Glossa Psycholinguistics

This study focuses on inter-individual convergence effects in the perception and categorization of speech sounds. We ask to what extent two listeners can come to establish a shared set of categorization criteria in a phoneme identification task that they accomplish together. Several hypotheses  are  laid  out  in  the  framework  of  a  Bayesian  model  of  speech  perception  that  we  have  developed  to  account  for  how  two  listeners  may  each  infer  the  parameters  that  govern  their partner’s responses. In our experimental paradigm, participants were asked to perform a joint  phoneme  identification  task  with  a  partner  that,  unbeknownst  to  them,  was  an  artificial  agent, whose responses we manipulated along two dimensions, the location of the categorical boundary and the slope of the identification function. Convergence was found to arise for bias but not for slope. Numerical simulations suggested that lack of convergence in slope may stem from the listeners’ prior level of confidence in the variance in VOT for the two phonemic categories. This study sheds new light on perceptual convergence between listeners in the categorization of speech sounds, a phenomenon that has received little attention so far in spite of its central importance for speech communication.

According to the Polarity Hypothesis, the presence or absence of a processing cost for Scalar Inferences (SIs) depends on their polarity. This hypothesis predicts, among other things, that the processing of lower-bounding SIs should not be affected by cognitive load the same way upper-bounding SIs are. To date, evidence in support of this prediction comes from the comparison between upper-bounding and lower-bounding SIs elicited by disparate scalar words. In this paper, we report on two dual-task experiments testing this prediction in a more controlled way by comparing upper-bounding and lower-bounding SIs arising from the same scalar words or scale-mates operating over the same dimension. Results show that, for these more minimal comparisons, lower-bounding SIs involve comparable cognitive demands as their upper-bounding counterparts. These findings challenge the idea that load effects are consistently modulated by SI polarity and suggest instead that these effects are relatively consistent across different types of SIs.

In this article, we look at sign duration and signing rate in corpora of three sign languages – British Sign Language (BSL), Dutch Sign Language (NGT), and Swedish Sign Language (STS). We investigate whether token frequency and sociolinguistic variables (e.g., age, gender, region) influence the production rate of signing. Following Zipf’s law of abbreviation, we see that a sign’s duration is negatively correlated with its frequency. Both sign duration and signing rate are found to correlate with signer age, in that older signers have longer durations and lower rates than younger signers. Signers' gender, family (deaf or hearing), and age of exposure have no effect on duration or signing rate. For NGT and STS, there is no effect of region on either duration or rate. However, in the BSL data, duration and signing rate vary with region. The overall findings align with previous work on spoken languages, particularly that frequency and aging are correlated with word length and production rate, thus demonstrating such patterns across modalities of language.

Previous research has shown that the computation of subject-verb number agreement can be derailed by the presence of syntactically illicit nouns, a phenomenon called agreement attraction. By contrast, the incidence of agreement attraction with anaphoric dependencies is less clear: previous work has mostly focused on reflexives and strong pronouns, which sometimes show attraction and other times do not. Meanwhile, research on clitics—a different class of pronominal anaphora—is scarcer. To expand the empirical record, we examined clitic pronouns in an under-researched language, Bulgarian. The results of a large sample eye-tracking study showed clear agreement attraction effects in fixation durations and regressive eye movements to the clitic pronoun and following words. These findings provide further evidence that the variable attraction profile of anaphoric dependencies might depend on the features of an anaphoric element, including its placement and the role of syntactic constraints in establishing the antecedent-pronoun dependency.

Linguists have often observed that the sense extensions in polysemous words follow patterns. Yet, these patterns have rarely been quantified, and it is unknown whether language users are sensitive to them. We developed four regularity metrics, focusing in this initial study on metaphor patterns that apply to nouns. We further tested adult English speakers’ capacity to understand new senses in an acceptability judgement task. We compared novel senses that followed a metaphor pattern against novel senses that did not respect any pattern. Our results showed that novel senses were judged as more acceptable when they were part of a polysemy pattern as opposed to when they were not. We also assessed whether acceptability judgements were influenced by the degree of regularity of the pattern that they follow. The results confirmed the psychological validity of degree of regularity as a measure: the more regular the polysemy pattern, the more acceptable the new sense following that pattern. Regularity metrics that captured the consistency with which a pattern is instantiated were more successful in predicting acceptability ratings than regularity metrics that captured the number of times a pattern is instantiated. These results motivate future psycholinguistic studies investigating the influence of regularity on learning, processing, and storage of polysemes in a more nuanced way than has been possible previously.

How do alternative forms of reference to individuals—first, last, and full names—guide inferences about the gender of the referent? Given distributional correspondences between English first names and gender, first names provide probabilistic information about an individual's gender. While English last names do not vary with gender, men are more likely to be referred to by last name alone. Across four experiments, we demonstrate that inferences about gender are shaped by a persistent bias to infer that people are male, along with probabilistic information carried by the first name. When an individual was introduced by last name alone, participants overwhelmingly used he to subsequently refer to the person, suggesting that participants inferred that the person was male. This bias was still present when the individual was introduced using a first or full name, with participants less likely to use she than the distributional characteristics of the first names would predict. When explicitly asked to recall an individual’s gender who was introduced by last name alone, participants preferentially responded that the person was male. This bias persisted even when the person was introduced using a first or full name. Repeated reference attenuated, but did not eliminate, this bias. We discuss implications for models of how world knowledge is linked to language use.

Within the psycholinguistic literature, there has been a longstanding debate regarding whether we resolve syntactic parsing ambiguities via universal or language-specific biases. The present study investigates attachment biases in the online parsing of ‘relative clause’ (RC) attachment in Italian with respect to pseudorelative (PR) availability. Following the PR account Grillo (2012), languages are assumed to universally prefer local attachment. When languages appear to prefer non-local attachment, this is due (at least partially) to the availability of PRs. Specifically, Grillo and Costa (2014) suggest that whenever a string is ambiguous between a PR and a RC, the parser will prefer the PR parse, resulting in apparent non-local attachment. Although there is growing evidence that PR availability indeed affects offline interpretations, few studies have explored this account from an online perspective. Hence, we conducted a self-paced reading task in Italian. In that task, we directly manipulated PR availability and attachment. Reading times for the critical and postcritical regions along with accuracy to comprehension questions were subjected to mixed-effect regressions. Consistent with the PR account, online results indicated a clear bias for local attachment with true RCs. When PRs were available, we observed a non-local bias. Additionally, the present study provides novel evidence in support of the PR-First Hypothesis, as results indicated that the initial preference for PRs may persist and affect the interpretation of even globally disambiguated items.

The general view in syntactic literature is that binding constraints can make antecedents syntactically inaccessible. However, several studies showed that antecedents which are ruled out by syntactic binding constraints still influence online processing of anaphora in some stages, suggesting that a cue-based retrieval mechanism plays a role during anaphora resolution. As in the syntactic literature, in semantic accounts like Discourse Representation Theory (DRT), formal constraints are formulated in terms of accessibility of the antecedent. We explore the discourse inaccessibility postulated in DRT by looking at its role in pronoun resolution of inter-sentential anaphoric relations in four off-line and two eye-tracking experiments. The results of the eye-tracking experiments suggest that accessibility has an effect on pronoun resolution from early on. The study quantifies evidence of inaccessible antecedents affecting pronoun resolution and shows that almost all evidence points to the conclusion that discourse-inaccessible antecedents are ruled out for pronoun resolution in processing. The only potential counter-example to this claim is also detected, but remains only as anecdotal evidence even after combining data from both eye-tracking studies. The findings in the study show that accessibility plays a significant role in the processing of pronoun resolution in a way which is potentially challenging for the cue-based retrieval mechanism. The paper argues that discourse accessibility can help expand the theories of retrieval beyond the syntactic and sentence-level domain and provides a window into the study of interference in discourse.

Language comprehension involves continuously making predictions about what will be mentioned next. If speakers take these predictions into account, one would expect that they try to be extra clear (e.g., by saying “the girl with the big earrings”) when they are going to say something less predictable. Conversely, speakers do not need to be as clear when the listener already expects the thing that they are about to mention, and can therefore suffice with a pronoun such as she. Previous research testing this hypothesis has found mixed results, with some studies finding that the referent’s predictability in discourse affects pronoun use and others finding that it does not. One explanation might be that speakers are more likely to take predictability into account when there is a co-present addressee who is predicting the next referent. To test this possibility, I conducted a language production experiment in which participants produced spoken continuations of narrative fragments. The fragments were accompanied by pictures that made clear how the story continued. Half of the participants performed the task without anyone else being present, while the other half told the stories to another person, who had to pick out the correct picture. Referent predictability was varied by manipulating the coherence relation in the narrative context. In addition, I calculated a surprisal score for each character in each narrative, as a more direct measure of its predictability. The results showed that with higher predictability, speakers were indeed more likely to use a pronoun than a definite NP to refer to the target character in their continuations. However, it did not matter whether the speaker was telling the stories to a co-present addressee or not. The results are discussed in light of accounts that distinguish between taking the perspective of a specific and that of a hypothetical listener.

Two experiments test how college students use nonbinary they to refer to a single and specific person whose pronouns are they/them, e.g., “Alex played basketball on the neighborhood court. At one point they made a basket,” compared to matched stories about characters with binary (she/her or he/him) pronouns. Experiment 1 shows that for both types of pronouns, people use pronouns more in a one-person than a two-person context. In both experiments, people produce nonbinary they at least as frequently as binary pronouns, suggesting that any difficulty does not result in pronoun avoidance in spoken language, even though it does in written language (Arnold et al., 2022). Nevertheless, there is evidence that nonbinary they is somewhat difficult, in that people made gender errors on about 9% of trials, and they used a more acoustically prominent and disfluent-sounding pronunciation for nonbinary pronouns than binary pronouns. However, exposure  to  they  in  the  context  of  the  experiment  had  no  effect  on  frequency,  accuracy,  or  pronunciation of pronouns. This provides the first evidence of how nonbinary they is used in a naturalistic storytelling context and shows that while it poses some minor difficulties, it can be used successfully in a supportive context. 

The study investigates the role of constituent order in structural priming. We report the results from a PO/DO priming experiment in German, in which we experimentally manipulated verb position in primes and targets. Significant structural priming effects occurred irrespective of whether verb position was the same in prime and target or not. However, additional similarity in constituent order was able to boost structural priming effects, with significantly stronger priming when the verb occurred in the same position in prime and target. We argue that existing one-stage and two-stage accounts of formulation struggle to account for the entire data pattern and propose an alternative account of formulation which can explain our results.

Sentence acceptability judgments are often affected by a pervasive phenomenon called satiation: native speakers give increasingly higher ratings to initially degraded sentences after repeated exposure. Various studies have investigated the satiation effect experimentally, the vast majority of which focused on different types of island-violating sentences in English (sentences with illicit long-distance syntactic movements). However, mixed findings are reported regarding which types of island violations are affected by satiation and which ones are not. This article presents a meta-analysis of past experimental studies on the satiation of island effects in English, with the aim of providing accurate estimates of the rate of satiation for each type of island, testing whether different island effects show different rates of satiation, exploring potential factors that contributed to the heterogeneity in past results, and spotting possible publication bias. The meta-analysis shows that adjunct islands, the Complex NP Constraint (CNPC), subject islands, the that-trace effect, the want-for construction, and whether-islands reliably exhibit satiation, albeit at different rates. No evidence for satiation is found for the Left Branch Condition (LBC). Whether context sentences were presented in the original acceptability judgment experiments predicts the differences in the rates of satiation reported across studies. Potential publication bias is found among studies testing the CNPC and whether-islands. These meta-analytic results can be used to inform debates regarding the nature of island effects and serve as a proof of concept that meta-analysis can be a valuable tool for linguistic research.

One kind of temporal inference in discourse operates over iconicity, such that inferred temporal order follows reported order. In two preregistered experiments (combined N = 930), we asked whether this temporal inference is predictably modulated by linguistic eventuality. Based on event-structural theories of temporal interpretation, stative descriptions, corresponding to cognitively less salient states in the world, should serve as backgrounds for eventive descriptions, locating states earlier in time. Participants read descriptions like Mary got/was married to John. She got/was pregnant and indicated which happened first. Eventuality type of both sentences and reported order were crossed. We find that states tend to be ordered before events, and longer states before shorter states. Our results support a model of discourse comprehension in which eventuality framing is crucial for (temporal) inferences.

The aim of the current study is to assess the impact of the wider discourse on pronoun interpretation. We specifically look at German demonstrative pronouns (dieser) in comparison to personal pronouns (er), investigating whether dieser-demonstratives are influenced only by factors in the preceding sentence (specifically, sentence topicality) or whether they are additionally influenced by cues from the wider discourse (i.e., discourse topicality). We found that discourse topicality competes with sentence topicality for prominence, when the two cues are not aligned to one and the same referent. This had an impact on referential interpretation of both personal and demonstrative pronouns, with weakened interpretive biases when sentence and discourse topic did not converge on the same referent (Exp. 3). Our data further indicate that the introduction of a protagonist from a well-known novel blocked the emergence of the discourse topic as a prominence-lending cue for personal pronouns (Exp. 1–2). We propose that reference to the famous protagonist triggers a protagonist layer, which introduces its own set of questions under discussion, which in turn invalidates the discourse topic. Crucially, the demonstrative pronoun dieser does not consider the protagonist layer and only relies on the discourse layer for interpretation.

In addition to referential uses, pronouns can have covarying interpretations, i.e., exhibit the behavior of a bound variable. The grammatical mechanism(s) behind such readings have been subject to longstanding debates: Some authors argue for a fairly flexible but unified semantic mechanism that is not tied closely to syntactic configurations. Others distinguish a core class of bona fidebinding with tight syntactic constraints from other mechanisms that give rise to ultimately parallel effects, but do so more indirectly. Psycholinguistic work has started to uncover the processing mechanisms involved in evaluating dependencies between covarying pronouns and (candidate) antecedents. Moulton and Han (2018) leverage the processing perspective to try to shed light on the theoretical question of what grammatical mechanism is at play for a given covarying pronoun. They argue that so-called Gender Mismatch Effects only arise for cases of bona fide binding, supporting the existence of distinct grammatical mechanisms. However, Kush and Eik (2019), looking at another construction involving the relevant other covariation mechanisms, do find Gender Mismatch Effects. These authors suggest that various contextual factors can make a covarying interpretation harder to obtain, and they propose adjustments to Moulton and Han’s stimuli that they think should lead to fast Gender Mismatch Effects even when no bona fide binding is involved. A series of self-paced reading experiments replicate the results from Moulton and Han, and then extend the paradigm to variations along the lines suggested by Kush and Eik. The adjustment of contextual factors indeed results in Gender Mismatch Effects for both environments. We discuss how the processing evidence informs the theoretical issues.

This paper surveys the linguistic diversity in psycholinguistic and neurolinguistic research by examining the languages under investigation in major international conferences from 2012 to 2023. The results showed that these studies are highly skewed towards English in particular and Indo-European languages in general. However, the overall number of languages, as well as the number and proportion of Indo-European (other than English) and non-Indo-European languages increased over time, indicating that language processing research is becoming more and more diversified. This typological bias was also found in the inspection of specific linguistic phenomena: (a) morphosyntactic alignment, richness of case morphology, canonical word order, and (b) temporal concepts. The analyses of typological bias at the general and specific levels indicate that there are gaps in various topics, and these can be filled by including more non-Indo-European languages in the investigation process. In addition, a sociolinguistic bias in language processing research emerges as the languages investigated are more often ‘Western’ languages with more than one million speakers and a shared written form. These results reflect the numerous challenges encountered when conducting experiments on less familiar languages, such as the geographic difficulty finding participants speaking these languages, the need of institutional support, as well as the difficulties in setting up collaborations with native speakers, among others.

The de re/de dicto ambiguity centers on the referential and/or attributive properties of noun phrases within the scope of intentional operators, such as belief reports. For instance, in the belief report "Julie believes Elizabeth’s poem will win the competition," a de re reading of the embedded referential noun phrase "Elizabeth’s poem" entails that the referential association between this noun phrase and the target poem is true from the speaker's perspective but may not be recognized as such in the belief holder’s (i.e., Julie’s) mind. In contrast, a de dicto reading describes Julie’s beliefs as she understands the referential association in her mind. While both de re and de dicto readings of definite noun phrases are considered acceptable, given different supporting contexts, we show that the acceptability of de re readings is vulnerable to contextual and pragmatic manipulations. One such case involves a context in which the belief holder, Julie, holds a mistaken belief about the identity of the poem, such as thinking that it was written by Nicole when, in reality, it was written by Elizabeth. This mistaken identity context introduces a de dicto reading of a competing noun phrase, "Nicole’s poem," in "Julie believes Nicole’s poem will win the competition." In this context, the speaker-oriented de re reading of "Elizabeth’s poem" has a roughly bimodal acceptability distribution, while the de dicto noun phrase was overall preferred. Our study is the first to systematically lay out the empirical landscape of de re/de dicto readings of definite noun phrases and highlights the vulnerability of the de re reading. This investigation solidifies the foundation for further theory development and endorses the practice of collecting reliable empirical judgment data for nuanced semantic phenomena.

Previous studies on the use of morphosyntactic gender cues for linguistic prediction show that non-native speakers’ use of grammatical gender information is influenced by various factors. In the present study, we examined the influence of differential cross-linguistic influence (DCLI), knowledge of L2 lexical gender, gender congruency, and L2 fluency. To this end, we investigated L1 Oromo L2 Amharic speakers as well as L1 Amharic speakers, using the Visual World Paradigm (VWP) and supplementary offline experiments. We investigated two groups of L2 Amharic speakers, i.e., L1 Eastern Oromo L2 Amharic and L1 Western Oromo L2 Amharic speakers. The Eastern Oromo dialect patterns with Amharic in terms of gender agreement unlike the Western Oromo dialect which does not have grammatical gender. Analyses of the participants’ proportion of eye fixations show that early exposure to the gendered Eastern Oromo dialect facilitates predictive L2 gender processing. L2 fluency, the speakers’ knowledge of L2 lexical gender, and specific properties of the gender cues modulate predictive L2 gender processing. However, there is no significant influence of lexical gender congruency. The study has ecological significance as it presents empirical data from understudied languages.

Language processing relies on memory. There exists a considerable body of literature on retrieval in sentence processing and, in particular, on cases involving recall of syntax-relevant information. There is no reason to doubt, however, that memory is involved in semantic aspects of language processing as well. In this work, we look at the case of additive presuppositions, such as those involved in interpreting the additive particle too. When one hears Mary went to the party, too, one should recall that someone other than Mary went to the party. We make the case that, as a starting hypothesis, it would be expected that the retrieval of this kind of information should share basic features of memory processes in language with the better-known cases of recall involved in syntactic parsing. In particular, we argue that, given certain assumptions and linking hypotheses, all prominent retrieval theories predict the existence of distance effects for the recall of previous information, independent of whether the recall is driven by syntactic or semantic, sentential or inter-sentential, considerations. As the distance increases, so does the difficulty of processing. We test this prediction in four experiments that investigate the role of retrieval in interpreting too. Using the Bayesian hierarchical modelling paradigm, we find evidence in two self-paced reading experiments that it takes more time to read sentences with too when the distance between the trigger and its antecedent is greater, compared to a baseline that lacks the presupposition trigger. This result shows that theories of the role of memory in language are just as relevant to the domain of discourse interpretation as they are to syntactic parsing. The fact that our evidence was relatively hard to find, however, suggests additionally that there are interesting differences to explore in the future.

The extraposition of a relative clause creates a discontinuous dependency between the relative clause and its host noun phrase, as in A man just entered the bank who claimed to have a gun. Since discontinuous dependencies are known to increase processing effort, a key question is why speakers produce them in the first place. Some factors known to affect extraposition – for example, the length of the relative clause and the main verb phrase – have received processing-based explanations, but others haven’t. We focus on two factors described by previous research: verb type and the grammatical function of the noun phrase hosting the relative clause. Specifically, extraposition from grammatical subjects is more common with unaccusative and passive verbs in English; further, extraposition is more common from grammatical objects than subjects in Dutch and German. We replicate these findings using corpus data from Persian. Further, we propose that verb type and grammatical function can be linked to a single underlying notion: argument structure. We demonstrate that argument structure modulates the likelihood of extraposition in Persian. We suggest that this occurs because speakers choose to extrapose relative clauses in order to keep the main clause verb close to its internal arguments. This explanation extends previous findings in psycholinguistics on the role of argument structure in speech planning during language production.

Repeated exposure to Subject Island violations can lead to increased acceptability ratings and faster reading times (Chaves & Dery, 2014, 2019; Clausen, 2011; Francom, 2009; Hiramatsu, 2000; Lu et al. 2021; Lu et al., 2022). However, it remains unclear what the nature of this effect and the driving mechanism is. The present paper describes a longitudinal investigation to test whether the effect of repeated exposure to Subject Island constructions is short-lived or whether it can spread over three weeks’ time, as measured by offline measures (Likert acceptability ratings) and online measures (self-paced reading). Using more observations and more sensitive methodologies, our work builds and improves on the only previous longitudinal study on such islands, Snyder (2022). We uncover evidence suggestive of gradual and strategic (by-construction and by-region) adaptation to Subject Island violations, indicated by faster response times, as well as higher acceptability ratings following repeated exposure, most consistent with a syntactic adaptation effect over and above task adaptation.

This paper investigates the relationship between morphological productivity and neological intuition, defined as the ability to identify novel words as such. It can be hypothesised that the more productive a word-formation process is, the less salient the neologisms it forms will be. We test this hypothesis experimentally on neologisms formed with prefixes and suffixes of variable productivity. Three experiments are conducted, involving lexical identification and reading tasks with eye tracking, to provide a comprehensive description of neological intuition. The negative correlation between productivity and neological salience is supported by experimental results, but only in the case of suffixed neologisms, as opposed to prefixed ones. The effect of affix type on neological intuition can be explained by differences in the grammatical nature of prefixes and suffixes. Broadly speaking, investigating the linguistic factors of neological intuition provides an original approach to both linguistic and psycholinguistic issues related to word structure and lexical processing.

Non-linguistic factors leave a distinct thumbprint on our speech production that is perceptible to listeners. A steadily growing line of research demonstrates that listeners can perceive a contrast between native and non-native (L2) speakers based on accents and further classify these speakers according to dialectal variation, even when they are not native speakers of a language. Most of these studies have focused on dialectal variation within US English speakers, a combination of US and International English dialects, or L2 speakers representing a wide range of languages. Most have also featured listeners who are monolingual native speakers of the target language coming from a homogenous background, or a contrast between these and a targeted set of L2 speakers. We therefore lack knowledge of how exposure to, or familiarity with, diverse accents and languages, or specific native language competence of the native language of L2 speakers, can guide listeners’ accent perception and categorization. In this research, we employed a free classification task, presenting listeners with speech samples of native speakers with accents representing multiple English dialects, and L2 speakers of nine Asian languages across three geographic regions speaking Asian-accented English. There were six groups of listeners: monolingual US English listeners in a diverse linguistic context, monolingual US English listeners in a homogeneous linguistic context, native speakers of a non-Asian language and English (bilinguals), and native speakers of each of the three target Asian language groups who are L2 speakers of English. The results reveal that nearly all listeners are sensitive to accents capturing native/L2 contrasts and dialectal variation in English. While regular exposure to a diversity of accents results in increased classification accuracy, classification of Asian L2-accented English speakers is best performed when there is alignment of similar language family and geographic area, as demonstrated by South Asian listeners.

Unlike the gender-based systems of noun categorization in many European languages, numerous semantic categories contribute to Bantu noun class systems. Kinyarwanda, the focus of our study, has a rich inventory of noun class prefixes, but it is unknown to what degree the semantic and morphological systems underlying these noun classes influence how speakers mentally categorize nominals in their language. To investigate this, speakers of Kinyarwanda (n = 46) were recruited to take part in an online triadic comparison experiment. Across 144 trials, participants were asked to identify the item most different from a written list of three nouns. These lists were constructed based on morphological similarity (from noun classes 3, 5, 7, or 9), semantic overlap (from the domains of ‘mammals’ and ‘tools’), or both. Results show an overall preference for semantic grouping in the triads, although the strength of these preferences differed across individuals. This variation turned out to be systematic and predictable: speakers of Kinyarwanda who spoke Kiswahili as an additional language generally preferred categorizing on the basis of noun class, while those who did not speak Kiswahili as an additional language were more likely to base their decisions on the shared semantic domains of the nouns. These data suggest that noun categorization choices in Kinyarwanda can be influenced by knowledge of other linguistic systems, highlighting the impact that learning additional languages may have on first-language lexical knowledge.

Research has shown that people anticipate upcoming linguistic content, but evidence regarding expectations of specific lexical markers is mixed. We use the Dutch pair of discourse markers "Aan de ene kant...Aan de andere kant" (“On the one hand...On the other hand”) and "Enerzijds...Anderzijds" (also equivalent to “On the one hand...On the other hand”) to test whether readers generate predictions of an upcoming upcoming contrast dependency based on the lexical marker for the first contrastive perspective, and whether such predictions focus on a lexical form or rather on a discourse relation. In a self-paced reading, we show that readers do generate expectations for upcoming discourse markers, but that these expectations are not specific to a lexical form. In an eye-tracking study, we replicate the facilitative effect of the first marker of a lexical pair on the processing of the second marker, and show that this effect occurs in immediate processing. These results establish expectation-driven effects at the discourse level in the earliest possible reading time measures, showing comprehenders’ awareness of the discourse dependency established by a discourse marker along with their flexibility in identifying and integrating discourse relations with different markers.

Deletion-based accounts of verb phrase ellipsis (VPE) predict that this construction requires a syntactically identical antecedent, but previous research shows that some antecedent-target mismatches are perceived as relatively acceptable in experiments (see e.g. Arregui et al., 2006; Miller & Hemforth, 2014). So far, the acceptability of these mismatches has been explained mostly by licensing conditions on VPE or by ellipsis-specific processing mechanisms. This article explores to what extent the acceptability of mismatches follows from the more general principles of an information-theoretic account of language use, which has been independently evidenced for other omission phenomena: To avoid under- or overutilizing the hearer’s processing resources, predictable VPs are more likely to be omitted, whereas unpredictable ones are more likely to be realized. This hypothesis is tested with three experiments that investigate a gradual acceptability cline between VPE mismatches which has been reported by Arregui et al. (2006). First, an acceptability rating study replicates the overall pattern found by Arregui et al. (2006) and confirms that the effect is specific to ellipsis. Second, a production task shows that the acceptability differences are indeed related to a gradual decrease in the predictability of the target VP, which is also reflected in the likelihood of participants producing VPE. Finally, a self-paced reading experiment shows that VPE is more acceptable when it is easier to process. Overall, the experimental results support the information-theoretic account and suggest that no specific syntactic constraints or reconstruction mechanisms might be required to account for the acceptability cline observed for the mismatches investigated.

Pronunciation variation is systematic, and provides listeners with cues to what the speaker is about to say. Shortened stems, for example, can indicate an upcoming suffix, while lengthened ones can indicate a word boundary follows. Previous work has shown that listeners draw on these cues to distinguish polysyllabic words, like rocket, from monosyllabic words, like rock. This strategy is useful in morphological processing, as additional morphological structure often adds additional syllables. The current study asks (i) whether listeners use these cues to distinguish words that differ only in morphological structure with no change in syllable count (e.g., rock/rocks); and (ii) how surrounding morphosyntactic context affects listeners’ ability to use these cues. Ideal observer models predict that listeners should be attentive to phonetic detail in all contexts regardless of how much new information it offers, while the strategic listener account allows listeners to dynamically adjust their attentiveness to phonetic detail based on its information value in context. In a visual-world eye-tracking study, English-speaking listeners were presented with utterances containing target nouns whose stem durations were manipulated to provide cues to the presence or absence of (a) a plural suffix (rock vs. rocks) or (b) a second, non-morphological syllable (rock vs. rocket). These words were embedded in two contexts: (i) preceded by agreeing determiners, which rendered stem duration cues redundant for predicting the presence or absence of a suffix (this rock/these rocks), and (ii) preceded by non-agreeing determiners (the rock(s)), where stem duration cues carried more information. The results are consistent with ideal observer models: listeners are highly attentive to all acoustic detail, and especially so when it is predictable (and hence redundant), as long as they have the cognitive resources to handle it.

While maximum standard absolute adjectives (such as straight) typically have a precise meaning (e.g., ‘perfectly straight’), they are also regularly used imprecisely (e.g., to mean ‘straight enough’). The current study investigates how contextual expectations of precision and a visual referent’s conceptual distance from an ideal maximum standard influence the processing effort of precise and imprecise interpretations of these adjectives. In three experiments, we showed native speakers of English images depicting objects that could be referred to precisely or imprecisely via an absolute adjective and asked them to select the image that best matched the written sentence (Experiments 1 and 2) or to read sentences containing maximum standard absolute adjectives (Experiment 3). Experiment 1 presented no discourse context, and participants accepted, on average, only a small degree of imprecision; and when they did, they took longer, relative to cases in which the same adjectives were used precisely, which is in line with existing empirical findings. Experiment 2 contrasted two kinds of discourse contexts (raising high or low expectations of precision) before the presentation of the test sentences. When expectations of precision were high, participants tolerated only a small degree of imprecision, and when they did, it came at a cost, as in Experiment 1. When expectations of precision were low, much larger degrees of imprecision were tolerated but, critically, participants were still, overall, faster to reach precise, relative to imprecise, interpretations in supporting contexts, suggesting that accessing the precise meaning is less effortful. Experiment 3 supported these findings by showing how the cost of understanding imprecision is also present in a self-paced reading task. Our results lend support to the view that maximum standards are part of the encoded meaning of these adjectives.

Recent research has highlighted that lexical scales vary in their likelihood of giving rise to a scalar inference—a finding labeled scalar diversity. The current paper examines the role of intonation for this phenomenon, which has thus far primarily been studied using written materials. A specific focus in this regard was on the so-called rise-fall-rise contour, which has been argued to (i) convey uncertainty, which could have an influence on scalar inference calculation, and (ii) be sensitive to properties of lexical scales, which could interact with factors driving scalar diversity. Experiment 1 combined production with an inference task to assess the likelihood of different intonational contours, as well as how a given contour affects scalar inference rates. Production of the rise-fall-rise varied across lexical scales, as expected, and led to an increase in scalar inference derivation relative to a fall. The latter finding was further confirmed in Experiment 2, which explicitly manipulated intonational contours in the inference task. The results, thus, show the importance of taking intonation into account when studying scalar diversity and scalar inference more generally, and they also have implications for theories of the rise-fall-rise contour. Additionally, the experiments revealed a contour that is prosodically similar to the so-called Contradiction Contour, but appears to serve a different pragmatic function.