Abstract
Reward and motivation have powerful effects on cognition and brain activity, but less is known about how they impact sustained attention and the neural mechanisms supporting accurate performance over time. Using a sensitive measure of sustained attention, the gradual onset continuous performance task (gradCPT), we recently showed that multiple types of motivators substantially improved accuracy and reduced variability in performance. To investigate how both transient and sustained activity in task-positive attention networks support these behavioral effects, participants completed multiple fMRI runs of the gradCPT, where minute-long blocks alternated between a rewarded (motivated) and unrewarded (unmotivated) condition. During motivated blocks, task-positive regions in attention networks exhibited sustained increases in activity relative to unmotivated blocks, and further, these increases predicted future performance from moment-to-moment. This suggests that reward elicits proactive, sustained cognitive control. During unmotivated blocks, activation in these task-positive attention regions was restricted to transient responses to errors and periods of more erratic and error-prone performance (“out of the zone”). This suggests that without reward, subjects use a more miserly strategy, in which task-positive control is only summoned reactively, when most needed. Together, this demonstrates that greater activation in task-positive attention networks can be both an indicator of optimal and suboptimal performance. In contrast, we found that basal ganglia activity best tracked optimal performance across both intrinsic, within-block fluctuations (in and out of the zone) and reward-driven between-block fluctuations (motivated vs. unmotivated). These results support a nuanced view of how task-positive attention networks and subcortical regions are engaged strategically during both intrinsic and reward-driven fluctuations in sustained attention.
Meeting abstract presented at VSS 2015