Self-regulation of tasks under dynamic conditions

In changing environments we need to dynamically re-organize our behavior in order to reach desired goals. This requires strategic switches between different tasks. However, little is known about the neural basis of such adaptation processes. We developed a novel, motivated choice paradigm and used multivariate pattern analysis of fMRI data (MVPA) to investigate this. Subjects chose between three different tasks on a trial-by-trial basis. Task difficulty varied independently for all three tasks, increasing for the chosen task set and decreasing for the non-chosen task sets. This created an exploration-exploitation dilemma for which subjects needed to find a solution. To keep difficulty and error rates low, they had to produce a motivated decision on which task to perform next, based on their estimation of the changing environmental variables. Using MVPA, we were able to predict subjects’ choices from medial PFC, and parietal cortex. Furthermore, we could predict the current difficulty level – the motivational factor driving choices – from reward related areas. Interestingly, the ventro-medial PFC (vmPFC) encoded both the choices and the diffi- culty. This area was previously associated with the processing of stimu- lus-values. Our results suggest that vmPFC is also involved in encoding motivated choices. Further planned analyses will investigate the interaction between areas encoding choices and areas encoding motivational factors in more detail. This should provide important clues on how humans endogenously adapt their behavior to cope with changing environments.