Test-Time Adaptation of Spiking Neural Networks for Intracortical Neural Decoding using Membrane Potential Alignment

2026-06-12

AI Summary

This work addresses the performance degradation of decoders in implanted brain–computer interfaces caused by day-to-day neural signal drift. The authors propose Membrane Potential Alignment (MPA), a novel test-time adaptation method for spiking neural networks that, for the first time, aligns membrane potential distributions between source and target domains. By minimizing the Kullback–Leibler divergence between these distributions, MPA updates only low-rank adaptation weights comprising less than 9% of the model parameters, substantially reducing computational overhead. Integrated with 4-millisecond high temporal resolution processing, the method achieves performance on par with the current state-of-the-art NoMAD approach in a month-long reaching task with non-human primates, while offering a simpler architecture and superior temporal precision.

Abstract

Intracortical brain-computer interfaces suffer from day-to-day neural signal shifts that degrade pretrained decoders. Existing unsupervised adaptation methods rely on deep recurrent or adversarial architectures that are too computationally expensive for implantable hardware. We propose Membrane Potential Alignment (MPA), a test-time adaptation method for spiking neural networks that realigns a pretrained decoder to shifted recordings by only matching membrane potential distributions via KL divergence. By restricting updates to low-rank (LoRA) weights, MPA adapts fewer than 9% of parameters. On a non-human primate reaching task spanning over one month, MPA achieves performance competitive with the state-of-the-art NoMAD method, while using a simpler architecture and finer temporal resolution (4 ms vs. 20 ms). These results show that efficient SNN-based test-time adaptation is a practical path toward long-term, recalibration-free brain-computer interfaces.

Problem

• intracortical neural decoding
• test-time adaptation
• neural signal shift
• spiking neural networks
• brain-computer interfaces

Innovation

• Spiking Neural Networks
• Test-Time Adaptation
• Membrane Potential Alignment
• Low-Rank Adaptation
• Brain-Computer Interface