An Adaptive Signal Detection Model applied to Perceptual Learning
- Percy Mistry, University of California Irvine, Irvine, California, United States
- Joshua Skewes,
- Michael Lee, UC Irvine, Irvine, California, United States
AbstractWe introduce a new model of adaptive criterion setting within a signal detection framework, and show how this provides psychological insights that allow us to segregate causes of sub-optimality in perceptual learning. We apply this to a perceptual learning task for both neurotypical and autistic participants. The model parameters provide a bridge between the mechanisms of an aberrant precision account of autism and resulting behavior that can be interpreted within a receiver operating characteristic framework. The model makes superior out-of-sample predictions compared to standard signal detection theory, about how people adapt to different environmental manipulations when asked to categorize audio-spatial stimuli. We find that accuracy of participants is more strongly correlated to the construct of persistence signals that inhibit response flexibility, than to the neuromodulatory gain. We also find evidence for individual differences in persistence that are correlated to scores on the autistic traits questionnaire.
Return to previous page