Prof Daniel Kaiser | Predictability and prediction in natural vision

This seminar is free to attend and is open to all, both within and outside the University. Attendance is possible both in-person and on Zoom, details of Zoom registration and physical location can be found above.

We are delighted to announce that the Centre for Human Brain Health (CHBH) will welcome Prof Daniel Kaiser, Professor of Neural Computation at JLU Gieβen, Germany, to present a hybrid CHBH Seminar, taking place on the date and time above. His full biography can be found below.

CHBH Event Host

Dr Clayton Hickey

Abstract

Natural environments are highly complex, with many individual objects present in our surroundings. Despite this complexity, humans are remarkably efficient in naturalistic visual tasks. I will argue that efficient naturalistic perception is linked to the statistical structure of our environments: Our visual system exploits the ways in which objects form meaningful arrangement and complex scenes in predictable ways. I will further argue that our brain actively generates predictions about such statistical regularities, which in turn guide the analysis of visual inputs. I will show how such predictions vary across individual observers, how they alter activity in visual cortex, and how they are carried in rhythmic neural signals.

Speaker Biography

Prof Daniel Kaiser is a trained psychologist and cognitive neuroscientist and currently Professor for Neural Computation at JLU Gießen. He is generally interested in naturalistic vision and wants to understand how we perceive the world around us in typical everyday situations.