Error Rate and Outcome Predictability Affect Neural Activation in Prefrontal Cortex and Anterior Cingulate during Decision-Making

Abstract

Decision-making in the presence of uncertainty is a complex process that involves both affective and cognitive factors. Both error rate and predictability have been implicated in the process of response selection during decision-making. This study examined the hypothesis that the rate of errors during decision-making differentially affects the activation in prefrontal and cingulate cortex. BOLD echo-planar signal intensity was obtained during a two-choice prediction task across 90-s blocks with 20, 50, or 80% error rates. This study yielded three main findings. First, at chance level error rates, activation of the right dorsolateral (BA 9, 46), inferior prefrontal (BA 44), and precuneus (BA 7) during the two-choice prediction task replicated the finding previously reported. Second, premotor (BA 6) and parahippocampal (BA 36) areas were relatively more active at high error rates, and dorsolateral (BA 9, 46) and inferior prefrontal cortex (BA 44) as well as parietal (BA 40) and cingulate cortex (BA 25, 32) were more active during low error rates. Third, the relationship between the frequency of the dominant strategy underlying decision-making (win–stay/lose–shift) and the activation in the dorsolateral prefrontal cortex and the anterior cingulate was dependent on error rate or outcome predictability. These results support the hypothesis that error rates and predictability affect the activation patterns in the neural systems underlying decision-making because these structures maintain a representation of the reinforcement history for the available response alternatives to select an “optimal strategy.”