Action selection based on numerical monitoring of self-action

Abstract Selection of action constitutes a central part of voluntary motor control. We here deal with action selection based on information generated in the brain, rather than follow external instructions. Specifically, we studied neural mechanisms involved in internal monitoring of self-action to judge the number of actions performed by subjects, when that information was necessary to select the next action. We recorded cellular activity in the frontal and parietal cortex of monkeys performing selection of action solely on the basis of the number of performed actions. We found that cellular activity in the superior parietal lobule reflected the number of self-motion executions, indicating participation of that area in accumulation of numerical information. Such activity was less common in the frontal cortex. On the other hand, when the animals finished the required number of actions and were preparing to select the next action, activity of prefrontal cells reflected that process. These observations suggest differential use of cortical areas: generation of numerical information about self-action in the parietal lobule and adaptive shifting of future actions based on that self-generated information prompted by the prefrontal cortex.