Incorporating cycle-consistency loss into a generative adversarial network creates a high-performance and robust 'aligner' of neural population activity, permitting a fixed intracortical brain computer interface to be used for months without recalibration and without requiring inference of a latent manifold.

New computer algorithms have allowed people with paralysis to achieve the highest reported typing performance using a brain-computer interface.

Neural populations in PPC dynamically represent motor-like and then sensory-like aspects of brain–computer interface finger movements with a representational structure that matches able-bodied individuals.

Increasing microstimulation frequency in some regions of the human somatosensory cortex decreased the perceived intensity and evoked specific percepts, providing insight into cortical organization and sensory feedback approaches for brain–computer interfaces.

The distribution of redundant neural activity is coupled with task-relevant activity, which may limit the extent to which redundancy can be exploited by the brain for computation.

Neurons in human dorsal motor cortex, an area involved in controlling arm and hand movements, are also active – and show similar ensemble dynamics – during speaking.

Interhemispheric inhibition can be manipulated by directly and bidirectionally modulating the bilateral sensorimotor excitabilities in a spatially bivariate Brain-Computer Interface-based neurofeedback paradigm.

A brain–computer interface for real-time identification of transient neural activity patterns enables causal inference of the role of these patterns in cognition through closed-loop manipulation.

Motor neurons adjust their sensitivity to direction of movement in a manner analogous to how neurons in the visual system adjust their sensitivity to light.

The human brain encodes coordinated patterns of joint movements – synergies – to increase the efficiency with which it can control complex hand movements.