Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics
This addresses the challenge of analyzing complex neural activity across brain areas for neuroscience research, though it appears incremental as it builds on existing transformer and latent variable models.
The paper tackled the problem of disentangling shared and region-specific neural dynamics from multi-region recordings by introducing the Coupled Transformer Autoencoder (CTAE), which better decodes behavioral variables compared to existing methods.
Simultaneous recordings from thousands of neurons across multiple brain areas reveal rich mixtures of activity that are shared between regions and dynamics that are unique to each region. Existing alignment or multi-view methods neglect temporal structure, whereas dynamical latent variable models capture temporal dependencies but are usually restricted to a single area, assume linear read-outs, or conflate shared and private signals. We introduce the Coupled Transformer Autoencoder (CTAE) - a sequence model that addresses both (i) non-stationary, non-linear dynamics and (ii) separation of shared versus region-specific structure in a single framework. CTAE employs transformer encoders and decoders to capture long-range neural dynamics and explicitly partitions each region's latent space into orthogonal shared and private subspaces. We demonstrate the effectiveness of CTAE on two high-density electrophysiology datasets with simultaneous recordings from multiple regions, one from motor cortical areas and the other from sensory areas. CTAE extracts meaningful representations that better decode behavioral variables compared to existing approaches.