Title: On sensing what's not there

Speaker: Dr. Andrew Welchman, Reader in Computational & Cognitive Neuroscience, Department of Psychology, University of Cambridge

Time: 15:00-16:00, October 24, 2018

Location: Room 1113, Wang Kezhen Building

Abstract:

How do we perceive the world around us? Traditionally, we conceive of sensory neurons as ‘feature detectors’ that indicate the presence of particular objects in the environment. Plucking an apple from a nearby tree, for instance, may involve a hierarchy of neural detectors that start with simple features (e.g., oriented edges in V1) and build up to complex feature conjunctions (e.g., ‘apple detectors’ in the inferotemporal cortex). In this talk, I will suggest an alternative approach, based on optimal information encoding, that mixes feature detection with proscription: actively ruling out unlikely interpretations of the scene. This involves the brain’s use of ‘what not’ sensors that respond best to stimulation that could not be caused by single objects in the environment. While this may seem paradoxical, I will review evidence from two domains (binocular vision and multisensory cue combination) where proscription accounts for neuronal responses that have long appeared puzzling.

In the talk, I will describe recent evidence from my lab consistent with proscription. In particular, I will describe human psychophysical, modelling and brain imaging (fMRI, MRS and tDCS) work that suggests that our perception of the surrounding 3D world is boosted by neural responses to what is not there.

Host PI: Dr. Sheng Li