Population rate-coding predicts correctly that human sound localization depends on sound intensity.

Antje Ihlefeld,Nima Alamatsaz,Robert M Shapley

doi:10.7554/elife.47027

Abstract

Human sound localization is an important computation performed by the brain. Models of sound localization commonly assume that sound lateralization from interaural time differences is level invariant. Here we observe that two prevalent theories of sound localization make opposing predictions. The labelled-line model encodes location through tuned representations of spatial location and predicts that perceived direction is level invariant. In contrast, the hemispheric-difference model encodes location through spike-rate and predicts that perceived direction becomes medially biased at low sound levels. Here, behavioral experiments find that softer sounds are perceived closer to midline than louder sounds, favoring rate-coding models of human sound localization. Analogously, visual depth perception, which is based on interocular disparity, depends on the contrast of the target. The similar results in hearing and vision suggest that the brain may use a canonical computation of location: encoding perceived location through population spike rate relative to baseline.

Highlights

A fundamental question of human perception is how we perceive target locations in space
For other sensory dimensions, including sound localization and visual depth perception, spatial locations must be computed by the brain
Analyzing how sound intensity affects perceived sound direction near sensation threshold offers an opportunity to disentangle whether our human auditory system relies on a place-based or rate-based population code for localizing sound based on ITD

Summary

Introduction

A fundamental question of human perception is how we perceive target locations in space. The shape of the MSO output firing rate curves as a function of ITD resembles that of a cross-correlation operation on the inputs to each ear (Batra and Yin, 2004) How this information is interpreted downstream of the MSO has led to the development of conflicting theories on the neural mechanisms of sound localization in humans. One prominent neural model for sound localization, originally proposed by Jeffress (1948), consists of a labelled line of coincidence detector neurons that are sensitive to the binaural synchronicity of neural inputs from each ear, with each neuron maximally sensitive to a specific magnitude of ITD (Figure 1A) This labelled-line model is computationally equivalent to a neural place-code based on bandlimited cross-correlations of the sounds reaching both ears (Domnitz and Colburn, 1977). Several studies support the existence of labelled-line neural place-code mechanisms in the avian brain (Carr and Konishi, 1988; Overholt et al, 1992), and versions of it have successfully been applied in many engineering applications predicting human localization performance

Methods

Results

Conclusion