AI Insight
This study introduces MOJO, a new training framework for neural decoding that combines self-supervised learning through masked autoencoding with traditional supervised learning. Testing across multiple datasets including monkey motor cortex, mouse multi-regional recordings, and human electrocorticography during speech, MOJO demonstrated superior performance compared to purely supervised models, particularly when labeled training data was limited. The approach also produced more interpretable neuronal representations and generalized effectively across different species and neural recording modalities.
Why it matters
This work addresses a critical bottleneck in developing brain-computer interfaces and neurotechnology: the need for large amounts of labeled data. By enabling effective use of unlabeled neural recordings, MOJO could accelerate development of more robust neural decoders and make neurotechnologies more practical and accessible across clinical and research applications.
Understand the Science
arXiv:2607.14086v1 Announce Type: cross
Abstract: Robust and accurate neural decoders are integral to neurotechnologies such as brain-computer interfaces and closed-loop experiments. Recent work has shown that tokenizing neural data at the spike level facilitates multi-session pretraining and delivers state-of-the-art decoding performance. However, current spike-based models are restricted to supervised learning (SL), limiting training to datasets with paired behavioural labels. To address this limitation, we introduce MOJO (Masked autOencoder-based JOint training), a training framework for spike-tokenizing models that jointly leverages self-supervised learning (SSL) via masked autoencoding and SL objectives. We evaluate MOJO on three spiking datasets spanning monkey motor cortex during reaching tasks and multi-regional mouse recordings during vision and decision making tasks, demonstrating superior performance over purely SL-trained models. This improvement is especially pronounced when training with limited labelled data, particularly in few-shot finetuning, where only a small amount of labelled data from a new session is available. Incorporating SSL also yields more interpretable neuronal representations, improving performance on brain region classification and spike-statistics prediction without explicit optimization for these tasks. We further show that MOJO generalizes beyond spiking data to human electrocorticography during speech, where it continues to outperform purely SL-trained models and achieves performance comparable to neuro-foundation models (NFMs) designed specifically for continuous signals. Overall, augmenting spike-tokenizing models with SSL improves performance in label-impoverished settings and enables the use of unlabelled data across various tasks and species, while generalizing to other neural modalities. These results suggest a path towards more flexible and scalable data usage when training NFMs.
Source: Leveraging unlabelled data for generalizable neural population decoding