Learning to segment self-generated from externally caused optic flow through sensorimotor mismatch circuits

Abstract

Efficient sensory detection requires the capacity to ignore task-irrelevant information, for example when optic flow patterns created by egomotion need to be disentangled from object perception. To investigate how this is achieved in the visual system, predictive coding with sensorimotor mismatch detection is an attractive starting point. Indeed, experimental evidence for sensorimotor mismatch signals in early visual areas exists, but it is not understood how they are integrated into cortical networks that perform input segmentation and categorization. Our model advances a biologically plausible solution by extending predictive coding models with the ability to distinguish self-generated from externally caused optic flow. We first show that a simple three neuron circuit produces experience-dependent sensorimotor mismatch responses, in agreement with calcium imaging data from mice. This microcircuit is then integrated into a neural network with two generative streams. The motor-to-visual stream consists of parallel microcircuits between motor and visual areas and learns to spatially predict optic flow resulting from self-motion. The second stream bidirectionally connects a motion-selective higher visual area (mHVA) to V1, assigning a crucial role to the abundant feedback connections: the maintenance of a generative model of externally caused optic flow. In the model, area mHVA learns to segment moving objects from the background, and facilitates object categorization. Based on shared neurocomputational principles across species, the model also maps onto primate vision. Our work extends the Hebbian predictive coding to sensorimotor settings, in which the agent actively moves - and learns to predict the consequences of its own movements. Significance statementThis research addresses a fundamental challenge in sensory perception: how the brain distinguishes between self-generated and externally caused visual motion. Using a computational model inspired by predictive coding and sensorimotor mismatch detection, the study proposes a biologically plausible solution. The model incorporates a neural microcircuit that generates sensorimotor mismatch responses, aligning with experimental data from mice. This microcircuit is integrated into a neural network with two streams: one predicting self-motion-induced optic flow and another maintaining a generative model for externally caused optic flow. The research advances our understanding of how the brain segments visual input into object and background, shedding light on the neural mechanisms underlying perception and categorization not only in rodents, but also in primates.