Prof. Pascal Mamassian, CNRS Research Director, CNRS and Ecole Normale Supérieure

时间: 2017-06-20 14:00 - 16:00

地点: Room 1113, Wang Kezhen Building

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 approach a full picture of perceptual decision making. However, established methods to measure visual confidence are prone to at least two major problems. First, they tend to rely on subjective and non-measurable variables, such as boundaries between confidence levels in confidence rating tasks or criteria to opt-out in opt-out paradigms. Second, they can often be accused of measuring perceptual performance with high-precision rather than genuinely meta-perceptual performance (the perception of our perception). It is notoriously difficult to resolve the second problem, as witnessed by the vast literature on animal cognition that attempts to decide which animal species have the ability to make confidence judgments. But it is simple to address the first problem by using a confidence forced-choice paradigm. In this paradigm, observers have to choose which of two perceptual decisions is more likely to be correct. I will review some results obtained with the confidence forced-choice paradigm, in particular on the origins of confidence inter-subject differences, the effects of task and sensory modality on confidence estimation, and the build-up of global confidence across multiple trials.

2017-06-20


2017-06-20