How sequential changes in reward expectation modulate cognitive control: Pupillometry as a tool to monitor dynamic changes in reward expectation

Much evidence suggests that positive affect associated with performance-contingent reward modulates cognitive flexibility and stability. For example, in voluntary task switching, it has been shown that unchanged high reward promotes cognitive stability whereas increases or decreases in reward prospect as well as unchanged low reward promote cognitive flexibility (higher voluntary switch rate, VSR). Pupil diameter has been shown to respond to reward prospect, to task switching manipulations, and more generally to effort in cognitive control tasks and thus appears to be the ideal tool to learn more about the processes underlying these reward-modulated decisions. Therefore, we measured pupillary activity in two voluntary task switching experiments with randomly changing reward magnitudes. Behaviorally, VSR was again lowest when reward remained high as compared to all other reward sequences. Baseline pupil diameter was generally higher following switch trials as compared to repetition trials. Furthermore, the pupil responded dynamically to the reward manipulation: Phasic cue- and target-locked pupil dilation was larger in the reward phase as compared to the non-reward baseline block. Most importantly, phasic pupil dilation in the target interval was highest when reward prospect increased and lowest when reward prospect decreased, suggesting that motivational arousal fluctuates in sync with changes in reward expectation. These results are discussed with respect to current theories on cognitive control as a form of affect regulation.