à 

W-5215
455, boulevard René-Lévesque Est
Montréal (QC) Canada  H2L 4Y2

Pour la prochaine séance de CRISCo, nous aurons le plaisir de recevoir Paul Cisek, professeur agrégé du département de Neurosciences de la faculté de médecine de l'Université de Montréal.

Il nous présentera une conférence intitulée : 

“Reconsidering the foundations of cognitive science”

Celle-ci aura lieu le jeudi 17 mars, à 14h00, au département de philosophie de l'UQAM, local W-5215.

Le résumé de sa présentation se trouve ci-dessous. Notez qu'elle sera en anglais.

Nous vous attendons en grand nombre pour cette conférence, ouverte à toutes et tous, et ne requiérant pas d'inscription.

-----------------

Résumé / Abstract

One of the central tenets of cognitive science is that the brain is an information processing system, which manipulates representations to build knowledge of the world, make decisions, and guide actions. While this is currently a dominant view in many areas of psychology, neuroscience, and philosophy, it is encountering increasing challenges when confronted with growing data on the neurophysiological organization of the brain. In my talk, I will explore the possibility that such discrepancies exist because the questions addressed by cognitive science were defined long before much was known about neurophysiology, and are simply not the right questions to ask to understand the brain mechanisms underlying behavior. I will propose that instead of seeking answers to questions inherited from the history of brain theories, we investigate new questions based on the history of brain evolution. In this approach, each conceptual distinction (physiology vs. behavior, sensation vs. action, etc.) is introduced in the context of the phylogenetic refinement of an existing ancestral system. This leads to a set of questions and proposed mechanisms that are very different from those usually encountered in mainstream brain science. Most notable is the proposal, made several times over the last 150 years, that the brain is a feedback control system whose fundamental role is interaction with the world, not understanding of that world. The progressive evolutionary elaboration of feedback systems suggests a highly parallel and dynamical functional organization of brain systems (as opposed to serial processing stages) and an emphasis on pragmatic representations (as opposed to descriptive ones). While these proposals are admittedly speculative, I will describe several examples from neurophysiological studies to illustrate how they can help to make sense of neural data that is difficult to explain using classical concepts of representation and computation.

Consulté 33 fois   ·   Modifier