| Last Updated:
Creating DOI. Please wait...
Human beings have the ability to express confidence judgements on their own decisions. In case of perceptual decisions, confidence has been thought to depend upon the quality of the sensory input. In a recent article, Steve Fleming and colleagues questioned this view (Fleming, S.M., Maniscalco, B., Ko, Y., Amendi, N., Ro, T. & Lau, H. (2015) Action-specific disruption of perceptual confidence. Psychological Science). They manipulated response-specific representations in the premotor cortex with transcranial magnetic stimulation (TMS) in a perceptive task. They show that stimulating the motor representation associated with the unchosen response reduced confidence in correct answer, thereby demonstrating that « action-specific information in the premotor cortex contributes to perceptual confidence ».
It has been shown that in conflict tasks (like Simon or Eriksen flankers tasks) subjects make «partial errors»: these trials correspond to sub-threshold (yet detectable by electromyographic recording) motor responses that are successfully inhibited before they turn into an overt error (Hasbroucq, T., Possamaï, C. A., Bonnet, M., & Vidal, F. (1999). Effect of the irrelevant location of the response signal on choice reaction time: An electromyographic study in humans. Psychophysiology). Interestingly, it has been found that the supplementary motor area (either the SMA or pre-SMA) provides cognitive control signals to the primary motor cortex to exert online inhibition and in turn rectify the course of initial (incorrect) action (partial errors) (Roger, C., Núñez Castellar, E., Pourtois, G., & Fias, W. (2014). Changing your mind before it is too late: The electrophysiological correlates of online error correction during response selection. Psychophysiology). Similar EMG patterns (partial errors) have been observed for discrimination. Crucially, the link between confidence and partial error has not been studied yet. The reason is probably that partial errors have mainly been studied in conflict tasks (Simon or Eriksen) for which there is no post-decision ambiguity about the correctness of the provided answers.
This leads us to the question we aim to study: does the presence of partial errors modulate confidence in a detection task?