Résumé : Extensive research has described how humans make decisions and compute the likelihood that their decisions are correct. This capacity to access and monitor one’s own mental states is usually referred to as metacognition. Here, I will present several experiments examining the contribution of sensorimotor signals to metacognition. In each experiment, participants were asked to take perceptual decisions and provide confidence estimates regarding the accuracy of each of them. Behavioral (e.g., response time, trajectory) and electrophysiological (e.g., cortical EEG, subcortical single units) signatures of sensorimotor activity were collected while participants made their decision. Confidence estimates were found to correlate with sensorimotor activity surrounding decisions, indicating that sensorimotor signals may serve as input to judge the accuracy of one’s own decision, in line with post-decisional origins of confidence. Based on these results, I will argue that metacognition does not operate in a vacuum but relies on the monitoring of signals from the body, and more specifically, on motor signals involved during the execution of an action when committing to a decision.
Lieu : BSHM A6