Is brain-computer interface training beneficial for improving movement and daily activities after stroke?

Key messages

• Brain–computer interface (BCI) training may slightly improve arm movement, but may make little to no difference to leg movement. We are very uncertain whether it helps with everyday activities.

• Compared with a 'sham' (fake) BCI, BCI training may make little to no difference to movement or everyday activities.

• Larger, well-designed studies are needed to confirm the true benefits and harms of BCI training.

What is a brain-computer interface (BCI)?

A brain–computer interface (BCI) is a technology that creates a direct link between the brain and an external device. It usually uses sensors placed on the scalp to detect brain signals when a person imagines or attempts a movement.

How is BCI used after stroke?

After a stroke, people often experience muscle weakness or loss of movement. BCI training reads the brain signals produced when a person intends to move, and turns them into real-time feedback. For example, when the patient thinks about moving their hand, the BCI detects this and triggers a robotic glove to move their hand, or delivers electrical stimulation to the muscles. This feedback happens at the same moment as the person's effort, which is thought to help "rewire" the brain and promote recovery of movement.

What did we want to find out?

We wanted to find out whether BCI training is better than conventional therapy, other active treatments, or a sham (fake) BCI at improving arm and leg movement, and the ability to carry out everyday activities in people with stroke. We also wanted to find out if BCI training is linked to any unwanted effects.

What did we do?

We searched for studies looking at the effects of BCI training in adults who had had a stroke and had difficulty moving as a result. The people in these studies had had either type of stroke (one caused by a blocked blood vessel or one caused by bleeding in the brain), and most were in the months following their stroke, once their condition had stabilised. We then compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as the methods used and the number of people studied.

What did we find?

We found 43 studies involving 1628 adults who had had a stroke. Here is what the evidence suggests:

• When BCI training was compared with conventional therapy, it may give arm movement a small boost (10 studies, 611 people), but may make little to no difference to leg movement (1 study, 64 people). We cannot tell whether it helps with everyday activities (5 studies, 282 people), as the evidence is very uncertain.

• When compared with other active treatments, BCI training may give balance a small boost (6 studies, 165 people), but may make little to no difference to leg movement (4 studies, 130 people). For arm movement (14 studies, 331 people) and everyday activities (5 studies, 163 people), we are not confident enough to say either way.

• When it was compared with a fake or 'sham' BCI, which looks and feels like the real thing but does not give genuine brain-signal feedback, BCI training did not seem to make much difference to arm (9 studies, 279 people) or leg movement (3 studies, 106 people), or to everyday activities (1 study, 28 people).

We also looked at a few other things, such as muscle tightness, muscle strength, and overall recovery of the nervous system. The evidence on these was too uncertain for us to draw conclusions.

We also wanted to know about unwanted effects, but many studies simply did not mention them. In the studies that did, the effects were usually mild, such as sore muscles or feeling tired. Because so few studies reported this properly, we cannot be sure whether BCI training carries any extra risk.

What are the limitations of this evidence?

We have little to very little confidence in the evidence, for several reasons:

• Most of the studies were small and involved only a few people.

• Many were carried out in ways that could have influenced their results. For example, the people taking part often knew which treatment they were receiving.

• The findings also varied widely from one study to another.

• Finally, some studies that found no benefit may not have been published, which could make BCI training appear more helpful than it really is.

How up to date is this evidence?

The evidence is up to date to October 2025.