As abstracts are sometimes too long to read, I have written one sentence summaries for all 3 CogSci papers that got accepted to the 2015 proceedings:
Assessing the perceived predictability of functions
When participants judge about how predictable different functions are, they treat smoothness as more important than noise as predicted by a rational function learning model (Gaussian Process regression).
Active learning as a means to distinguish among prominent decision strategies
Using active learning as a tool to distinguish between models, we found that people’s active learning behaviour is better described by a heuristic (Take-The-Best) than by an integrative strategy (Logistic Regression).
Learning and decisions in contextual multi-armed bandit tasks
Within contextual bandit tasks participants have to learn and make decisions simultaneously and they seem to do so by probability matching outcomes of rational function learning models.