Computational-mathematical modeling of decision-making using Iowa Gambling Task based on Cognitive Inputs

Investigating how individuals' decision-making is influenced by other cognitive elements and using computational modeling of decision can help us to better appreciate this cognitive function, as well as better decision-making quality.  Thepresent study aimed to present a new cognitive model in the field of decision-making and to examine the model’s efficacy in predicting the decisions of those participating in the Iowa decision test compared to other classical decision- making models.

In this study, 56 subjects, including 20 men and 36 womenwith an average age of 43.52, were asked to participate in the Iowa gambling task. The results of the model used were then compared with the results of the three well-known decision- making models, including expected value, expected utility, and prospect model.. The flexibility of this model by calculating the impact factor of each concept for each individual, allows us to model each person's decision-making individually and identify the concepts, which are most effective in each individual's decision-making.

The obtained results revealed that the expected value, expected utility, prospect, and the proposed cognitive connectionist models predicted 36.04%, 42.46%, 49.18%, and 73.02% of subjects’ decisions, respectively.

As a result, it can be argued that the proposed cognitive connectionist model has more potential for modeling examinee’s behavior in the Iowa test. It also provided a reasonable way to study the essential cognitive elements that can affect the participant’s decisions.