Uncertainty coding in the reward system

This chapter provides an overview of the neural mechanisms implementing decisions under risk and ambiguity. Risk refers to situations where probabilities of the outcomes are known and ambiguity refers to situations where the exact probability is unknown but has to be estimated. We review two major approaches to formalize the decision variables for decisions under risk: probability-weighting of valued outcomes vs. combining the moments of the distribution of outcomes (mean-variance-skewness). The neural literature suggests that cortico-striatal systems receiving dopaminergic input such as the striatum and medial prefrontal cortex process components from both approaches. Common processing of both risk and ambiguity by these brain regions suggests that they play a general role in the valuation of choices under uncertainty. However, theoretical and empirical advances are required to account for the continuum of uncertainty (from completely unknown to explicitly represented probability) and for how our brains allow us to move along this continuum.