Skip to main content
Dryad

Data and code from: Human learning of noninvasive brain-computer interfaces via manifold geometry

Busch, Erica1; Fincke, E. Chandra1; Lajoie, Guillaume2; Krishnaswamy, Smita1; Turk-Browne, Nicholas1

Published Apr 01, 2026 on Dryad. https://doi.org/10.5061/dryad.9cnp5hr0w

Data files

Apr 01, 2026 version files 38.82 GB

Click names to download individual files Select up to 11 GB of files for zip download

Abstract

Brain-computer interfaces (BCIs) promise to restore and enhance a wide range of human capabilities. However, a barrier to the adoption of BCIs is that learning to control them can be difficult and slow, with inconsistent results across users. We hypothesized that human BCI learning could be accelerated by leveraging the naturally occurring geometric structure of brain activity, or its intrinsic manifold, extracted using a data-diffusion process. We trained participants on a noninvasive BCI that allowed them to gain real-time control of an avatar in a virtual reality game by modulating functional magnetic resonance imaging (fMRI) activity in brain regions that support spatial navigation. We then perturbed the mapping between fMRI activity patterns and the movement of the avatar to test our manifold hypothesis. When the new mapping relied on directions of significant variance along the intrinsic manifold, participants succeeded in regaining control of the BCI by aligning their brain activity within the manifold. When the new mapping did not rely on the intrinsic manifold, participants could not learn to control the avatar again. These findings show that the manifold geometry of brain activity constrains human learning of a complex cognitive task in higher-order brain regions. Manifold geometry may be a critical ingredient for unlocking the potential of future human neurotechnologies.