Performance monitoring for brain-computer-interface actions.
Journal article

Performance monitoring for brain-computer-interface actions.

  • Schurger A Laboratory of Cognitive Neuroscience, Brain-Mind Institute, Department of Life Sciences, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Defitech Chair in Non-Invasive Brain-Machine Interface, Center for Neuroprosthetics, School of Engineering, Ecole Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland. Electronic address: aaron.schurger@gmail.com.
  • Gale S Laboratory of Cognitive Neuroscience, Brain-Mind Institute, Department of Life Sciences, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland.
  • Gozel O Laboratory of Cognitive Neuroscience, Brain-Mind Institute, Department of Life Sciences, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland.
  • Blanke O Laboratory of Cognitive Neuroscience, Brain-Mind Institute, Department of Life Sciences, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne, Campus Biotech, 9 Chemin des Mines, 1202 Genève, Switzerland; Department of Neurology, University Hospital of Geneva, Rue Micheli-du-Crest 24, 1205 Geneva, Switzerland.
  • 2016-11-07
Published in:
  • Brain and cognition. - 2017
English When presented with a difficult perceptual decision, human observers are able to make metacognitive judgements of subjective certainty. Such judgements can be made independently of and prior to any overt response to a sensory stimulus, presumably via internal monitoring. Retrospective judgements about one's own task performance, on the other hand, require first that the subject perform a task and thus could potentially be made based on motor processes, proprioceptive, and other sensory feedback rather than internal monitoring. With this dichotomy in mind, we set out to study performance monitoring using a brain-computer interface (BCI), with which subjects could voluntarily perform an action - moving a cursor on a computer screen - without any movement of the body, and thus without somatosensory feedback. Real-time visual feedback was available to subjects during training, but not during the experiment where the true final position of the cursor was only revealed after the subject had estimated where s/he thought it had ended up after 6s of BCI-based cursor control. During the first half of the experiment subjects based their assessments primarily on the prior probability of the end position of the cursor on previous trials. However, during the second half of the experiment subjects' judgements moved significantly closer to the true end position of the cursor, and away from the prior. This suggests that subjects can monitor task performance when the task is performed without overt movement of the body.
Language
  • English
Open access status
closed
Identifiers
Persistent URL
https://sonar.ch/global/documents/46586
Statistics

Document views: 35 File downloads: