Modeling emotional influences on human decision making under risk

We model choices in the Iowa gambling task, developed by Bechara and Damasio to diagnose decision making deficits of patients with damage to orbital prefrontal cortex. In this task, the subjects choose at each trial between four decks of cards: two decks yield high short-term payoffs but long-term losses, and the other two yield lower short-term payoffs but long-term gains. Normals but not patients learn to choose advantageous decks. We simulate this effect with a model based on connections between prefrontal cortex (orbital, medial, and anterior cingulate), basal ganglia, and thalamus, and emotional signals from amygdala.