Human decisions can be shaped by predictions of emotions that ensue after choosing advantageously or disadvantageously. Indeed, anticipating regret is a powerful predictor of future choices. We measured brain activity using functional magnetic resonance imaging (fMRI) while subjects selected between two gambles wherein regret was induced by providing information about the outcome of the unchosen gamble. Increasing regret enhanced activity in the medial orbitofrontal region, the anterior cingulate cortex and the hippocampus. Notably, across the experiment, subjects became increasingly regret-aversive, a cumulative effect reflected in enhanced activity within medial orbitofrontal cortex and amygdala. This pattern of activity reoccurred just before making a choice, suggesting that the same neural circuitry mediates direct experience of regret and its anticipation. These results demonstrate that medial orbitofrontal cortex modulates the gain of adaptive emotions in a manner that may provide a substrate for the influence of high-level emotions on decision making.

Compared with reward seeking, punishment avoidance learning is less clearly understood at both the computational and neurobiological levels. Here we demonstrate, using computational modelling and fMRI in humans, that learning option values in a relative—context-dependent—scale offers a simple computational solution for avoidance learning. The context (or state) value sets the reference point to which an outcome should be compared before updating the option value. Consequently, in contexts with an overall negative expected value, successful punishment avoidance acquires a positive value, thus reinforcing the response. As revealed by post-learning assessment of options values, contextual influences are enhanced when subjects are informed about the result of the forgone alternative (counterfactual information). This is mirrored at the neural level by a shift in negative outcome encoding from the anterior insula to the ventral striatum, suggesting that value contextualization also limits the need to mobilize an opponent punishment learning system.