Technical Program

Paper Detail

Paper: RS-1B.1
Session: Late Breaking Research 1B
Location: Late-Breaking Research Area
Session Time: Thursday, September 6, 18:45 - 20:45
Presentation Time:Thursday, September 6, 18:45 - 20:45
Presentation: Poster
Publication: 2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania
Paper Title: NMDA receptor inactivation destabilizes neural representations of decision signals during probabilistic reasoning
Authors: Valentin Wyart, Ecole Normale Superieure, France; Alexandre Salvador, Centre Hospitalier Sainte-Anne, France; Luc Arnal, Universite de Geneve, Switzerland; Philippe Domenech, Institut du Cerveau et de la Moelle Epiniere, France; Raphael Gaillard, Centre Hospitalier Sainte-Anne, France
Abstract: In schizophrenia, early stages of psychosis are characterized by a state of pathological uncertainty (or ‘strangeness’) which is difficult to study in patients due to its transitional nature. Therefore, the alterations of decision processes which lead to full-blown delusions remain largely unknown. Here we studied a pharmacological model of this condition using ketamine, an NMDA receptor antagonist, in healthy subjects performing a probabilistic reasoning task. Computational modeling of human behavior revealed a selective decrease in the precision of probabilistic inference under ketamine. The analysis of simultaneously recorded electroencephalographic (EEG) signals provided support for this selective alteration, by showing unstable neural representations of decision evidence under ketamine. In particular, upcoming decisions could be decoded earlier from brain signals under ketamine, and evidence inconsistent with upcoming decisions were progressively discarded in the inference process. Together, these findings indicate a premature and irrational commitment to uncertain decisions under ketamine, a selective alteration of decision-making which may explain the stabilization of aberrant beliefs characteristic of psychosis.