Intelligence Moderates Neural Responses to Monetary Reward and Punishment

Daniel R. Hawes (University of Minnesota)
Colin G. DeYoung (University of Minnesota)
Jeremy R. Gray (Michigan State University)
Aldo Rustichini (University of Minnesota)

Abstract: The nature of the relation between cognitive skills and economic preferences is a key question in psychology and economics. Examining subject’s intelligence (IQ) and responses to probabilistic feedback during a simple decision task allows us to investigate neural correlates of cognitive ability at the foundational level of gain/loss processing. Our neuroimaging results for 94 subjects show that typical declines in striatal BOLD signal after monetary punishment are significantly less pronounced for subjects with higher IQ. This finding strongly implicates IQ in the ex post processing of decision outcomes, thereby opening up the hitherto unconsidered possibility that cognitive mechanisms underlying outcome evaluation may be central to how intelligence influences preferences and decision, especially attitudes towards risk. We further investigate the role of IQ for outcome evaluation on subject’s behavior on our decision task to demonstrate a correlation between IQ and the extent to which past decision outcomes influence future choices. Specifically we find that larger IQ predicts behavior to be more strongly correlated with an extended period of previously experienced decision outcomes, while lower IQ predicts behavior to be correlated exclusively to the most recent decision outcomes. In addition to showing the moderating effect of intelligence on neural responses to gains and losses, our findings illustrate the existence of a link between intelligence and choice behavior that extends beyond the ex-ante comparison of decision options to include ex-post evaluation of decision outcomes. Importantly, this identified link suggests that observed correlations between intelligence and preferences may be rooted in a unitary process of how decision outcomes are experienced.