Edinburgh Research Explorer

Maladaptive decision-making in a rat version of the Iowa Gambling Task

Research output: Contribution to conferencePoster

Original languageEnglish
Number of pages1
Publication statusPublished - Feb 2012
Event9th Annual Annual Computational and Systems Neurscience Meeting (COSYNE 2012) - Salt Lake City, United States
Duration: 23 Feb 201226 Feb 2012

Conference

Conference9th Annual Annual Computational and Systems Neurscience Meeting (COSYNE 2012)
CountryUnited States
CitySalt Lake City
Period23/02/1226/02/12

Abstract

Deficits in decision-making have been repeatedly observed in various psychiatric disorders (e.g. ADHD, Mania, OCD) and are often assessed using the Iowa Gambling Task (IGT). The IGT represents a realistic decisionmaking task where subjects have to choose between targets associated with rewards and penalties of varying likelihood and amplitude. Previous studies have shown that a third of healthy subjects perform poorly in the IGT, as observed in psychiatric patients [1]. Recently, the IGT was adapted for rodents (the Rat Gambling Task, RGT).
As in human studies, a third of healthy rats were found to exhibit poor decision-making [2]. These rats were then run on a battery of tests to extract measures of impulsivity, reward sensitivity, behavioral inflexibility and riskseeking.
Poor decision-makers were always characterized by high scores for a combination of these behavioral traits. We modified the TD-learning algorithm to model learning and decision-making in the RGT and include reward sensitivity, inflexibility and risk-seeking. This novel model was then used to assess: (1) how the behavioral traits influence learning (2) Whether they can they explain different performances in healthy subjects. The model was able to account for the performances of good and poor decision-makers. The model was fitted to individual rat performances to describe their levels of reward sensitivity, inflexibility and risk-seeking. The parameters correlated significantly with the scores obtained from experiments assessing these behavioral traits. This suggests that the mathematical description of the traits is valid. This work supports the hypothesis that a combination of high scores for reward sensitivity, inflexibility and risk-seeking affects the rats’ learning by altering reward prediction and their ability to reverse their initial estimations. Biased perception and representation of the environment lead to aberrant decisions according to the real outcome of the task but optimal according to the rat’s internal model.

Event

9th Annual Annual Computational and Systems Neurscience Meeting (COSYNE 2012)

23/02/1226/02/12

Salt Lake City, United States

Event: Conference

ID: 15113664