Robust alignment of cross-session recordings of neural population activity by behaviour via unsupervised domain adaptation
This enables more effective and flexible use of brain-computer interface technologies by allowing stable behavior prediction across sessions without recalibration.
The paper tackled the problem of neural decoder performance degradation over time due to neural activity drift and probe movement, by introducing an unsupervised domain adaptation model with a sequential variational autoencoder that achieves good generalization to unseen data and correctly predicts behavior where conventional methods fail.
Neural population activity relating to behaviour is assumed to be inherently low-dimensional despite the observed high dimensionality of data recorded using multi-electrode arrays. Therefore, predicting behaviour from neural population recordings has been shown to be most effective when using latent variable models. Over time however, the activity of single neurons can drift, and different neurons will be recorded due to movement of implanted neural probes. This means that a decoder trained to predict behaviour on one day performs worse when tested on a different day. On the other hand, evidence suggests that the latent dynamics underlying behaviour may be stable even over months and years. Based on this idea, we introduce a model capable of inferring behaviourally relevant latent dynamics from previously unseen data recorded from the same animal, without any need for decoder recalibration. We show that unsupervised domain adaptation combined with a sequential variational autoencoder, trained on several sessions, can achieve good generalisation to unseen data and correctly predict behaviour where conventional methods fail. Our results further support the hypothesis that behaviour-related neural dynamics are low-dimensional and stable over time, and will enable more effective and flexible use of brain computer interface technologies.