This research has both been driven by pragmatic insights into the practicalities of coordinating computational processes and resulted in formal theories of distributed systems. We believe that advances in distributed computation research can be used to refine understanding of distributed cognition by establishing new metrics for evaluating cognitive systems and new methods for modeling cognitive ecologies. To illustrate this we revisit classical distributed cognition scenarios from a distributed systems perspective

Many conceptual change theories posit that change occurs due to a variety of cognitive, social, and emotional factors (Dole & Sinatra, 1998; Ohlsson, 2011), however, few theories have tested these claims via computational models of conceptual change. In this paper, we present a hierarchical Bayesian model that addresses change processes and their effects on re-categorization, a form of concept change. Human data from a study using the re-categorization paradigm (Ramsburg & Ohlsson, 2013) are compared to the computational model. The structure of the human data suggests the ‘non-monotonic’ nature of conceptual change (Ohlsson, 2011) as indicated by the best-fit learning curves. For several such curves, model comparisons suggest good fits between the computational simulations and human data. The nonlinear form of the model’s update functions lends additional support to concept change as a non-monotonic process. The model is discussed as a “proof of concept” for future conceptual change modeling endeavors.

The Cognitive Systems Institute Group (CSIG) is a relatively new initiative of IBM Global University Programs to better link IBMers and the cognitive systems community around grand challenge problems in cognitive science and artificial intelligence. For example, the mission of the CSIG is to augment and scale human expertise with cognitive assistants for all occupations in smart service systems. This poster will introduce CSIG and invite the CogSci community to help shape its evolution.

We present a series of propositions that explains why people find sitting quietly in a dark room strongly aversive (Wilson et al., 2014). (i) Conflict-monitoring is an essential cognitive function; likely performed at the level of information processing conflicts (Botvinick et al, 2001) (ii) Memory is sensitized to processing conflicts; if a conflict has not been resolved in real-time, it is recalled when the mind is disengaged (iii) This is mind-wandering (Smallwood et al, 2003) (iv) Since mind-wandering privileges conflict recall for resolution, and resolving conflicts requires effort, mind-wandering becomes aversive (v) To avoid mind-wandering, a common strategy is to increase intensity of activity, so mind has no time to wander (vi) But increasing density of activity increases the number of possible information conflicts, which further deepens aversion to sitting quietly (vii) This is anxiety Understanding the cognitive mechanics of this spiral of anxiety may help break it

How does network topology affect neural coding? We approached this question with a large parametric study simulating clustered network topologies of cortical excitatory spiking neurons with inhibitory interneurons, while taking into account variance in axonal length and spike propagation times. To evaluate the stability of rate coded information, we systematically varied within cluster conduction delay means, variances, and connection densities, as well as between cluster conduction delay means, variances, and connection densities. Networks received rate coded stimulation from one cluster, and we varied frequency and spike jitter of this input. Networks contained 960 excitatory and 240 inhibitory neurons, divided evenly between 6 recurrently connected clusters. We found that variances of inter-spike intervals in the presence of rate coded stimulation were greatly increased with the introduction of even small variances in between cluster conduction delays and by changes in inter-cluster and within-cluster connection density, identifying topologies that resist stable rate coding.

It is well known that listeners of probably all languages give verbal and non-verbal signals, called backchannels, to their interlocutors. However, it is not well understood what drives listeners to backchannel. To what degree are they an indicator of listener attention? Are backchannels semantically motivated, performed when the message has been parsed and comprehended? Or are they automatic responses triggered by overt cues from the speaker (such as eye contact, gestures, or prosodic information), requiring minimal comprehension? An important first step in answering this question is identifying what overt speaker cues trigger backchannels, and to what degree. This preliminary study looks at storytelling data from conversational dyads. We find that the speaker cue most likely to ‘successfully’ trigger a backchannel is making eye contact. Interestingly, however, other cues are more likely to trigger different kinds of backchannels: gestural cues trigger more head nods, while prosodic cues trigger more verbal backchannels.

The development of representational competence was investigated by using 3D (concrete and virtual) models as feedback in teaching organic chemistry students to translate between 2D diagrams. In two experiments, students translated between diagrams of molecules and received verbal feedback in one of three intervention conditions: with concrete models, with virtual models, or without models. Learning was measured in three posttests (with models, without models, and after a 7-day delay). Virtual models had either low (Study 1) or high (Study 2) congruence between actions performed with the input device and resulting movements of the virtual model. In terms of learning outcomes, model-based feedback was superior to verbal-feedback alone, models functioned as a scaffold rather than as a crutch, and learning with model-based feedback was resilient over a 7-day delay. Finally, concrete and virtual models were equivalent in promoting learning, and action congruence did not affect learning.

Newcomb’s paradox is a famous thought experiment in the field of decision theory. There are two paradoxical, yet “rational”, strategies to approach this decision-making problem. We addressed this debate by testing the paradox in a real world experiment. Analyzing participants’ mouse movements allowed us to reveal the internal cognitive dynamics of their thought process during the task explanation as well as the actual decision. This knowledge of internal processes helped us to accurately (73%) predict their decision before it was made. Moreover, the consistency of mouse movements before and during the actual decision significantly interacted with RT. This suggests that subjects were revealing their indecision in the mouse movement, and that this indecision weighed on both possibilities. This work has implications for exploring human decision-making, as well as predicting consumers’ choice in online setups.

This paper discusses the processing of topic meaning in metaphor comprehension, which consists not only of the acceptance of metaphor-relevant meaning but also the rejection of metaphor-irrelevant meaning. Our study examined the rejection process of topic word (e.g., “lawyer”) that include an irrelevant meaning (e.g., “helps people”) in metaphor comprehension (e.g., “a lawyer is like a shark”) using a priming paradigm and meaningfulness decision task: an experimental study reveals that the metaphor-irrelevant meaning is activated prior to the rejection process of the irrelevant meaning in metaphor comprehension. Our results suggest that the processing of metaphor-irrelevant meaning is composed of two different stages.

Imagery is an important feature of mental simulation, which is central to human cognitive functions from decision making to joint action to language production. Imagery is often used as a mental rehearsal strategy in areas of expertise, such as music, athletics and surgery, but also in movement rehabilitation. Individual imagery abilities may vary by general, modalityunspecific imagery capacity, as well as by imagery types. Within the literature, multiple tests have been used to measure imagery ability in various modalities such as visual, auditory, motor and spatial imagery. Participants (n=301) completed common imagery questionnaires (MIQ3, VMIQ2, BQMI, VVIQ, MASMI, OSIVQ, BAIS, CAIS). Findings suggest that greater reported dance, video game or music experience is related to increased kinesthetic, spatial or auditory imagery ability respectively. Other individual differences were found across subscales of the same modality, suggesting issues with reliability between questionnaires. Further factor analyses may reveal commonalities between imagery types. 2998

Self-regulated learning (SRL) can be measured in different ways (e.g., eye-tracking) and can be impacted by individual differences (e.g., prior knowledge) as college students learn with MetaTutor, an intelligent hypermedia system. In this study (N = 30), we examined fixation and duration data on interface-related areas of interest (AOI)-pairs as indicators of cognitive and metacognitive SRL strategies, and whether the frequencies of fixations and proportion of time spent on these AOI-pairs differed between prior knowledge groups. Results indicated that high prior knowledge learners selected significantly more cognitive (e.g., summarize) SRL strategies than learners with low prior knowledge. Additionally, learners with low prior knowledge spent a significantly higher proportion of time engaging in help seeking behavior, compared to high prior knowledge learners. These results have implications for designing advanced learning technologies capable of detecting real-time eye-tracking data used to adapt to fluctuations in learners’ SRL processes and foster effective learning.

Recent research has shown that learning technology combining adaptive and perceptual learning (PL) methods can improve pattern recognition, transfer, and fluency in complex learning domains (e.g., Mettler & Kellman, 2014). Both classic research and recent work suggest the benefit of paired comparisons in PL, but no previous work has used adaptive techniques to trigger comparisons. We asked whether PL can be enhanced by adaptively triggered comparison trials, in which erroneous responses led to comparisons designed to distinguish confusable categories. Undergraduates learned to interpret basic patterns from electrocardiograms (ECGs) with either: (1) adaptive PL based on single category exemplars, (2) adaptive PL combined with adaptively triggered comparisons, (3) adaptive PL combined with non-adaptive comparisons. Results showed strong learning in all conditions. Comparison conditions produced the strongest learning gains and showed smaller performance declines over a one-week delay. The results also suggested that adaptively triggered comparisons may enhance training efficiency

Recent work suggests that people think more systematically when using a second language because second languages are less automatic or emotionally valenced. Here, we use a different population (from India) to further investigate this possibility. We also test whether nuanced factors like language proficiency, usage context, and age of acquisition affect the degree to which people show a foreign language effect. We do not find a strong difference between native and second language speakers. However, we do find a more nuanced effect of language proficiency: people who are more proficient in the target language show more loss aversion. We also find that proficient English speakers are more willing to take on risk in both experiments, suggesting that English, itself, may lead people to think differently – possibly because it is a highly agentive language or because it is associated with individualistic cultural values.

Do people use a single strategy or sample from multiple strategies when estimating the conjunction and disjunction of two independent events? Here we address this tension directly by comparing individual level Bayesian simulations of multi and single strategy models using data from a frequency estimation experiment. Participants were shown two statements describing attributes and asked to estimate how many people had either one attribute, conjunction, or disjunction of attributes. In our Bayesian simulations we compare models in which participants either adopt a single strategy or sample from a set of strategies when forming estimates of both conjunctions and disjunctions. We compared every permutation of models in which a participant is responding based on a single component, a weighted average of the two events, probability theory, or combination of strategies. Our findings show that people sample from multiple strategies and are sampling from the normative strategy some of the time.

Previous studies have investigated the differences between concrete concepts and abstract concepts. Nevertheless there’s still no research probing emotional words to that font. The concepts behind emotional words have both affective and cognitive components, hence emotional concepts might have unique pattern of semantic properties. The present study then strives to compare the semantic properties of the three kinds of concepts. Concrete, abstract, and emotional words were selected, and participants had to report the property freely. Collected properties were categorized according to the semantic frame proposed by Wu and Barsalou (2009), and distributions of properties across three kinds of concepts were examined. It was found that the introspective types of word property reveal most dominant for emotional words. In contrast, concrete concepts were reported with more entity properties. Further research can explore different semantic properties within category in order to delineate the semantic continuity and boundaries of human concept structure.

Suppose you are told that in an alien language the strings AA, MM, DD, RR are all valid words, whereas BF, QG, CE, TM are not. You are then asked if you think EE is a valid word. Most people identify EE as a valid word; they are sensitive to the fact that all the valid words consist of two identical letters, whereas the invalid words do not. This is known as an identity effect, and has been observed in artificial language learning experiments and in a diverse range of natural languages. I give a formal proof that many popular learning framework, including methods for training neural network of arbitrary number of layers, cannot learn such identity effects. The proof exploits symmetries in the architectures and their training regimes to show that such learners cannot perform with human-like behaviour on these grammatical judgment tasks.

Accounts of induction development suggest that young children’s inferences are based either on object kind knowledge (Gelman & Markman, 1986), or on perceptual similarity (e.g., Sloutsky & Fisher, 2004). However, both accounts suggest that inferences with familiar and newly-learned categories engage a common set of psychological processes (determination of object kind or perceptual similarity). Alternately, young children may perform similarity-based induction with newly-learned categories, but use prior knowledge to make inferences with familiar categories. In this study, children complete two versions of a task in which a property attributed to a target can be extended to a category match or a perceptual match. In one version, items belong to familiar biological categories; in the other, items belong to two novel pseudo-biological categories. Preliminary findings indicate that although Kindergarten-age children learn to accurately categorize the novel items, they make similarity-based inferences with newly-learned categories, and category-consistent inferences with familiar categories.

We carried out two category norming studies (repeating MacRae 2005) to empirically examine Barrett’s (2008) notion of intuitive expectation sets as a coherent set of categorical expectations that are strongly correlated with each other. The studies validate some aspects of Barrett’s handcrafted list of intuitive sets and suggest removal as well as addition of some properties. The revised table of intuitive expectation sets is presented below: Solid Objects (a) are hard, rigid, and firm (b) are heavy (c) have a mass (d) are tangible (e) are visible Living Things (a) breathe (b) eat food (c) reproduce (d) move (e) grow (f) vulnerable to injury & death Animals (a) have limbs (b) have blood & heart (c) have a mind (d) have a mind (e) have emotions Mental Beings (a) think (b) are human (c) are animals (d) are intelligent (e) perceive (f) self-aware (g) talk to others (h) understand language.

Multiple studies (Galati & Brennan, 2014; Hoetjes, Koolen, Goudbeek, Krahmer, & Swerts, 2001; Jacobs & Garnham, 2007) have shown reduction in co-speech gesture across repetition. Reduction can be interpreted in terms of demonstrating automatization of effort (Vajrabhaya & Pederson, 2014). This study specifically investigates a special type of gesture used when speakers are accessing a low-familiarity lexeme. These gestures are assumed to aid lexical retrieval (Rauscher, Krauss, & Chen, 1996) and might be expected to not reduce as long as lexical access remains difficult. Unexpectedly, results show that these gestures also reduce across repetition like any other gesture and independent of lexical access difficulties. This suggests gesture automatization is a broad phenomenon across diverse gesture functions.

Some theories of language see it as a complex and highly adaptive system. For example, language may adapt to certain social or demographic variables of a linguistic community. If so, language may be used as an indication of certain social influences. Studies have begun to explore how the structure of social-networks contribute to language use. Until recently, datasets large enough to test how subtle effects of socio-cultural properties—spanning vast amounts of time and space—influence language change have been difficult to obtain. We analyzed over one million online business reviews using network analyses and information theory to quantify social connectivity and language structure. Results indicate that sometimes a surprisingly high proportion of variance in individual language use can be accounted for by differences in social structure. We consider how big data can be used as an arena for testing the influence of social variables on language use.

We argue that the acquisition of chemistry expertise requires considerable conceptual changes, which among other things involve a change from reliance on visual-spatial thinking to the employment of analytic strategies. We also argue that this shift in chemistry and specifically in knowledge about molecular structure is related to the acquisition of expertise and not to individual differences in visual-spatial thinking. In this presentation we will present an experiment designed to investigate the visual-analytic shift in knowledge about molecular structure in 132 11th graders. The results showed that the students, who were novices in chemistry, could solve the items requiring visual strategies but not those requiring analytic strategies, suggesting that they had not achieved the visual/analytic shift. Additional studies are needed to compare novices and experts in order to further test our hypothesis regarding the visual-analytic shift in chemistry.

A robust finding in category-based induction tasks is for positive observations to raise the willingness to generalize to other categories while negative observations lower the willingness to generalize. This pattern is referred to as monotonic generalization. In earlier work, we have found evidence for non-monotonic generalization when negative observations are involved. For example, we presented participants with the information that Mozart’s music has a certain property, asking to judge the likelihood of the conclusion that Metallica’s music too has the property. We found people more willing to accept the conclusion if they were additionally informed that the sound of a falling rock does not have the property. Here, we test the hypothesis that non-monotonic generalization following negative observations crucially depends on the reasoner’s assumptions regarding the way the arguments were constructed. We find that people may generalize non-monotonically when they assume the observations were intended to be helpful.

The present study investigates the relation between the reading process and text comprehension during naturalistic text reading. To that end, participants read easy and difficult texts while their eye movements were recorded. After each reading, participants filled-in comprehension questionnaires. We investigated classical measures of the reading process related to comprehension (fixation duration, regressive eye movements), as well as power-law scaling in eye movements that are indicative of degree of cognitive coordination during reading. The results show that text difficulty led to longer fixation durations and stronger power-law scaling in eye movements. Moreover, the degree of power-law scaling in eye movements was predictive of text comprehension. In line with previous research on natural text reading that utilized the self-paced reading method, powerlaw scaling turned out to be a superior predictor of reading comprehension compared to standard measures, suggesting that it is an effective measure of cognitive performance in complex reading tasks

Face recognition is widely believed to rely on two distinct mechanisms, the configural (e.g., the distance between two eyes or between mouth and nose) and featural (e.g., the shape of the eyes or mouth) face processing. However, little is known about whether the two processing types are affected by spatial attention. In our study, spatial attention was manipulated by asking participants to attend to the left or right visual field. We found that configural face processing elicited a larger P1 compared to featural face processing when they were attended. In contrast, the P2 was larger for featural relative to configural face processing when they were attended. Therefore, the results suggested that configural and featural face processing are differentially affected by spatial attention at different time windows.

Word learning is a fundamental part of language acquisition. Learning words from cross-situational statistics (Yu & Smith 2007) has been argued to be critical for lexical acquisition, but the resulting representations are not well understood. Here, we examine the claim from Hamrick & Rebuschat (2014) that cross-situational learning results in implicit representations. Three experiments provide evidence to the contrary. First, we establish that confidence ratings positively correlate with accuracy. By using a cover story where participants were not told to infer word meaning, only highly confident answers were above chance, contrary to what accounts of implicit memory would predict. In addition, using a deadline procedure (Voss, Bayem & Paller 2008), we found that participants performed no differently than without a deadline, contrary to predictions from implicit memory representations. In sum, we conclude that representations from cross-situational word learning are explicit.

Most of Chinese corpus were created for single-character words with indexes, such as frequency, stroke number, and phonetic information, for the purposes of basic research. However, multi-character Chinese words are recognized of referring alterations of meaning and more useful for investigating reading processes and comprehension. Therefore, for studying complete relations between words and lexicons of Chinese, a corpus requires statistics based on more than single-character words with valid and reliable indexes. In this study, we illustrate a corpus of Traditional Chinese providing five word indexes, including word sound, word position, word form, semantics, and competence of forming multi-character words by integrating current credible corpus. The integration approach of the present study is beneficial not only for minimizing inconsistencies of word entities between corpus, but also for calculating quantitative properties of character-to-character relationship. The utilization of the present corpus will significantly impact the studies of Chinese words and reading comprehension.

Performance on symbolic and non-symbolic numeric order determination tasks was examined as predictors of Woodcock Johnson calculation (exact) and computation estimation (approximate) scores among university aged adults. For Woodcock Johnson scores, only the symbolic task variant was found to be a significant predictor of performance outcomes after entering both task variants into a multiple regression. For the computational estimation task, both symbolic and non-symbolic task variants were significant predictors of performance outcomes. However, when controlling for general math ability (using Woodcock Johnson scores in the first step of a multiple regression), only the non-symbolic task variant remained predictive of computational estimation scores. Predictors remained significant for each outcome measure after controlling for non-numerical (luminance) order determination tasks through regression. These findings suggest that 1) the relations are due to numeric (not general order) judgements, and 2) both symbolic and non-symbolic task variants are related to specific mathematical outcome measures.

This is a cross-cultural replication of Seiver, Gopnik & Goodman (2013) and compares the development of social causal attributions in Chinese and U.S. children. In this study, Chinese (n=110) 4-and 6-year-olds were directly compared to the U.S. children in Seiver, Gopnik & Goodman (2013). Children were shown covariation evidence that varied across conditions to imply that a person, situation, or neither was the cause of a person’s actions. Following observation, children were asked to explain why the person engaged in the actions. Findings indicate that U.S. children significantly increased the amount of person attributions they made with age, while Chinese 4-and 6-year-olds answered comparably. Children from both cultures were sensitive to covariation manipulations when they suggested a person was the cause of an action. Only U.S. children were sensitive to evidence when it favored the situation.

In an ERP study we investigate the time course of the interaction between the lexically specified telic role of a noun and the contextually provided ad-hoc affordance induced by a linguistic discourse. If preceded by a neutral discourse context, a verb inconsistent with the telic role in a sentence elicits an enhanced N400 compared to a congruent verb. However, if the preceding discourse context induces an ad-hoc affordance for the object that conflicts with the lexically specified telic role of the referring noun, we observe a crossing-over: compared to the neutral context, the N400 elicited by the inconsistent verb is significantly reduced in this context, whereas the N400 elicited by the congruent verb is significantly enhanced. We interpret these results as a consequence of the immediate functional replacement of the telic role with the contextually triggered ad-hoc affordance, thus supporting a single-step model of sentence meaning composition.

Bodily self-consciousness (BS) refers to self-knowledge like body shape, position and ownership. BS relies on the integration of multisensory bodily signals and can be experimentally manipulated with the Rubber Hand Illusion (RHI). In this illusion, the incorporation of a rubber hand into the BS is facilitated by visual similarity to the participant’s hand. Neurophysiological research indicates that tool use alters neural networks mapping body shape and posture for coordinating motor actions. The present experiment sought to unify perceptual and motoric BS accounts using a modified RHI. We found that synchronous multisensory stimulation induced perceptual embodiment of an external tool. The RHI was stronger if multisensory stimulation was preceded by tool use, highlighting the motor system’s role in embodiment. The illusion, as measured by proprioceptive drift and questionnaires, was stronger for skilled individuals but also occurred for untrained participants. This experiment helps to clarify the role of perceptual and motoric embodiment.

It is widely believed that semantic activation from print is not capacity limited (i.e., that it does not need attentional resources). Prior research has tested this assumption by examining the Stroop effect in the context of the psychological refractory period (PRP) paradigm. These studies yielded additivity of the Stroop effect and SOA on RT, consistent with the hypothesis that semantic activation is itself capacity limited (given demonstrations that prior processes are not capacity limited). There is, however, an alternative explanation for such additivity: performance optimization (Miller and colleagues, 2009). Given that participants in PRP experiments are told to respond as quickly as possible, they may opt to process serially to improve performance. We investigated whether additivity of the Stroop effect (standard and semantic) and SOA in the context of PRP is best explained in terms of a structural bottleneck or performance optimization.

We assessed whether students’ reasoning about historical causality is biased by accessibility and previous knowledge. Undergraduates in Canada provided explanations for historical events: the attacks on Pearl Harbor and September 11, 2001. Participants were then given several explanations for these events, including conventional historical and alternative conspiracist explanations, and reported how satisfying they found them. Overall, historical explanations were more satisfying than alternative explanations. Historical explanations were more satisfying for Pearl Harbor than 9/11, and alternative explanations were more satisfying for 9/11 than Pearl Harbor. Similarly, ease of generating an explanation was associated with satisfaction with historical explanations for Pearl Harbor, and with satisfaction with alternative explanations for 9/11. Participants’ general conspiracist beliefs were also associated with acceptance of alternative and historical explanations. Students learn about Pearl Harbor in history class, whereas information about 9/11 comes primarily from the media, leaving the door open to a multiplicity of ahistorical explanations.

Pointing movements can serve instrumental goals (‘pointing to press a button’) or communicative goals (‘pointing to indicate to someone which button to press’). Previous work has shown that communicative pointing follows different trajectories, and has different end points than instrumental pointing movements, depending on the addressee’s spatial location. This suggests that motor control processes are affected by communicative intentions, but the nature of this interface remains unknown. Using optimal feedback control theory, we construct a model of instrumental pointing, and explore how this model can be adjusted to reproduce the dependency between communicative trajectories and addressees’ locations. Our results show that the variations in end points cannot account for those trajectories. Instead, the kinematic data are best explained by ‘perspective taking’ on the part of the communicator, i.e., communicative pointing movements seem to be planned in a frame of reference that is adjusted to the addressee’s point of view.

Past research suggests that people think about numbers in terms of multiple spatial mappings. For instance, they think of “more” as “bigger” or “higher” (e.g., Andres et al., 2008; Sell & Kaschak, 2012). Here, we investigate mental number space by looking at naturally occurring co-speech gestures. Using the TV News Archive (archive.org/details/tv), we selected a random sample of 1,320 videos containing the phrases “tiny number,” “huge number,” “high number” and “low number,” of which 314 had associated manual gestures. Our analysis shows that with “tiny number,” speakers produce pinching gestures; with “huge number,” they move their hands outward; with “high numbers,” the palms are oriented and move upward; and with “low numbers,” the palms are oriented and move downward. Speakers did not gesture systematically along the horizontal axis. This work provides new insights into how people’s spatial conceptualizations of numbers shape communicative practice in naturally occurring discourse.

Children can recover beliefs from knowledge of an agent’s actions and desires, or desires from her actions and beliefs. However, both beliefs and desires are sometimes unknown and underdetermined by agents’ actions. Here we ask how emotional expressions might support mental-state inferences that are otherwise ambiguous given agents’ actions. We present two experiments, showing that children (mean: 5.9 years) can infer beliefs and desires simultaneously from change or stability in the valence of emotional expressions between when agents expect and observe outcomes. However, children did not infer that surprised expressions indicated false beliefs, or the absence of surprise, true beliefs, in the context of joint belief-desire reasoning. Rather they inferred that positively valenced responses to outcomes indicate true beliefs and negative responses indicated false beliefs. We propose that emotional expressions support rational inferences about mental states, consistent with the expectation that agents act on their desires given their (probabilistic) beliefs

Since Johanson (1973), it is well known that people perceive human behavior, not only familiar ones such as walking and dancing but also unfamiliar behaviors in point-light display. While categorization among well-known behaviors has been well studied, it is not clear yet how people extract the cues that are useful for perceiving unfamiliar behaviors. We hypothesized that the hierarchical information in human structure was playing an important part for perception of unfamiliar behaviors, and examined this hypothesis by comparing performances of three kinds of models in categorization tasks of unfamiliar arm movements. While the first model was based only on local motion of the point lights, the second model used hierarchical ordering information to analyze the local motion. The third model used strict positioning following the hierarchical structure in addition to the ordering information. The comparison suggested that hierarchical ordering was important information to producing close performance to human performance.

Readers and listeners rely upon previous experience to generate predictions about multiple aspects of an unfolding linguistic signal. Error signals elicited by unexpected input feed forward to higher-level units, serving in the adjustment of expectancies and thus increasing the precision of predictions in that context. When a syntactic ambiguity is resolved with a dis-preferred continuation, a garden-path effect occurs, but decreases in magnitude as a function of exposure to the unexpected event. But, can readers adjust lower-level expectations about word forms in contexts that do not permit overt higher-level ambiguity? We monitored eye-movements as participants read expected or unexpected words in highly-constraining sentences. Half of items contained the predicted word and half contained a plausible but unexpected word. Adaptation—in the form of decreased fixation duration on unexpected words—was observed on first fixation duration but nowhere else, suggesting that adaptation occurs at different levels of a multilayered processing system.

Episodic memory involves a mechanism that binds information into a coherent representation structure. Especially, more complex memory structures are required when there are more overlapping elements among different episodes (Humphreys, Bain & Pike, 1989), and the ability to use memory structures of different complexities increases throughout development (Yim, Dennis, & Sloutsky, 2013). Although the major neural structures considered to underlie the development of episodic memory are the prefrontal cortex (PFC) and the hippocampus, it is possible that the development of episodic memory involves multiple brain mechanisms interacting with different types memory structure. The current study tries to examine the division of labor between the PFC and hippocampus when forming different types of binding structure in episodic memory development. We utilized a multinomial processing tree (MPT) model, and Single Nucleotide Polymorphisms (SNP) genotyping approach to elucidate the division of labor between different brain areas involved in forming different memory structures.

The present study examined whether Cantonese-English bilingual listeners made use of their L1 probabilistic phonotactics in the segmentation process of English continuous speech (L2). Previous research in different languages demonstrated that probabilistic phonotactics could serve as a useful cue to locate the possible word boundary in continuous speech. The use of these kinds of information in L1 can also be easily transferred to deal with L2 listening for bilingual listeners. In the present study, a word-spotting experiment was conducted to examine this issue and the results revealed that the Cantonese-English bilingual listeners might not make use of all their L1 probabilistic phonotactics in segmenting the L2 speech. 3027

Theory of Mind (TOM) is an ability to simulate mental activities, attribute intention, and predict behaviors of others. According to Ickes’s study (1990), there are strong correlation between empathy and TOM. However, whether they have any causal-effect relationship is still unclear. It is possible to differentiate the two processes from a developmental perspective since TOM ability develops later than empathy. This study aimed to study the influence of empathy on TOM processes. In the experiment (N=28), 4-year-olds children received a traditional false belief task first; and then after a week, they were divided into two different empathic situations: bullying condition and bully-victim condition. The study found a trend that the TOM score is lower in the bullying condition but higher in the bully-victim condition compared to the baseline condition despite of the insufficient subjects. It suggests that empathic states indeed impact TOM processing.

The ‘sense of agency’—the feeling of doing—is a phenomenological experience that cannot be taken for granted in infants. Researching the development of this phenomenon hinges on its conceptualization. Although it may seem natural to ask “When do infants have sense of agency?”, this binary view on the presence (or absence) of a sense of agency seems conceptually problematic. Cognitive phenomenological research reveals that sense of agency is a complex set of agentive experiences with different contents (e.g. ‘the experience of acting’, ‘the experience of volition’) that are often conflated. Given that different contents presuppose different representational capacities, developmental psychology may better target the question “What senses of agency can infants have at different developmental stages?” We show how this re-conceptualization impacts the interpretation of existing empirical findings in infant research, and how it can improve our understanding of the developmental trajectory of infants’ experiences of agency.

Previous research has demonstrated categorical perception (CP) of facial expressions from American Sign Language (ASL) in hearing, English-speaking non-signers. Notably, CP was observed even for faces with no obvious linguistic labels in English, suggesting the existence of covert categories based on nonlinguistic facial properties. However, in the earlier work, CP was assessed using memory procedures, leaving open the possibility that such categories have no impact on the discrimination of simultaneously presented faces. Here, we used a visual search task with no memory component to test for CP for both labeled (happy/sad) and unlabeled (adverbial expressions with no lexical signs) categories of ASL facial expressions. CP was observed for both sets of categories, suggesting that unlabeled covert categories——not just labeled ones——are accessed even when stimuli are readily available to perception. We interpret these findings in light of competing accounts of the interrelations of language, categories, and perception. 3030

Metacognitive processes such as monitoring and control and sense-making processes such as self-explanation and analogical comparison are each hypothesized to result in effective problem solving, learning and transfer (Chi et al., 1994; Alfieri, et al., 2013; Zepeda et al., 2015). However, little work has examined how these processes relate to one another and their associated learning outcomes. In this study we explored two methodologies to measure and relate these processes to one another by investigating students verbal protocols and task-based self-reports during a learning task. We investigated (1) whether specific metacognitive skills are more likely to occur before, during, or after self-explanation and analogical comparison, (2) individual differences in students’ use of these skills, and (3) whether specific skills results in better learning outcomes. Results from these analyses and their implications will be discussed.

Learning the meanings of words involves not only forming connections between individual words and concepts but also building a network of connections across objects and words. Previous studies reveal that infants and adults can learn word-referent links across multiple ambiguous training instances by tracking the statistical co-occurrence of labels and objects (Smith & Yu, 2008; Dautriche & Chemla, 2014). We asked whether adults are sensitive to multiple types of statistical structure in these learning instances by manipulating the frequency with which objects co-occurred with each other during training trials. Across several studies (n=150), we find that adults not only learned to disambiguate label referents, but simultaneously formed connections both between the frequently co-occurring objects themselves and between the labels of frequently co-occurring objects. These findings indicate that learners exploit statistical regularities to form multiple types of associations during word learning.

Most accounts of statistical learning (e.g., Saffran et al., 1996) assume that the learner computes cooccurrence statistics over units, such as syllables, that abstract away from the physical features of the input. This assumption need not hold when the units are underlearned, unfamiliar, or uncategorizable. We tested statistical learning with variable, unfamiliar units and show that if the featural variation is small, adults treat words differently from part-words and non-words, as if learning were occurring over abstract units. When the featural variation is large, and the categorical boundaries are unclear, participants can still learn statistical regularities defined over trajectories through this space: Words, part words and even non-words that follow trajectories consistent with the familiarization set are rated as equally familiar, whereas non-words that take trajectories opposite the familiarized direction are rated as less familiar. Conceiving of statistical learning over trajectories through perceptual space explains the results under both conditions.

Previous visual world eye-tracking studies have shown that when a sentential verb can refer (via tense information on the verb and on a following time adverb) to either a recent and a future action event performed by an actor, people inspected the target of the recent event more often than the (different) target of the future event. This ’recent event preference’ replicated even when the frequency of future events within the experiment greatly exceeded the frequency of recent events (e.g., 75% vs 25%). The recent event preference may arise because the past action is situation-immediate and thus more relevant at the particular point in time when the sentence is processed (at that point participants have seen the past action performed and will not see the future action until after the sentence). If the situation-immediate relevance of a cue is responsible for the recent event preference, then we should be able to “overwrite” the effect of the recent action with another situation-immediate cue. Accordingly, two current eye-tracking experiments pitted the recent event preference against a situation-immediate cue, the shift in the actor’s gaze to the target object. Given that interlocutors’ gaze has been shown to be a powerful cue in guiding listeners’ attention to objects in the visual context, we hypothesized that the actor’s gaze to the future target should rapidly guide a listener’s attention to it. Analyses revealed indeed that listeners’ visual attention was rapidly guided to the target by the actor’s gaze; crucially the gaze cue was particularly helpful in guiding looks to the future target. Importantly, however, we still replicated the overall preference to look at the recent target regardless of tense and gaze; and even for future gaze conditions, the preference was not immediately reversed, suggesting it is surprisingly robust in competition with a situation-specific future-biasing cue.

The aim of this study was to clarify how the sense of heaviness changes our cognition. According to recent studies in cognitive science, intelligent human behaviors ranging from perception to inference are not closed mental processes; rather, they are affected by body and action (Wilson, 2002; Gibbs, 2005; Proffitt, 2006). In previous studies, the sense of heaviness activated concepts metaphorically related to heaviness, and changed impressions accordingly. However, previous studies have not distinguished between subjective heaviness and physical weight. The purpose of this study was to clarify whether changes in impressions are due to subjective heaviness or physical weight. To examine this issue, a psychological experiment using a tasting task was conducted. The results confirmed that subjective heaviness influences evaluations of price and value.

The ability to extract meaningful relationships from sequences is crucial to many aspects of perception and cognition, such as speech and music. This paper explores how leading computational techniques may be used to model how humans learn abstract musical relationships, namely, tonality and octave equivalence. Rather than hard-coding musical rules, this model uses an unsupervised learning approach to glean tonal relationships from a musical corpus. We develop and test a novel input representation technique, using a perceptually-inspired harmonics-based representation, to bootstrap the model’s learning of tonal structure. The results are compared with behavioral data from listeners’ performance on a standard music perception task: the model effectively encodes tonal relationships from musical data, simulating expert performance on the listening task. Lastly, the results are contrasted with previous findings from a computational model that uses a more simple symbolic input representation of pitch.

Current evidence provides support for the idea that time is mentally represented by spatial means, as a lateral mental time line. However, available studies have tested only factual events, i.e., those which have occurred in the past or will occur in the future. In the present study we tested whether past and future potential events are also represented along the lateral mental timeline. In Experiment 1 participants categorized the temporal reference (past or future) of either factual or potential events and responded by means of a lateralized (left or right) keypress. Factual events showed a space-time congruency effect that replicated prior findings: participants were faster to categorize past events with the left hand and future events with the right hand than when using the opposite mapping. More importantly, this also ocurred for potential events. Experiment 2 replicated this finding using blocks comprising only potential events. In order to assess the degree of automaticity of the activation of the mental timeline in these two kinds of events, Experiment 3 asked participants to judge whether the expressions referred to factual or potential events. In this case, there was no space-time congruency effect, showing that the lateral timeline is active only when relevant to the task. Moreover, participants were faster to categorize potential events with the left hand and factual events with the right hand than when using the opposite mapping, suggesting for the first time a link between the mental representations of space and potentiality.

Human preference choice suffers curious contextual effects: the relative preference between two multi-attribute options (e.g. cars with differing safety and economy ratings) can dramatically shift, depending on the presence/absence of additional options. This phenomenon defies any simple utilitybased account of choice, and has been taken to imply irrationalities/sub-optimalities in human decision-making, or reflect idiosyncrasies in neural processing. Recently, we used a Bayesian model to show that these contextual effects are normative consequences of observers using options to learn about the “market”. However, it had an unsavory implication that all decision-makers asymptotically converge to the same beliefs/ behavior. Here, we propose a new model that uses both market and personal utilities to make choices. This model still captures the contextual effects, while also allowing asymptotic differences in individual preferences and providing a general framework for explaining how consumption informs one’s beliefs and preferences.

How do people conceptualize motion events and talk about them? The current study examines how gestural representations of motion events arise from linguistic expressions in Persian, which has characteristics of both Talmy’s satellite- and verb-framed languages. We examined native Persian speakers’ speech and gestures in describing 20 motion events. We focused on two motion event components: path (trajectory of motion like up) and manner (how the action is performed like jumping). Results indicated that when expressing motion, Persian speakers produced path in both speech and gesture, whereas manner was conveyed only through speech (mostly as adverbs). Additionally, dynamic gestures tended to occur in the same order they were uttered. The difference between path and manner findings asks for further research to examine language-gesture interaction in detail among different languages. Results also suggest refinement in gesture theories that argue for one-to-one correspondence between speech and gesture.

In English, many words can be used flexibly to label artifacts, as nouns, or functional uses of those artifacts, as verbs: We can shovel snow with a shovel and comb our hair with a comb. Here, we examine whether young children form generalizations about flexibility from early in life and use such generalizations to predict new word meanings. When children learn a new word for an artifact, do they also expect it to label its functional use, and vice versa? In Experiment 1, we show that when four- and five-year-olds are taught a first novel word to label a familiar action—e.g., that bucking means shoveling—they exclude the artifact involved in this action— i.e., the shovel—as the meaning of a second novel word (e.g., gork). This suggests that children spontaneously expected the first novel word—which referred to the action—to also refer to the artifact. In Experiment 2, we show that this pattern extends to words that label novel actions involving novel artifacts, suggesting that children expect any word for an action to label the artifact that helps carry out that action. Experiment 3 traces how such generalizations may arise in development. In particular, we show that while four- and five-year-olds each expect words to label artifacts and their functional uses, three-yearolds may not.

Models of spatial reasoning often assume distinct visual and spatial representations. In particular, the visual impedance effect – slower response time when more visual details are represented in three-term series spatial reasoning tasks – has been taken as evidence for the distinctive roles of visual and spatial representations. In this paper, we show that a memory model of spreading activation based on the ACT-R architecture can explain the visual impedance effect without the assumption of distinct visual and spatial representations. Using the same memory representation, varying levels of visual features associated with an object are represented in the model. The visual impedance effect is explained by the spreading activation mechanism of ACT-R. The model not only provides a more parsimonious explanation to the visual impedance effect, but also leads to testable predictions of a wide range of memory effects in spatial reasoning.

We investigated the differential effects of self-explaining a refutational text, compared to thinking aloud or rereading. Undergraduate students (n = 105) read a refutational text about natural selection and were asked to either self-explain, think-aloud, or re-read the text. Then they completed a posttest that assessed general knowledge of natural selection. Students who self-explained the refutational text subsequently outperformed their peers on a test of their knowledge of natural selection. Additionally, the results suggest that both instructional and performance differences were significantly linked to the degree of causal cohesion present within students’ natural language responses to the text (i.e., selfexplanations and think-alouds).

Many social judgments hinge on assigning responsibility to individuals for their role in a group’s success or failure. Often the group’s success depends on every team member acting in a rational way. When someone does not conform to what others expect of them, cooperation breaks down. We present a computational model of responsibility judgments for individuals in a cooperative setting. We test the model in two behavioral experiments where participants were asked to evaluate agents acting in a cooperative, one-shot game. In Experiment 1, we show that participants’ action predictions are consistent with a recursive reasoning model. In Experiment 2, we show that people’s assignments of blame are influenced by both an agent’s presumed rationality, or adherence to an expected policy, as well as the pivotality of the agent’s actions, or how close the situation was to one in which the action would have made a difference to the outcome.

We study the projection of cognitive representations into continuous motor (reaching) responses with a computational model that unifies three influential approaches: accumulation of evidence, statistical inference, and optimal feedback control. We modeled a number comparison task that asked participants to respond with a reaching gesture which of two side had more dots. The model successfully reproduced subjects’ pattern of reach and performance across varying difficulties of numerical comparison. Our model parameterized several potentially relevant cognitive variables, including a threshold, memory decay, and mental sampling rate. Remarkably, a threshold for movement was not needed for modeling human behavior when statistical inference is combined with optimal motor planning. Overall, the model indicates that the motorsystem positions the effectors optimally, both biomechanically through an optimal feedback controller, and cognitively by means of continuous statistical inference on the available evidence.

Healthy development leads to a fluid integration of competing constraints. A marker of such behavior is hysteresis, reflecting a multi-stable system that takes into account its immediate history. The current study investigates patterns of hysteresis in typically developing children (TD) and those diagnosed with Autism Spectrum Disorder (ASD). The task was to grasp and lift objects that increased in size, either from smallest to largest, or from largest to smallest. The objects could be picked up with one or two hands, marking a range of bi-stable behavior. Results of the grasping task showed hysteresis in TD children, whether or not the task was situated in the social context. In contrast, children with ASD showed hysteresis only in the non-social context. For both diagnostic groups, perseveration did not correlate to the degree of hysteresis, regardless of the presence or absence of social cues.

This is the first study to examine the influence of gender-sex congruence (match or mismatch between grammatical gender markers and participant sex) on the embodied processing of first-person sentences and images with either an internal or an external perspective in a picture-sentence verification task in Bulgarian modeled on Bruny e et al.’s (2009) exp erimental paradigm. Participants were shown not to discriminate between perspectives when the grammatical gender was congruent with their own sex, thus allowing for an agentive interpretation by the reader. However, in the gender-sex incongruent condition, a significant 83 ms effect of image perspective was observed indicating large processing costs for attempting to adopt an internal perspective when the p articip ant’s sex was incomp atible with the first -person gender marking, hence with action simulation from an egocentric perspective. These results are discussed in terms of embodiment specificity accounts and the experiential basis of grammar processing.

The ability to reason about another person’s mental states, such as belief, desires and knowledge – first-order theory of mind – develops between the ages three and four. On the other hand, children need one or two more years to reason about a person who reasons about another person – secondorder theory of mind. Is it possible to accelerate the development of theory of mind? There are several training studies that showed that it is possible to teach preschool children to pass first-order false belief tasks. However, the literature is missing analogous training effects for school-age children with respect to second-order false belief tasks. In this study, we focus on the role of feedback in the development of second-order false belief reasoning in two different conditions in children between the ages five and six: (i) feedback with explanation, (ii) feedback without explanation. Children’s performance improved in both conditions. Previous theories suggest either that children’s development of second-order theory of mind requires conceptual changes or that 4-5 year old children have cognitive constraints that need to be overcome in order for them to be able to apply second-order theory of mind. In line with our findings, however, we argue that five-year-old children who cannot yet pass the secondorder false belief task reason about the false belief questions based on the reasoning strategy that they most frequently use in daily life (i.e. first-order or zero-order theory of mind). Moreover, we argue that most of the time children can revise their wrong reasoning strategy and change to the correct second-order reasoning strategy based on repeated exposure to the feedback “Correct/Wrong” together with the correct answer.

The London Underground Diagram (LUD) is a cognitive artifact and a well-known example of representational efficiency, having been copied by urban transportation systems worldwide. Here we describe the design of the LUD as an example of cognitive niche construction happening through iconic meaning of a problem space. We argue that the LUD's meaning is grounded on the offer of opportunities for action through diagrammaticity. Our examination suggest that iconicity is at the core the cognitive niche construction.

Cross-situational learning is a basic mechanism that enables people to infer the correct referent for a novel word by tracking multiple hypotheses simultaneously across exposures. Previous research has shown that adults are capable of exploiting cross-situational information, but recently this gradual statistical learning mechanism has been put under debate by researchers who argue that people learn via a fast mapping procedure. We compared the performance of adult participants on a word learning task in which information was manipulated cross-situationally with the performance of simulated learning strategies. Experimental evidence indicates that adults use cross-situational learning, which appears to be a robust mechanism that facilitates word learning even under cognitively demanding circumstances.

The pattern-based sequence classification system (PBSC) identifies regularly occurring patterns in collections of sequences and uses these patterns to predict meta-information. This automated system has been proven useful in identifying patterns in written language and musical notations. To illustrate the wide applicability of this approach, we classify symbolic representations of speech-accompanying gestures produced by adults in order to predict their level of empathy. Previous research that focused on isolated gestures has shown that the frequency and salience with which individuals produce certain speech-accompanying gestures are related to empathy. The current research extends these analyses of single gestures by investigating the relationship between the frequency of multi-gesture sequences of speech-accompanying gestures and empathy. The results show that patterns found in multi-gesture sequences prove to be more useful for predicting empathy levels in adults than patterns found in single gestures. This paper thus demonstrates that sequences of gestures contain additional information compared to gestures in isolation, suggesting that empathic people structure their gestural sequences differently than less empathic people. More importantly, this study introduces PBSC as an innovative, effective method to incorporate time as an extra dimension in gestural communication, which can be extended to a wide range of sequential modalities.

Implicit bias has recently gained much attention in scholarly attempts to understand and explain different forms of social injustice by identifying causally relevant mental states in individual’ minds. Here we question the explanatory power of implicit bias in a particular type of injustice, testimonial injustice, and more generally in what we call speech injustice. Testimonial injustice occurs when the audience deflates a speaker’s credibility due to the speaker’s perceived social identity (Fricker, 2007). We identify two drawbacks of a widely accepted explanation attributing testimonial injustice to prejudices (e.g. implicit bias) in the mind of the hearer, and argue that further understanding of this phenomenon can be gained from a structural explanation that appeals to discursive conventions and interlocutors’ positions in the communicative exchange.

In this study we examined the effectiveness of self-regulated learning (SRL) training in facilitating college students’ science learning with hypermedia. Sixty (N = 60) undergraduate students were randomly assigned to either a training condition or a control condition and used a hypermedia environment to learn about the circulatory system. On Day 1, all participants were administered a pretest and a self-report measure of SRL. On Days 2–4, participants in the experimental group underwent 3-day training on the use of specific, empirically based cognitive and metacognitive SRL processes (e.g., judgment of learning, making inferences) designed to foster their conceptual understanding; control students received no training. Three weeks later (on Day 5), all participants were administered a pretest on the science topic and a self-report measure of SRL, and then used a different version of the system to learn about another science topic (i.e., the central nervous system). Verbal protocol data were collected from both groups on Days 2–5. Overall, there were no significant differences on several learning outcome measures between conditions. However, those in the training condition remembered significantly more declarative knowledge of cognitive and metacognitive strategies. Lastly, think-aloud protocol data showed significant differences in the use of the SRL processes immediately following training, but not following a 3-week interval on a hypermedia transfer task.

In the present study, we investigate if and when speakers refer to moving entities in route directions (RDs). On the one hand, there is a general agreement that landmarks should be perceptually salient and stable objects. On the other hand, animated movement attracts visual attention, making entities intrinsically salient. In two experiments, we tested to what extent people are prepared to use moving entities. Our results show that participants mention moving entities when the communicative setting affords such references (route directions in a joint communicative setting) and when the movement is informative for the place where a turn should be taken.

Computational models of category learning and attention have historically focused on capturing trial and experiment level interactions between attention and decision. However, evidence has been accumulating that suggests that the moment-to-moment attentional dynamics of an individual affects both their immediate decision-making processes as well as their overall learning performance. To extend the scope of these formal theories requires a modeling approach that can index fine-grained decision-making at millisecond time scales. Here we implement a model of eye movements during category learning using concepts from Dynamic Neural Field Theory research. Our model uses a combination of timing signals, spatial competition and Hebbian association to simultaneously account for a number of foundational attentional efficiency results from eye tracking and category learning. We report the results of fitting this model to accuracy, fixation probabilities, fixation counts and fixation duration data in 42 subjects from a standard category learning experiment.

Recent research on moral dynamics (the processes and phenomena –collective or individual– by which moral behavior and moral attitudes emerge, evolve, spread, erode or disappear) shows that an individual’s ethical mind-set (i.e., outcome-based vs. rule-based) moderates the impact of an initial ethical or unethical act regarding the likelihood of behaving ethically on a subsequent occasion. More specifically, an outcome-based mind-set facilitates Moral Balancing (behaving ethically or unethically decreases the likelihood of engaging in the same type of behavior later), whereas a rule-based mind-set facilitates Moral Consistency (engaging in an ethical or unethical behavior increases the likelihood of engaging in the same type of behavior later). Our objective was to look at the evolution of moral choice across a series of scenarios and so explore if these moral patterns (Balancing vs. Consistency) are maintained over time. The results of three studies showed that Moral Balancing is not maintained over time. On the other hand, Moral Consistency could be maintained over time, if the mind-set was reinforced before making a new moral judgment (but not otherwise).

The purpose of this study was to characterize the ways in which psychologists address research hypothesis risk in academic articles, and to support undergraduates in learning to write about such risk using argument diagramming prewriting activities. First, 90 articles recently published in top social, developmental, and cognitive psychology journals were examined for their presentation of research hypothesis ‘risk’ – an element of the intellectual merit of a research study denoting the novelty and importance of the study being conducted. Second, an experimental study was conducted involving 82 students in undergraduate research methods classes. They were assigned to either argument diagram or traditional instruction conditions. Research reports were coded for explicit discussion of risk. Students using argument diagramming were significantly more likely to write about risk when compared to matched classes given no diagramming support

What constitutes good teaching, and what factors do learners consider when evaluating teachers? Prior developmental work suggests that even young children accurately recognize and evaluate under-informativeness. Building on prior work, we propose a Bayesian model of teacher evaluation that infers the teacher’s quality from how carefully he selected demonstrations given what he knew. We test the predictions of our model across 15 conditions in which participants saw a teacher who demonstrated all or a subset of functions of a novel device and rated his helpfulness. Our results suggest that human adults seamlessly integrate information about the number of functions taught, their values, as well as what the teacher knew, to make nuanced judgments about the quality of teaching; the quantitative pattern is well predicted by our model

Liquids can splash, squirt, gush, slosh, soak, drip, drain, trickle, pool, and be poured–complex behaviors that we can easily distinguish, imagine, describe, and, crucially, predict, despite tremendous diversity among different liquids’ material and dynamical characteristics. This proficiency suggests the brain has a sophisticated cognitive mechanism for reasoning about liquids, yet to date there has been little effort to study this mechanism quantitatively or describe it computationally. Here we find evidence that people’s reasoning about how liquids move is consistent with a computational cognitive model based on approximate probabilistic simulation. In a psychophysical experiment, participants predicted how different liquids would flow around solid obstacles, and their judgments agreed with those of a family of models in which volumes of liquid are represented as collections of interacting particles, within a dynamical fluid simulation. Our model explains people’s accuracy, and their predictions’ sensitivity to liquids of different viscosity. We also explored several models that did not involve simulation, and found they could not account for the experimental data as well. Our results are consistent with previous reports that people’s physical understanding of solid objects is based on simulation, but extends this thesis to the more complex and unexplored domain of reasoning about liquids

In current approaches to pragmatic reasoning the comprehension and production of referring expressions is modeled as a result of the interlocutors’ mutual perspective-taking. While such models of pragmatic reasoning have been empirically validated in referential language games experiments, empirical (and computational) work on the generation of referring expressions has shown that speakers do not always take the listener’s perspective into account, but instead produce referring expressions according to their own preferences. One particularly well studied example is color: speakers often include color terms in their referring expressions even if they do not help identify the intended referent. We show that like speakers, listeners treat color differently from other properties like e.g. size. Our results suggest that listeners do not seem to perform much pragmatic reasoning when the referring expression only expresses color, but instead follow a simple saliencebased heuristic.

Existing research has emphasized the importance of normality judgments in many aspects of cognition and life (e.g., causal cognition, gradable adjectives, cooperative behavior). Yet little work has explored how people actually come to understand what sorts of things are normal. We argue that people’s normality intuitions reflect a mixture of statistical and evaluative considerations. Specifically, we suggest that people’s intuitions about what is normal can be influenced by representations both of the average and of the ideal. We test this idea in three experiments. Experiment 1a demonstrates that explicit judgments of normality reflect this mixture of statistical and evaluative considerations. Experiments 1b and 2 then show that the hybrid notion that comes out in these explicit judgments can also explain people’s judgments about gradable adjectives. Taken together, these findings have potential implications not only for normality judgments themselves, but also for the many other mental activities that these judgments impact.

We examine the nature of phonological and semantic similarity in early language learning. We consider how the use of this information might change over the course of development. To this end, we represent the lexicon as either a phonological or semantic network and model the growth of this network. Constructing normative vocabularies from the Communicative Development Inventory norms, we utilize a preferential attachment growth algorithm. We predict and quantify the words which will be learned next, comparing the two network representations. We consider the effect of age, total vocabulary size and language ability as measured through CDI percentile. Our findings suggest that the semantic representation does not outperform the baseline bag-of-words model, whereas the phonological representation conditionally does. More generally, we show that the network representation influences the ability of a model to capture vocabulary growth. We further offer a method of analysis for testing representational assumptions in network models.

How does a child's vocabulary production change over time? Past research has often focused on characterizing population statistics of vocabulary growth. In this work, we develop models that attempt to predict when a specific word will be learned by a particular child. The models are based on two qualitatively different sources of information: a representation describing the child (age, sex, and quantifiers of vocabulary skill) and a representation describing the specific words a child knows. Using longitudinal data from children aged 15-36 months collected at the University of Colorado, we constructed logistic regression models to predict each month whether a word would be learned in the coming month. Models based on either the child representation or the word representation outperform a baseline model that utilizes population acquisition norms. Although the child- and word-representation models perform comparably, an ensemble that averages the predictions of the two separate models obtains significantly higher accuracy, indicating that the two sources of information are complementary. Through the exploration of such models, we gain an understanding of the factors that influence language learning, and this understanding should inform cognitive theories of development. On a practical level, these models may support the development of interventions to boost language acquisition.

In semantic typology, it is desirable to have quick and easy access to crosslinguistic elicitations describing stimuli from a semantic domain. We explore the use of crowdsourcing for obtaining such data, and compare it with fieldwork data obtained through in-person elicitations. Despite potential concerns about the quality of crowdsourced data, we find no difference in the amount of between-language variation and can replicate a cognitive modeling experiment using the crowdsourced data in place of the fieldwork data. Both results suggest that crowdsourcing elicitations is a viable method for gathering data for semantic typology and cognitive modeling

Comparing analogs is a key recommendation in mathematics instruction, but successful structure-mapping may impose high demands on children’s executive functions (EF). We examine the role of individual differences in EF resources on learning from an everyday mathematics video-lesson placing a particular strain on children’s cognitive resources: comparing three analogs presented sequentially. Specifically, we examine the separate contributions of working memory (WM) and inhibitory control (IC) on successful schemaformation. Overall, WM and IC explained distinct variance for predicting improvements in procedural knowledge, procedural flexibility, and conceptual knowledge after a 1- week delay. WM & IC are less predictive at immediate posttest, suggesting that these functions are not simply correlated with mathematics skill, but may be particularly important in the process of structure-mapping for durable schemaformation. These results inform the literature on both analogy and mathematics learning, extending previous findings implicating EFs as key for successful structure-mapping to an ecologically valid learning context.

Current psychological accounts of causal representation and reasoning do not capture phenomena related to causation by omission (e.g., “The absence of breathing causes death”), with one exception (Wolff, Barbey, & Hausknecht, 2010). We describe a novel theory of omissive causation that posits that people build discrete mental simulations – mental models – of causal relations (Goldvarg & Johnson-Laird 2001). The theory states that causes by omission refer to a set of temporally ordered models of possibilities. Reasoners tend to focus on only one of those models, i.e., the possibility in which breathing does not occur and death subsequently does. Likewise, the theory posits that reasoners distinguish between omission in the context of causation, enabling conditions, and prevention. We describe some initial predictions made by the model-based account, contrast it with an alternative psychological theory based on the transmission of causal forces, and set out directions for further research.

We describe a computational model of humans' ability to provide a detailed interpretation of a scene’s components. Humans can identify in an image meaningful components almost everywhere, and identifying these components is an essential part of the visual process, and of understanding the surrounding scene and its potential meaning to the viewer. Detailed interpretation is beyond the scope of current models of visual recognition. Our model suggests that this is a fundamental limitation, related to the fact that existing models rely on feed-forward but limited top-down processing. In our model, a first recognition stage leads to the initial activation of class candidates, which is incomplete and with limited accuracy. This stage then triggers the application of class-specific interpretation and validation processes, which recover richer and more accurate interpretation of the visible scene. We discuss implications of the model for visual interpretation by humans and by computer vision models

We show how the wide range in strengths of intensifying degree adverbs (e.g. very and extremely) could be explained by pragmatic inference based on differing cost, rather than differing semantics. This predicts a linear relationship between the meaning of intensifiers and their length and log-frequency. We test this prediction in two studies, using two different dependent measures, finding that higher cost does predict stronger meanings. We discuss the implications for adverbial meaning and the more general question of how extensive non-arbitrary form-meaning association may be in language.

We present a computational model of the representativeness heuristic. This model is trained on the entire English language Wikipedia corpus, and is able to use representativeness to answer questions spanning a very large domain of knowledge. Our trained model mimics human behavior by generating the probabilistic fallacies associated with the representativeness heuristic. It also, however, achieves a high rate of accuracy on unstructured judgment problems, obtained from large quiz databases and from the popular game show Who Wants to be a Millionaire?. Our results show how highly simplistic cognitive processes, known to be responsible for some of the most robust and pervasive judgment biases, can be used to generate the type of flexible, sophisticated, high-level cognition observed in human decision makers.

The idea that optimization plays a key role in linguistic cognition is supported by an increasingly large body of research. Building on this research, we describe a new approach to parsing distributed representations via optimization over a set of soft constraints on the wellformedness of parse trees. This work extends previous research involving the use of constraintbased or “harmonic”’ grammars by suggesting how parsing can be accomplished using fully distributed representations that preserve their dimensionality with arbitrary increases in structural complexity. We demonstrate that this method can be used to correctly evaluate the wellformedness of linguistic structures generated by a simple context-free grammar, and discuss a number of extensions concerning the neural implementation of the method and its application to complex parsing tasks.

It has been shown that prior knowledge and information are organized according to categories, and that also background knowledge plays an important role in classification. The purpose of this study is first, to investigate the relationship between background knowledge and text classification, and second, to incorporate this relationship in a computational model. Our behavioral results demonstrate that participants with access to background knowledge (experts), overall performed significantly better than those without access to this knowledge (novices). More importantly, we show that experts rely more on relational features than surface features, an aspect that bagof- words methods fail to capture. We then propose a computational model for text classification which incorporates background knowledge. This model is built upon vector-based representation methods and achieves significantly more accurate results over other models that were tested.

The present study examined how differences in onset (cohort) and offset (rhyme) neighborhood density influence the types of spoken word recognition errors made by listeners. Simulations of the TRACE model were used to derive preliminary predictions. Younger (N=15) and older (N=15) adults identified spoken words presented in moderate noise. Participants exhibited the standard inhibitory effect of phonological neighborhood density: slower recognition of spoken words from denser neighborhoods, with a larger effect for older adults. Most errors were phonological neighbors with few unrelated errors. However, the manipulation of cohort and rhyme density produced an unexpected reversal: the relative proportion of cohort vs. rhyme errors was biased toward cohorts when cohort density was low or when rhyme density was high, and toward rhymes when cohort density was high or rhyme density was low. These results are not consistent with the TRACE simulations and suggest a more complex pattern of lexical competition

How does abstract structure emerge during language learning? On some accounts, children’s early syntax emerges from direct generalizations from particular lexical items, while on others, syntactic structure is acquired independently and follows its own timetable. Progress on differentiating these views requires detailed developmental data. Using parental reports of vocabulary and grammar abilities, previous analyses have shown that early syntactic abstraction strongly depends on the growth of the lexicon, providing support for lexicalist and emergentist theories. Leveraging a large cross-linguistic dataset, we replicate and extend these findings, demonstrating similar patterns in each of four languages. Moreover, the power of our dataset reveals that there are measurable effects of age over and above those attributable to vocabulary size, and that these effects are greater for aspects of language ability more closely tied to syntax than morphology. These findings suggest non-lexical contributions to the growth of syntactic abstraction that all theories must address.

Causal models are key to flexible and efficient exploitation of the environment. However, learning causal structure is hard, with massive spaces of possible models, hard-to-compute marginals and the need to integrate diverse evidence over many instances. We report on two experiments in which participants learnt about probabilistic causal systems involving three and four variables from sequences of interventions. Participants were broadly successful, albeit exhibiting sequential dependence and floundering under high background noise. We capture their behavior with a simple model, based on the “Neurath’s ship” metaphor for scientific progress, that neither maintains a probability distribution, nor computes exact likelihoods.

The role of probabilistic reasoning in moral decision making has seen relatively little research, despite having potentially profound consequences for our models of moral cognition. To rectify this, two experiments were undertaken in which participants were presented with moral dilemmas with additional information designed to anchor judgements about how likely the dilemma’s outcomes were. It was found that these anchoring values significantly altered how permissible the dilemmas were found when they were presented both explicitly and implicitly. This was the case even for dilemmas typically seen as eliciting deontological judgements. Implications of this finding for cognitive models of moral decision making are discussed.

The present paper is a follow-up to the journal paper (Bra¨uner, 2014) which in turn is a revised and extended version of the conference paper (Bra¨uner, 2013). These papers were concerned with formalizations of the reasoning when giving correct responses to the psychological tests called the Sally-Anne task and the Smarties task, testing children’s capacity to ascribe false beliefs to others. In the present paper we give an analysis of what goes wrong when incorrect answers are given. Our analysis corroborates the claim that children under four and autistic children have difficulties shifting to a perspective different from their own.

With few exceptions, architectural approaches to modeling cognition have historically emphasized what happens in the mind following the transduction of environmental signals into percepts. To our knowledge, none of these architectures implements a sophisticated, general theory of human attention. In this paper we summarize progress to date on a new cognitive architecture called ARCADIA that gives a central role to attention in both perception and cognition. First, we give an overview of the architecture, comparing it to other approaches when appropriate. Second, we present a model of incremental object construction and property binding in ARCADIA using the well known change blindness phenomena to illustrate the time course of object perception and its dependence on attention. Finally, we discuss near-term challenges and future plans

The ability to form complex linguistic units from simpler ones lies at the center of many explanations of the communicative success and robustness of natural language. A closely related ability is that to generalize knowledge about such constructions to novel ones. The present investigation addresses the question what the minimal conditions for the emergence of such productive compositional communication are. Two features are argued to be required for this: relations between elements and classes over their relations. Using signaling games with reinforcement learning we show that a learning bias involving both aspects can lead to the emergence of such generalizable structure.

We propose a new model of risk preferences that integrates theoretical principles relevant to mental representation, metacognitive monitoring and editing, and individual differences in risk-taking propensity. Our model is based on fuzzy-trace theory, a theory of decision-making under risk. The theory posits that decision-makers use fuzzy gist representations of the meaning of decision information, in parallel with precise verbatim representations of the exact wording of that information. We account for core phenomena in decision theory, such as shifts in risk preference when logically equivalent gambles are described in terms of gains rather than losses—framing effects—and also extend fuzzytrace theory beyond these phenomena to encompass research on affect and personality

Existing neuroimaging studies in decision making predominantly employ the fMRI method. Despite its superior spatial resolution, fMRI is an expensive and impractical neuroimaging technology for purchasing behavior studies in the field. This study aims to explore the role of prefrontal cortex during purchasing behavior by utilizing functional nearinfrared (fNIR) spectroscopy; a low-cost, non-invasive and portable optical brain imaging methodology. The findings suggest that fNIRS can be effectively used for developing a neuro-physiologically informed, predictive model of purchasing behavior based on multivariate effects of activations in frontopolar, dorso-medial and dorso-lateral prefrontal cortex.

Why do languages parcel human experience into categories in the ways they do? Languages vary widely in their category systems but not arbitrarily, and one possibility is that this constrained variation reflects universal communicative needs. Consistent with this idea, it has been shown that attested category systems tend to support highly informative communication. However it is not yet known what process produces these informative systems. Here we show that human simulation of cultural transmission in the lab produces systems of semantic categories that converge toward greater informativeness, in the domains of color and spatial relations. These findings suggest that larger-scale cultural transmission over historical time could have produced the diverse yet informative category systems found in the world’s languages.

Study sequence can have a profound impact on learning. Previous research has often shown advantages for interleaved over blocked study, though the reverse has also been found. Learners typically prefer blocking even in situations for which interleaving is superior. The present study investigated learner regulation of study sequence, and its effects on learning in an ecologically valid context – university students using an online tutorial relevant to an exam that counted toward their course grades. The majority of participants blocked study by problem category, and this tendency was positively associated with subsequent exam performance. The results suggest that preference for blocked study may be adaptive under some circumstances, and highlight the importance of identifying task environments under which different study sequences are most effective.

Speaker transitions in conversation are often brief, with minimal vocal overlap. Signed languages appear to defy this pattern with frequent, long spans of simultaneous signing. But recent evidence suggests that turn boundaries in signed language may only include the content-bearing parts of the turn (from the first stroke to the last), and not all turn-related movement (from first preparation to final retraction). We tested whether signers were able to anticipate “stroke-to-stroke” turn boundaries with only minimal conversational context. We found that, indeed, signers anticipated turn boundaries at the ends of turnfinal strokes. Signers often responded early, especially when the turn was long or contained multiple possible end points. Early responses for long turns were especially apparent for interrogatives—long interrogative turns showed much greater anticipation compared to short ones.

The present study investigated the relationship between eye movement pattern in face recognition and cognitive performance during natural aging through modeling and comparing eye movement of young (18-24 years) and older (65-81 years) adults using Hidden Markov Model (HMM) based approach. Young adults recognized faces better than older adults, particularly when measured by the false alarm rate. Older adults’ recognition performance, on the other hand, correlated with their cognitive status assessed by the Montreal Cognitive Assessment (MoCA). Eye movement analysis with HMM revealed two different strategies, namely “analytic” and “holistic”. Participants using the analytic strategy had better recognition performance (particularly in the false alarm rate) than those using the holistic strategy. Significantly more young adults adopted the analytic strategy; whereas more older adults adopted the holistic strategy. Interestingly, older adults with lower cognitive status were associated with higher likelihood of using the holistic strategy. These results suggest an association between holistic eye movement patterns and cognitive decline in the elderly.

Studies show that people can recognize their own movements, such as their own walking (presented in silhouette using point lights), their own drawing (presented as a moving point light), own clapping, and their own piano playing. We extend this result to proprioceptive control, showing that people can recognize their own eye movements, when presented as just a point moving against a black background. Eye movements were recorded using a wearable eye tracking glass, while participants executed four tasks. A week later, participants were shown these videos, alongside another person's videos, for each task, and asked to recognize their own movements. Males recognized their own eye movements significantly above chance, but only for tasks with large and familiar body movements. Females performed below chance in these tasks. We argue that the standard common coding/motor simulation model does not account for this result, and propose an extension where eye movements and body movements are strongly coupled. In this model, eye movements automatically trigger covert motor activation, and thus participate directly in motor planning, simulations and the sense of agency

Short-term memory and working memory are two distinct concepts that have been measured in simple and complex span tasks respectively. A new span task was designed to manipulate a chunking factor while using a procedure similar to simple span tasks. This span task allowed studying the interaction between storage and processing in working memory, when processing is fully dedicated to optimizing storage. The main hypothesis was that the storage × processing interaction that can be induced by the chunking factor is an excellent indicator of intelligence because both working memory and intelligence depend on optimizing storage. Two experiments used an adaptation of the SIMON® game in which chunking opportunities were estimated using an algorithmic complexity metric. The results show that the metric can be used to predict memory performance and that intelligence is well predicted by the new chunking span task in comparison to other simple and complex span tasks

Recent work has begun to investigate how structured relations can be learned from non-relational and distributed input representations. A difficult challenge is to capture the human ability to evaluate relations between items drawn from distinct categories (e.g., deciding whether a truck is larger than a horse), given that different features may be relevant to assessing the relation for different categories. We describe an extension of Bayesian Analogy with Relational Transformations (BART; Lu, Chen & Holyoak, 2012) that can learn cross-category comparative relations from autonomously-generated and distributed input representations. BART first learns separate representations of a relation for different categories and creates second-order features based on these category-specific representations. BART then learns weights on these second-order features, resulting in a category-general representation of the relation. This hierarchical learning model successfully generalizes the relation to novel pairs of items (including items from different categories), outperforming a flat version of the learning model.

Mathematical competence can be evaluated by analyzing pauses between strokes that occurring whilst an individual copies equations. These pauses provide a temporal signal that reflects the cognitive effort to process chunks. ‘Centre-click’ interaction with a mouse and response grid on a computer display is introduced as new technique for measuring the temporal chunk signal. Alternative pause measures and forms of normalization are explored. It is shown that centre-click copying can be used to measure mathematical competence

Directional comparisons are often used to express similarity (e.g., “North Korea is like China”). These statements, however, frame the subject as the less typical figure and the complement as the more typical or prominent ground. Thus, despite expressing similarity, directional comparisons may imply that the ground is more representative. In Study 1, we analyze Twitter to show that directional comparisons occur in everyday conversation about gender; that men are the ground more often than women; and that only males frequently serve as the ground for positive traits (e.g., “Girls are as smart as boys”), suggesting that positive traits are considered typical of males, but not females. In Study 2, we show that directional comparisons intended to express equivalent ability (e.g., “Boys are as good as girls at a game called gorp”) cause adults to infer that the ground has more natural skill and that the figure has to work harder.

The mental number line (MNL) hypothesis is that numbers are mentally represented in spatial format, particularly in leftto- right orientation among Westerners. The MNL has received support from various paradigms, but it remains controversial as it is challenged by alternative models. Here we used an individual differences approach to assess spatialnumerical associations (SNAs) across a variety of tasks. The MNL hypothesis predicts correlations across SNA tasks because they should tap a common MNL representation. Control tasks were included to account for effects not specific to SNAs. Correlation analyses revealed significant associations across several SNA tasks, even when controlling for general cognitive abilities or individual differences in response time (RT). These findings provide unique support for the MNL hypothesis, and begin to shed insight on potential explanations that may contribute to variation in the strength of the correlations among SNA tasks.

Recent imaging studies have found that in simple arithmetic processing, addition is lateralized to the right hemisphere, whereas multiplication to the left. Here we aimed to investigate the cognitive mechanism underlying complicated arithmetic processing with a dual task paradigm. Participants were asked to complete a calculation task (addition or multiplication) and a letter judgment task (rhyme or shape judgment) simultaneously. We found that participants’ performance in addition and multiplication was interfered more by the simultaneous shape judgment task than the rhyme judgment task. This effect suggested that both complicated addition and multiplication relied more on right-lateralized visuospatial than left-lateralized phonological/verbal processing. The shift from left- to more right-lateralized processing in complicated multiplication suggests that participants may have adopted a visuospatial strategy to approximate numerosity when the calculation involved large numbers. These results suggest that the cognitive mechanism involved in arithmetic processing depends on both the operation and the context.

The current study asks when children begin to understand that when an object is added to a set, the numerosity of the set has increased regardless of set size. This knowledge can be expressed algebraically as ‘x + 1 > x’. In Experiment 1, 3- to 5-year-old children were asked to reason about transformations (i.e., addition, subtraction, rearrangement) performed on a visible set of objects. We found that 5-yearolds were able to reason about how each transformation affected numerosity, and 4-year-olds showed limited understanding. In Experiment 2, children were asked to reason about transformations performed on a hidden set of objects. Similar results were found. Together, we showed that the ability to reason about number algebraically develops gradually between the ages of 3 and 5. Implications for number word acquisition were discussed

Understanding and acting on online health information is increasingly a pre-requisite for patient self-care. Therefore, inadequate health literacy is a barrier to self-care among older adults with chronic illness. The goal of our study was to improve older adults’ comprehension of online health information. We extracted typical health texts from multiple credible health websites, and systematically improved the texts in terms of, content, language, organization and design. Results showed that older adults better understood the revised passages than the typical ones, in terms of their reading efficiency (time per unit of information uptake). Intervention benefits were greater for older adults with more domainspecific health knowledge, suggesting that knowledge facilitated the comprehension of health information in the revised texts. Implications for promoting older adults’ comprehension of health information are discussed.

Problem presentation can influence students’ understanding, choice of strategy, and accuracy. For example, the presence and type of external representation can alter performance. In this study, we examined the effect of diagrams on students’ performance in a symbolic problem domain. Sixty-one seventh-grade students solved algebraic equations with or without an accompanying diagram. The presence of diagrams increased accuracy and use of informal strategies. Overall, the benefits of diagrams found for word problems generalized to symbolic problems

In an attempt to assist proctors to prevent test takers from academic dishonesty in remotely administrated exams, this study investigated the ability of test takers’ behaviors during online assessments to predict their cheating decisions. Specifically, this experimental study focused on the role of students’ time delay and certainty rating during lab based online testing sessions. The analysis of hierarchical logistic regression indicated that not only time delay but also certainty rating had significantly statistical relation to test takers’ cheating decisions. The importance of the two proposed factors during online assessments was discussed and the prospects of the improvements of online proctoring systems were addressed.

Here we explored eye movement strategies that lead to better performance in face recognition with hidden Markov models (HMMs). Participants performed a standard face recognition memory task with eye movements recorded. The durations and locations of the fixations were analyzed using HMMs for both the study and the test phases. Results showed that in the study phase, the participants who looked more often at the eyes and shifted between different regions on the face with long fixation durations had better performances. The test phase analyses revealed that an efficient, short first orienting fixation followed by a more analytic pattern focusing mainly on the eyes led to better performances. These strategies could not be revealed by analysis methods that do not take individual differences in both temporal and spatial dimensions of eye movements into account, demonstrating the power of the HMM approach.

Holistic processing (HP) has been proposed to be a characteristic of right hemisphere (RH) processing. Here we test this claim using the divided visual field paradigm with Chinese character stimuli. HP is assessed through the composite paradigm, which is commonly used in perceptual expertise research. We found that in novice Chinese readers, a standard HP pattern emerged only in the left visual field/RH but not in the right visual field/left hemisphere, consistent with the analytic/ holistic hemispheric dichotomy in the literature. However, in expert Chinese readers, neither visual field showed the HP pattern, consistent with the finding that reduced HP is an expertise maker for Chinese character recognition. Thus, the RH does not always employ holistic processing; it depends on the perceivers’ experience with the stimuli. This is the first study demonstrating that expertise with a visual object type can modulate hemispheric difference in HP.

Visual memory for naturalistic scenes is mediated by: amount of exposure, semantic content, and type of encoding. These factors might interactively contribute to scene memorability. Thus, we tracked computer-mouse movements during an encoding phase where participants verified the congruency of sentence and scene pairs, which varied in plausibility. The presentation time of the scenes was also manipulated. Subsequently, in an unexpected recognition phase, participants had to indicate whether they remembered scenes (old and new). Recognition improved when correct verifications were made during encoding especially: when the scene was implausible, the stimuli pair congruent, and for longer presentation times. When comparing the trajectories between encoding and recognition, we found greater hesitancy during encoding, especially for implausible scenes in incongruent pairs, decreasing as presentation time increased. These results provide novel insights into the factors modulating the memorability of naturalistic scenes.

People often make decisions with the goal of gaining information which can help reduce their uncertainty. However, recent work has suggested that people sometimes do not select the most diagnostic information queries available to them. A critical aspect of information search decisions is evaluating how obtaining a piece of information will alter a learner’s beliefs (e.g., a piece of information that is redundant with what is already known is useless). This suggests a close relationship between information seeking decisions on one hand, and belief updating on the other. This paper explores the deeper relationship between these two constructs in a causal intervention learning task. We find that patterns in belief updating biases are predictive of decision making patterns in tasks where people must make interventions learn about the structure of a causal system.

“Upfixes” are graphic representations originating in the visual vocabulary used in comics where objects float above a character’s head, such as lightbulbs to mean inspiration. We posited that these graphic signs use an abstract schema stored in memory. This schema constrains upfixes to their position above the head and requires them to “agree” with the expression of their associated face. We asked participants to rate and interpret upfix-face pairs where the upfix was either above the head or beside the head, and/or agreed or disagreed with the face. Our stimuli also contrasted conventional and novel upfixes. Overall, both position and agreement impacted the rating and interpretations of both conventional and unconventional upfixes, and such understanding is modulated by experience reading comics. These findings support that these graphic signs extend beyond memorized individual items, and use a learned abstract schema stored in long-term memory, governed by particular constraints

Much of what we believe to know, we know through the testimony of others (Coady, 1994). Whether the resultant beliefs constitute knowledge or erroneous beliefs consequently rests directly on the reliability of our sources. While there has been long-standing evidence that people are sensitive to source characteristics, for example in the context of persuasion, exploration of the wider implications of source reliability considerations for the nature of our beliefs has begun only fairly recently. Likewise, much remains to be established concerning what factors influence source reliability. In this paper, we examine, both theoretically and empirically, the implications of using message content as a cue to source reliability.

Despite the impressive amount of financial resources invested in carrying out large-scale brain simulations, it is controversial what the payoffs are of pursuing this project. The present paper argues that in some cases, from designing, building, and running a large-scale neural simulation, scientists acquire useful knowledge about the computational performance of the simulating system, rather than about the neurobiological system represented in the simulation. What this means, why it is not a trivial lesson, and how it advances the literature on the epistemology of computer simulation are the three preoccupations addressed by the paper

Research in category learning has been dominated by a ‘reference point’ view in which items are classified based on attention-weighted similarity to reference points (e.g., prototypes, exemplars, clusters) in a multidimensional space. Although much work has attempted to distinguish between particular types of reference point models, they share a core design principle that items will be classified as belonging to the category of the most proximal reference point(s). In this paper, we present an original experiment challenging this distance assumption. After classification training on a modified XOR category structure, we find that many learners generalize their category knowledge to novel exemplars in a manner that violates the distance assumption. This pattern of performance reveals a fundamental limitation in the reference point framework and suggests that stimulus generalization is not a reliable foundation for explaining human category learning.

When speakers gesture, their gestures shape their thoughts, but how this happens remains unclear. What kinds of feedback from gesture—visual, proprioceptive, or both— drive these cognitive effects? Here we address this question using a test bed previously employed to explore gesture’s cognitive effects (Beilock & Goldin-Meadow, 2010). Participants solved the Tower of Hanoi puzzle, explained their solutions in speech and gesture, and solved the puzzle a second time. Previous studies using this paradigm have demonstrated that the gestures participants produce during the explanation phase affect their ability to solve the problem the second time. Unlike these prior studies, however, participants in the present study were blocked from seeing their hands while they gestured. Despite this absence of visual feedback, our results replicate previous studies in which visual feedback was available. These findings suggest that gesture may shape thought through proprioceptive feedback alone.

Many blind and low-vision individuals are unable to access digital media visually. Currently, the solution to this accessibility problem is to produce text descriptions of visual graphics, which are then translated via text-to-speech screen reader technology. However, if a text description can accurately convey the meaning intended by an author of a visualization, then why did the author create the visualization in the first place? This essay critically examines this problem by comparing the so-called graphic–linguistic distinction to similar distinctions between the properties of sound and speech. It also presents a provisional model for identifying visual properties of graphics that are not conveyed via text-tospeech translations, with the goal of informing the design of more effective sonic translations of visual graphics.

This study explores the relation between pain sensitivity and the cognitive processing of words. 130 participants evaluated the pain-relatedness of a total of 600 two-syllabic nouns, and subsequently reported on their own pain sensitivity. The results demonstrate that pain-sensitive people (based on their self-report) associate words more strongly with pain than less sensitive people. In particular, concrete nouns like syringe, wound, knife, and cactus, are considered to be more painrelated for those who are more pain-sensitive. We discuss our results in the light of three theoretical frameworks – cognitive bias, prototype theory, embodied account. We argue that the latter is best suited to explain the results of this study in the sense in which it implies the principle of body specificity, according to which different bodily characteristics lead to corresponding differences in the way in which people construct concepts and word meanings

We aim at examining how communication mode influences the production of gestures under specific contextual environments. Twenty-four participants were asked to present a topic of their choice under three instructional settings: a blackboard, paper-and-pencil, and a tablet. Participants’ gestures were investigated in three groups: deictic gestures that point to entities, representational gestures that present picturable aspects of semantic content, and beat gestures that are speech-related rhythmic hand movements. The results indicated that gesture production of the participants was influenced by the mode of instruction (i.e., board, paper-and-pencil, tablet).

To investigate the real time processing of metaphoric adjectives, we measured participants looking behavior as they listened to sentences such as The little boy was shocked as a result of the [electrical socket/report card] in the context of a display with four images. Displays included two Unrelated pictures, a Literal picture consistent with the literal interpretation of the adjective (an electrical socket), and a Metaphor picture consistent with the metaphorical interpretation (a report card). Sentences were divided into those with a preferred literal versus metaphorical reading of the adjective based on a norming study involving sentence fragments without the disambiguating information. Although conducted with different participants, those preferences were predictive of looking behavior during the eye tracking study. During the 1s interval before the onset of the disambiguating word, participants were more likely to fixate the image consistent with the preferred interpretation of the adjective than the unrelated pictures. That is, they were more likely to fixate the Literal picture in Literal biased sentences, and the Metaphor picture in Metaphor biased sentences. After the disambiguating information, participants showed an increased probability to fixate the actual target item, regardless of the preferred reading of the adjective. Results argue against models of metaphor comprehension that posit parallel activation of literal and metaphoric meaning.

Associative recognition—the ability to discriminate between studied and novel associations—has been attributed to the operation of a recall-like process that is not engaged during recognition of single items. An alternative mechanism for associative recognition is the formation of a compound memory cue that incorporates relational information between the two elements of the association. These alternatives make different predictions about the dynamics of associative recognition as revealed by speed-accuracy trade-off (SAT) functions: if recall were operating, SAT functions should approach asymptotic performance at a faster rate for stronger associations, whereas a compound cue mechanism predicts that only asymptotic performance, not rate, should be affected by strength. In a review of the literature, we find that only asymptotic performance, not rate, is affected by the strength of studied associations, supporting the operation of a compound cue mechanism. We present a formal model of this mechanism as a direct outgrowth of a model of single-item recognition (Cox & Shiffrin, 2012) and use it to predict observed SAT curves for both single-item and associative recognition in a variety of experiments

How early in life, and in what situations, are children sensitive to speakers' accents? Some researchers have suggested that accent is an early-developing, perhaps intrinsic, signal of group membership. However, other studies find little sensitivity to or awareness of accent in young children. Three experiments reported here examine 3-5-yearolds' comprehension of, and social decision-making with, a familiar (US English) accent and a foreign (Dutch) accent. Dutch accents were comprehended less well, particularly when salient phonological competitors were present, but social sensitivity was fairly weak until age 6-7 years. The latter finding contrasts with accounts positing early (perhaps innate) social sensitivity to accents.

Odor naming is difficult for people, but recent cross-cultural research suggests this difficulty is culture-specific. Jahai speakers (hunter-gatherers from the Malay Peninsula) name odors as consistently as colors, and much better than English speakers (Majid & Burenhult, 2014). In Jahai the linguistic advantage for smells correlates with a cultural interest in odors. Here we ask whether sub-cultures in the West with odor expertise also show superior odor naming. We tested wine and coffee experts (who have specialized odor training) in an odor naming task. Both wine and coffee experts were no more accurate or consistent than novices when naming odors. Although there were small differences in naming strategies, experts and non-experts alike relied overwhelmingly on source-based descriptions. So the specific language experts speak continues to constrain their ability to express odors. This suggests expertise alone is not sufficient to overcome the limits of language in the domain of smell.

We examined the effects of individual versus joint action and racial similarity and dissimilarity on a Simon task using mouse tracking to explore the implicit cognitive dynamics underlying responses. Participants were slower to respond when working with a partner than when working alone, and their mouse movements also differed across conditions. Participants paired with a different-race partner took longer to respond than participants paired with a same-race partner. We argue that, in the joint conditions, participants’ longer responses were the result of automatic inhibitory processes that arise within the social context.

In the current study, hand motions captured by a mousetracking system were used to index listener’s cognitive processes while making commitments to different choice alternatives during the processing of English passive and active structures. Fifty-seven second language (L2) speakers and 19 first language (L1) speakers of English carried out an aural forced-choice picture identification task comprised of 75 items. The findings indicate that although L1 participants have quicker response times for both active and passive structures than L2 participants, both L1 and L2 participants demonstrate similar difficulties in processing passive constructions.

Decision-bound models of categorization like General Recognition Theory (GRT: Ashby & Townsend, 1986) assume that people divide a stimulus space into different response regions, associated with different categorization decisions. These models have traditionally been applied to empirical data using standard model-fitting methods like maximum likelihood estimation. We implement the GRT as a Bayesian latent mixture model to infer both qualitative individual differences in the types of decision bounds people use, and quantitative differences in where they place the bounds. We apply this approach to a previous data set with two category structures tested under different cognitive loads. Our results show that different participants categorize by applying diagonal, vertical, or horizontal decision bounds. Various types of contaminant behavior are also found, depending on the category structures and presence or absence of load. We argue that our Bayesian latent mixture framework offers a powerful approach to studying individual differences in categorization.

The developmental relationship between linguistic semantic processing and non-linguistic semantic processing (interpreting pictures) is investigated in a longitudinal neuroimaging study of two deaf individuals who did not begin acquiring their first language until the age of 14. 1-2 years after they began learning language, the two case studies performed a picture-sign priming task. Magnetoencephalography was localized to the cortical surface, showing that picture processing was initially bilateral or focused in the canonical left hemisphere language network, while single sign processing was initially focused in the right hemisphere. After 15 months of additional language experience, the neural responses too both pirctures and single signs reversed in lateralization, becoming more similar to those observed in a control group of native signers. The results shed new light on the interdependence between linguistic and non-linguistic semantics in cognitive development, as well as the neural underpinnings of semantic processing.

We introduce a novel class of visual illusion -- motion pareidolia -- in which sequential presentations of random textures can trigger percepts of coherent apparent motion. In two experiments we presented observers with sequences of random 140x140 pixel arrays refreshing at 2.5Hz. In Experiment 1, observers were primed with a coherent motion pattern, such as fixed texture shifting up-and-down across frames. After 8 priming frames, the textures became completely random from frame to frame. Participants were instructed to indicate when they could no longer perceive the primed motion pattern. Participants' responses were delayed by an average of 6 frames (or 2.4 seconds). In Experiment 2, observers detected motion patterns in 6-frame sequences under different noise levels and falsely identified coherent motion in 39% of the purely random sequences. To account for this phenomenon, we propose a selective visual attention process that is biased to detect expected motion.

Semantic structure in the mental lexicon is often assumed to follow a taxonomic structure grouping similar items. This study uses a network clustering analysis of a massive word association dataset that does not primarily focus on concrete noun categories, but includes the majority of the words used in daily life. At this scale, we found widespread overlap between thematically organized clusters, arguing against a discrete categoric view of the lexicon. An empirical analysis focusing on taxonomic categories confirmed the widespread thematic structure even for concrete noun categories in the animal domain. Overall, this suggests that applying network clustering to word association data provides valuable insight into how large-scale semantic information is represented. This analysis leads to a different, more thematic topology than the one inferred from idealized small-scale approaches that sample only specific parts of the lexicon

Moral judgment depends upon inferences about agents’ beliefs, desires, and intentions. Here, we argue that in addition to these factors, people take into account the moral optimality of an action. Three experiments show that even agents who are ignorant about the nature of their moral decisions are held accountable for the quality of their decision—a kind of behaviorist thinking, in that such reasoning bypasses the agent’s mental states. In particular, whereas optimal choices are seen as more praiseworthy than suboptimal choices, decision quality has no further effect on moral judgments—a highly suboptimal choice is seen as no worse than a marginally suboptimal choice. These effects held up for judgments of wrongness and punishment (Experiment 1), positive and negative outcomes (Experiment 2), and agents with positive and negative intentions (Experiment 3). We argue that these results reflect a broader tendency to irresistibly apply the Efficiency Principle when explaining behavior

We present evidence that successful chunk formation during a statistical learning task depends on how well the perceiver is able to parse the information that is presented between successive presentations of the to-be-learned chunk. First, we show that learners acquire a chunk better when the surrounding information is also chunk-able in a visual statistical learning task. We tested three process models of chunk formation, TRACX, PARSER, and MDLChunker, on our two different experimental conditions, and found that only PARSER and MDLChunker matched the observed result. These two models share the common principle of a memory capacity that is expanded as a result of learning. Though implemented in very different ways, both models effectively remember more individual items (the atomic components of a sequence) as additional chunks are formed. The ability to remember more information directly impacts learning in the models, suggesting that there is a positive-feedback loop in chunk learning.

The evidence concerning the level at which learned CP effects occur is complex. The goal of this study was to use a different approach to this question by manipulating the abstractness of the information available for distinguishing pairs of items in an XAB task, and the presence or absence of a short task-filled delay between X and AB. Participants engaged in XAB trials containing a mixture of trials with and without the delay task before and after standard training to classify visual texture stimuli into two categories. Training improved discrimination of pairs differing on the category-relevant dimension whether within- or betweencategory, but not on pairs differing only on non-category relevant low level features. In addition, only successful learners in the post-training trials avoided decreased discrimination accuracy due to the delay task, suggesting that they formed more stable representations. However, this effect was not limited to pairs varying in category-relevant ways.

In social settings, people often reason about unobservable mental content of other people, such as their beliefs, goals, or intentions. This ability helps them to understand and predict the behavior of others. People can even take this ability further, and use higher-order theory of mind to reason about the way others use theory of mind, for example in ’Alice believes that Bob does not know about the surprise’. However, empirical evidence suggests that people do not spontaneously use higher-order theory of mind in strategic games. In this paper, we let participants negotiate with computational theory of mind agents in the setting of Colored Trails. We find that even though participants are unaware of the level of sophistication of their trading partner, within a few rounds of play, participants offers are more indicative of second-order theory of mind reasoning when their trading partner was using second-order theory of mind as well.

World knowledge enters into pragmatic utterance interpretation in complex ways, and may be defeasible in light of speakers’ utterances. Yet there is to date a surprising lack of systematic investigation into the role of world knowledge in pragmatic inference. In this paper, we show that a state-of-the-art model of pragmatic interpretation greatly overestimates the influence of world knowledge on the interpretation of utterances like Some of the marbles sank. We extend the model to capture the idea that the listener is uncertain about the background knowledge the speaker is bringing to the conversation. This extension greatly improves model predictions of listeners’ interpretation and also makes good qualitative predictions about listeners’ judgments of how ‘normal’ the world is in light of a speaker’s statement. Theoretical and methodological implications are discussed.

The concepts expressed by social role terms such as artist and scientist are unique. In a series of experiments, Knobe, Prasada, and Newman (2013) show that these terms allow two independent criteria for categorisation, one of which is inherently normative. This paper presents and tests a novel account of the content of these ‘dual character concepts’. We argue that the normative dimension of dual character concepts represents commitments to fulfill certain idealized functions. We then present evidence that the normative dimension is a central dimension in the conceptual structure of dual character concepts. Finally, we show that our account is both descriptively and explanatorily adequate.

Previous research on multiplicative reasoning has shown that for whole numbers, understanding of division is intimately linked to multiplication, as retrieval of division facts is often accomplished through reverse multiplication. We recently extended this research to rational numbers, and found that inverse multiplication problems can serve as primes for one another (e.g., a √ó b/a = a primes b √ó a/b = b) when the second multiplier is expressed as a fraction, but not when it is expressed as a decimal. In the current paper we propose a process model of how such relational priming takes place, and report two experiments that test the limits of this priming effect. The first varies the format of the equations as fractions or a total division equation, and shows that priming is only observed using the fraction format; the second varies the multiplicative complexity of the factors in the equations, and shows that priming requires a common factor linking the successive problems.

Recent work has shown that undergraduates at a major public university demonstrate implicit understanding of inverse relations between multiplication problems with fractions, as evidenced by the fact that solving one problem facilitates solving its inverse. The present study investigated whether such implicit understanding of mathematical relations is related to overall math ability. We found that low performers showed relational facilitation only when it was supported by perceptual similarity, whereas high performers showed relational facilitation on both perceptually similar and dissimilar problems. These findings are interpreted in terms of novice-expert differences in the representation of mathematical relations.

Various studies have recently shown that the long-held claim that the relation between the sound of a word and its meaning is arbitrary needs to be revisited. In two computational studies we investigated whether word valence can be derived from sound features in English, Dutch and German. In Study 1, we identified the extent to which individual phonological features explained valence scores per language separately. In Study 2, we aimed to determine the optimal combination of cues that can predict valence scores across the three languages using two statistical classifiers and four machine learning classifiers. Our results showed that frequency and word complexity were the most reliable shared cues to predict valence for all three languages, obtaining a correct valence classification of about 60%. This percentage could be enhanced for individual or pairs of languages using additional relevant cues. These findings demonstrated that the claim that the relation between the sound of a word and its meaning is arbitrary is too strong.

A key aspect of theory of mind is the ability to reason about other people's desires. As adults, we know that desires and preferences are subjective and specific to the individual. However, research in cognitive development suggests that a significant conceptual shift occurs in desire-based reasoning between 14 and 18 months of age, allowing 18- but not 14- month-olds to understand that different people can have different preferences (Lucas et al., 2014; Ma & Xu 2011; Repacholi & Gopnik, 1997). The present research investigates the kind of evidence that is relevant for inducing this shift and whether younger infants can be trained to learn about the diversity of preferences. In Experiment 1, infants younger than 18 months of age were shown demonstrations in which two experimenters either liked the same objects as each other (in one training condition) or different objects (in another training condition). Following training, all infants were asked to share one of two foods with one of the experimenters – they could either share a food that the experimenter showed disgust towards (and the infants themselves liked) or a food that the experimenter showed happiness towards (and the infants themselves did not like). We found that infants who observed two different experimenters liking different objects during training later provided the experimenter with the food she liked, even if it was something they disliked themselves. However, when infants observed two experimenters liking the same objects, they later incorrectly shared the food that they themselves liked with the experimenter. Experiment 2 controlled for an alternative interpretation of these findings. Our results suggest that training allows infants to overturn an initial theory in the domain of Theory of Mind for a more advanced one.

Some languages describe musical pitch in terms of spatial height; others in terms of thickness. Differences in pitch metaphors also shape adults’ nonlinguistic space-pitch representations. At the same time, 4-month-old infants have both types of space-pitch mappings available. This tension between prelinguistic space-pitch associations and their subsequent linguistic mediation raises questions about the acquisition of space-pitch metaphors. To address this issue, 5-year-old Dutch children were tested on their linguistic knowledge of pitch metaphors, and nonlinguistic spacepitch associations. Our results suggest 5-year-olds understand height-pitch metaphors in a reversed fashion (high pitch = low). Children displayed good comprehension of a thickness-pitch metaphor, despite its absence in Dutch. In nonlinguistic tasks, however, children did not show consistent space-pitch associations. Overall, pitch representations do not seem to be influenced by linguistic metaphors in 5-year-olds, suggesting that effects of language on musical pitch arise rather late during development.

Recent research has identified key differences in the way individuals make decisions in predictive learning tasks, including the use of feature- and rule-based strategies in causal learning and model-based versus model-free choices in reinforcement learning. These results suggest that people rely to varying degrees on separable psychological processes. However, the relationship between these types of learning strategies has not been explored in any depth. This study investigated the relationship between feature- vs rule-based strategies in a causal learning task and indices of model-free and model-based choice in a two-step reinforcement learning procedure. We found that rule-based transfer was associated with the use of model-based, but not model-free responding in a two-step task.

A great deal of research has compared monolingual and bilinguals on conflict resolution tasks, with inconsistent findings: Some studies reveal a bilingual advantage on global RTs, some reveal a bilingual advantage on interference cost, and some show no advantage. We report a meta-analysis of 73 comparisons (N = 5538), with estimates of global RTs and interference cost for each study. Results revealed a moderately significant effect size that was not moderated by type of cost (global RT or interference cost) or task. Age interacted with type of cost, showing a pattern difficult to reconcile with theories of bilingualism and executive control. Additionally there was a significant main effect of lab, which might be due to sociolinguistic differences in samples, data treatment and methodology, or Hawthorne effects.

Research into sound-symbolism has shown that people can consistently associate certain pseudo-words with certain referents; for instance, pseudo-words with rounded vowels and sonorant consonants are linked to round shapes, while pseudowords with unrounded vowels and obstruents (with a noncontinuous airflow), are associated with sharp shapes. Such sound-symbolic associations have been proposed to arise from cross-modal abstraction processes. Here we assess the link between sound-symbolism and cross-modal abstraction by testing dyslexic individuals’ ability to make sound-symbolic associations. Dyslexic individuals are known to have deficiencies in cross-modal processing. We find that dyslexic individuals are impaired in their ability to make sound-symbolic associations relative to the controls. Our results shed light on the cognitive underpinnings of sound-symbolism by providing novel evidence for the role —and disruptability— of cross-modal abstraction processes in sound-symbolic e?ects.

Despite widespread political conspiracy theories about Presidents Barack Obama and George W. Bush, a majority of partisans continue to distance themselves from such beliefs. Even so, the ideological biases that drive conspiratorial thinking may be hard to overcome. In this study, we examine the unintentional endorsement of conspiratorial beliefs as revealed in movement dynamics. We track the cursor movements of Republicans and Democrats as they click target regions on their computer screens, ostensibly providing bias-free opinions (e.g., clicking “FALSE” upon reading “Barack Obama was born in Kenya”). However, during these response movements, we find inhibition and movement attraction to regions of the screen where a competitor response is located (e.g., “TRUE” for the “birther” conspiracy). These dynamics are not present for general conspiracies or political knowledge items. Though both Republicans and Democrats show evidence of implicit biases, changes in the strength of competition also reveal key asymmetrical differences.

How do people decide between several options presented to them? Normative accounts suggest the utilities of options are fixed, but subjective accounts suggest utilities depend on context. In the current paper, we propose a novel model of choice that may help reconcile these accounts. We propose that choice behavior may depend on an “Intentional Selection Assumption”: when people are presented with multiple options, they assume the options were intentionally selected by a person with specific questions in mind. Inferences about the intentional selection of options inform the chooser about the features that are intended to be most relevant. In this way, context can affect the desirability of a particular option, without requiring shifting utilities over features. Two behavioral experiments support the claim that participants are sensitive to intentional selection. We discuss the importance of taking choosers’ assumptions about intentional selection into account in future investigations of choice behavior.

Science education aims at developing students’ knowledge of scientific concepts and principles. However, students differ in their prior knowledge and cognitive skills and thus follow different learning pathways. We examined whether and how experimentation skills predict elementary students’ pathways of conceptual knowledge development in science education. First to sixth grade students (N = 1275) received 15 units of inquiry-based classroom instruction on the topic “floating and sinking”. Students’ experimentation skills were assessed before instruction. Their conceptual knowledge about floating and sinking was assessed before and after instruction. Latent profile transition analysis, a markov chain mixture model for continuous longitudinal data, revealed that students with higher experimentation skills were more likely to develop proficient and consistent knowledge of floating and sinking. We discuss theoretical implications of this finding, advantages of mixture models to examine conceptual knowledge development, and implications for science education in elementary school.

Minimizing cognitive resources while executing wellpracticed motor tasks has been shown to increase automaticity and enhance performance (e.g., Beilock, Carr, Macmahon, & Starkes, 2002). Based on this principle, we examined whether more fluent speech production could be induced through a dual task paradigm that engaged working memory (WM) while speech was produced. We also considered whether effects varied for speakers who differed in their habitual degree of attentional control during speech production. Twenty fluent adults and 19 adults who stutter performed (1) a baseline speaking task, (2) a baseline WM task with manipulations of domain, load, and inter-stimulus interval (ISI), and (3) a series of dual tasks in which the speaking task was combined with each unique set of WM conditions. Results indicated a fluency benefit under dual task conditions, which was specific to atypical forms of disfluency but comparable across speaker types and manipulations of the WM task. Findings suggest that WM is associated with atypical forms of disfluency and that suppressing these resources enhances speech fluency, although further research is needed to specify the cognitive mechanism involved in this effect and clarify the nature of this association.

The current project examined how changes to task constraints impacted the behavioral dynamics of an interpersonal collision avoidance task previously examined and modeled by Richardson and colleagues (2015). Overall, the results demonstrate that decreasing the cost associated with colliding influences the stability and symmetry of the movement dynamics observed between co-actors in a manner consistent with those predicted by the Richardson et al. (2015), collision avoidance model. The current study therefore provides evidence that the behavioral dynamics that shape interpersonal or joint-action behavior are not only defined by the physical and informational properties of a task, but also by the strength and importance of the shared task goal.

Humans are able to successfully detect characteristics about others that serve to guide interaction, yet the source of this information is unclear. We hypothesized that biological motion specifies sex and race as these invariant categorical characteristics often guide interaction. Results indicated that movement kinematics are necessary but not sufficient for sex detection and that race is detectable when movement is produced by Caucasians but not African Americans, and only when kinematic information is embedded in body structure. These results imply that social psychological perspectives on person perception should be integrated with ecological psychological perspectives on affordances in order to understand social cognition.

This paper describes how a cognitive architecture builds a spatial model and navigates from it without a map. Each constructed model is a collage of spatial affordances that describes how the environment has been sensed and traversed. The system exploits the evolving model while it directs an agent to explore the environment. Effective models are learned quickly during travel. Moreover, when combined with simple heuristics, the learned spatial model supports effective navigation. In three simple environments, these learned models describe space in ways familiar to people, and often produce near-optimal travel times

What is the relationship between recognizing objects and drawing objects? We examine the possibility that both functions are supported by a common internal representation. First, we show that a model of ventral visual cortex only optimized to recognize objects in photographs generalizes to drawings of objects, suggesting that the capacity for visual abstraction is rooted in the functional architecture of the visual system. Next, we tested whether practice drawing objects might alter how those and other objects are represented. On each trial, participants sketched an object. The model then guessed the identity of the sketched object, providing realtime feedback. We found that repeatedly sketched objects were better recognized after training, while sketches of unpracticed but similar objects worsened. These results show that visual production can reshape the representational space for objects: by differentiating trained objects and merging other nearby objects in the space.

Patterns of object naming often differ between languages, but bilingual speakers develop convergent naming patterns in their two languages that are distinct from those of monolingual speakers of each language. This convergence appears to reflect dynamic interactions between lexical representations for the two languages. In this study, we present a self-organizing neural network model to simulate semantic convergence in the bilingual lexicon and investigate mechanisms underlying semantic convergence. Our results demonstrate that connections between two languages can be established through the simultaneous activations of related words in both languages, and these connections between two languages pull the two lexicons toward each other. These results suggest that connections between words in the bilingual lexicon play a major role in bilinguals’ semantic convergence. The model provides a foundation for exploring how various input variables will affect bilingual patterns of output.

Artificial agents are becoming more than human-computer interfaces: they are becoming artificial companions, interacting on a long-term basis and building a relationship with the user. This evolution brought new challenges, such as designing agents with personalities to the benefits of users. We endow artificial agents with regulatory focus, taking a sociocognitive approach of personality, by using machine-learning techniques. We test whether this personality can be perceived by users and if there is a regulatory fit effect on the users credibility judgement of the agent (i.e. is the agent perceived as more credible if its regulatory focus is the same as that of the user?). Our results show agents regulatory focus can be adequatly perceived by users playing a board game against an agent expressing its regulatory focus via machine-learned strategies. A regulatory fit effect was found on the likeability judgment for prevention focus users but not for promotion focus users.

Recent evidence shows that, contrary to what is commonly assumed, people who are pressured to think fast are also less likely to provide a heuristic judgment when heuristic and logical considerations point to conflicting answers in a conjunction fallacy task (Vallée-Tourangeau & Faure- Bloom, under review). The present study explores this finding using an eye-tracking methodology. Eye movements from 41 participants were recorded while they read a thumbnail description and made a judgment on a statement comparing the probability of a single-event and that of a conjunctive event. Results showed participants focused more on the comparative probability statement under logicoheuristic conflict while they focused more the task description in the absence of conflict. Additionally, longer judgment latencies predicted higher rates of heuristic responding, which contradicts the original dual-process assumption that heuristic thinking in conjunction fallacy tasks is fast.

The role of processing constraints on sentence structure has been a topic of central interest in cognitive science. One proposal (Hawkins, 2004) suggests that language production system is organized to facilitate efficient parsing. We experimentally test this hypothesis using a miniature artificial language learning paradigm. Our findings support this account. Even though the input languages did not favor early placement of cues to grammatical function assignment (case and word order), participants used these cues in their own productions significantly more often in such a way as to allow early correct parsing commitments. This preference interacted with a bias to mark the less expected: Participants tended to use more case-marking in non-English OSV sentences. Our results underscore the potential of miniature artificial learning for language production research.

Mind wandering has been investigated in a variety of sustained attention tasks. In the present research, we investigated the role of mind wandering while listening to familiar or unfamiliar musical excerpts, and its effects on linguistic processing. Participants performed a lexical congruity task involving judging the semantic relatedness of a list of word pairs while listening to familiar classical music, unfamiliar classical music, or non-music environmental sound clips. Mind wandering episodes were probed randomly and intermittently for participants to self-report their mind wandering episodes during the task. Results showed that listening to familiar music is associated with faster response times and lower frequency of mind wandering. Whereas mind wandering episodes tend to be more frequent when participants listened to unfamiliar music. Implications from previous attention models and theories of music familiarity suggest that familiar music might increase task enjoyment without compromising behavioral performance.

The perception of time is distorted by many factors, but is it possible that causality would affect our perception of time? We investigate timing changes in the temporal binding effect, which refers to a subjective shortening of the interval between actions and their outcomes. Two experiments investigated whether binding may be due to variations in the rate of an internal clock. Specifically, we asked whether clock processes in binding reflect a general timing system, or a dedicated clock unique to causal sequences. We developed a novel experimental paradigm in which participants made temporal judgments of either causal and non causal intervals, or the duration of an event embedded within that interval. While we replicated the temporal binding effect, we found no evidence for commensurate changes to time perception of the embedded event, suggesting that temporal binding is effected by changes in a specific and dedicated, rather than a general clock system.

Infants’ ability to detect and generalize abstract rules (e.g., ABB, ABA) in auditory stimuli has been well documented, however their ability to do so from visual stimuli has received considerably less attention. Moreover, the few studies reported suggest that this kind of learning is especially sensitive to details of the experimental design. Here, we focus on 3- to 4-month-old infants (N=40) to identify both the origins of visual abstract rule learning in infancy and the conditions that best support it. Our results provide the earliest evidence to date, documenting that by 3 months of age, infants successfully learn and generalize rules in the visual modality. They also reveal that providing infants with an opportunity to examine the stimuli simultaneously may be instrumental to their success.

The present study explores learning phonological alternations that contain exceptions. Participants were exposed to a back/round vowel harmony pattern in which a regular suffix followed harmony, varying between /e/ and /o/ depending on the back/round phonetic features of the stem, and an exceptional suffix that was always /o/ regardless of the features of the stem vowel. Participants in Experiment 1 learned the behavior of both suffixes, but performance for the non-alternating suffix was higher when the suffix happened to adhere to vowel harmony. In Experiment 2, participants were exposed only to the same suffixes as Experiment 1, but the non-alternating suffix only appeared in harmonic contexts, creating ambiguity between exceptionality and alternation. Participants only correctly selected the non-alternating suffix when it appeared in a harmonic context. This suggests that learners are biased towards alternating harmony patterns, but require concrete evidence of non-alternation to learn the nonalternating suffix.

The present study explores the effects of frequency in learning to parse novel morphological patterns. In two experiments, suffixes were divided into three classes: high, medium and low frequency, based on the proportion of stems in the input that each suffix attached to (high frequency = 12/12, medium frequency = 6/12, and low frequency = 2/12). In Experiment 1, learners were better at segmenting words containing high frequency suffixes compared to low frequency suffixes, even when the stems were novel. In Experiment 2, token frequency was controlled for across all three suffix frequency classes, but learners were still better at segmenting high frequency suffixes, even when words containing high frequency suffixes were less frequent. These results suggest that learners are sensitive to the frequency distributions of the morphemes in their language, supporting work suggesting that a Zipfian distribution may be ideal for language learning

Adaptive Control of Thought-Rational (ACT-R) and the Linear Ballistic Accumulator (LBA) were compared in a model mimicry simulation of the Psychomotor Vigilance Task (PVT), a simple, reaction time (RT) task requiring sustained attention. The models use different formalisms to capture the full response profile of the PVT. The parameters were varied systematically to illustrate the ranges of the models’ predictions, to assess the models’ estimation properties, and to determine which parameters in the models correspond with each other. Both models produced skewed RT distributions typical of empirical data, including false starts and lapses. The simulation study demonstrated that both models and their parameters are recoverable. Lastly, isolated parameters in the LBA model captured the effects of varying parameters in the ACT-R model, but the reverse was not always true. These interesting correspondences across different modeling formalisms suggest the possibility of integrating ACT-R and the LBA in future work

An experiment tested the hypothesis that people sometimes take physical actions to help themselves solve problems. The task was to generate all possible words that could be formed from seven Scrabble letters. In one condition, participants could use their hands to manipulate the letters, and in another condition, they could not. Quantitative results show that more words were generated and lower frequency words were generated with physical manipulation than without. Qualitative results suggest that participants who could manipulate the letters tended to subdivide the task into smaller tasks (focusing on fewer letters at a time). Overall, our results can be explained in terms of an interactive search process in which external, physical activity effectively complements internal, cognitive activity, providing a reliable way to simplify search, explore the space of letter combinations, and identify potential words

Clinical disorders are multidimensional phenomena that are important to both clinicians and the lay public, as well as to cognitive scientists interested in understanding how people think and reason about complex domains. To date, however, little work has examined the factors that influence the folk psychiatry of addiction. Participants in the present study read a brief paragraph about addiction pitched at either an abstract or personal level, followed by a series of questions about the causes and treatment of drug addiction. We further manipulated whether addiction was described using a medical or psychological label. Results revealed that liberals and conservatives varied dramatically with respect to their folk psychiatric reasoning, with liberals preferring a more biological/medical view, which is associated increased support for medical interventions, reduced feelings of personal responsibility, and elevated feelings of stigma. Framing addiction using medical labels and at an abstract level pushed people towards this biological view, suggesting that media reports and messaging campaigns may influence how people conceptualize addiction.

The current study investigated eye-hand coordination in natural reaching. We asked whether the speed of reaching related to the quality of visual information obtained by young children and adults. Participants played with objects on a table while their eye and hand movements were recorded. We developed new techniques to find reaching events in natural activity and to determine how closely participants aligned gaze to objects while reaching. Reaching speed and eye alignment were related for adults but not for children. These results suggest that adults but not children adapt reaching movements according to the quality of visual information (or vice-versa) during natural activity. We discuss possibilities for why this coordination was not observed in children

Information provided by the linguistic context has been shown to have a strong facilitatory effect on the activation and processing of upcoming words. The studies described in this paper aim to model the relation between context and target words using a distributional semantic model. We report three modelling studies in which we show that this model can successfully capture context effects in human-generated data (reading times and association scores).

The processing of Navon figures (Navon, 1977), i.e., hierarchical letter stimuli, has been studied in experimental settings for many years. In particular, they have been studied in the context of visual hemifield studies and yielded an interaction between hemifield and whether a target is at the local or global level, with a right hemisphere advantage for the global level, and a left hemisphere advantage for the targets at the local level (Sergent, 1982). This is a ventral stream process, however, and we were interested in whether there might be a similar interaction for hierarchical motion stimuli, presumably a dorsal stream process. Hence we developed a series of dynamic geometric Navon figures in order to study global/local rotation processing. These figures consist of a global figure (a triangle or a square) made up of local figures (also triangles or squares). Both global and local figures can rotate in either clockwise or counterclockwise directions independently. We found that there is no right or left visual field perceptual advantage for either the global or local levels of these figures. However, curiously enough, we found that there is a significant processing advantage for clockwise motion compared to counterclockwise motion. We also found a highly significant interaction between the detection of a particular rotational motion and the presence or absence of that motion in the figure being examined. Finally, our data strongly support the Global Precedence Hypothesis which says that people generally tend to focus on the global properties of an object before local properties and that processing proceeds in a global-to-local direction.

This paper describes an extension to the MOSAIC model which aims to increase MOSAIC’s fit to the cross-linguistic occurrence of Optional Infinitive (OI) errors. While previous versions of MOSAIC have successfully simulated these errors as truncated compound finites with missing modals or auxiliaries, they have tended to underestimate the rate of OI errors in (some) obligatory subject languages. Here, we explore defaulting effects, where the most frequent form of a given verb is substituted for less frequent forms, as an additional source of OI errors. It is shown that defaulting in English tends to result in the production of bare forms that are indistinguishable from the infinitive, while defaulting in Spanish is less pronounced, and tends to result in the production of 3rd person singular forms. Dutch verb forms are dominated by the stem in corpus-wide statistics, and the infinitive in utterance-final position, suggesting defaulting in Dutch may change qualitatively across development. Defaulting is shown to increase MOSAIC’s fit to English and Dutch without affecting its already good fit to Spanish, and provides a potential way of simulating the cross-linguistic pattern of verb-marking errors in children with SLI.

Interpersonal physiological entrainment is increasingly argued to underlie rapport, empathy and even team performance. We investigate the presence, temporal development, possible mechanisms and impact of interpersonal heart rate entrainment during collective creative LEGO construction tasks. We observe a statistically significant presence and increase over time of structured entrainment, which relates to the actual unfolding and development of behavioral coordination. Heart rate entrainment does not predict rapport and perceived group competence, but behavioral coordination does. Physiological entrainment, thus, should not be considered a universal unmediated proxy for shared emotions, empathy and collective performance. Behavioral coordination – at least in tasks requiring forms of joint action – seems to be a more informative proxy for both physiological entrainment and collective experience.

This paper takes a two-pronged approach to investigate cross-domain influence on creativity. We present a study in which creative individuals were asked to list influences on their creative work. More than half the listed influences were unrelated to their creative domain, thus demonstrating empirically that crossdomain influence is widespread. We then present a preliminary model of exaptation, a form of crossdomain influence on creativity in which a different context suggests a new use for an existing item, using as an example waste recycling of petroleum byproducts

How do listeners integrate multiple sources of information in order to accurately anticipate turn endings? In two experiments using synthesised speech and a virtual agent we examined the role of verbal and gaze information in a turnend anticipation task. Listeners were as good at anticipating the synthesised voice as they were with human speakers (Experiment 1). However, the direction and timing of the agent’s gaze had little influence on their accuracy (Experiment 2). Overall, these findings support the idea that anticipation of turn ends relies primarily, but not exclusively, on verbal content.

Human conditional reasoning is defeasible: people withdraw logically valid conclusions if they are aware of situations (i.e., exceptions) that prevent the consequent of the rule to happen although the antecedent is given. In this paper we investigate defeasible reasoning with quantified rules. In two experiments we rephrased conditionals from the literature (Experiment 1) and rules from penal code (Experiment 2) as either universal or existential rules and embedded them into Modus Ponens and Modus Tollens inference problems. We show that defeasible reasoning also exists for quantified rules. However, the kind of quantifier (universal vs. existential) did not affect inferences. This last finding conflicts with theories highlighting the importance of logic in human reasoning.

Computational models of semantics have emerged as powerful tools for natural language processing. Recent work has developed models to handle compositionality, but these models have typically been evaluated on large, uncontrolled corpora. In this paper, we constructed a controlled set of phrase pairs and collected phrase similarity judgments, revealing novel insights into hu- man semantic representation. None of the computa- tional models that we considered were able to capture the pattern of human judgments. The results of a sec- ond experiment, using the same stimuli with a trans- formational judgment task, support a transformational account of similarity, according to which the similarity between phrases is inversely related to the number of ed- its required to transform one mental model into another. Taken together, our results indicate that popular mod- els of compositional semantics do not capture important facets of human semantic representation.

How do people make causal judgments? Here, we propose a counterfactual simulation model (CSM) of causal judgment that unifies different views on causation. The CSM predicts that people’s causal judgments are influenced by whether a candidate cause made a difference to whether the outcome occurred as well as to how it occurred. We show how whethercausation and how-causation can be implemented in terms of different counterfactual contrasts defined over the same intuitive generative model of the domain. We test the model in an intuitive physics domain where people make judgments about colliding billiard balls. Experiment 1 shows that participants’ counterfactual judgments about what would have happened if one of the balls had been removed, are well-explained by an approximately Newtonian model of physics. In Experiment 2, participants judged to what extent two balls were causally responsible for a third ball going through a gate or missing the gate. As predicted by the CSM, participants’ judgments increased with their belief that a ball was a whether-cause, a how-cause, as well as sufficient for bringing about the outcome.

How do people assign responsibility for the outcome of an election? In previous work, we have shown that responsibility judgments in achievement contexts are affected by the probability that a person’s contribution is necessary, and by how close it was to being pivotal (Lagnado, Gerstenberg, & Zultan, 2013). Here we focus on responsibility judgments in voting scenarios. We varied the number of people in different voting committees, their political affiliations, the number of votes required for a policy to pass, which party supports the policy, and the pattern of votes (creating 170 different situations). As expected, we found that participants’ responsibility judgments increased the closer the voter was to being pivotal. Further, judgments increased the more unexpected a vote was. Voters were assigned more responsibility when they voted against the majority in the committee, and when they voted against their party affiliation.

In a categorization task involving both labeled and unlabeled data, it has been shown that humans make use of the underlying distribution of the unlabeled examples. It has also been shown that humans are sensitive to shifts in this distribution, and will change predicted classifications based on these shifts. It is not immediately obvious what causes these shifts – what specific properties of these distributions humans are sensitive to. Assuming a parametric model of human categorization learning, we can ask which parameters or sets of parameters humans fix after exposure to labeled data and which are adjustable to fit subsequent unlabeled data. We formulate models to describe different parameter sets which humans may be sensitive to and a dataset which optimally discriminates among these models. Experimental results indicate that humans are sensitive to all parameters, with the closest model fit being an unconstrained version of semi-supervised learning using expectation maximization.

A perpetrator’s mental state – whether she had mens rea or a “guilty mind” – typically plays an important role in evaluating wrongness and assigning punishment. In two experiments, we find that this role for mental states is weaker in evaluating conventional violations relative to moral violations. We also find that this diminished role for mental states may be associated with the fact that conventional violations are wrong by virtue of having violated a (potentially arbitrary) rule, whereas moral violations are also wrong inherently

There is a growing body of research experimentally demonstrating a relationship between selective sustained attention and young children’s learning outcomes. Collectively, this work has documented that as selective sustained attention decreases children’s learning also declines. However, a precise understanding of how disrupted attention negatively impacts learning is lacking. The present experiment expands upon the existing work and explores three potential mechanisms by which inattention may impede learning: 1) inattention may disrupt encoding of the individual features of the stimulus, 2) inattention may impede children from binding the features together, or 3) inattention may disrupt both feature encoding and binding

We present a computational model performing the n-back task. This task requires a number of cognitive processes including rapid binding, updating, and retrieval of items in working memory. The model is implemented in spiking leakyintegrate- and-fire neurons with physiologically constrained parameters, and anatomically constrained organization. The methods of the Semantic Pointer Architecture (SPA) are used to construct the model. Accuracies and reaction times produced by the model are shown to match human data. Namely, characteristic decline in accuracy and response speed with increase of n is reproduced. Furthermore, the model provides evidence, contrary to some past proposals, that an active removal process of items in working memory is not necessary for an accurate performance on the n-back task

The nature of the quantities involved in arithmetic problems promotes semantic encodings that affect the strategy chosen to solve them (Gamo, Sander, & Richard, 2010). Such encoding effects might prevent positive transfer to problems sharing the same formal mathematical structure (Bassok, Wu, & Olseth, 1995). In this study with 5th and 6th graders, we investigated the conditions promoting positive and negative transfer in arithmetic problems that could be solved with two distinct strategies. We showed that basic training do not overcome the initial impact of semantic encodings, and we provided evidence that a poor semantic encoding of the training problems leads to transfer errors. This suggests the existence of ontological restrictions on the representation mechanisms involved in word arithmetic problem solving

In optimal stopping problems, people are asked to choose the maximum out of a sequence of values, under the constraint that a number can only be chosen when it is presented. We present a series of threshold models of human decision making on optimal stopping problems, including a new hierarchical model that assumes individual differences in threshold setting are controlled by deviations or biases from optimality associated with risk propensity, and is applicable to optimal stopping problems of any length. Using Bayesian graphical modeling methods, we apply the models to previous data involving 101 participants with large individual differences who completed sets of length 5 and length 10 problems. Our results demonstrate the effectiveness of the bias-from-optimal hierarchical model, find individual differences in thresholds that people use, but also find that these individual differences are stable across the two optimal stopping tasks.

In this study, the influence of two types of language on memory for object location was investigated: demonstratives (this, that) and possessives (my, your). Participants read instructions (containing this/that/my/your/the) to place objects at different locations. They then had to recall those object locations. Experiments 1 and 2 tested the contrasting predictions of two possible accounts of language on memory: the expectation model (Coventry, Griffiths, & Hamilton, 2014) and the congruence account (Bonfiglioli, Finocchiaro, Gesierich, Rositani, & Vescovi, 2009). In Experiment 3, the role of attention as a possible mechanism was investigated. The results across all three experiments show striking effects of language on object location memory; objects in the “that” and “your” condition were misremembered to be further away than objects in the “this” and “my” condition. The data favored the expectation model: expected location cued by language and actual location are concatenated leading to (mis)memory for object location.

This work investigates the intuitive and deliberative cognitive processes underlying risky decision-making by manipulating time pressure. A recent fMRI study by De Martino et al. (2006) found greater activation of the amygdala when exhibiting framing effects suggesting that they may be driven by System 1. Because this system is characterized as being fast, we expect more pronounced framing effects under time pressure. In our experiment, we manipulated time pressure and accuracy and use a dynamic dual-process model to explain our results. The model we develop is a sequential sampling model in which the drift rates and boundaries vary in accordance with the thinking modes, frames, and time pressure

Participants completed long single digit sums in two interactivity contexts. In a low interactivity condition sums were solved with hands down. In a second, high interactivity condition participants used moveable tokens. As expected accuracy and efficiency was greater in the high compared to the low interactivity condition. In addition, participants were profiled in terms of working memory capacity, numeracy, math anxiety and expertise in math. All of these measures predicted calculation errors in the low interactivity conditions; however, in the high interactivity condition, participants’ performance was not determined by any of these variables. We also developed a scale to measure task engagement: Participants were significantly more engaged with the task when they completed the sums in the high interactivity condition. However engagement level did not correlate with calculation error, suggesting improvement in performance with tokens was not the result of greater task engagement. Interactivity transformed the deployment of arithmetic skills, ameliorated performance, and helped to reduce the difference in performance between individuals of low and high math expertise.

What underlies young children’s failure in evaluating underinformative teachers? We explore the hypothesis that children have difficulty representing relevant alternatives; knowing what the teacher could have done. Children rated two teachers who demonstrated toys to a na¨ıve learner. One group first observed a fully informative teacher and then an underinformative teacher, while the other group saw the reverse order. Six- and seven-year-olds successfully rated the underinformative teacher lower than the fully-informative teacher regardless of the order (Exp.1). However, four- and five-yearolds showed this pattern only when they saw the fully informative teacher first (Exp.2). Given a binary choice after seeing both teachers, four-year-olds showed a preference for the fully informative teacher (Exp.3). We discuss these results in light of recent literature on children’s understanding of pragmatic violations in linguistic communication; the contrast between the fully informative vs. under-informative teachers might help children understand what the teacher could have shown.

This paper reports a construction of typologies of players based on their strategic reasoning in turn-taking games. Classifications have been done based on latent class analysis and according to different orders of theory of mind, and exploratory validations have been provided for the resulting classifications. Finally, interaction of the typologies described by these classifications is discussed towards achieving a common perspective of typologies of players originating from various aspects of strategic thinking.

In a series of two experiments we investigated the influence of an allocentric and egocentric perspective on landmarkbased wayfinding and finding the according return path. Participants had to learn a route consisting of twelve intersections with four different verbal landmarks at each intersection. They were asked to memorize at least one of the landmarks for providing a route description after the learning phase, either in the learning direction (initial path) or in the opposite direction (return path). In the allocentric experiment, a clear preference and higher performance was demonstrated for landmarks located at the position before the intersection and in the direction of turn, while in the egocentric perspective landmarks in the direction of turn were better remembered and used more frequently, independent of the position before or behind the intersection. These results will be discussed with respect to current research on structural salience in landmark-based wayfinding.

In this paper, we investigate how people use mental simulations: do people vary the number of simulations that they run in order to optimally balance speed and accuracy? We combined a model of noisy physical simulation with a decision making strategy called the sequential probability ratio test, or SPRT (Wald, 1947). Our model predicted that people should use more samples when it is harder to make an accurate prediction due to higher simulation uncertainty. We tested this through a task in which people had to judge whether a ball bouncing in a box would go through a hole or not. We varied the uncertainty across trials by changing the size of the holes and the margin by which the ball went through or missed the hole. Both people’s judgments and response times were well-predicted by our model, demonstrating that people have a systematic strategy to allocate resources for mental simulation

Real-world time-series data can show substantial short-term variability as well as underlying long-term trends. Verbal descriptions from a pilot study, in which participants interpreted a real-world line graph about climate change, revealed that trend interpretation might be problematic (Experiment 1). The effect of providing a graph interpretation strategy, via a linguistic warning, on the encoding of longterm trends was then tested using eye tracking (Experiment 2). The linguistic warning was found to direct visual attention to task-relevant information thus enabling more detailed internal representations of the data to be formed. Language may therefore be an effective tool to support users in making appropriate spatial inferences about data

What makes a question useful? What makes an answer appropriate? In this paper, we formulate a family of increasingly sophisticated models of question-answer behavior within the Rational Speech Act framework. We compare these models based on three different pieces of evidence: first, we demonstrate how our answerer models capture a classic effect in psycholinguistics showing that an answerer’s level of informativeness varies with the inferred questioner goal, while keeping the question constant. Second, we jointly test the questioner and answerer components of our model based on empirical evidence from a question-answer reasoning game. Third, we examine a special case of this game to further distinguish among the questioner models. We find that sophisticated pragmatic reasoning is needed to account for some of the data. People can use questions to provide cues to the answerer about their interest, and can select answers that are informative about inferred interests.

In standard decision theory, rational agents are objective, keeping their beliefs independent from their desires (Berger, 1985). Such agents are the basis for current computational models of Theory of Mind (ToM), but this fundamental as- sumption of the theory remains untested. Do people think that others’ beliefs are objective, or do they think that others’ de- sires color their beliefs? We describe a Bayesian framework for exploring this relationship and its implications. Motivated by this analysis, we conducted two experiments testing the a priori independence of beliefs and desires in people’s ToM and find that, contrary to fully-normative accounts, people think that others engage in wishful thinking. In the first ex- periment, we found that people think others believe both that desirable events are more likely to happen, and that undesir- able ones are less likely to happen. In the second experiment, we found that social learning leverages this intuitive under- standing of wishful thinking: participants learned more from the beliefs of an informant whose desires were contrary to his beliefs. People’s ToM therefore appears to be more nuanced than the current rational accounts, but consistent with a model in which desire directly affects the subjective probability of an event.

Two experiments examined theoretical accounts of why people fail to consider alternative hypotheses in judgments under uncertainty. Experiment 1 found that a majority of participants failed to spontaneously search for information about an alternative hypothesis, even when this required minimal effort. This bias was reduced when a specific alternative was mentioned before search. Experiment 2 showed that when participants were given the likelihoods of the data given a focal hypothesis p(D|H) and an alternative hypothesis p(D|¬H), they gave estimates of p(H|D) that were consistent with Bayesian principles. The results show that neglect of the alternative hypothesis typically occurs at the initial stage of problem representation. However judgments are more consistent with Bayesian norms when they involve utilizing information about a given alternative

Knowledge of color has strong individual, environmental, and cultural differences that may systematically influence performance in cognitive tasks. For example, color knowledge has been shown to influence recall of color (Persaud & Hemmer, 2014). This manifests as a systematic regression to the mean effect, where memory is biased towards the mean hue of each universal color category. What remains unclear is whether differences, such as culture and environment, might differentially influence memory. We tested recognition memory for color in the Tsimane’ of Bolivia; an indigenous population with little or no modern schooling, whose environment is very different from industrialized societies. We found that recognition regressed towards the mean of some universal color categories, but for others was systematically biased toward neighboring categories. A cluster analysis suggested that the Tsimane’ use five underlying color categories—not the standard universals. This might be shaped by education, language and the environment

Can learning new linguistic metaphors foster new nonlinguistic representations? We describe a set of studies in which we trained English-speaking participants to talk about time using vertical spatial metaphors that are novel to English. One group learned a mapping that placed earlier events above and the other a mapping that placed earlier events below. After mastering the new metaphors, participants were tested in a non-linguistic space-time implicit association task – the Orly task. This task has been used previously to document cross-linguistic differences in representations of time (Boroditsky et. al 2010; Fuhrman et al 2011). Some participants completed temporal judgments in the Orly task without any other secondary task, while others did so under either verbal or visual interference. Finally, we report data from a serendipitous sample of Chinese-English bilinguals on the same task.

Does the similarity between two items change over time? Previous studies (Goldstone & Medin, 1994; Gentner & Brem, 1999) have found suggestive results but have relied on interpreting complex interaction effects from “deadline” decision tasks in which the decision making process is not well understood (Luce, 1986). Using a self-paced simple decision task in which the similarity between two items can be isolated from strategic decision processes using computational modeling techniques (Ratcliff, 1978), we show strong evidence that the similarity between two items changes over time and shifts in systematic ways. The change in similarity from early to late processing in Experiment 1 is consistent with the theory of structural alignment (Gentner, 1983; Goldstone & Medin, 1994), and Experiment 2 demonstrates evidence for a stronger influence of thematic knowledge than taxonomic knowledge in early processing of word associations (Lin & Murphy, 2001).

We present a computational cognitive model of mind wandering, an important cognitive phenomenon whose mechanisms are involved in insight, problem-solving, and creativity. The model posits that mind wandering begins when one is not engaged in goal-oriented cognition, whether when between tasks or when in the middle of a task but not actively thinking about one’s goal. At such times, the model thinks about other, highly-activated thoughts in memory. This model sheds light on both how task-oriented and more basic cognitive processes interact, as well as how mind wandering content is generated; both unresolved questions for mind wandering research. We compare our model against data presented by McVay and Kane (2013), who induced mind wandering in a laboratory setting by embedding participants’ personal goals and concerns in a lexical SART task. Overall, our model matched the data’s mind wandering rates very well. We discuss implications and future work on the model.

Teaching with evaluative feedback involves expectations about how a learner will interpret rewards and punishments. We formalize two hypotheses of how a teacher implicitly expects a learner to interpret feedback – a reward-maximizing model based on standard reinforcement learning and an action-feedback model based on research on communicative intent – and describe a virtual animal-training task that distinguishes the two. The results of two experiments in which people gave learners feedback for isolated actions (Exp. 1) or while learning over time (Exp. 2) support the action-feedback model over the reward-maximizing model

Pragmatic implicatures—inferences that weak statements imply that stronger ones could not be used—are a popular case study of children’s pragmatic development. A growing literature suggests that children make implicatures under certain conditions, but their performance varies widely across tasks, and few datasets allow direct comparisons between implicature types. We designed a simple paradigm to address these issues. In Experiment 1, we included both ad-hoc (contextual) and scalar (quantifier) descriptions and found that 4-year-olds were at ceiling in ad-hoc trials but had difficulty with scalar implicatures. In Experiment 2, 4-year-olds’ performance increased when we included only scalar trials, but was still low. Across both datasets, performance for “some” and “none” quantifiers was positively correlated. Our work provides more precise developmental data on the emergence of different implicature computations and illustrates that preschoolers’ recognition of implicatures relates both to their comprehension of particular lexical items and also their recognition of relevant alternatives

The capacity coefficient function is a well-established, modelbased measure comparing performance with multiple sources of information together to performance on each of those information sources in isolation. Because it is a function across time, it may contain a large amount of information about a participant. In many applications, this information has been ignored, either by using qualitative assessment of the function or by using a single summary statistic. Recent work has demonstrated the efficacy of functional principal components analysis for extracting important information about the capacity function. We extend this work by applying clustering techniques to examine individual capacity differences in configural learning.

Much problem-solving research has investigated if and why ‘two heads are better than one’, but typically posits that if there is any process gain observed it is because of the exposure to the ideas provided by another person’s attempted solutions. This work fails to acknowledge or investigate what the interaction itself contributes to joint problem solving. Using an online version of the Alternative Uses Task, we compare situations in which people are passively exposed to what is said in a dialogue with situations in which people are actively engaged in the dialogue, thus varying the interactivity independently of the informational content that participants were exposed to. Interacting participants produce more turns overall, but they do not come up with more ideas. Interacting participants were also more likely to build on each other’s ideas and produce more complex ideas when a turn is linked to a previous idea; following leads to elaboration – but only if there is genuine interactivity. These results indicate that conversational mechanisms promote the exploration of a problem space and that merely counting the number of ideas produced would miss the importance of the interaction itself.

Analogical comparison has been found to promote learning across many conceptual domains. Here, we ask whether this mechanism can facilitate children’s understanding of others’ mental states. In Experiment 1, children carried out comparisons between characters’ thoughts and reality and between characters with true beliefs vs. those with false beliefs. Children given this training improved from pre- to post-test. In Experiment 2, we used a more minimal comparison technique. Children saw a series of three stories involving true or false beliefs. There were two betweensubjects conditions that either facilitated (High Alignability) or impeded (Low Alignability) comparison across stories. We found that children made more gains from pre- to post-test in the High Alignability condition than in the Low Alignability condition. We also found effects of production of mental state verbs, as assessed in an Elicitation Task. These results provide evidence for the role of analogical comparison in theory of mind development.

One way to optimize social learning is to be selective when choosing from what sources to accept information. Preschoolers prefer to learn from previously accurate or competent sources, rather than from unreliable ones (e.g., Koenig, Clément, & Harris, 2004). The current study extends this work by comparing the ability to monitor an actor’s success in two species: children and chimpanzees. Members of both species saw two actors try to open containers, with different outcomes. Then, a forced-choice response was used to determine whether participants would pair the container with the previously successful actor. While preschoolers correctly elicited help from a previously successful actor, chimpanzees did not reliably select the type of object the actor could open. The current findings suggest a difference between humans and chimpanzees’ use of past source reliability, which may reflect or result from differences in their use of social learning

We investigate how children balance the quality of informants’ knowledge with the number of endorsements when deciding which of two boxes contains the better option. When group numbers are equal, children choose boxes endorsed by informants with visual access over informants with hearsay (Experiment 1), but are at chance when group size conflicts with quality of knowledge (Experiments 2 and 3). This suggests that children tend to conform to a majority opinion, compared to adults (Experiment 4) and a normative computational model. These studies suggest that preschoolers consider the testimony of multiple informants and evaluate their knowledge sources, but may assume that informants are more individually informative than they are

The causal frame problem is an epistemological puzzle about how the mind is able to disregard seemingly irrelevant causal knowledge, and focus on those factors that promise to be useful in making an inference or coming to a decision. Taking a subject’s causal knowledge to be (implicitly) represented in terms of directed graphical models, the causal frame problem can be construed as the question of how to determine a reasonable “submodel” of one’s “full model” of the world, so as to optimize the balance between accuracy in prediction on the one hand, and computational costs on the other. We propose a framework for addressing this problem, and provide several illustrative examples based on HMMs and Bayes nets. We also show that our framework can account for some of the recent empirical phenomena associated with alternative neglect.

Knowledgeable agents always choose what they like best, thus revealing their preferences. But naïve agents only choose what they believe they like best, and may end up disliking their choice. As such, sensitivity to an agent’s prior experience is critical for interpreting their behavior. Here we show that four- and five-year-olds expect knowledgeable agents, as compared to naïve agents, to have stable choices that lead to higher rewards (Experiments 1 and 2). Additionally, we show that four- and five-year-olds can infer which of two agents is naïve given information about the rewards they obtained and the stability of their choices (Experiments 3 and 4). These results show that young children understand that beliefs and desires are interconnected and that, in addition to having uncertainty about the world, agents can also be uncertain about their own desires

The understanding that agents have goals, and the ability to infer them, is fundamental in social cognition. However, much of our social understanding goes beyond goal attribution. Drawing on both behavioral studies throughout development, and on the limitations of past models, we propose that humans have a naïve utility calculus to reason about the costs and rewards underlying agents’ goals. We show that the naïve utility calculus model, embedded in a Bayesian framework, can jointly infer the costs and rewards of agents navigating in complex scenarios. Using this model we test humans’ ability to make quantitative cost-reward inferences in scenarios with various sources of costs and rewards. Our results suggest the naïve utility calculus model fits human inferences better than simple goal inference models.

The similarity between objects is judged in a wide variety of contexts from visual search to categorization to face recognition. There is a correspondingly rich history of similarity research and many known behavioral trends and models of similarity. Nevertheless, most similarity behaviors have been identified and tested only in a comparatively narrow set of unique contexts. This leaves open the question of the extent to which similarity judgments rely on common processes or resources and the specific nature of those processes if so. We tested three diverse yet well-established measures of object similarity using identical, psychometrically controlled stimuli and identical analyses across tasks. We found several consistent behavioral effects across tasks that provide clues as to the nature of task-general similarity processes and serve as diagnostic targets for computational models of similarity

In this paper I argue that the prioritization of sensory input by top-down attention is constitutive of and essential to conscious perception. Specifically, I argue that top-down attention is required to provide informational integration at the level of the subject, which can be contrasted with integration at the level of features and objects. Since the informational content of conscious perception requires integration at the level of the subject, top-down attention is necessary for conscious perception as we know it. I present this argument through a theory, which I call the “Standard Theory.” According to this theory, top-down attention brings about subject-level integration for sensory input by prioritizing that input with respect to a subject-level standard

We examine a “mere newness bias,” a preference for novelty purely due to recentness of release. In a series of studies, we show that, for newer and older products of identical quality, people prefer newly released goods over older goods across a range of domains. This bias translates to a higher willing to pay, greater anticipated excitement, and higher likelihood of purchase for products perceived to be newer. The mere newness bias persists even for die rolls, where there cannot be any difference in quality and where there is no social benefit to newness.

Languages of the world universally encode spatial relationships between objects. However, speakers employ a variety of different language-specific expressions, which may encode culture-specific information about objects and/or different spatial concepts. We ask whether aspects of the encoding of spatial relations across languages nevertheless show common underlying spatial concepts as reflected in the distributions of spatial expressions over spatial sub-types. We examine a set of hypothesized distinctions within the spatial relational concepts of Containment and Support across three typologically distinct languages: English, Hindi, and Mandarin. We find support for two related hypotheses concerning common patterns of variation in (a) speakers' use of select "basic" spatial expressions, and (b) languages' inventory and distribution of expressions across hypothesized Containment and Support subtypes. The results underscore the presence of strong universal similarities in both the extension of basic spatial expressions across relations and in the principles governing the diversity of expressions available for encoding particular relations

According to probabilistic theories of higher cognition, beliefs come in degrees. Here, we test this idea by studying how people make predictions from uncertain beliefs. According to the degrees-of-belief theory, people should take account of both high- and low-probability beliefs when making predictions that depend on which of those beliefs are true. In contrast, according to the all-or-none theory, people only take account of the most likely belief, ignoring other potential beliefs. Experiments 1 and 2 tested these theories in explanatory reasoning, and found that people ignore all but the best explanation when making subsequent inferences. Experiment 3A extended these results to beliefs fixed only by prior probabilities, while Experiment 3B found that people can perform the probability calculations when the needed probabilities are explicitly given. Thus, people’s intuitive belief system appears to represent beliefs in a ‘digital’ (true or false) manner, rather than taking uncertainty into account

Our beliefs guide our actions. But do potential actions also guide our beliefs? Three experiments tested whether people use pragmatist principles in fixing their beliefs, by examining situations in which the evidence is indeterminate between an innocuous and a dire explanation that necessitate different actions. According to classical decision theory, a person should favor a prudent course of action in such cases, but should nonetheless be agnostic in belief between the two explanations. Contradicting this position, participants believed the dire explanation to be more probable when the evidence was ambiguous. Further, when the evidence favored either an innocuous or a dire explanation, evidence favoring the dire explanation led to stronger beliefs compared to evidence favoring the innocuous explanation. These results challenge classic theories of the relationship between belief and action, suggesting that our system for belief fixation is sensitive to the utility of potential beliefs for taking subsequent action

Our ability to induce the general from the specific is a hallmark of human cognition. Inductive reasoning tasks ask participants to determine how strongly a set of premises (e.g., Collies have sesamoid bones) imply a conclusion (Dogs have sesamoid bones). Here, we present evidence for an abductive theory of inductive reasoning, according to which inductive strength is determined by treating the conclusion as an explanation of the premises, and evaluating the quality of that explanation. Two inductive reasoning studies found two signatures of explanatory reasoning, previously observed in other studies: (1) an evidential asymmetry between positive and negative evidence, with observations casting doubt on a hypothesis given more weight than observations in support; and (2) a latent scope effect, with ignorance about potential evidence counting against a hypothesis. These results suggest that inductive reasoning relies on the same hypothesis evaluation mechanisms as explanatory reasoning

Like scientists, children must find ways to explain causal systems in the world. The Bayesian approach to cognitive development holds that children evaluate explanations by applying a normative set of statistical learning and hypothesis-testing mechanisms to the evidence they observe. Here, we argue for certain supplements to this approach. In particular, we demonstrate in two studies that children, like adults, have a robust latent scope bias that conflicts with the laws of probability. When faced with two explanations equally consistent with observed data, where one explanation made an unverified prediction, children consistently preferred the explanation that did not make this prediction (Experiment 1). The bias can be overridden by strong prior odds, indicating that children can integrate cues from multiple sources of evidence (Experiment 2). We argue that children, like adults, rely on heuristics for making explanatory judgments which often lead to normative responses, but can lead to systematic error.

Metacognitive skills are critical in learning but difficult to teach. Thus the question becomes how can we facilitate metacognitive tutoring? We present an exploratory learning environment called MILA-T with embedded metacognitive tutors imitating five functional roles of teachers in classrooms. We tested MILA–T in a controlled experiment with 237 middle school students. We examine the impact of MILA–T on the models of a natural phenomenon constructed by the students. We find that students with access to MILA–T wrote better evidential justifications for their models, and thus, deliver better-justified models for the phenomenon. We also find that these improvements persisted during a transfer task. These results lend support for organizing metacognitive tutoring around the functional roles of teachers for supporting inquiry-driven modeling

This study investigates the relationship between sentence prominence and the predictability of word-specific statistical descriptors of prosody. We extend from an earlier wordinvariant model by studying a model that marks words as prominent if the acoustic prosodic features differ from their expected values during the lexemes. To test the approach, the most common acoustic features associated with the perception of prominence are extracted and several lexeme-specific statistical measures are computed for each feature. Simulations are conducted on a corpus of continuous English speech and the algorithm output is compared to manually assigned prominence labels. The results show that the deviant prosodic descriptors of the words correlate with the perception of prominence. However, this effect is much smaller than that obtained by modeling the prosodic predictability at the utterance level, suggesting that contextindependent lexeme-specific models are unable to capture relevant aspects of sentence prominence

This study examined 4-year-old children’s understanding of the successor function, the concept that for every positive integer there is a unique next integer. Children were tested in the context of cardinal numbers and ordinal numbers. The results suggest that knowledge of the successor for cardinal numbers precedes that for ordinal numbers. In addition, for both cardinal and ordinal numbers, children generally failed to demonstrate understanding that the successor of a given number is unique

Verbal irony plays an important role in how we communicate and express opinions about the world. While there exist many theories and empirical findings about how people use and understand verbal irony, there is to our knowledge no formal model of how people incorporate shared background knowledge and linguistic information to communicate ironically. Here we argue that a computational approach previously shown to model hyperbole (Kao, Wu, Bergen, & Goodman, 2014) can also explain irony once we extend it to a richer space of affective subtext. We then describe two behavioral experiments that examine people’s interpretations of utterances in contexts that afford irony. We show that by minimally extending the hyperbole model to account for two dimensions of affect—valence and arousal—our model produces interpretations that closely match humans’. We discuss the implications of our model on informal theories of irony and its relationship to other types of nonliteral language understanding

Cognitive scientists sometimes find themselves embroiled in debates over the precise definitions of high-level concepts in their fields – COGNITION, EMOTION, SENSE, and so on. The idea behind these debates seems to be that achieving a precise definition of these concepts will be a boon to scientific inquiry. We argue that these efforts of conceptual analysis would benefit from greater appreciation of the importance of such high-level concepts in supporting association or semantic priming, as opposed to deduction. In this associative role, they provide the basis for making connections between related concepts, connections that can then be explored by empirical methods, which in turn yield more precise, but often quite novel, concepts. In combination with well-established work in cognitive psychology on the non-classical structure of natural concepts, this perspective suggests that researchers should be cautious about investing substantial time and energy in attempts to precisely define concepts like COGNITION.

A standard assumption in linguistic, psycholinguistic, and computational research on pronoun use is that production and interpretation are guided by the same set of contextual factors. Kehler et al. (2008) and Kehler & Rohde (2013) have argued instead for a Bayesian model, one in which pronoun production is insensitive to a class of semantically- and pragmaticallydriven contextual biases that have been shown to influence pronoun interpretation. Here we evaluate the model using a passage completion study that employs a subtle contextual manipulation to which traditional analyses are insensitive, specifically by varying whether or not a relative clause that modifies the direct object in the context sentence invites the inference of a cause of the event that the sentence denotes. The results support the claim that pronoun interpretation biases, but not production biases, are sensitive to this pragmatic enrichment, revealing precisely the asymmetry predicted by the Bayesian analysis. A correlation analysis further establishes that the model provides better estimates of measured pronoun interpretation biases than two competing models from the literature.

Culture is considered an evolutionary adaptation that enhances human reproductive fitness. A common explanation is that social learning, the learning mechanism underlying cultural transmission, enhances fitness by avoiding the extra costs of individual learning. This explanation was disproved by a mathematical model of individual and social learning, showing that social learners can invade a population but do not enhance its fitness. We extend this model to include a more complex environment, limited cognitive resources, and hybrid learners that combine social and individual learning. In this extended model, we show that social learning evolves and enhances population fitness via hybrid learners capable of specializing their individual learning.

Reasoners distinguish between different types of causal relations, such as causes, enabling conditions, and preventions. Psychologists disagree about the representations that give rise to the different relations, but agree that mental simulations play a central role in inferring them. We explore how causal relations are extracted from mental simulations. The theory of mental models posits that people use a kinematic simulation to infer possibilities. It predicts that causes should be easier to infer than enabling conditions, and that the time it takes to infer a causal relation should correlate with the number of moves in a mental simulation. To test these two predictions, we adapted a railway domain designed to elicit mental simulations, and we devised problems in which reasoners had to infer causal relations from simulations of the movements of cars in this domain. Two studies corroborated the model theory's predictions. We discuss the results in light of recent theories of causation and mental simulation.

Young children’s estimates of numerosity increase approximately logarithmically with actual set size. The conventional interpretation of this finding is that children’s estimates reflect an innate logarithmic encoding of number. A recent set of findings, however, suggest logarithmic number-line estimates could emerge via a dynamic encoding mechanism that is sensitive to the prior distribution of stimuli. Here we test this idea by examining trial-to-trial changes in logarithmicity of numerosity estimates. Against the dynamic encoding hypothesis, first trial estimates in both adults (Study 1) and adults and children (Study 2) were strongly logarithmic, despite there being zero previous stimuli. Additionally, although numerosity of a previous trial affected adult estimates of numerosity, the nature of this effect varied across experiments, yet always resulted in a logarithmic-to-linear shift from trial-to-trial. These results suggest that a dynamic encoding mechanism is neither necessary nor sufficient to elicit logarithmic estimates of numerosity

We investigated how the semantic properties of verbs influence the way in which language users process sentences and how well they remember the verb. In particular, our study focused on the frequency of motion repetition, that is, how many times actions generally repeat in a row. The experimental sentences contained action verbs, such as sneezing, knocking on a door, clapping, and bouncing a ball. Half of the target sentences contained verbs that refer to actions that generally repeat once or twice in a row in the real world (determined by norming), such as sneezing, coughing, and knocking on a door. The other half contained verbs referring to actions that typically repeat many times in row, such as hiccupping, clapping, and bouncing a ball. Native Korean speakers performed a sensicality judgment task where they decided whether given Korean sentences were sensical or not. We also tested how well participants remember the verbs in target sentences. The results show an effect of action repetition frequency: Participants judged sentences with low repetition frequency verbs more accurately than sentences containing high repetition frequency verbs. We propose that verbs describing multiple repetitions may place a greater processing load than verbs involving fewer repetitions

We use an auditory-visual semantic priming paradigm to investigate the effect of phonetically-cued emotional information (emotional prosody) on semantic activation of a lexical carrier. In two experiments, we show that words uttered in emotional prosody, although infrequent and atypical, do not necessarily hinder lexical access nor hamper subsequent semantic spreading, and that effects of emotional prosody on word processing crucially depend on the global context in which different types of prosody are presented. These results illustrate the complex nature of spoken word recognition and raise questions about how listeners incorporate multi-faceted information from spoken words.

Although young children often rely on salient perceptual cues, such as shape, when categorizing novel objects, children shift towards deeper relational reasoning when they compare category members or attend to functional properties. In this study, we investigated the independent and combined effects of comparison and function in children’s categorization of novel objects. Across two experiments, we found that comparing two perceptually similar category members led children to discover non-obvious relational features that supported their categorization of novel objects. Together, these findings underscore the difficulty in categorizing novel objects but demonstrate that comparison may aid in this process by rendering less obvious relational structures more salient, thus inducing a shift towards a categorical rather than perceptual response.

Benefits to lexical access are provided by acoustically-cued speaker characteristics (such as gender and age), but little work has investigated these effects in meaning-based tasks. Word recognition is affected both by a word’s base-level activation and by associative spread of activation among words, and is correlated with speed of lexical access. In a free association task and a semantic priming task, we find off-line and on-line evidence of speaker-specific relationships between words. Our results suggest the need to extend existing models of spoken word recognition to include interactions between linguistic information and social information that is cued by variation in speech.

The three-space theory of problem solving predicts that the quality of a learner’s model and the goal specificity of a task interact on knowledge acquisition: Learners having a good model should learn more with a nonspecific than a specific goal, which should not apply to learners having a poor model. This study tested this prediction using a computer based learning task on torques. Participants (N = 77 psychology students) either had to test hypotheses with a simulation of a lever system (nonspecific goal), or to produce given values for variables in this simulation (specific goal). In the good model condition but not in the poor model condition they saw the torque depicted as an area. Results revealed the predicted interaction. A nonspecific goal only resulted in better learning when a good model of torques was provided but not with a poor model. Our findings support the three-space theory. They emphasize the importance of understanding in studying problem solving and stress the need to study underlying processes.

The actions of a rational agent reveal information about its mental states. These inferred mental states, particularly the agent’s intentions, play an important role in the evaluation of moral permissibility. While previous computational models have shown that beliefs and desires can be inferred from behavior under the assumption of rational action they have critically lacked a third mental state, intentions. In this work, we develop a novel formalism for intentions and show how they can be inferred as counterfactual contrasts over influence diagrams. This model is used to quantitatively explain judgments about intention and moral permissibility in classic and novel trolley problems.

Speech perception requires ongoing perceptual category learning. Each talker speaks differently, and listeners need to learn each talker’s particular acoustic cue distributions in order to comprehend speech robustly from multiple talkers. This phonetic adaptation is a semi-supervised learning problem, because sometimes a particular cue value occurs with information that labels the talker’s intended category for the listener, but other times no such labels are available. Previous work has shown that adaptation can occur in both purely supervised (all labeled) and purely unsupervised (all unlabeled) settings, but the interaction between them has not been investigated. We compare unsupervised with (semi-) supervised phonetic adaptation and find, surprisingly, that adult listeners do not take advantage of labeling information to adapt more quickly or effectively, even though the labels affect their categorization. This suggests that, like language acquisition, phonetic adaptation in adults is dominated by unsupervised, distributional learning.

Verbs may refer to the means (I bumped into the lamp) or outcome (I broke the lamp) of an action (cf. Rappaport Hovav & Levin, 2010; Talmy, 1985). Do young children expect language to encode this distinction? Children’s imitation patterns suggest that they analyze nonlinguistic events in these terms. When a head-touch is the simplest action available, toddlers include just the outcome, not the means, in their own imitation (Gergely, Bekkering, & Király, 2002). We ask whether syntax influences this inference. An experimenter with her hands occupied made a toy activate with a headtouch, using either Means-focused (I’m daxing to my toy) or Outcome-focused language (I’m daxing my toy). Toddlers then imitated the action. Means- but not Outcome-focus language encouraged children to include the distinctive headtouch, overriding the standard ‘rational imitation’ effect. This suggests that toddlers’ knowledge of argument structure includes an understanding of a means/outcome divide in verb meaning.

The effect of face inversion has been heavily studied, whereas fewer studies have investigated inversion in scenes. We investigated the influence of scene inversion on decisions and contextual guidance of eye movements during visual search. A saccade contingent display termination paradigm was used to assess the temporal dynamics of the effect. Observers searched for a computer mouse in office scenes and performed a yes/no detection task. Observers’ sensitivity (d’) was lower for inverted images relative to upright. Observers’ false positive rate decreased with additional eye movements when they viewed upright images, but remained constant during the first three eye movements when viewing inverted images. The average distance of observers’ eye movements to the target location was greater for inverted than upright scenes. We interpret that inverting an image disrupts the rapid extraction of scene gist, subsequently disrupting guidance in eye movement behavior and slowing the process of rejecting false positives.

Cognitive Task Analysis (CTA) interview technique is commonly used to elicit knowledge of subject-matter experts and to design instruction better focused on what experts don’t know they know. However, the knowledge of how to conduct an effective interview is, itself, largely implicit. In this study we performed protocol analysis on a set of interview transcripts from an expert CTA practitioner to elicit the cognitive processes of conducting CTA interviews. We also consulted expert CTA practitioners to identify the strategies that they used during the interviews. We present key strategies that were employed by the expert CTA practitioners to ensure comprehensiveness and accuracy in the information collected, such as looking for perceptual cues (e.g. considering verbs such as “determine”) to ascertain adequacy of SME’s responses and selection of follow-up questions. We present a production rule model as a detailed description of the cognitive processes underlying expert CTA interviewing

We present an indoor guidance study to explore the interplay between spoken instructions and listeners’ eye movements. The study involves a remote speaker to verbally guide a listener and together they solved nine tasks. We collected a multi-modal dataset consisting of the videos from the listeners’ perspective, their gaze data, and instructors’ utterances. We analyse the changes in instructions and listener gaze when the speaker can see 1) only the video, 2) the video and the gaze cursor, or 3) the video and manipulated gaze cursor. Our results show that listener visual behaviour mainly depends on utterance presence but also varies significantly before and after instructions. Additionally, more negative feedback occurred in 2). While piloting a new experimental setup, our results provide indication for gaze reflecting both: a symptom of language comprehension and a signal that listeners employ when it appears useful and which therefore adapts to our manipulation.

Up until now, several studies have shown that a speech interface system giving verbal suggestions with beeping sounds that decrease in pitch conveyed a low system confidence level to users intuitively, and these beeping sounds were named “artificial subtle expressions” (ASEs). However, all participants in these studies were only Japanese, so if the participants’ mother tongue has different sensitivity to variations in pitch compared with Japanese, the interpretations of the ASEs might be different. We then investigated whether the ASEs are interpreted in the same way as with Japanese regardless of the users’ mother tongues; specifically we focused on three language categories in traditional phonological typology. We conducted a web-based experiment to investigate whether the ways speakers of German, Portuguese (stress accent language), Mandarin Chinese (tone language) and Japanese (pitch accent language) interpret the ASEs are different or not. The results of this experiment showed that the ways of interpreting did not differ, so this suggests that these ways are language-independent

For cases in which precise information is practically or actually unknowable, certainty and precision can indicate a lack of competence, while expressions of ignorance may indicate greater expertise. In two experiments, we investigated whether children and adults are able to use this “virtuous ignorance” as a cue to expertise. Experiment 1 found that adults and children older than 9 years selected confident informants for knowable information and ignorant informants for unknowable information. However, 5-7-yearolds overwhelmingly favored a confident informant, even when such precision was completely implausible. In Experiment 2, we demonstrated that 5-8-year-olds and adults are both able to distinguish between knowable and unknowable items when asked how difficult the information would be to acquire, but those same children still failed to reject the precise and confident informant for unknowable items. We suggest that children have difficulty integrating information about the knowability of particular facts into their evaluations of expertise

We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.

One frequent piece of advice is not to “put all our eggs in one basket” and opt for multiple alternatives in order to minimize risk and uncertainty in our decisions. In a behavioral study involving decisions-from-experience, Ashby, Konstantinidis, and Gonzalez (2015) showed that participants follow an “irrational” strategy in choice selection which departs from maximization. As structural complexity (number of available options) increased, participants diversified their choices more, proportional to rank ordering options based on their expected value. The current work explores the underlying cognitive mechanisms through a reinforcement-learning model and shows that people’s choices can be explained by a singular strategy (diversification in choice), which originates from similar cognitive processes regardless of structural complexity

People’s memory for an event is known to be affected by their verbal descriptions prior to memory assessment. The present experiment investigated whether the computational difficulty of production itself, which is known to affect what people say, can shape descriptions and subsequent event memory. Participants viewed simple scenes and were asked to describe them using either speech or silent gesture, the latter being a much more difficult task. We hypothesized that gesturing participants would over-use action pantomimes, which would yield poorer Inaction (vs. Action) scene memory. Following scene descriptions, participants were given a forced-choice recognition task to discriminate previously presented scenes from foils. Patterns of gesturing showed that gesturers used action pantomimes for Inaction scenes, and they performed reliably worse on Inaction scene memory. Increased production of action pantomimes predicted increased guesses for Action scenes at test, independent of the correct response. Implications for memory and production are discussed

Three experiments explore differences between blocked and interleaved study with and without item repetition. In the first experiment we find that when items are repeated during study, blocked study results in higher test performance than interleaved study. In the second experiment we find that when there is no item repetition, interleaved and blocked study result in equivalent performance during the test phase. In the third experiment we find that when the study is passive and includes no item repetition, interleaved study results in higher test performance. We propose that learners create associations between items of the same category during blocked study and item repetition strengthens these associations. Interleaved study leads to weaker associations between items of the same category and therefore results in worse performance during test when there are item repetitions

The paper briefly reviews the contribution of recent neuroscience findings to our understanding of our human nature – more exactly, to the understanding of the three properties that we conceive of as highly-specifically human: consciousness, freedom, and language. The analysis yields rather surprising results. Self-consciousness is possibly not the highpoint of our sophisticated cognitive functions, but rather the basic pre-reflective self-other distinction intimately related to body control and affective states, within whose limits cognitive processes become possible. Freedom is not a violation of natural (biological) laws, but, in contrast, a necessary attribute of complex behavior; it roots in the fundamental biomechanical freedom of biological movements. Language comprehension is neither an instinct nor a set of complex inferences, but a behavior based on learnt hierarchy of predictive, anticipatory processes. Thus the answer to the question formulated in the title is positive: yes, it can change. From the author’s viewpoint, these changes emphasize embodied, enacted nature of the specifically human functions.

Despite its importance, human collective intelligence remains enigmatic. We know what features are predictive of collective intelligence in human groups, but we do not understand the specific mechanisms that lead to the emergence of this distributed information processing ability. In contrast, there is a well-developed literature of experiments that have exposed the mechanisms of collective intelligence in nonhuman animal species. We adapt a recent experiment designed to study collective sensing in groups of fish in order to better understand the mechanisms that may underly the emergence of collective intelligence in human groups. We find that humans in our experiments act at a high level like fish but with two additional behaviors: independent exploration and targeted copying. These distinctively human activities may partially explain the emergence of collective sensing in our task environment at group sizes and on times scales orders of magnitudes smaller than were observed in fish.

Most theories of pragmatics and language processing predict that speakers avoid informationally redundant utterances. From a processing standpoint, it remains unclear what happens when listeners encounter such utterances, and how they interpret them. We argue that uninformative utterances can trigger pragmatic inferences, which increase utterance utility in line with listener expectations. In this study, we look at utterances that refer to stereotyped event sequences describing common activities (scripts). Literature on processing of event sequences shows that people automatically infer component actions, once a script is ‘invoked.’ We demonstrate that when comprehenders encounter utterances describing events that can be easily inferred from prior context, they interpret them as signifying that the event conveys new, unstated information. We also suggest that formal models of language comprehension would have difficulty in accurately estimating the predictability or potential processing cost incurred by such utterances.

Research on analogical problem solving has found that people often fail to spontaneously notice the relevance of a source analog when solving a target problem, although they are able to form mappings and derive inferences when given a hint to recall the source. To investigate the determinants of spontaneous analogical transfer, the present study systematically compared the effect of augmenting verbal descriptions of the source with animations or static diagrams. Solution rates to Duncker’s radiation problem were measured across varying source presentation conditions, and participants’ understanding of the relevant source material was assessed. Supplemental animations increased both comprehension of the source analog and spontaneous transfer to the radiation problem. Supplemental diagrams yielded lesser improvement in participants’ understanding of source material and did not increase solution rates to the target problem. To investigate individual differences in spontaneous transfer, fluid intelligence was measured for each participant using an abridged version of the Raven’s Progressive Matrices (RPM) test. Animated source depictions were most beneficial in facilitating spontaneous transfer for those participants with low scores on the fluid intelligence measure.

We revisit a transfer-of-training study and analyze its data using a cognitive modeling approach. Fitting a diffusion model to participant behavior over sessions al- lows conclusions as to the underlying causes of behav- ioral changes—be they changes in cognitive strategies, adaptation to the paradigm, increasing familiarity with the stimuli, or speed of information processing. Our dif- fusion model analysis revealed that participants simul- taneously adapt speed-accuracy trade-off, increase their non-decisional response speed, and increase their speed of information processing. All three of these adaptations transferred to a similar, non-trained outcome task.

Most research on adaptive decision-making takes a strategyfirst approach, proposing a method of solving a problem and then examining whether it can be implemented in the brain and in what environments it succeeds. We present a method for studying strategy development based on computational evolution that takes the opposite approach, allowing strategies to develop in response to the decision-making environment via Darwinian evolution. We apply this approach to a dynamic decision-making problem where artificial agents make decisions about the source of incoming information. In doing so, we show that the complexity of the brains and strategies of evolved agents are a function of the environment in which they develop. More difficult environments lead to larger brains and more information use, resulting in strategies resembling a sequential sampling approach. Less difficult environments drive evolution toward smaller brains and less information use, resulting in simpler heuristic-like strategies.

We examine the fan effect in overlapping data sets and logical inference. Three experiments are presented and modeled using the ACT-R cognitive architecture. The results raise issues over the scope of the memories that determine the fan effect and the use of search strategies to retrieve from memory

A growing body of research investigates how humans learn complex hierarchical structures with center-embedded recursion (Bahlmann, Schubotz, & Friderici, 2008; Poletiek & Lai, 2012). Increasing evidence indicates that properties of the learning input have an impact on learning this type of recursion. For instance, recent studies found that staged input, fewer unique exemplars and unequal repetition facilitate learning (e.g. Lai, Krahmer, & Sprenger, 2014; Lai & Poletiek, 2011, 2013). Most of these studies investigated learning center-embedded recursion through visual input, whereas few studies examined the processing of auditory input. In the current study, we test: 1) whether participants are able to learn center-embedded recursive structure from exclusively auditory input; 2) whether the facilitative cues (ordering and frequency distribution) are attuned to the auditory modality. Our results successfully demonstrate the learning of auditory sequences with center-embedded recursion, and replicated the effect with visual input in the previous study (Lai et al., 2014).

The latest generation of neural networks has made major performance advances in object categorization from raw images. In particular, deep convolutional neural networks currently outperform alternative approaches on standard benchmarks by wide margins and achieve human-like accuracy on some tasks. These engineering successes present an opportunity to explore long-standing questions about the nature of human concepts by putting psychological theories to test at an unprecedented scale. This paper evaluates deep convolutional networks trained for classification on their ability to predict category typicality – a variable of paramount importance in the psychology of concepts – from the raw pixels of naturalistic images of objects. We find that these models have substantial predictive power, unlike simpler features computed from the same massive dataset, showing how typicality might emerge as a byproduct of a complex model trained to maximize classification performance

When learning from others, it is important to take a critical stance—evaluating both the informants themselves as well as the content of their claims. In addition to accuracy, one can evaluate claims based on quality. The current study investigates developmental change in learners’ evaluations of evidence that varies in quality—inductive strength based on typicality or diversity. We found that while younger children track which informant provides which examples, they do not have clear preferences for the informant who provides stronger examples. Older children, on the other hand, are in the middle of a developmental transition. They rate informants who provide inductively strong examples as more trustworthy, but only reliably choose the informant who provides diverse examples

In the linguistic domain, conceptual metaphors have been shown to structure grammar, the lexicon, and abstract reasoning. Much recent research on conceptual metaphor comes from corpus examination, which is increasingly focused on developing quantificational tools to reveal cooccurrence patterns indicative of source and target domain associations. Some mappings between source and target are transparent. However, other metaphors, especially those that structure abstract processes, are more complex because the target domain is lexically divorced from the source. This study introduces new techniques directed at the quantitative evaluation of metaphorical salience when target and source relationships are nonobvious. Constellations of sourcedomain triggers are identified in the data and shown to disproportionately emerge in topic specific discourse. This measurement can be taken as one indicator of conceptual salience among the target speech community.

Previous work has shown that adults in the United States selectively use fractions and decimals to model discrete and continuous entities, respectively. However, it is unclear whether this apparent semantic alignment between the format of rational numbers and quantitative ontology is specific to the American education system, the English language, or measuring conventions (primarily imperial measures). Here we test whether similar alignments hold for Korean college students who differ from American students in educational background, language, and measurement conventions. Across three experiments, we found that the alignments found in the United States were generally replicated in South Korea. Relative to Americans, Korean students showed an overall bias towards using continuous representations, perhaps related to their exclusive use of the metric measurement system and to differences in instructional practice identified in a textbook analysis.

We consider the role that memory and knowledge play in the accuracy of people’s generation of top-10 lists. We report data from an experiment in which people answered questions like “list the top 10 most watched TV shows in the US”, with and without the help of a memory aid that provided the true top 50 items. Our analyses examine the changes in accuracy resulting from the availability of the memory aid, the patterns with which people modify their lists when the aid is provided, and the stability of individual differences in the memory and decision-making processes involved. We find clear evidence that, for those involving large number of potentially relevant items, memory retrieval plays a central role in determining the accuracy of the list. We discuss implications of these findings for the development of models for aggregating rank orders produced by people when not given the relevant items.

The Monty Hall Dilemma (MHD) is a classic brain teaser that even mathematicians appear to consistently answer incorrectly, and when the correct solution is presented people remain unconvinced. We examined how convincing were three solution types: a simple statement of the solution, a guided diagram solution, or simulated trials. Participants were given the MHD, followed by one of the three types of solutions, then we measured their level of conviction and their numeracy, Cognitive Reflection (CR), Need for Cognition (NFC), and Openness. Overall, both guided diagrams and simulated trials led to higher conviction compared to a simple solution statement. Higher numeracy and higher CR were associated with lower conviction after the simple solution; furthermore higher numeracy tended to help more in the simulation condition, whereas higher CR helped more in the guided condition. Therefore the persuasiveness of a solution depended both on its nature and characteristics of individual reasoners.

The overlap of numerical and non-numerical properties in concrete object arrays raises the question of how these input dimensions interact. Two studies were conducted to address this question and showed that changing the object identity (while retaining the numerosity) and changing the numerosity (while retaining the object identity) both resulted in attenuated recognition of object arrays. However, this interference differed across development. In adults interference was asymmetrical (i.e. changing the object identity has greater effect on memory for numerosity than changed numerosity had on object identity). In contrast, children showed a symmetrical pattern of interference. These results imply that for adults, processing numerosity might be an attention-demanding process compared to a spontaneous object perception. Children, however, processed neither the object identity nor numerosity independently.

Gestures related to time can reveal implicit representations of the TIME is SPACE metaphor (Núñez & Sweetser, 2006). While past research has shown that gestures illustrate the direction of future and past on timelines, no detailed analysis of timelines has been possible. Using the Kinect depth camera and body tracking technology, we tracked participants’ co-speech gestures while explaining time-related concepts. We present data collected with novel, relatively unsupervised Kinect-based methods that offer evidence similar to traditional gesture-coding methods and could provide the opportunity for novel theoretical findings.

Perseverance, above and beyond IQ, predicts academic outcomes in school age children. Yet, little is known about how very young children learn the contexts in which persistence is valuable (or not). Here, we explore how young children and infants learn about the rational deployment of effort through observing adults’ persistent behavior. Results from Experiments 1 and 2 indicate that preschoolers persist more after watching an adult persist, but only if the adult is successful at reaching their goal. Experiment 3 extends these findings, showing that even infants use adult models to modulate their persistence, and can generalize this inference to novel situations. Thus, both preschoolers and infants are sensitive to adult persistence and use it to calibrate their own tenacity.

Limitations in visual working memory (VWM) have been extensively studied in psychophysical tasks, but not well understood in terms of how memory limits translate to performance in more natural domains. For example, in reaching to grasp an object based on a spatial memory representation, overshooting the intended target may be more costly than undershooting, such as when reaching for a cup of hot coffee. The current body of literature lacks a detailed account of how the physical consequences and costs of memory error influence what we encode in visual memory, and how we act on the basis of remembered information. Here, we study whether externallyimposed monetary costs influence behavior in a task that involves motor planning based on information recalled from VWM. Our results indicate that subjects accounted for the uncertainty in their visual memory, showing a significant difference in their motor planning when monetary costs were imposed for memory errors. However, our findings indicate that subjects’ memory representations per se were not biased by the imposed costs, but rather subjects adopted a near-optimal post-mnemonic decision strategy

Epistemologists have argued that there are three basic sources of belief: perception, testimony and inference. These three belief sources correspond directly to the way in which many languages mark statements morphologically for sources of evidence for the statements (evidentiality). In this paper, we connect generalizations from the fields of epistemology and evidentiality. We also introduce a new method for investigating how reliable people find different types of evidence to be. A study based on this method indicates that speakers of English rank different sources of evidence according to the same criteria that govern the use of grammaticalized evidential marking

Previous research has suggested the presence of a cognitive goal bias, which favors the linguistic and non-linguistic (mnemonic) encoding of goals of motion over sources. The present corpus-based study tests the limits of the goal bias by comparing the path-encoding tendencies of English come and go in young children’s and adults’ naturalistic speech. Both verbs can occur with source and goal adjuncts; however, they differ in their presuppositional structure, such that come presupposes a goal while go presupposes a source. This difference in presupposition might lead adult speakers to inhibit goal-encoding for come via the Gricean maxim of quantity. As input, this might lead young children to acquire different path structures for go and come, even before they have mastered conversational-pragmatic abilities. Descriptive statistics replicate earlier findings of a general goal bias for both verbs. However, the results of more detailed regression analyses suggest that go exhibits a stronger goal bias than come for children and adults. Moreover, children from ages 2- 3 persist in inhibiting goal mentioning for come at rates similar to adult usage. This effect holds even while goal expressions for go are becoming more complex. These findings suggest that statistical patterns in the input can override non-linguistic biases, even during early lexical acquisition.

David Gil has argued that Riau Indonesian (Sumatra. Indonesia) has no syntax, or at least not much. This controversial analysis undermines all current models of grammar, especially those describing acquisition and on-line processing. To test the strength of this analysis, we computed the information gain holding between unigram and bigram models of regular and randomized samples of English and Riau Indonesian. English samples were included as a relatively syntax-heavy baseline. We then correlated information gain values with language (English vs. Riau Indonesian), text type (original vs. randomized), and their interaction within a linear mixed-effects regression. The results suggest (a) that English and Riau Indonesian have the same amount of bigram informativity and (b) that randomization eliminates this effect in both languages. These findings do not support Gil’s syntax-free analysis; rather, they point to some kind of productive constraints on Riau Indonesian word order

Previous research on morphological processing suggests that the probability distribution of a word across its inflected variants influences the recognition of that word. Recently, similar effects have been reported for relations between prepositions and definite-noun-phrase heads in English trigrams (e.g., in the bucket). In the present study, we test whether both effects could be accounted for in terms of string proximity and/or semantic similarity alone, or whether the findings for English trigrams should be attributed to syntactic paradigm effects. We ‘fake’ a case system for English using syntactic positions and prepositions as proxies for the relational meanings expressed inflectionally in other languages. Based on these syntactic factors, we define a syntactic inflectional entropy to parallel the morphological entropy measures used in prior studies. We found that this new measure correlates negatively with visual lexical decision RTs. However, unlike prior studies, we did not find a semantic priming effect between nouns with similar distributions in our paradigmatic vectors. This finding suggests that abstract constructional distributions facilitate lexical access while obscuring semantic relations between similarly distributed words

People can systematically match information from different senses, and these matches are known as crossmodal correspondences. Most work on these correspondences has explored how they might arise through neural mechanisms, statistical covariance in the environment, or semantic associations (e.g., Spence, 2011). Recently, Palmer, Schloss, Xu, & Prado-León (2013) demonstrated that at least some color-music correspondences can be explained by emotional mediation. The present study investigates the emotion mediation hypothesis for correspondences between odor and music, testing whether the strength of odor-music matches for particular odors and musical selections can be predicted by the similarity of the emotional associations with the odors and music. We found that perceived matches were higher when the emotional responses were similar and that a model including emotional dimensions captured a significant amount of the variance of match scores. These results provide new evidence that crossmodal correspondences are mediated by emotions.

What hierarchical structures do people use to encode visual displays? We examined visual working memory’s priors for locations by asking participants to recall the locations of objects in an iterated learning task. We designed a nonparametric clustering algorithm that infers the clustering structure of objects and encodes individual items within this structure. Over many iterations, participants recalled objects with more similar displacement errors, especially for objects our clustering algorithm grouped together, suggesting that subjects grouped objects in memory. Additionally, participants increasingly remembered objects as lines with similar orientations and lengths, consistent with the Gestalt grouping principles of continuity and similarity. Furthermore, the increasing tendency of participants to remember objects as components of hierarchically organized lines rather than individual objects or clusters suggests that these priors aid the perception of higher-level structures from ensemble statistics

Although natural languages are generally arbitrary in their mapping of forms to meanings, there are some detectable biases in these mappings. For example, longer words tend to refer to meanings that are more conceptually complex (what we refer to as a complexity bias; Lewis, Sugarman, & Frank, 2014). The origins of this bias remain an open question, however. One hypothesis is that this lexical regularity is the product of a complexity bias in individual speakers, and that it emerges in the lexicon over the course of language evolution. In the present work, we use an iterated learning paradigm to explore this proposal. Speakers learned labels of varying lengths for objects of varying complexity, and then were asked to recall the learned labels. We then presented the labels that participants produced to a new set of speakers, iterating this procedure across generations. The results suggest the presence of a complexity bias that guides language change but that interacts with pressures for simplicity

Recent research has shown that expertise in English and music reading both rely more on left hemisphere (LH) processing whereas Chinese character processing is more bilateral. Accordingly, music-reading expertise may influence hemispheric lateralization in English word processing more than in Chinese processing due to stronger competition in LH processing. Here we recruited musicians and non-musicians in a divided visual field study of English word and Chinese character naming. In English word processing, whereas non-musicians showed a typical right visual field (RVF)/LH advantage, musicians showed a left visual field (LVF)/right hemisphere (RH) advantage and responded significantly faster than nonmusicians in both the LVF and the center position. This effect may be due to competition for LH processing between music and English reading expertise, making musicians’ English word processing more right-lateralized. In contrast, in Chinese character naming, both musicians and non-musicians showed a similar bilateral pattern. This result suggests that music reading experience may have differential influences on the processing of different languages, depending on their similarities in the cognitive processes involved

Non-adjacent dependencies (NADs) refer to dependencies between items that are not adjacent in a sequence. Peña et al. (2002) discovered adult participants could learn the NADs of syllables in an artificial language when there were 25ms pauses before and after the NADs. Studies using videos of human body movements showed similar learning outcomes (Endress & Wood, 2011). However, participants failed to learn the NADs with respect to non-linguistic acoustic stimuli, such as tones or noises (Gebhart, Newport, & Aslin, 2009). Four experiments in this study examined the constraints on learning the NADs of visual stimuli. We propose that acquisition of the NADs requires the sequences be packed into a coherent unit, and the motor system provides the require packaging for stimuli that can be mapped onto motor representation. Implications on the acquisition of syllable NADs are discussed.

Containers are ubiquitous in daily life. By container, we consider any physical object that can contain other objects, such as bowls, bottles, baskets, trash cans, refrigerators, etc. In this paper, we are interested in following questions: What is a container? Will an object contain another object? How many objects will a container hold? We study those problems by evaluating human cognition of containers and containing relations with physical simulation. In the experiments, we analyze human judgments with respect to results of physical simulation under different scenarios. We conclude that the physical simulation is a good approximation to the human cognition of container and containing relations.

The human mind appears to be equipped with a toolbox full of cognitive strategies, but how do people decide when to use which strategy? We leverage rational metareasoning to derive a rational solution to this problem and apply it to decision making under uncertainty. The resulting theory reconciles the two poles of the debate about human rationality by proposing that people gradually learn to make rational use of fallible heuristics. We evaluate this theory against empirical data and existing accounts of strategy selection (i.e. SSL and RELACS). Our results suggest that while SSL and RELACS can explain people’s ability to adapt to homogeneous environments in which all decision problems are of the same type, rational metareasoning can additionally explain people’s ability to adapt to heterogeneous environments and flexibly switch strategies from one decision to the next.

Adults and children rely heavily on other people’s testimony. However, domains of knowledge where there is no consensus on the truth are likely to result in conflicting testimonies. Previous research has demonstrated that in these cases, learners look towards the majority opinion to make decisions. However, it remains unclear how learners evaluate social information, given that considering either the overall valence, or the number of testimonies, or both may lead to different conclusions. We therefore formalized several social learning strategies and compared them to the performance of adults and children. We find that children use different strategies than adults. This suggests that the development of social learning may involve the acquisition of cognitive strategies.

At the core of every decision-making task are two simple features; outcome values and probabilities. Over the past few decades, many models have developed from von Neumann’ and Morgenstern’s (1945) Expected Utility Theory to provide a thorough account of people’s subjective value and probability weighting functions. In particular, one such model that has been largely successful in both Psychology and Economics is Cumulative Prospect Theory (CPT; Tversky & Kahneman, 1992). While these models do fit people’s choice behavior well, few models have attempted to provide a psychological account for subjective value, probability weighting, and resulting choice behavior. In this paper, we focus on a memory confusion process as described in Hawkins et al.’s (2014) exemplar-based model for decisions from experience, the Exemplar Confusion (ExCon) model, and adapt it to account for biased probability estimates in decisions from description. Using Bayesian model selection techniques, we demonstrate that it is able to account for real choice data from a Rieskamp (2008) study using gains, losses, and mixed description-based gambles, and performs at least as well as CPT.

The discourse of laymen and professionals reveals the dependence of cognition on the interaction between participants, and the limitations of studying expertise by examining isolated individual behavior. This paper examines distributed cognition in the management of Multiple Sclerosis (MS). By varying the level of patient experience with the management of MS, we demonstrate the dependence of physician cognition on the patient’s contribution in four doctor-patient interactions. Experienced patients actively constructed clinical representations and presented initial evaluations for the doctor to refine and validate. Conversations between newly diagnosed patients and doctors demonstrated the physician work to establish a common understanding of the problem and acceptable interventions. Our analysis focuses on the complementary participant roles, and challenges the notion that medical cognition equals physician cognition

How does modality affect our ability to create a new communication system? This paper describes two experiments that address this question, and extend prior related findings by drawing from a significantly more extensive list of concepts (over 1000) than has been used previously. In Experiment 1, participants communicated concepts to a partner using either gestures or non-linguistic vocalizations (sounds that are not words). Experiment 1 confirmed that participants who gesture 1) produce more strongly ‘motivated’ signs that physically resemble the concepts they represent (i.e., are iconic), 2) are better able to correctly guess the meaning of a partner’s signs, and 3) show stronger alignment on a shared inventory of signs. Experiment 2 addressed a limitation of Experiment 1 (concurrent feedback only in the gesture condition). In Experiment 2 concurrent feedback was eliminated from the gesture and vocal conditions. Gesture again outperformed vocalization on communication effectiveness and sign alignment.

Different linguistic modalities (speech or sign) offer different levels at which signals can iconically represent the world. One hypothesis argues that this iconicity has an effect on how linguistic structure emerges. However, exactly how and why these effects might come about is in need of empirical investigation. In this contribution, we present a signal creation experiment in which both the signalling space and the meaning space are manipulated so that different levels and types of iconicity are available between the signals and meanings. Signals are produced using an infrared sensor that detects the hand position of participants to generate auditory feedback. We find evidence that iconicity may be maladaptive for the discrimination of created signals. Further, we implemented Hidden Markov Models to characterise the structure within signals, which was also used to inform a metric for iconicity

Holistic processing (HP) is an expertise marker in visual perception; nevertheless, it can be modulated by writing experience (Tso, Au, & Hsiao, 2014). We have recently found that HP also indicates expertise in Cantonese speech perception (Liu & Hsiao, 2014). Nevertheless, Cantonese has a logographic writing system where one syllable corresponds to one character, whereas in alphabetic languages, each syllable can be decomposed into phonemes that correspond to letters. This distinction between logographic and alphabetic languages may also modulate HP effects in speech perception. Here we tested HP effects through the composite paradigm with Korean syllables. In contrast to Cantonese speech perception, native Korean speakers were less holistic than novices in Korean syllable perception. Thus, experience with an alphabetic language may promote analytic processing of its spoken syllables. Similar to visual perception, our results suggest that HP as an expertise marker in speech perception depends on the listeners’ learning experience.

Symbolic and non-symbolic number representations are thought to share common neural substrates. However, recent studies have shown that the two numerical systems are more distinct than previously thought. These disparate findings may be explained by the use of sequential presentations of symbolic and non-symbolic quantities, the use of magnitude-reliant tasks, or the use of limited number ranges. We investigated whether adults integrate symbolic and non-symbolic numerical information during a non-magnitude-based task in which symbolic and non-symbolic double-digit numerical information is shown simultaneously. Participants viewed images in which symbolic numerals or letter pairs were superimposed on non-symbolic numerical stimuli and were asked to determine whether the text was a numeral or letter, ignoring the dots. After perceptual biases were taken into account, participants were more accurate and faster in their judgments when symbolic and non-symbolic information matched than when information mismatched, suggesting that adults can integrate symbolic and non-symbolic numerical information

Pezzulo (2008, 2011) and Grush (2004) contend for embodied cognitive science but interpret representations of motor behaviors as grounded on predictive internal models; those representations are referring-based. By contrast, the present paper contends that the motor control is ground on both referring-based representation (as manifest in forward models) and non-referring-based representation (as manifest in inverse models); the latter is pragmatic representation with the following six characteristics: perspective, changing perspectives, normativity of the goal, planning, coordination, and motor learning by refining inverse models. This is an internal version of the action-oriented representation.

Relational recognition is the process by which relational representations get recognized (i.e., representations that specify an actor and a patient, and are role sensitive). This process is currently poorly understood, but is an important aspect of relational cognition (Livins & Doumas, 2014). This paper presents two experiments that investigate the degree to which visuospatial factors influence it. The first is an exploratory eye-tracking study that shows that first fixations are correlated with what object gets bound to the actor role, while the second uses priming to show that such fixations can alter which relation is recognized.

Creating a plausible Unified Theory of Cognition (UTC) requires considerable effort from large, potentially distributed, teams. Computational Cognitive Architectures (CCAs) provide researchers with a concrete medium for connecting different cognitive theories to facilitate development of a robust, unambiguous UTC. However, due to wide dissemination of research effort, and broad scope of cognition as a psychological science, keeping track of CCA contributions is difficult. We compare the structuring of long-term memory (LTM) in two CCAs: ACT-R and CHREST. LTM structuring is considered in particular since it is an essential component of CCAs and underpins most of their operations. We aim to consolidate knowledge regarding LTM structuring for these CCA’s and identify similarities and differences between their approaches. We find that, whilst the architectures are similar in a number of ways, providing consensus for some concepts to be included in a UTC, their differences highlight important questions and development opportunities

Previous research has shown that perceptual relations, social affiliations, and geographical locations can be predicted using distributional semantics. We investigated whether this extends to chronological relations. In several computational studies we demonstrated that the chronological order of days, months, years, and the chronological sequence of historical figures can be predicted using language statistics. In fact, both the leaders of the Soviet Union and the presidents of the United States can be ordered chronologically based on the cooccurrences of their names in language. An experiment also showed that the bigram frequency of US president names predicted the response time of participants in their evaluation of the chronology of these presidents. These findings are explained by the Symbol Interdependency Hypothesis which predicts that as a function of language use, language encodes relations in the world around us. Language users can then use language as a cognitive short-cut for mental representations

The traditional clustering method utilized to partition neuronal firing patterns, including K-means and FCM algorithm, require specification of clusters numbers as priori knowledge. A new approach to analyze groups of firing patterns of neuronal spike trains based on community structure partitioning analysis and modularity function Q is examined in this study. This approach is able to automatically identify the optimal number of groups in neuronal firing patterns, realizing the true unsupervised analysis, and identify groups of neurons with similar firing patterns. The method was tested on a surrogate data set and a testing data set with firing patterns known in advance. The method was also applied to multi-electrode recording spike trains with previously unknown patterns. Results indicate this method can effectively self-determine number of pattern groups and locate firing patterns of neuronal populations based on community modularity Q.

Utility based models are common in both the risky and intertemporal choice literatures. Recently there have been efforts to formulate models of choices which involve both risks and time delays. An important question then is whether the concept of utility is the same for risky and inter-temporal choices. We address this question by fitting versions of two popular utility based models, Cumulative Prospect Theory for risky choice, and Hyperbolic Discounting for inter-temporal choice, to data from three experiments which involved both choice types. The models were fit assuming either the same concept of utility for both, by way of a common value function, or different utilities with separate value functions. Our results show that while many participants seem to require the flexibility of different value functions, an approximately equal number do not suggesting they may have a single concept of utility. Furthermore for both choice types value functions were concave

Response time (RT) is an oft-used but ”noisy” behavioral measure in psychology. Here, we combine modeling and psychophysics to examine the hypothesis that RT variability may reflect ongoing statistical learning and consequent adjustment of behavioral strategy. We utilize the stop-signal task, in which subjects respond to a go stimulus on each trial, unless instructed not to by a subsequent, rare stop signal. We model across-trial learning of stop signal frequency (P(stop)) and stop-signal onset time (SSD) with a Bayesian hidden Markov model, and within-trial decision-making as optimal stochastic control. The model predicts that RT should increase with expected P(stop) and SSD, a prediction borne out by our human data. Thus, it appears that humans continuously monitor environmental statistics and adjust behavioral strategy accordingly. More broadly, our approach exemplifies the use of ”noisy” RT measures for extracting insights about cognitive and neural processing.

Tracking word-object co-occurrence statistics can reduce referential uncertainty during word learning. But human learners are constrained by limits on attention and memory, and therefore must store a subset of the information available—how do they select what information to store? We hypothesize that the presence of referential cues like eye gaze guides how learners allocate their attention. In three large-scale experiments with adults, we test how the presence of referential cues affects cross-situational word learning. Referential cues shift learners away from multiple hypothesis tracking towards storing only a single hypothesis (Experiments 1 and 2). In addition, learners are sensitive to the reliability of a referential cue and when it is less reliable, they are less likely to use the cue and more likely to store multiple hypotheses (Experiment 3). Together, the data suggest a rational tradeoff: In conditions of greater uncertainty, learners tend to store a broader range of information.

In the present study, we test the theory that humans selectively encode incoming sensory information on an as-necessary basis, when the information would not be accessible otherwise, in order to compensate for cognitive limitations on the quantity of new information that they can encode. We investigate whether external informational sources—much like high-level knowledge obtained from previous experiences—can spare learners from having to encode all new information in fine-grained detail. If this is true, we would expect to observe differences between the way human learners encode new information that they know to be easily available through external informational resources (e.g., names of actors in a movie, the date of a historical event) and those that they know are not (e.g., names of new acquaintances, the date of a wedding anniversary). Specifically, we would expect learners to encode far less detail for information that is available through known external informational resources than for information that is not. We present evidence from a study run on Amazon Mechanical Turk that human memory preferentially encodes information that is not expected to be available from external informational resources.

How does the brain identify stimuli that are relevant for predicting important events and how does it distinguish spurious relationships from truly predictive ones? We examined two contrasting theoretical frameworks: in the first, learning proceeds by considering a fixed hypothesis of the environment’s statistical structure (the set of predictive and causal relationships) and adjusting strength parameters for these relationships to optimize predictions. In contrast, the second approach directly assesses ambiguity in predictive relationships by evaluating multiple hypothesis of the environment’s statistical structure. We compared these frameworks in an animal model of aversive conditioning, allowing us to also manipulate the underlying brain systems. We show that when facing novel predictive stimuli, rats initially adopt a structure learning strategy, but switch to updating parameters during subsequent learning

A study in political psychology identifies four item-based factors of political trustworthiness in the USA: capability, consistency and closeness, egotism and opportunism, and communal commitment. Additionally, a list of items describe epistemic expertise. Together, these elements make up a description of political source credibility in the USA. The current study examines the power of these elements to predict source credibility. Eliciting estimations of likelihood and importance of each item on a Likert-type scale as well as overall estimations of trustworthiness and expertise, the paper presents weighted as well as non-weighted models that predict the likelihood that election candidates are trustworthy, have expertise, and are credible sources for individual respondents. Multiple regression analyses show that non-weighted scales have slightly better predictive power than weighted scales. The findings further provide an example of a datadriven method for applying a general cognitive models of source credibility to specific domains.

In this study based on ethnographic methods, we investigated how using a three-dimensional (3D) printed liver influenced doctors during liver resection surgery. Results of the analy- ses implied that using the 3D printed liver enhanced the con- struction of elaborate mental models of patients’ livers, the mental simulation of liver resections, and the construction of shared mental models of patients’ livers among doctors. Based on these results, we compared the advantages of using a 3D printed liver over a two-dimensional (2D) and a 3D image dur- ing surgery.

Many beliefs about oneself are constructed through experience, but the kinds of evidence that inform these beliefs in early childhood are not well understood. One source of information that affects adults and older children’s appraisals of themselves is social comparison. We found that even preschoolers (mean=57 months) spontaneously use social comparisons to guide their behavior. In Experiment 1, children who saw they out-performed peers on a task subsequently persisted less than children in other conditions. Children who saw evidence suggesting they performed either better or worse than peers on the task were more likely to choose an easy (versus difficult) novel task relative to those who saw neutral or no evidence. In Experiment 2 children who saw peers perform better were inclined to persist more than children in other conditions. This suggests preschoolers use social comparison to draw inferences about themselves without explicit cues, and this affects their motivation.

The human part of chance-discovery is usually analyzed as an effect of the agent’s knowledge of herself and of her environment. In this paper, setting off from the importance of “understanding the meaning of an impending phenomenon as a chance,” we will analyze how chance-discovery activities are affected and driven by the agent’s ignorance, and the relationship she entertains with the latter. More specifically, we will spell out two kinds of ignorance that are relevant for chancediscovery, also considering which abductive chance-discovery processes they can be related to.

A classical conjecture in generative linguistics is that universal restrictions on determiners in Natural Language (e.g. monotonicity, invariance, and conservativity) serve the purpose of simplifying the language acquisition task. This paper formalizes this conjecture within the PAC-learnability framework

Greenberg’s (1963) Universal 34 states that “No language has a trial number unless it has a dual. No language has a dual unless it has a plural.” We present an associative model of the acquisition of grammatical number based on the Rescorla- Wagner learning theory (Rescorla & Wagner, 1972) that predicts this generalization. Number as a real-world category is inherently structured: higher numerosity sets are mentioned less frequently than lower numerosity sets, and higher numerosity sets always contain lower numerosity sets. Using simulations, we demonstrate that these facts, along with general principles of probabilistic learning, lead to the emergence of Greenberg’s Number Hierarchy

Mathematical expertise builds on a foundation of space, especially the ability to map exact numbers to linear space. This “mental number-line” is known to vary cross-culturally, but there is debate about the mechanisms responsible for its cultural elaboration. We investigated the role of co-speech gesture, a ubiquitous cultural activity, in stabilizing and entrenching the mental number-line within a community. Imitating culture-specific gestures systematically shaped gesturers’ mental number-line. Moreover, gestures were used spontaneously to infer speakers’ spatial understanding of number, and merely observing these gestures was sufficient to shape the observer’s own mental number-line. These findings establish co-speech gesture as one mechanism for propagating and perpetuating the number-line

In decisions from experience (DFE), people sample from two or more lotteries prior to making a consequential choice. Although existing models can account for how sampled experiences relate to choice, they don’t explain decisions about how to search (in particular, when to stop sampling information). We propose that both choice and search behavior in this context can be understood as a sequential sampling process whereby decision makers sequentially accumulate outcome information from each option to form a preference for one alternative over the other. We formalize this process in a new model, Choice from Accumulated Samples of Experience (CHASE). The model provides a good account of choice behavior and goes beyond existing models by explaining variations in sample size under different task conditions. This approach offers a process-level framework for understanding how interactions between the choice environment and properties of the decision maker give rise to decisions from experience.

Building on earlier neuropsychological work, we adopted a novel individual differences approach to examine the relationship between spatial language and a wide range of both verbal and nonverbal abilities. Three new measures were developed for the assessment of spatial language processing: spatial naming, spatial verbal memory, and verbal comprehension in spatial perspective taking. Results from a sample of young adults revealed significant correlations between performance on the spatial language tasks and performance on both the analogous (non-spatial) verbal measures as well as on the (non-verbal) visual-spatial measures. Visual-spatial abilities, however, were more predictive of spatial language processing than verbal abilities. Furthermore, results from a sample of older adults revealed impairments in visual-spatial tasks and on spatial verbal memory. The results support dual process accounts of meaning, and provide further evidence of the close connection between the language of space and non-linguistic visualspatial cognition.

Intellectualism – the thesis that know-how is a kind of knowthat – has proved difficult to assess by the traditional philosophical method of conceptual analysis. Recently, some authors have argued that we should instead look to results in psychology – specifically whether all procedural knowledge is declarative knowledge. I argue that such an approach is unsatisfactory, since the concepts employed in psychology do not map onto our concepts of knowledge in any neat way. There is no straightforward psychological interpretation of the intellectualist thesis.

In two experiments we tested people’s naïve beliefs about where interventions act in real-world causal systems. We provided people with a description of a novel health condition that could be treated by two different treatments, a medication and a lifestyle modification. Participants judged a medication as acting on the symptoms of a disorder instead of the cause of the disorder, while a lifestyle modification was seen as acting on both the cause and the symptoms of a health condition (Experiment 1). These results held despite participants rating both treatments as effective. Providing information about the specific causal mechanism by which a treatment could work did not increase beliefs about a medication’s ability to target the cause of a disorder (Experiment 2). Implications for understanding of everyday causal interventions and health treatments is discussed.

Using Spanish-English bilingual Latino-Americans, this study tested whether priming Latino or American cultural representations facilitated the accessibility of the Spanish meaning or English meaning of Spanish-English homographs. Seventy-four participants were randomly assigned to a Latino prime, American prime, or no prime condition. After being primed, subjects performed an English lexical decision task wherein they indicated whether a letter string formed an English word. Homographs, English controls, and non-words were included in the array. As predicted, there was a significant prime condition by word type interaction, F(2, 70)= 5.48, p= .006, partial eta squared = .136, suggesting that prime condition modulated reaction times to homographs. Planned contrasts showed that participants in the Latino prime condition had slower reaction times to homographs than English controls, F(1, 22)= 4.84, p= .039, partial eta squared = .180, suggesting that the Latino prime facilitated access to homographs’ Spanish meaning.

Work in the mental model tradition has shown that human reasoners are subject to fallacious inferences from very simple premises that have been described as tantamount to cognitive illusions (Walsh & Johnson-Laird, 2004; Khemlani & Johnson-Laird, 2009). We present four experiments that show that these phenomena are much more general and systematic than has previously been thought. Among other results, we find that premises using ‘some’ mirror premises using ‘or’ in generating fallacious inferences, showing that there are interesting facts about reasoning with quantifiers beyond syllogisms that have been the main focus in the literature. Neither mental model theory nor other familiar theories of reasoning account for the results we present. However, the novel illusory inferences we present are predicted by the erotetic theory of reasoning (Koralus and Mascarenhas, 2013). The key idea is that, by default, we reason by interpreting successive premises as questions and maximally strong answers to those questions, which generates the observed fallacies

Learning argument structure constructions is believed to depend on input properties. In particular, in a cued production task, verb production within each construction has been shown to depend on three input factors: frequency of a verb in a construction, contingency of verb–construction mapping, and verb semantic prototypicality. Earlier studies have estimated these values from a language corpus, without accounting for variation in the input to individual learners. We use a computational model to control for such variation, and our results replicate those reported for human learners. The second issue that we address relates to different ways of representing constructions: while the earlier studies employ form-only representations, we run an additional analysis for form–meaning representations. Again, the results show the impact of all three input properties on the verb production, but their relative impact depends on the representations used.

There are considerable differences in language processing skill among the normal population. A key question for cognitive science is whether these differences can be ascribed to variations in domain-general cognitive abilities, hypothesized to play a role in language, such as working memory and statistical learning. In this paper, we present experimental evidence pointing to a fundamental memory skill—chunking—as an important predictor of crossindividual variation in complex language processing. Specifically, we demonstrate that chunking ability reflects experience with language, as measured by a standard serial recall task involving consonant combinations drawn from naturally occurring text. Our results reveal considerable individual differences in participants’ ability to use chunk frequency information to facilitate sequence recall. Strikingly, these differences predict variations across participants in the on-line processing of complex sentences involving relative clauses. Our study thus presents the first evidence tying the fundamental ability for chunking to sentence processing skill, providing empirical support for construction-based approaches to language.

Information in the visual environment is largely accessed through a series of fixations punctuated by saccades. Changes in fixation patterns in response to learning are well documented in studies of categorization, but the properties of the saccades that precede them and the role of visual salience in effecting eye movements remains poorly understood. This eye tracking study examines oculomotor changes in a categorization task with salient distractors. The design examines high-level, goal-directed attention that serves the purpose of learning, and making decisions based on that learned knowledge in the presence of salient distractors. We find that salient distractors draw fixation durations and saccade velocities that display similar properties to eye movements directed to task relevant items, challenging existing accounts that salience draws rapid saccades.

Sound to meaning correspondences in spoken language are assumed to be largely arbitrary. However, research has identified a number of exceptions to the arbitrariness assumption. In particular, non-arbitrary mappings between sound and shape, the bouba/kiki effect, have been documented across diverse languages and both children and adults are sensitive to this type of sound symbolic mapping. The cognitive basis for the associations between nonword labels and particular shapes remains poorly understood making it difficult to predict how findings generalize beyond the limited stimuli tested. To identify systematic bases for sound-to-shape mappings, we collected ratings of roundedness and pointedness for a large database of pseudowords. We find that attributes of both consonants and vowels are systematically related to judged shape meanings of pseudowords, and offer hypotheses as to the cognitive mechanisms underlying the observed patterns

Many high-level cognitive tasks involve understanding – the mechanisms by which an agent attempts to construct accurate mental representations of its world. In this paper, we discuss two such processes: explanation and question answering. We propose four theoretical assumptions about representation and processing that arise in these tasks: both involve inference, this inference requires making default assumptions, it occurs in an incremental manner, and it produces structures that can be expressed as directed graphs of conceptual ground literals. We analyze two models of explanation and question answering in terms of these commonalities and evaluate experimental claims about them using reading comprehension passages. In closing, we discuss our findings in light of related research

People often incorporate the opinions of others to make predictions about the world, including their preferences for novel experiences and items. In two experiments, we explored how people use the opinions of dissimilar others in making such predictions. While social cognition research has found that similar others tend to influence our judgments more than dissimilar others, the diversity principle from category-based induction argues that we value evidence from diverse sources. Our results suggest that people seek and use information from dissimilar others differently when predicting their own preferences than when making predictions with more verifiable values. For self-relevant predictions, participants were less likely to seek the opinion of dissimilar advisors (Experiment 1) and more likely to contrast their judgments away from these advisors’ opinions (Experiment 2).

Current computational models of word learning make use of correspondences between words and observed referents, but as of yet cannot—as human learners do—leverage information regarding the meaning of other words in the lexicon. Here we develop a Bayesian framework for word learning that learns a lexicon from multiword utterances. In a set of three simulations we demonstrate this framework’s functionality, consistency with experimental work, and superior performance in certain learning tasks with respect to a Bayesian word leaning model that treats word learning as inferring the meaning of each word independently. This framework represents the first step in modeling the potential synergies between referential and distributional cues in word learning

We develop a quantum probability model that can account for situations where people’s causal judgments violate the properties of causal Bayes nets and demonstrate how the parameters of our model can be interpreted to provide information about underlying cognitive processes. We implement this model within a hierarchical Bayesian inference framework that allows us to systematically identify individual differences and also provide a latent classification of individuals into categories of causal and associative reasoners. Finally, we implement a basic normative causal Bayes net within the same inference framework that allows us to directly compare quantum and classical probability models using Bayes factors

We show how dynamic (changing) environments can affect choice behavior, and highlight the challenges that recent models face in explaining the learning and selection of heuristic strategies under such conditions, especially when decisions are made using only a small subset of the available information. We propose an enhanced modeling framework that includes a trial-by-trial implementation of a Bayesian adaptive toolbox, redefinition of heuristic strategies, and incorporation of intricate learning rate mechanisms into a strategy learning model. We use data from a new empirical study to show how this improves the quality of inference

We assessed the influence of instrumental outcome divergence – the extent to which actions differ in terms of their outcome probability distributions – on behavioral preference in a two-alternative forced choice task. We found that participants preferred a pair of available actions with high divergence to a pair with low divergence. The effect of outcome divergence, dissociated here from that of other motivational and information theoretic factors, potentially reveals the value of flexible control.

Current models of word learning focus on the mapping between words and their referents and remain mute with regard to conceptual representation. We develop a cross-situational model of word learning that captures word-concept mapping by jointly inferring the referents and underlying concepts for each word. We also develop a variant of our model that incorporates recursion, which entertains the idea that children can use learned words to aid future learning. We demonstrate both models’ ability to learn kinship terms and show that adding recursion into the model speeds acquisition

Context plays a ubiquitous role in language processing. For the most part, work in language processing investigates the effects of context without investigating questions about what determines a context. For example, interpretation of any referential expression must take into account the notion of a referential domain. Here we investigate the influence of perceptual cues in establishing a referential domain, or linguistic context. We demonstrate that people use perceptual cues to establish a linguistic context; the influence of perceptual cues is gradient with respect to cue magnitude; and the contribution of a perceptual cue in constructing a linguistic context is not an effect of attention or salience. We provide these results as a first step toward developing a formal model for the establishment of linguistic context.

The current research examines what types of change are perceived as allowable versus disallowable in the self while still maintaining a sense of personal continuity. We find that overall, improvements are seen as more allowable than worsening or unspecified change, although this difference varies in magnitude based on the centrality of the trait being considered. Additionally, valence interacts with expectations of change, such that the differential impact of positive versus negative change on self-continuity is largest when positive change is expected, but is attenuated when negative change is expected.

Duration of an event tends to be underestimated as it becomes temporally distant (Spetch & Wilkie, 1983). The current study investigated this so-called choose-short effect in time and length in order to reevaluate the claim that the choose-short effect is special to temporal memory (Wearden, Parry, & Stamp, 2002). Participants made discrimination judgments in time or length on a pair of line stimuli separated by a delay. The stimulus presented during delay was varied in time or length. A length manipulation intended to be an analogue of temporal delay induced the choose-short effect in length discrimination. We developed a computational model based on ACT-R memory mechanisms (Anderson et al., 2004) to account for the main results in both time and length. The current results indicate that domain-general memory principles could account for the seemingly unique temporal phenomenon.

In this paper, we study the antecedents of moments of particularly successful learning while students use a Cognitive Tutor for geometry. Students used the Cognitive Tutor as part of their regular classroom activities and data was collected automatically. Learning moments were operationalized as when the probability that the student just learned was extremely high, as determined by a probabilistic model: the moment-by-moment learning model. The results indicate that while self-explanation is weakly predictive of learning moments, contextual guessing and several other factors are even better predictors of learning moments. These results suggest that unexpected events in student behavior may be good predictors of changes in knowledge.

Previous research has shown sketching to be useful to students solving math problems. The present study examines which aspects of middle school students’ sketching are related to, or predict, successfully answering math problems. The effects of individual differences in cognitive factors – working memory, spatial ability, and prior math knowledge – on answer accuracy are also analyzed. Stepwise regression analysis indicates that prior math knowledge and the inclusion of numerical representations of key problem relationships in sketches positively predict answer accuracy, whereas including irrelevant relationships in a sketch is associated with lower answer accuracy. Methodological implications for future research are discussed.

The Baldwin Effect is a proposed mechanism by which plasticity facilitates adaptive phenotypic and genetic evolution. In particular it has been proposed to be involved in the evolution of language. Here we investigate three factors affecting the extent to which plastic traits are fixed by selection: (i) the difficulty with which traits can be acquired through plasticity, (ii) the importance of traits to fitness, and (iii) the nature of dependencies between different traits. We find that selection preferentially fixes traits that are difficult to acquire through plasticity, traits that have larger fitness benefits, and traits that affect the acquisition of, or benefits from, other traits. We conclude by discussing the implications of these findings for the evolution of language as well as non-human behaviors and reconsider the evolutionary significance of the Baldwin Effect

Most models of choice in language focus on broadly applicable generative knowledge, treating item-specific variation as noise. Focusing on word order preferences in binomial expressions (e.g. bread and butter), we find meaning in the item-specific variation: more frequent expressions have more polarized (i.e. frozen) preferences. Of many models considered, only one that takes expression frequency into account can predict the language-wide distribution of preference strengths seen in corpus data. Our results support a gradient trade-off in language processing between generative knowledge and itemspecific knowledge as a function of frequency

We present an analysis of a treebank of spontaneous English dyadic conversations, investigating whether the degree of syntactic priming found across speakers is a function of the degrees of a?ective alignment and over- all positivity of the speakers. We use information theory to measure the proportion of overlap between the syn- tactic structures of the speakers. The a?ective state of the speakers is indexed by aggregated measures of the a?ective valences of the words they use. We ?nd that there is a positive relation between syntactic priming and a?ective alignment, over and above any lexical rep- etition e?ects. This constitutes evidence for the percola- tion of inter-speaker alignment across multiple levels of representation. This also illustrates the indexical value of syntactic alignment, as has been proposed in modern functional theories of grammar such as Dialogic Syntax.

Pedagogical agents (PAs) have the ability to scaffold and regulate students’ learning about complex topics while using intelligent tutoring systems (ITSs). Research on ITSs predominantly focuses on the impact that these systems have on overall learning, while the specific components of human- ITS interaction, such as student-PA dialogue within the system, are given little attention. One hundred undergraduate students interacted with MetaTutor, a multiagent hypermedia ITS, to learn about the human circulatory system. Data from these interactions were drawn from questionnaires and logfiles to determine the extent to which a specific agent from MetaTutor, Sam the Strategizer, impacted students’ overall emotions while using the system. Results indicated that Sam negatively impacted students’ experiences of enjoyment, in relation to the other agents of MetaTutor, and the frequency of Sam’s interactions with students significantly predicted their reports of boredom while using the system. Implications for the design of affect-sensitive multiagent ITSs are discussed.

We extend universal moral grammar theory (UMGT) with nonmonotonic logic. Our experiment shows that such revision is necessary as it allows to account for the effects of alleviations and aggravations in moral reasoning. Our new theory updates UMGT from classical to non-monotonic logic, which reflects the incompleteness of information and uncertainty in actual human reasoning. In addition, it provides an explanation of the paradoxical findings in the moral dilemma of the Trolley problem and the Knobe effect

Two visual world eye-tracking studies investigated the effect of emotions and actions on sentence processing. Positively emotionally valenced German non-canonical object-verb-subject (OVS) sentences were paired with a scene depicting three characters (agent-patient-distractor) as either performing the action described by the sentence, or not performing any actions. These scene-sentence pairs were preceded by a positive prime in the form of a happy looking smiley (vs. no smiley) in experiment 1 and in the form of a natural positive facial expression (vs. a negative facial expression) in experiment 2. Previous research has demonstrated the effect of action depiction on sentence processing of German OVS sentences (Knoeferle, Crocker, Scheepers, & Pickering, 2005). Moreover, emotional priming facilitates sentence processing for older and younger adults (Carminati & Knoeferle, 2013). However, up to date there is no evidence as to whether schematic faces such as smileys are as effective as natural faces in facilitating sentence processing. These insights lead to the hypotheses that participants would not only profit from depicted events, but that processing of OVS sentences might also be positively affected by emotional cues. Plus, we assessed the degree of naturalness the emotional face needs to possess to affect sentence processing. Results replicate the predicted effect of action depiction (vs. no action depiction). The expected facilitatory effect of emotional prime is trending in both experiments. However, the effect is more pronounced in the natural face version (exp. 2) than in the smiley version (exp. 1).

Previous studies on numeric cognition have focused primarily on magnitude, based on its role as a core feature of number knowledge. In this paper, we report the results of three experiments investigating adults’ sensitivity to properties of number apart from magnitude. In Experiment 1, we use a triadic judgment task to replicate a classic study of number properties. In Experiment 2, we compare these representations among expert and non-expert groups. In Experiment 3, we examine whether instruction can tune representation of number properties. Results indicate that the triadic comparison task is a reliable method of assessing sensitivity to number properties. We found that magnitude is difficult to suppress among non-experts, who are primarily attuned to magnitude and parity. Mathematically sophisticated participants were sensitive to a range of number properties compared with the non-expert group. We discuss implications for theories of number concepts and their relation to special populations

People with autism spectrum disorder (ASD) tend to detect lo- cal patterns of visual stimuli more quickly than global patterns, which is opposite to the behavior of typically developing peo- ple. We hypothesized that the imbalance between excitation and inhibition neurons in the visual cortex causes the local pro- cessing bias observed in ASD. Stronger inhibitory connections could diminish the neural activities and thus prevent global fea- ture integration, whereas properly balanced connections would enable the cortex to detect features of any size. We verified our hypothesis by employing a computational neural network called a neocognitron. Our experimental results demonstrated that the network with stronger inhibitory connections exhib- ited a local processing bias, whereas the network with properly adjusted connections showed a moderate global bias. More- over, the networks with extremely strong or weak inhibitions revealed no perception bias. These results suggest that an ex- citation/inhibition imbalance causes multiple types of atypical perception in ASD.

We investigate how learning that an established type-level causal relationship is implemented by human agency affects people’s conceptualization of this relationship. In particular, we ask under what conditions subjects continue to perceive the original root cause as appropriate explanation for the resulting effect, and under what conditions they perceive the mediating intentional action as alternative explanation instead. Using a new experimental paradigm, we demonstrate that mechanisms involving intentional action lead to intuitions of causal intransitivity, but only when these actions are norm-violating. Potential generalizations and implications for scientific theory construction are discussed.

The foreign language effect (Costa et al., 2014) refers to a phenomenon in which the response to a moral dilemma depends on whether it is asked in a native or second language. This study explored this effect with Japanese participants using various types of moral dilemmas. Study 1 adopted twelve variations of trolley dilemmas from Mikhail (2007). Study 2 used seven types of moral dilemmas from Greene et al. (2001). The dilemmas required permissibility and understandability judgments. Results of the two studies demonstrated the following two points. (1) Interactions between types of dilemmas (switch/footbridge) and language (native/foreign) were significant in both studies, indicating that the foreign language effects were replicated consistent with Costa et al. (2014). (2) Evidence that contradicts the theoretical explanation of the foreign language effect was also found.

Previous research investigating the dynamical processes supporting coordinated joint action has typically used nongoal- directed tasks. The present study expands on this research by investigating the coordination that emerges among pairs in a complex, goal-directed task of herding virtual sheep to the center of a field. The results revealed that the majority of pairs converged on the same stable movement coordination strategy in order to complete the task. This strategy involved pairs moving in an in-phase or anti-phase oscillatory pattern around the sheep. By adopting this strategy pairs formed an interpersonal synergy. Interestingly, the strength of this synergy was modulated by the number of sheep being herded. More specifically, more dimensional compression was observed among pairs when herding the 7- sheep compared to herding 3 or 5 sheep. The implications of these results for understanding how task difficulty and mutually defined environmental co-regulation influenced the behavioral dynamics of coordinated joint-action are discussed.

The present study investigated the diverse cognitive profiles of children learning mathematics in early elementary school. Unlike other types of learning difficulties, mathematics impairments are not characterized by a single underlying cognitive deficit, instead multiple general and numeracyspecific cognitive skills have been proposed to underlie mathematics ability. Combining theory- and data-driven approaches, the study investigated cognitive mathematics profiles. Participants for this study were 97 children tracked from senior kindergarten to grade two, as part of the Count Me In Study. Using numeracy, working memory, receptive language, and phonological awareness factors, a two-step cluster analysis revealed a three-cluster solution. The groups were characterized as (1) above average overall, (2) average overall with weak visuospatial working memory, (3) poor overall with strong visuospatial working memory. Cluster 1 demonstrated strengths in mathematics and reading, compared to clusters 2 and 3. Developmental trends and potential interventions are discussed.

Metaphors are commonly used by individuals to represent and reason about time in daily conversations. These metaphors are often paired with gestures that reveal the possible axes along which our internal conceptualisation of time may be aligned against. The present study attempts to use such gestures as temporal primes to investigate how individuals conceptualize time. Results revealed effects of congruency along the sagittal axis, but not the lateral. This suggests that individuals primarily represent time most strongly along the sagittal axis. Implications for models of how individuals represent time as well as methods of investigating how time is represented in the mind are discussed.

Children understand early in development that different people know different things, and they are adept at using this information to select appropriate sources of information (Lutz & Keil, 2002). However, in the current digital age, information may be gathered from both humans and technological sources that select and present information as humans do. Using methods designed to study epistemic trust in human informants (e.g., Koenig, Clement, & Harris, 2004), the current study investigates children’s and adults’ selective trust in a technological and human informant. Children (ages 4 and 5) and adults were presented with queries designed to probe their willingness to seek out and accept information from human versus technological informants. The results demonstrate that 4-year-olds prefer to seek information from a human informant, but by age 5, children show an increasing preference for the technological informant. The relationship between children’s trust and their experience with technology is also discussed.

This paper describes an evolutionarily plausible description of of a specialized working memory system involved in information management for high-order cognitive tasks through its capability for controlled maintenance and schematic access to symbolic representations. Along a volatile serially accessible symbolic storage that serves a basic maintenance function the system utilizes other accessory volatile memory systems along long-term memory (LTM) and learning systems for execution of schematic access to its content. Accessory systems can help encode the episodic information including the current state of the task and more importantly provide a means for address-based access to the content of symbolic storage. LTM and learning systems help map the current state of the task onto execution programs and thus help render schematic access and process of the retained symbolic information. Implications of this feature of the model are examined for the case if concurrent-counting task.

This paper presents an evidence for involvement of long-term memory (LTM) resources along volatile memory (VM) resources in active management of information in a working memory (WM) task that features schematic processing ofWM content. It was observed that in rehearsing frequently changing WM items in a self-paced concurrent-counting task when subjects learn and use a fixed rehearsing order across different episodes of the task they make significantly less error compared to when they adopt different rehearsing order for different episodes. This finding suggests that while retaining information in this task practically draws on volatile resources such as the phonological loop (PL), access to the corresponding item in WM relies on learning and retaining data structures in LTM. It is discussed that in this role learning and LTM resources help render schematic access to episodic information stored in less structured storage units such as PL. In this role LTM and learning plays a crucial role in execution ofWM tasks that employ complex process schemas.

Why do some negative sentences sound strange, even when they are both true and grammatical? We explore the pragmatics of negation by examining adults’ explicit felicity judgments of negative sentences in context. In Experiment 1, we found that a pragmatically supportive context elicited higher felicity ratings for negative sentences, and that negative sentences expressing nonexistence were rated higher than negative sentences referring to an alternative object. In Experiment 2, we used a within-subjects design to compare three context types, and found that negative sentences were rated more felicitous in a context where most of the characters possessed the negated object, compared to contexts where the other characters possessed an alternative object or nothing. We discuss the pragmatics of negation in light of these results, arguing that the felicity of negative sentences is influenced by changes in the informativeness of these sentences in different contexts

We studied changes to the pattern of speech errors as a function of selectively attending to one word in a sequence to learn how attention is implemented in language production. Three hypotheses were tested: (1) attention specifically inhibits the past, (2) attention enhances the activation of the present without affecting the past or the future, and (3) attention decreases priming of the future. In Experiment 1, using a model of sequential word production, we simulated the pattern of anticipatory and perseveratory errors on the attended words, and compared them to empirical error data. Our findings support a model in which attention only affects the present. Experiment 2 tested the prediction of this model regarding the error patterns on the word following the attended word. These results were also compatible with a transient enhancement in the activation of present that does not affect the production of the future words.

While the P600 is generally presumed to be a uniform response elicited consistently across individuals in specific syntactic contexts, Tanner and Van Hell (2014) showed evidence of distinct response profiles (N400 or P600 dominant) for syntactic violations across individuals. The current analysis used Tanner and Van Hell’s response-dominance index (RDI) to examine the impact of response dominance on comprehension of garden-path sentences. P600 dominant individuals showed enhanced comprehension of garden-path sentences, even when controlling for working memory capacity. Response dominance as an individual difference measure has the potential to enhance understanding of the neurocognitive basis of sentence processing and greater cognition in general

People often base judgments on stereotypes, even when contradictory base-rate information is provided. It has been suggested this occurs because people fail to engage or complete deliberative reasoning needed to process numerical base-rate information, and instead rely on intuitive reasoning. However, recent research indicates people have some access to this base-rate information even when they make stereotype judgments. Here we tested several hypotheses regarding these phenomena: A) People may believe stereotype information is more diagnostic; B) People may find stereotype information more salient; C) People have some intuitive access to baserate information, but must engage in deliberation to make full use of it. Aligning with account C, and counter to account A, we found inducing deliberation generally increased the use of base-rate information. Counter to account B, inducing deliberation about stereotype information decreased use of stereotype information. Additionally, more numerate participants were more likely to make use of base-rate information

The present research investigates cultural variation in conceptual frameworks for interpreting agency. A mind perception measure (Gray, Gray, & Wegner, 2007) was adapted for interviews with Indigenous Ngöbe adults in Panama and US college students. Participants ranked the agency capacities of various entities and provided explanations. Rankings varied systematically, with Ngöbe more likely to ascribe agency to nonhuman natural kinds than US participants. Analysis of explanations indicated that agency concepts are organized under different folktheories: US participants construed agency as a hierarchical, prototypically human capacity requiring consciousness, whereas Ngöbe construed agency as a multidimensional relational capacity expressed in directed interactions. An emphasis on psychological agency as distinct from other (biological, physical) forms of agency is widely assumed to be a conceptual prior, but these findings suggest it may instead be a feature of Western cultural epistemologies

The literature on the wisdom of crowds argues that in most situations, the aggregated judgments of a large crowd perform well relative to the average individual. There are, however, many real-world cases where crowds perform poorly. A small crowd literature has since developed, finding that better performing small crowds often exist within whole crowds. We compare previously proposed small crowd selection methods based on absolute or relative group performance to a new sequential search method and find that it selects better performing small crowds more consistently for forecasts of real gross domestic product (GDP) growth, inflation (measured by consumer price index, CPI), and unemployment rate made by US and Euro-zone surveys of professional forecasters.

Observers often judge agents who miss a desired outcome by a small, compared to a large, margin to be less happy. This nearmiss effect has typically been examined in situations where the agents have control over outcomes (e.g., missing a flight). Here, we extend this work in three ways. First, we show that near-miss effects play into observers’ intuitive theories of emotion even for randomly-determined outcomes over which agents demonstrably have no control. Second, we find data consistent with a hypothesis in which—even in randomly determined cases—near-miss effects reflect an illusion of control over those events. Finally, we integrate near-miss effects into a broader model of affective cognition, and quantify the psychological cost of a missing a desired outcome by relatively little distance, relative to winning or losing that outcome

Statistical learning (SL) is the ability to implicitly extract regularities in the environment, and likely supports various higher-order behaviors, from language to music and vision. While specific patterns experience are likely to influence SL outcomes, this ability is tacitly conceptualized as a fixed construct, and few studies to date have investigated how experience may shape statistical learning. We report one experiment that directly tested whether SL can be modulated by previous experience. We used a prepost treatment design allowing us to pinpoint what specific aspects of “previous experience” matter for SL. The results show that performance on an artificial grammar learning task at post-test depends on whether the grammar to be learned at post-test matches the underlying grammar structures learned during treatment. Our study is the first to adopt a pre-post test design to directly modulate the effects of learning on learning itself.

Correctly assessing the consequences of events is essential for a successful interaction with the world. It not only requires a causal understanding of the world but also the ability to distinguish whether a given event is the result of an agent’s own action (intervention) or simply the consequence of the world being in action (observation). Previous studies have shown that humans can learn causal structures, and that they can distinguish interventions from observations. These studies almost exclusively focused on structures where interventions led to a simple forward conditioned inference problem. We tested human subjects in a prediction game that required the integration over hidden causes, using a betting mechanism that allowed us to monitor subjects’ beliefs. Subjects learned the causal structure and the conditional probabilities with appropriate feedback. Once learned, all but one were immediately able to correctly predict the causal effects of their interventions according to optimal causal reasoning.

We describe an interactive mathematics technology intervention From Here to There! (FH2T) that was developed by our research team. This dynamic program allows users to manipulate and transform mathematical expressions. In this paper, we present initial findings from a classroom study that investigates whether using FH2T improves learning. We compare learning gains from two different instantiations of FH2T (retrieval practice and fluid visualizations), as well as a control group, and investigate the role of prior knowledge and content exposure in FH2T as possible moderators of learning. Findings, as well as implications for research and practice are discussed

From choosing a book to picking a restaurant, most choices people encounter are about “matters of taste” and thus no universal, objective criterion about the options’ quality exists. Tapping into the knowledge of individuals with similar tastes who have already experienced and evaluated options—as harnessed by recommender system algorithms—helps people select options that they will enjoy. Although recommender systems are available in some domains, for most everyday decisions there is neither an algorithm nor “big data” at hand. We mapped recommender system algorithms to models of human judgment and decision making about “matters of fact” and then recast the latter as social recommender strategies for “matters of taste”. This allowed us to investigate how people can leverage the experiences of other individuals to make better decisions when no machine recommender systems are available. Using computer simulations on a widely used data set from the recommender systems literature, we show that experienced individuals can benefit from relying on only the opinions of seemingly similar people. Inexperienced individuals, in contrast, are often well-advised to pick the mainstream option (i.e., the one with the highest average evaluation) even if there are interindividual differences in taste; this is because reliable estimation of similarity requires considerable experience

Data continuously stream into our minds, guiding our learn- ing and inference with no trial delimiters to parse our experi- ence. These data can take on a variety of forms, but research on causal learning has emphasized discrete contingency data over continuous sequences of events. We present a formal framework for modeling causal inferences about sequences of point events, based on Bayesian inference over nonhomo- geneous Poisson processes (NHPPs). We show how to apply this framework to successfully model data from an experiment by Lagnado and Speekenbrink (2010) which examined human learning from sequences of point events.

Although gaze can be directed at any location, different locations in the visual environment vary in terms of how likely they are to draw another person’s attention. One could therefore weigh incoming perceptual signals (e.g., eye cues) against this prior knowledge (the relative visual saliency of locations in the scene) in order to infer the true target of another person’s gaze. This Bayesian approach to modeling gaze perception has informed computer vision techniques, but we assess whether it is a good model for human performance. We present subjects with a “gazer” fixating his eyes on various locations on a 2-dimensional surface, and project an arbitrary photographic image onto that surface. Subjects judge where the gazer is looking in the image. A full Bayesian model, which takes image saliency information into account, fits subjects’ gaze judgments better than a reduced model that only considers the perceived direction of the gazer’s eyes.

This work focuses on the impact of sharing contexts on consumers’ decision processes and purchase-amount decisions. Four studies, using both hypothetical and real (incentive-compatible) choices, find that people regularly purchase more in sharing (vs. non-sharing) contexts. Evidence is presented suggesting that a significant portion of this effect is driven by a cognitive bias arising in sharing contexts that focuses people on what they will give to others, and away from what they will receive from others. Consequences of this bias include the noted surplus in purchase amounts, over-consumption, and waste.

Using a choice blindness paradigm, it is possible to switch decisions and outcomes in simple choice tasks. Such switches have been found to carry over into later choices, hypothesized to be mediated by beliefs about earlier decisions. Here we investigated participants’ memories for stimuli in a simple choice blindness task involving preferential choices between pairs of faces. We probed participants’ recognition and source memory following a round of choices where on some trials participants were presented with the opposite face to the one they actually selected. We found no effect on recognition memory accuracy. Source memory was impaired such that participants failing to detect the manipulation later misremembered recognized non-chosen faces as being previously chosen. The findings are discussed in the light of self-perception theory and previous work on how beliefs affect memories for choices

A long-standing debate in decision making has been whether people rely on very little information for making choices, or weigh and add all available information. We propose a new method to determine whether a non-compensatory (Take-The- Best) or compensatory strategy (Logistic Regression) is more psychologically plausible: by looking at peoples active learning queries. This method goes beyond traditional model selection techniques as it reveals the information people choose to learn early on, which subsequently drives their decisions. We developed active learning algorithms for both Take-The-Best and Logistic Regression, and designed an active learning experiment to distinguish between these models. By letting both models and humans actively learn, we could compare their queries, and found that people follow a rank-based learning strategy in non-compensatory environments, but prefer more certainty-based queries in compensatory environments. We argue that active learning studies provide a promising new methodology to distinguish among decision models

Self-directed learning, defined as the ability to choose what to learn about, represents a unique educational opportunity. We test the effect of self-direction on learning outcomes in children (N=32, age range=3-5 years) in a novel word-learning task conducted via touchscreen tablets. Study participants were randomly assigned to one of two learning conditions: one in which learning was self-directed and one in which it was not. Children in the self-directed condition performed better on a recognition task, controlling for subject and item effects. Our results suggest that self-directed learning facilitates information retention in children, in line with previous work that has found improved information retention using self-directed learning paradigms in adults (e.g., Markant, DuBrow, Davachi, & Gureckis, 2014).

An important goal in the study of higher-order cognition is to understand how relational categories are acquired and applied. Previous work has explored the potential of withincategory comparison opportunities to promote relational category learning and transfer. This follows from predictions of structure mapping theory (Gentner, 1983, 2003) that alignment leads to highlighting and abstraction of common relational structure. However, a straightforward merging of traditional classification learning with comparison (i.e., trials presenting two same-category items) has not been effective. We explore the hypothesis that classification and comparison have an unforeseen incompatibility. In a 3x2 betweensubjects design we tested three presentation conditions (unconstrained item pairs, category-matched items pairs, single items) in two supervised category learning modes: classification and observation. The major finding is an interaction driven by highly accurate categorization for the observational learners with same-category pairs. The introduction of the observational mode yielded the predicted, but elusive result of an advantage for within-category pairs over twice as many single-item trials. We conclude that within-category comparison can be an effective means to promote relational category learning and discuss the apparent impediment of the guess-and-correct cycle

Although negotiating joint action through dialogue can be difficult, dyads may be able to improve collaborative performance by managing communicative efficiency in language production, balancing effort (words per turn) with output (turn-level success). Comparing dyads with high, medium, and low levels of accuracy in communication, growth curve modeling revealed a negative relationship between success and excessive variability in levels of efficiency. Dyads performed better by maintaining moderately fluid efficiency (seen in high-success dyads) or minimizing efficiency variability (seen in medium-success dyads), rather than scrambling for efficiency only as needed (seen in low-success dyads). Balancing efficiency variability in language production may create flexible but relatively stable interaction structures, laying the groundwork for successful communication

From an early age, people exhibit strong links between certain visual (e.g. size) and acoustic (e.g. duration) dimensions. Do people instinctively extend these crossmodal correspondences to vocalization? We examine the ability of congenitally deaf Chinese children and young adults (age M = 12.4 years, SD = 3.7 years) to generate iconic vocalizations to distinguish items with contrasting magnitude (e.g., big vs. small ball). Both deaf and hearing (M = 10.1 years, SD = 0.83 years) participants produced longer, louder vocalizations for greater magnitude items. However, only hearing participants used pitch—higher pitch for greater magnitude – which counters the hypothesized, innate size “frequency code”, but fits with Mandarin language and culture. Thus our results show that the translation of visible magnitude into the duration and intensity of vocalization transcends auditory experience, whereas the use of pitch appears more malleable to linguistic and cultural influence.

Scholars have documented substantial classes of iconic vocabulary in many non-Indo-European languages. In comparison, Indo-European languages like English are assumed to be arbitrary outside of a small number of onomatopoeic words. In three experiments, we asked English speakers to rate the iconicity of words from the MacArthur- Bates Communicative Developmental Inventory. We found English—contrary to common belief—exhibits iconicity that correlates with age of acquisition and differs across lexical classes. Words judged as most iconic are learned earlier, in accord with findings that iconic words are easier to learn. We also find that adjectives and verbs are more iconic than nouns, supporting the idea that iconicity provides an extra cue in learning more difficult abstract meanings. Our results provide new evidence for a relationship between iconicity and word learning and suggest iconicity may be a more pervasive property of spoken languages than previously thought.

Interpreting sentences spoken in a second language can be demanding and plagued with uncertainty, especially for lower proficiency listeners. While native language listeners use numerous information sources to anticipate upcoming words accurately, the pattern of anticipation may be different for second language users. We explore this issue in bilinguals with varying English proficiency by recording anticipatory eyemovements as participants listened to sentences (e.g., “The pirate chases the ship”) for which the object and three distractors (agent-related, action-related, unrelated) appeared in the concurrently presented images. Higher proficiency participants were faster than lower proficiency participants. Fixations to action-related distractors after onset of the action also varied by proficiency, with lower proficiency participants showing greater tendency to fixate this locally coherent actionrelated distractor. This final effect is supported by a trial level analysis, but appears to be unrelated to the effect of proficiency on anticipation speed.

Navigating in a novel environment can serve as an applied insight problem solving task, since many people gain a sudden, clear understanding (Aha-moment) of the spatial relations after being lost. With a unique design, we transformed the city center of a medieval German city into a virtual maze. The aim of the study was to test whether a spatial decision making task simulating real navigation would be feasible for investigating insight problem solving. Participants learned two pathways which they subsequently had to restructure to find their way to the navigation targets. We found evidence for the restructuring of participants’ prior knowledge during the solution attempts. 73% of all problem solvers reported an Aha-moment and there was an error drop at the critical intersection by those who had insight. The slope of the learning curve was established as a measurement of insightful experiences.

Systematicity commonly means that having certain cognitive capacities entails having certain other cognitive capacities. Learning is a cognitive capacity central to cognitive science, but systematic learning of cognitive capacities—second-order systematicity—has received little investigation. We proposed associative learning as an instance of second-order systematic- ity that poses a paradox for classical theory, because this form of systematicity involves the kinds of associative constructions that were explicitly rejected by the classical explanation. In fact, both first and second-order forms of systematicity can be derived from the formal, category-theoretic concept of uni- versal morphisms to address this problem. In this paper, we derived a model of systematic associative learning based on (co)recursion, which is another kind of universal construction. This result increases the extent to which category theory pro- vides a foundation for cognitive architecture.

There are often multiple explanations for someone’s behavior, but people generally find some behavior explanations more satisfying than others. We hypothesized that people prefer behavior explanations that are simple and rational. We present a computational account of behavior explanation that captures these two principles. Our computational account is based on decision networks. Decision networks allow us to formally capture what it means for an explanation to be simple and rational. We tested our account by asking people to rate how satisfying several behavior explanations were (Experiment 1) or to generate their own explanations (Experiment 2). We found that people’s responses were well predicted by our account.

To navigate many-to-many mappings between referents and linguistic expressions, listeners need to calibrate likelihood estimates for different referential expressions taking into account both the context and speaker-specific variation. Focusing on speaker variation, we present three experiments. Experiment 1 establishes that listeners generalize speakerspecific patterns of pre-nominal modification use across different adjective types. Experiment 2 examined a) the dimension of generalization (form-based or informativitybased); b) effects of the strength of the evidence (implicit or explicit); and c) individual differences in dimensions of generalization. Experiment 3 asked parallel questions for exposure to over-specified utterances; we predicted more conservative generalizations because in spontaneous utterances, speakers are more likely to over-modify than under-modify.

In language-mediated visual search, memory and attentional resources must be allocated to simultaneously process verbal instructions while navigating a visual scene to locate linguistically specified targets. We investigate when and how listeners use object names in visual-search strategies across three visual world experiments, varying the presence and location of an added visual memory demand. The results suggest that as long as objects in the display can be visually inspected throughout the trial, participants do not linguistically encode those objects. We suggest that instead they use the visual environment as an external memory, mapping the spoken word onto potential referents and using perceptual visual routines automatically triggered by the spoken word. The results are discussed in terms of flexible and efficient allocation of memory resources in natural tasks that combine language and vision.

People look at emptied spatial locations where information has been presented during encoding. There is evidence that this socalled ‘looking at nothing’ behaviour plays a functional role in memory retrieval of visuospatial and verbal information. However, it is unclear whether this effect is caused by the oculomotor movement of the eyes per se or if covertly shifting attention is sufficient to cause the observed differences in retrieval performance. In an experimental study (N = 26), participants were manipulated in being able to shift either their eyes or their focus of attention to a blank spatial location whilst retrieving verbal information that was associated with the location during a preceding encoding phase. Results indicate that it is not the oculomotor movement of the eyes that causes the facilitation while retrieving verbal materials, but rather covert shifts of attention are sufficient to promote differences in retrieval performance.

In Dung’s abstract argumentation, arguments are either acceptable or unacceptable, given a chosen notion of acceptability. This gives a coarse way to compare arguments. In this paper, we propose a counting approach for a more fine-gained assessment to arguments by counting the number of their respective attackers and defenders based on argument graph and argument game. An argument is more acceptable if the proponent puts forward more number of defenders for it and the opponent puts forward less number of attackers against it. We show that our counting model has two well-behaved properties: normalization and convergence. Then, we define a counting semantics based on this model, and investigate some general properties of the semantics.

To adapt in an ever-changing world, people infer what basic units should be used to form concepts and guide generalizations. While recent computational models of human representation learning have successfully predicted how people discover features from high-dimensional input in a number of domains (Austerweil & Griffiths, 2013), the learned features are assumed to be additive. However, this assumption is not always true in the real world. Sometimes a basic unit is substitutive (Garner, 1978), which means it can only be one value out of a set of discrete values. For example, a cat is either furry or hairless, but not both. In this paper, we explore how people form representations for substitutive features, and what computational principles guide such behavior. In a behavioral experiment, we show that not only are people capable of forming substitutive feature representations, but they also infer whether a feature should be additive or substitutive depending on the observed input. This learning behavior is predicted by our novel extension to the Austerweil and Griffiths (2011, 2013)’s feature construction framework, but not their original model. Our work contributes to the continuing effort to understand how people form representations of the world.

Readers misinterpret garden-path sentences such as While the man hunted the deer that was brown and graceful ran into the woods as meaning The man hunted the deer that was brown and graceful and the deer ran into the woods. The “Goodenough” processing account proposes that misinterpretation occurs when readers are satisfied with the interpretation derived from the first-pass parse, and thus do not bother to fully reanalyze the sentence (Ferreira et al., 2001; Christianson et al., 2001). Such an account predicts that there should be more evidence of reanalysis at the disambiguating verb (ran) on trials with correct responses to the question Did the man hunt the deer?, than on those with incorrect responses. The present study tested this prediction using separate self-paced reading and event-related brain potential (ERP) experiments. Results from Experiment 1 (self-paced reading) showed no difference in the reading time at the disambiguating verb between trials that were answered correctly and those that were answered incorrectly. Experiment 2 (ERP) corroborated this finding by showing no difference in the amplitude of the P600 component elicited by the disambiguating verb in trials with correct responses and those with incorrect responses. However, results from a norming experiment showed that plausibility information significantly predicted question accuracy in both experiments. Overall, these results suggest that responses to questions intended to probe whether garden-path sentences are fully reanalyzed do not always answer that question, but can instead be determined primarily by the plausibility of the events described in that question.

There is growing support for Temporal Difference (TD) Learning as a formal account of the role of the midbrain dopamine system and the basal ganglia in learning from reinforcement. This account is challenged, however, by the fact that realistic implementations of TD Learning have been shown to fail on some fairly simple learning tasks — tasks well within the capabilities of humans and non-human animals. We hypothesize that such failures do not arise from natural learning systems because of the ubiquitous appearance of lateral inhibition in the cortex, producing sparse conjunctive internal representations that support the learning of predictions of future reward. We provide support for this conjecture through computational simulations that compare TD Learning systems with and without lateral inhibition, demonstrating the benefits of sparse conjunctive codes for reinforcement learning.

Interpreting causal relations plays an important role in everyday life, for example in scientific inquiries and text comprehension. Errors in causal reasoning can be a recipe for disaster. Despite vast literature on the psychology of human causal reasoning, there are few investigations into preferred inferences in relational three-term problems. Based on a previous formal investigation about relevant causal relations we develop a cognitive modeling approach with mental models. The key principle for this approach proves to be the prediction of preferred inferences by model operations and the process of sub model integration. Subsequent experiments test preferred inferences, the number of model operations, and if concrete or generic problems make a difference in causal reasoning performance. Implications of the model are discussed.

Hyperdimensional computing (HDC) refers to the representation and manipulation of data in a very high dimensional space using random vectors. Due to the high dimensionality, vectors of the space can code large amounts of information in a distributed manner, are robust to variation, and are easily distinguished from random noise. More importantly, HDC can be used to represent compositional and hierarchical relationships and recursive operations between entities using fixed-size representations, making it intriguing from a cognitive modeling point of view. However, the majority of the existing work in this area has focused on modeling discrete categorical data. This paper presents a new method for mapping continuous-valued multivariate data into hypervectors, enabling construction of compositional representations from non-categorical data. The mapping is studied in a word classification task, showing how rich distributed representations of spoken words can be encoded using HDC-based representations

Existing models of infant word learning have mainly assumed that the learner is capable of segmenting words from speech before grounding them to their referential meaning, while segmentation itself has been treated relatively independently of meaning acquisition. In this paper, we argue that situated cues such as visually perceived concrete objects or actions are not just important for word-to-meaning mapping, but that they are useful in pre-linguistic word segmentation, thereby helping the learner to bootstrap the language learning process. We present a model where joint acquisition of proto-lexical segments and their meanings maximizes the referential quality of the lexicon, and where learning can occur without any a priori knowledge of the language or its linguistically relevant units. We investigate the behavior of the model using a computational implementation of statistical learning, showing successful word segmentation under varying degrees of referential uncertainty.

Human infants learn meanings for words in interaction with their environment. Individual learning scenarios can be ambiguous due to the presence of several words and possible meanings. One possible way to overcome ambiguity is called cross-situational learning (XSL), where information is gathered over several learning trials. Experimental studies of human XSL have shown that cognitive constraints, such as attention and memory limitations, decrease human performance when compared to computer models that can store all available information. In this paper, we approach modeling of human performance with a novel computational XSL algorithm, FAMM (Familiarity preference, Associative learning, Mutual exclusivity, Memory decay), equipped with the four main components motivated by experimental research. The model is evaluated based on a number of earlier XSL experiments that probe different aspects of learning. FAMM is shown to provide a better fit to the behavioral data than the earlier proposed model of Kachergis et al. (2012

When children encounter objects, design constrains and affords action and cognition. An observational study in the wild revealed how manipulable objects afforded greater complexity of cognitive outcomes, including testing causeand- effect and expressing abstract ideas about phenomena in the natural world. Evidence comes from video analysis of children’s speech, gesture, and action when using a wide range of natural history exhibits. In the museum—an environment expressly designed for learning—children sought information with their moving bodies, eyes and hands. They explored sensorimotor contingencies, looking while touching, pushing, and pulling; they probed the perceptual affordances of different types of museum media, including graphic panels, specimens, models, and interactive exhibits. Children spoke more about the museum’s content when they touched the exhibits, but the content of their speech changed depending on the object’s affordances for interaction. With static specimens and models, children most often referred to objects’ concrete properties. With interactive exhibits, children’s speech involved references to dynamic relations among exhibit elements. Use of abstract speech and iconic gestures also suggests that they perceived interactive exhibits as representations of objects and phenomena beyond the hereand- now. In summary, when children used interactive exhibits, the content of their speech was relational, representational, and at times, both representational and relational; they employed modes of conceptualization not seen when using non-interactive exhibits.

Although the relationship between sound and meaning in language is arbitrary, reliable correspondences between sound and meaning have been found in natural language. These sound symbolic relationships affect word learning, but less is known about how sound symbolism affects online processing during learning or for well-learned stimuli. We use the visual world paradigm and an artificial lexicon featuring carefully controlled sound symbolic correspondences to examine the effects of sound symbolism on the online processing of novel and well-learned stimuli. Initially, participants chose novel shapes matching the sound symbolic properties of the word above chance, reliably fixating consistent shapes around word offset. As learning approached ceiling, accuracy and reaction time differences between matching and mismatching stimuli disappeared but a disadvantage in the online processing of mismatching stimuli persisted in the form of lagging target fixations. This suggests that sound symbolism affects the online processing of spoken stimuli even for well-learned words.

People often make repeated decisions from experience. In such scenarios, persistent biases of choice can develop, most notably the “hot stove effect” (Denrell & March, 2001) in which a prospect that is mistakenly believed to be negative is avoided and thus belief-correcting information is never obtained. In the existing literature, the hot stove effect is generally thought of as developing through interaction with a single, stochastic prospect. Here, we show how a similar bias can develop due to people’s tendency to selectively attend to a subset of features during categorization. We first explore the bias through model simulation, then report on an experiment in which we find evidence of a decisional bias linked to selective attention. Finally, we use these computational models to design novel interventions to “de-bias” decision-makers, some of which may have practical application

Listeners draw on their knowledge of phonetic categories when identifying speech sounds, extracting meaningful structural features from auditory cues. We use a Bayesian model to investigate the extent to which their perceptions of linguistic content incorporate their full knowledge of the phonetic category structure, or only certain aspects of this knowledge. Simulations show that listeners are best modeled as attending primarily to the most salient phonetic feature of a category when interpreting a cue, possibly attending to other features only in cases of high ambiguity. These results support the conclusion that listeners ignore potentially informative correlations in favor of efficient communication.