Search papers, labs, and topics across Lattice.
NeuroLens, a self-supervised JEPA-based model, learns denoised, semantically informative latent embeddings from chronic neural recordings by employing an adaptive encoder and a temporal predictor. This approach effectively disentangles representational plasticity from recording instability, enabling more robust analysis of neural dynamics. The learned representations significantly improve decoding of cognitive variables and demonstrate superior generalization and adaptation capabilities compared to existing methods.
Chronic neural recordings can now reveal stable, semantically rich representations of cognition by disentangling plasticity from noise.
Understanding how neural activity represents higher-order cognition and how these representations evolve over time has long been a central pursuit in neuroscience. However, current analytical tools cannot easily distinguish representational plasticity from recording instability in chronic neural recordings. Here, we introduce NeuroLens (Latent Embeddings of Neural Semantics), a self-supervised model based on the Joint-Embedding Predictive Architecture (JEPA) framework that learns denoised, semantically informative latents from chronic neural recordings. An adaptive encoder maps changing neural populations into a common latent space, while a temporal predictor learns structure that supports prediction of future latent states. By predicting in latent space, NeuroLens captures temporally predictable structure and reduces sensitivity to transient, recording-specific variability. Across chronic intracortical data in mice and humans, the learned representations improve decoding of decision-making and semantic task variables. Multi-day pretraining enables generalization to future sessions, rapid few-shot adaptation to unseen neural populations, and more stable decoding over time than state-of-the-art baselines. Together, these results establish NeuroLens as a new paradigm for studying how neural representations change during learning and over long timescales.
