A Light Switch for the Brain

Continued from page 1

Both epilepsy and Parkinson's disease can be treated with electrodes implanted into the brain. But the electricity delivered by the electrode stimulates all nearby cells rather than just the diseased ones, increasing side effects and potentially decreasing the effectiveness of the treatment. "It's been the source of incredible frustration," says Deisseroth, a practicing psychiatrist who is testing electrical stimulation as a treatment for severe depression. "We know we can get treatment benefits by sticking electrodes in the brain, but we don't really know what the target cell type is."

Deisseroth and Boyden are now using the light switches to study animal models of these diseases in order to find out exactly which cells need to be turned on or off. Their findings could be used to develop new drugs targeted to only those cells or, one day, to replace electrodes with more-precise light-activated implants.

The switch could also help decode the language of the brain by helping neuroscientists determine how different patterns of neural activity give rise to complex thoughts and actions. For example, recent research has suggested that rhythmic electrical patterns in our brain are important to our ability to pay attention. Scientists could use the switch to disrupt these patterns in animals and see if it wipes out their capacity to pay attention. Or they could try to induce these patterns and see if this improves the animals' focus. "This has been a dream of neuroscientists for a long time," says Hausser. "To be able to manipulate the spatiotemporal pattern of activity in a network and find the code that is linked to a particular kind of behavior."

In addition, scientists can manipulate the specific units of the neural code--the pulses, or spikes, of electrical activity that are transmitted between cells. "We've shown we can push spikes around, block them, delay them," says Boyden. "We can really alter neural coding at a millisecond time scale." That should allow scientists to determine which aspect of the code--the spikes' timing or the spikes' rate--encodes information in the brain, a debate that has raged for decades.