Population Code Dynamics in Categorical Perception.

Chihiro I Tajima,Hidehiko Komatsu,Kazuyuki Aihara,Satohiro Tajima,Kowa Koida,Hideyuki Suzuki

doi:10.1038/srep22536

Abstract

Categorical perception is a ubiquitous function in sensory information processing, and is reported to have important influences on the recognition of presented and/or memorized stimuli. However, such complex interactions among categorical perception and other aspects of sensory processing have not been explained well in a unified manner. Here, we propose a recurrent neural network model to process categorical information of stimuli, which approximately realizes a hierarchical Bayesian estimation on stimuli. The model accounts for a wide variety of neurophysiological and cognitive phenomena in a consistent framework. In particular, the reported complexity of categorical effects, including (i) task-dependent modulation of neural response, (ii) clustering of neural population representation, (iii) temporal evolution of perceptual color memory, and (iv) a non-uniform discrimination threshold, are explained as different aspects of a single model. Moreover, we directly examine key model behaviors in the monkey visual cortex by analyzing neural population dynamics during categorization and discrimination of color stimuli. We find that the categorical task causes temporally-evolving biases in the neuronal population representations toward the focal colors, which supports the proposed model. These results suggest that categorical perception can be achieved by recurrent neural dynamics that approximates optimal probabilistic inference in the changing environment.

Highlights

We perceive sensory stimuli in two different ways: fine, or coarse as groups
The probabilistic population code has been proposed as an effective scheme to represent probabilistic distribution of stimulus with neural ensemble activity[26,27,28,29] its biological plausibility in dynamic and high-dimensional inference problems is a recent topic of debate[30]
We show that the probabilistic population code can approximate a dynamic hierarchical inference of stimulus and category when it is equipped with a recurrent interaction between neural populations at different levels

Summary

Introduction

We perceive sensory stimuli in two different ways: fine, or coarse as groups. For example, we can dissociate a group of red berries from the background green grass-field because of the rough color differences between them; at the same time, we can judge their maturities by discriminating between the slight differences in their colors. In color perception, color selective neurons in the ventral visual areas [including V1 and V4 and the inferior temporal (IT) cortex] show relatively smooth, continuous preference functions over the color space (called hue)[1,2,3,4,5,6,7,8] These areas are known to relate to fine color discrimination[1,2,4]. Recalled colors are gradually attracted toward the nearest categorical centers as time elapses after the stimulus offset[23,24,25] These facts imply that the sensory percepts are shaped by dynamic and interactive mechanisms between the low-level sensory processes and the high-level, symbolic www.nature.com/scientificreports/. Based on new electrophysiological data analyses, we confirm the key model dynamics in behaving monkey visual cortex

Methods

Results

Conclusion