Know Thyself by Knowing Others: Learning Neuron Identity from Population Context

Neurons process information in ways that depend on their cell type, connectivity, and the brain region in which they are embedded. However, inferring these factors from neural activity remains a significant challenge. To build general-purpose representations that allow for resolving information about a neuron's identity, we introduce NuCLR, a self-supervised framework that aims to learn representations of neural activity that allow for differentiating one neuron from the rest. NuCLR brings together views of the same neuron observed at different times and across different stimuli and uses a contrastive objective to pull these representations together. To capture population context without assuming any fixed neuron ordering, we build a spatiotemporal transformer that integrates activity in a permutation-equivariant manner. Across multiple electrophysiology and calcium imaging datasets, a linear decoding evaluation on top of NuCLR representations achieves a new state-of-theart for both cell type and brain region decoding tasks, and demonstrates strong zero-shot generalization to unseen animals. We present the first systematic scaling analysis for neuron-level representation learning, showing that increasing the number of animals used during pretraining consistently improves downstream performance. The learned representations are also label-efficient, requiring only a small fraction of labeled samples to achieve competitive performance. These results highlight how large, diverse neural datasets enable models to recover information about neuron identity that generalize across animals. Code is available at https://github.com/nerdslab/nuclr.

Our core contributions are:

• We introduce NuCLR, a self-supervised framework for learning neuron-level representations from neural population activity. After unsupervised pretraining, these representations support linear decoding of cell type and brain region and set a new state-of-the-art across multiple electrophysiology and calcium imaging datasets.

• We show that NuCLR generalizes in a zero-shot manner: the same pretrained model and embeddings transfer robustly to entirely new sessions and animals without retraining or requiring additional metadata, enabling out-of-the-box decoding of cell type and brain region.

• We provide the first systematic scaling analysis for neuron-level representation learning, demonstrating that increasing the number of animals used during pretraining yields consistent gains in zero-shot cell-type and region decoding. This underscores the value of large, diverse unlabeled neural corpora for inferring neuronal identity from activity alone.