Distinct spatial coordinate of visual and vestibular heading signals in macaque FEFsem and MSTd.

Lihua Yang,Yong Gu

doi:10.7554/elife.29809

Abstract

Precise heading estimate requires integration of visual optic flow and vestibular inertial motion originating from distinct spatial coordinates (eye- and head-centered, respectively). To explore whether the two heading signals may share a common reference frame along the hierarchy of cortical stages, we explored two multisensory areas in macaques: the smooth pursuit area of the frontal eye field (FEFsem) closer to the motor side, and the dorsal portion of medial superior temporal area (MSTd) closer to the sensory side. In both areas, vestibular signals are head-centered, whereas visual signals are mainly eye-centered. However, visual signals in FEFsem are more shifted towards the head coordinate compared to MSTd. These results are robust being largely independent on: (1) smooth pursuit eye movement, (2) motion parallax cue, and (3) behavioral context for active heading estimation, indicating that the visual and vestibular heading signals may be represented in distinct spatial coordinate in sensory cortices.

Highlights

To navigate effectively through the environment, we usually combine multiple sensory inputs to precisely estimate our direction of self-motion
A critical problem for the brain to combine cues is that the two heading signals originate from different spatial reference frames: visual signals arise from the retina and are represented in an eye-centered coordinate, whereas vestibular signals arise from the peripheral organs in the inner ears that are represented in a head-centered coordinate
Heading stimuli were delivered through a virtual reality apparatus (Figure 1A) that allowed independent control of visual optic flow and vestibular inertial motion cues

Summary

Introduction

To navigate effectively through the environment, we usually combine multiple sensory inputs to precisely estimate our direction of self-motion (i.e, heading). A critical problem for the brain to combine cues is that the two heading signals originate from different spatial reference frames: visual signals arise from the retina and are represented in an eye-centered coordinate, whereas vestibular signals arise from the peripheral organs in the inner ears that are represented in a head-centered coordinate In this case, the vision information about heading is often confounded by changes of the eye ball in orbits, for example, when subjects vary gaze eccentrically or pursue moving objects (Royden et al, 1992; Royden, 1994; Royden et al, 1994; Warren and Saunders, 1995; Banks et al, 1996; Royden and Hildreth, 1996; Crowell et al, 1998). How the brain exactly compensates these eye movements to recover true heading and correctly integrate it with the vestibular cues is unclear

Methods

Results

Conclusion