A normative account of the sense of confidence during probabilistic learning

Date

Jan 26, 2017

Time

4:00 PM - 5:00 PM

Speaker

Dr. Florent Meyniel

Affiliation

Neurospin, Paris

Series

TUD NIC Kolloquium

Language

en

Main Topic

Psychologie

Other Topics

Medizin, Informatik, Psychologie

Description

A normative account of the sense of confidence during probabilistic learning” Learning is inherently uncertain when the world fluctuates randomly and ceaselessly. However, many classical models reduce learning to the updating of parameter estimates and neglect the fact that learning is often accompanied by a sense of confidence in what has been learned. We developed a probabilistic learning task to probe the characteristics, the origin and the functional role of subjective confidence during learning. Subjects estimated non-stationary probabilities and reported confidence in their estimates. We computed the optimal solution for this learning problem and analyzed the data from this normative viewpoint. Behavioral data showed that humans not only infer a model of their environment, but they also accurately track the likelihood that their inferences are correct. Learning and estimating confidence in what has been learned seemed to derive from the same, close-to-optimum probabilistic inference. MRI data showed functional signals that conformed to a hierarchical, confidence-weighted learning process. Together, our results provide evidence that the sense of confidence is an essential ingredient of probabilistic learning in the human brain.

Links

• Livestream

Last modified: Apr 28, 2017, 6:15:58 PM