Yelin Dong1,2 · Hongzhi You3  · Yuxiu Shao1  · Yong Gu4  · KongFatt Wong‑Lin5  · Da‑Hui Wang1

1 School of Systems Science and State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing 100875, China 

2 Department of Brain and Cognitive Sciences, Center for Visual Science, University of Rochester, Rochester, New York 14627, USA 

3 School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu 611731, China 

4 Center for Excellence in Brain Science and Intelligence Technology, Institute of Neuroscience, Chinese Academy of Sciences, Shanghai 200031, China 

5 Intelligent Systems Research Centre, School of Computing, Engineering and Intelligent Systems, Ulster University, Londonderry, Northern Ireland BT48 7JL, UK

Abstract

Bayesian integration is posited as a fundamental computational mechanism underlying multisensory integration, and feedforward neural networks have been proposed to instantiate optimal Bayesian integration (OBI). However, empirical and theoretical research highlights the prevalence of neural feedback projections, raising questions about how recurrent neural networks might contribute to multisensory OBI. We simulated a two-layer neural circuit computational model with reciprocal projections performing a perceptual discrimination task, in which sensory inputs comprise single or dual modalities. The model with reciprocal projections between sensory and decision-making modules can match, underperform, or outperform OBI, depending on feedforward–feedback interplay. This model performance variability accords with prior experimental data. In addition, our theoretical analysis reveals the importance of non-linear interactions within neuronal assemblies in mediating such multisensory integration behaviors. Our work suggests that sensory modalities can be entangled through top-down feedback, challenging the traditional view of their independence, while explaining deviations from OBI.

Keywords

Neural network; Multisensory integration; Bayesian optimal integration; Non-Bayesian integration

[SpringerLink]