Document Type


Date of Award


Degree Name

Master of Science (MS)



First Advisor

Donovick, Peter J.

Second Advisor

Johnson, Matthew D.

Third Advisor

Kurtz, Kenneth J.


This study used the Valence-Expectancy Learning model, a mathematical cognitive model, to dissect young adult performance on the Iowa Gambling Task (IGT). Drinking behavior and monetary incentive were examined as predictors of performance on the IGT. No differences were found among groups when data were analyzed using traditional behavioral analyses. However, when the Expectancy-Valence Learning model was applied to the data, differences between groups were found related to attention and choice consistency. Importantly, the cognitive model was not a good fit for fifty-seven percent of the data, meaning that it did not succeed in explaining how the participants' choices were dependent on trial-by-trial feedback. This may indicate that many young adults in this study were not invested in the task.