Brain–machine interface articles from across Nature Portfolio

A convolutional neural network that maps electrical signals from a brain–computer interface to neural features improved the performance of three participants with tetraplegia on a cursor-control task.

Candrea et al. develop a brain-computer interface click detection algorithm using electrocorticographic signals. Using this click detector, a clinical trial participant with amyotrophic lateral sclerosis was able to control a switch-scanning spelling application and achieve high rates of spelling without model retraining.

Detection of neural data instability offers a strategy for optimizing brain-computer interfaces.

Information on clinical trial progress and outcomes of implanted brain–computer interfaces (iBCIs) remains limited. This Review provides a comprehensive survey of all identified clinical trials of iBCIs for communication, motor control or restoration of tactile perception conducted from the first chronic implantation in 1998 to December 2023.

Advanced sensory feedback from upper limb prostheses would provide multiple benefits to people with upper limb amputations, but achieving functional and natural-feeling sensation is technologically challenging. Advances are being made with invasive and non-invasive stimulation approaches, but considerable challenges need to be addressed with technological innovation.

We show that nonlinear latent factors and structures in neural population activity can be modelled in a manner that allows for flexible dynamical inference, causally, non-causally and in the presence of missing neural observations. Further, the developed neural network model improves the prediction of neural activity, behaviour and latent neural structures.

Two papers report the development of high performance brain–computer interfaces that can decode speech from cortical activity.