The learning rate and convergence of connectionist learning algorithms are often dependent on their parameters. Most algorithms, if their parameters have been optimized at all, have been optimized by hand. This leads to absolute and relative performance problems. In absolute terms, researchers may not be getting optima] performance from their networks. In relative terms, comparisons of unoptimized or hand optimized algorithms may not be fair. (Sometimes even one is optimized and the other not.) This paper reports data suggesting that comparisons done in this manner are suspect. An example algorithm is presented that finds better parameter sets more quickly and fairly. Use of this algorithm (or similar techniques) would improve performance in absolute terms, provide fair comparisons between algorithms, and encourage the inclusion of parameter set behavior in algorithmic comparisons.

In order to ground our understanding of cognitive development we have started to create a model of how children and adults solve a well-studied three-dimensional puzzle. We started with a model that fits the adult behaviour on the puzzle. We then modified the model's cognitive architecture (ACT-R) and its perceptual/motor architecture (the Nottingham Interaction Architecture) in three ways to simulate a younger problem solver by: (a) reducing the accuracy of vision, (b) reducing working memory, and (c) doing both. The modifications, particularly reduced working memory (and its combination with reduced visual accuracy), allow the model to approximate, on some measures, the behaviour of seven year olds on the puzzle. The results suggest that cognitive models and their architectures can help answer the question of "What develops?".

In this study, we investigate how high- and low-performance learners (N=12) act differently while using a cognitive tutoring system. We examine three research questions: (1) Can we predict learners’ performance using only their visual attention (eye movement data)? (2) Can we predict learners’ performance from visual attention data and initial performance? (3) Are age, gender, first language, where they look, and the sequence of Areas of Interests (AOIs) significant factors in the learners’ performance? Learners more correctly answer questions taken from larger rather than smaller AOIs. Our results show that high-performance learners pay more attention to the content that contains answers to later questions. Surprisingly, the tutor did not change the learners’ visual search to a goal-oriented search. Our analyses can help instructional designers create a more productive learning experience because visual search behavior as part of a learner model with acceptable accuracy in early stages can be used in adaptive tutors. Additionally, we trained a classifier on the eye movement data to predict learners’ performance for each question. Its results provide a list of suggestions for designing more productive learning experiences, such as enticing user attention by increasing the size of the content that contains answers and changing the order of contents.

Using a process model of skill acquisition allo- wed us to examine the microstructure of subjects' performance of a scheduling task. The model, im- plemented in the Soar-architecture, fits many qua- litative (e.g., learning rate) and quantitative (e.g., solution time) effects found in previously collec- ted data. T h e model's predictions were tested with data from a new study where the identical task was given to the model and to 14 subjects. Again a general fitof the model was found with the restrictions that the task is easier for the m o - del than for subjects and its performance impro- ves more quickly. T h e episodic memory chunks it learns while scheduling tasks show h o w acquisition of general rules can be performed without resort to explicit declarative rule generation. T h e model also provides an explanation of the noise typically found when fittinga set of data to a power law — it is the result of chunking over actual knowledge rather than "average" knowledge. Only when the data are averaged (over subjects here) does the smooth power law appear.

How do covert emotional stimuli affect decision-making? We investigated this question by exposing par-ticipants to subliminal visual stimuli during a computer-ized version of the Iowa Gambling Task (IGT) to assesswhether different categories of images (negative, neutral,or positive emotional evaluations) would influence deci-sion-making behavior. Results did show sex-group inter-actions for IGT scores. In decision learning model simu-lations, it was found that different models were more ap-propriate to explain the task performance for differentsex-group pairs. Overall, women showed more of anability to integrate the additive negative signals from thestimuli to make more advantageous decisions than themen; consequently, this made the men more resilient tothe negative effects of the positive stimuli on task-performance. When taken with existing research, the re-sults indicate that subliminal emotional stimuli can havesubtle, potentially sex-dependent, effects on behaviorduring the decision-making process.