"It's something that has tremendous potential," says Leonardo Cohen, a neurologist and chief of the Human Cortical Physiology Section at the National Institute for Neurological Disorders and Stroke, in Bethesda, MD, who was not involved in the study. Researchers now need to figure out the optimal brain-stimulation parameters to trigger healing, he says.

Schlaug's study was unique in that his team applied stimulation to both hemispheres of the brain, using one direction of current to increase brain activity on the damaged side, and the reverse current to inhibit activity on the healthy side. "Maybe with the new technique, the magnitude of the effect may be better," says Cohen.

It's still unclear how long the benefits of tDCS can last, and what magnitude of improvement they can provide. In Schlaug's study, the benefits lasted for at least a week, which was as long as the scientists followed the patients. A previous study in healthy patients found lasting benefits after three months.

Most tDCS studies of stroke have focused on patients whose stroke occurred two or more years previously, when the course of spontaneous recovery has come to an end. "The question for the future is, what happens when this is applied in the subacute period [soon after a stroke]?" says Cohen. Assessing the effectiveness of treatments applied during this period is challenging because it's difficult to determine whether improvements are triggered by the treatment or would have happened naturally.

Schlaug's team is also trying to better understand the changes that take place in the brain as patients recover. They are using a brain-imaging method called diffusion tensor imaging to track the brain's white matter--the neural wires that connect brain cells--before and after treatment. "We are very interested in seeing if something changes in the tracts that go from the brain to the spinal cord," says Schlaug. "And based on those changes, can we figure out how to make the treatment better, or try to predict which patients will benefit?"