Introspection and decision-making: value-based decision-making and confidence computations

National Institute Of Mental Health

Investigators

Abstract

During the fiscal year, in this project (protocol MH000944) we have conducted an initial study with a total of 151 online participants. Each participant has completed a series of decision-making tasks that span the domains of perceptual and value-based choice, and reported their confidence level after every decision. We have applied computational modeling techniques to derive individual-level parameters for preferences for risk, delay, decision sensitivity, and confidence quality metrics. Our preliminary results have been presented at a number of intra-NIH and outside conferences as talks and poster presentations. They have informed the design of a new study also to be conducted online but that applies state-of-the-art techniques in adaptive trial design to improve the modeling of these behaviors.