Colloquium Cognitive Systems

Visual Confidence

Dr. Pascal Mamssian  (CNRS and Ecole Normale Supérieure)

Abstract: Visual confidence refers to our ability to predict the correctness of our perceptual decisions. Knowing the limits of this ability, both in terms of biases (e.g. overconfidence) and sensitivity (e.g. blindsight), is clearly important to get a full picture of perceptual decision making. But what are the mechanisms that allow us to re-evaluate the quality of our perceptions? Is confidence relying on the same information as the one that was used for perceptual decisions, or is it using a parallel system that can monitor and control our perception? I wll present some experimental paradigms and models that can distinguish serial from parallel confidence processing. I will then review some behavioural results that reflect the richness of our visual confidence judgments.

About: Dr. Pascal Mamassian received his PhD in Psychology from the University of Minnesota in 1995. After two fellowships at the Max-Planck Institute in Tubingen (Germany) and New York University (USA), he moved as a lecturer to the University of Glasgow (Scotland, UK), before taking a CNRS researcher position in Paris (France). His research interests focus on visual perception, in particular 3D and motion perception, time perception, history effects on perception, and meta-perception. He is using mostly behavioural (psychophysics) and modelling (Bayesian inference) techniques. He was the president of the Vision Sciences Society in 2011, and is currently serving on the editorial board of 5 journals. He is the founding director of the Laboratoire des Systèmes Perceptifs at the Ecole Normale Supérieure, Paris.