Neural encoding of three-dimensional space - properties and constraints


May 1, 2018 - 12:00pm
Northwest Building, Room B103
About the Speaker
Kate Jeffery
Speaker Title: 
Speaker Affiliation: 
University College London

Studies of mammalian self-localization and navigation have elucidated a network of brain structures whose neurons respond to specific aspects of spatial encoding such as position (place cells), head direction (head direction cells) and distance (grid cells). These studies have mostly taken place in typical two-dimensional laboratory settings, but the real world is complex and three dimensional. Studies of how navigation works in three dimensions are just beginning, and a picture is slowly emerging of how the spatial neurons might link information from horizontal movement together with that from vertical. This talk will explore these studies and propose that for surface-travelling animals, the brain selects the locomotor surface as a reference plane over which to perform distance calculations, and then uses gravity to determine how that plane relates to other, differently oriented planes. The result is a complex surface-applied map made from stitched-together planar fragments, which we call “multi-planar.” An open question concerns whether a similar encoding scheme applies when animals can move through a volume (by flying or swimming) or whether the brain now constructs a true, volumetric map. The theoretical properties of such a map will be discussed, and some preliminary data presented.