Short-Term Postsynaptic Plasticity Facilitates Predictive Tracking in Continuous Attractors

TL;DR

This paper demonstrates that short-term postsynaptic plasticity (STPP) in continuous attractor neural networks enables predictive tracking of moving stimuli, revealing a new mechanism for sensory prediction inspired by neural dynamics.

Contribution

The study introduces the implementation of STPP in CANNs, showing how it destabilizes network states to facilitate predictive stimulus tracking, a novel insight into neural computation.

Findings

STPP increases network mobility, enabling predictive tracking.

STPP destabilizes the network state, promoting anticipation of stimulus movement.

The mechanism offers a new perspective on sensory prediction in neural systems.

Abstract

The N-methyl-D-aspartate receptor (NMDAR) is a crucial component of synaptic transmission, and its dysfunction is implicated in many neurological diseases and psychiatric conditions. NMDAR-based short-term postsynaptic plasticity (STPP) is a newly discovered postsynaptic response facilitation mechanism. Our group has suggested that long-lasting glutamate binding of NMDAR allows input information to be held for up to 500 ms or longer in brain slices, which contributes to response facilitation. However, the implications of STPP in the dynamics of neuronal populations remain unknown. In this study, we implemented STPP in a continuous attractor neural network (CANN) model to describe the neural information encoded in neuronal populations. Unlike short-term facilitation, which is a kind of presynaptic plasticity, the temporally enhanced synaptic efficacy induced by STPP destabilizes the network state of the CANN by increasing the mobility of the system. This nontrivial dynamical effect enables a CANN with STPP to track a moving stimulus predictively, i.e., the network state responds to the anticipated stimulus. Our findings reveal a novel STPP-based mechanism for sensory prediction that can help develop brain-inspired computational algorithms for prediction.