Before making a reward-based choice, we must evaluate each option. Some theories propose that prospective evaluation involves a reactivation of the neural response to the outcome. Others propose that it calls upon a response pattern that is specific to each underlying associative structure. We hypothesize that these views are reconcilable: during prospective evaluation, offers reactivate neural responses to outcomes that are unique to each associative structure; when the outcome occurs, this pattern is activated, simultaneously, with a general response to the reward. We recorded single-units from macaque orbitofrontal cortex (Area 13) in a riskless choice task with interleaved described and experienced offer trials. Here we report that neural activations to offers and their outcomes overlap, as do neural activations to the outcomes on the two trial types. Neural activations to experienced and described offers are unrelated even though they predict the same outcomes. Our reactivation theory parsimoniously explains these results.
Bibliographical noteFunding Information:
We thank M. Mancarella and M. Castagno for helping with data collection and R. Akaishi for useful comments on the manuscript. This research was supported by a grant to B.Y.H. from the Klingenstein-Simons Foundation and NIH R01 DA037229: Neural Basis of Reward-based Choice.
© 2017, The Author(s).