The past decade has seen a revolution in our understanding of the brain. Functional magnetic resonance imaging (fMRI) gives scientists a view of our deepest thoughts and hidden anxieties. Tiny arrays of electrodes that record neural signals from the different parts of the brain reveal clues to the way our neurons encode and send information. But what discoveries will the next generation of technologies bring to neuroscience? Scientists at MIT hope to hurry along that answer with the McGovern Institute Neurotechnology Program, a new program dedicated to the development of novel neurotechnologies. Charles Jennings, newly hired director of the program, talks with Technology Review about his vision for the future.
Technology Review: Why start a specific program to develop neurotechnologies?
Charles Jennings: Neuroscience has always been both driven and limited by the technologies available to study it. The brain is so challenging–you need ways to record from and stimulate it. The power with which you can do these things determines the pace of the research and of the eventual clinical applications.
TR: What are the limitations of existing technologies for studying neuroscience?
CJ: For the most part, recording from the brain involves peering through the skull in ways that are fundamentally limited. FMRI, which was a great advance and one of the most powerful of these techniques, measures blood flow. So you can never get a better resolution than the speed of blood flow. You can never get down to the level of a single cell.
On the other end of the spectrum, we can stick an electrode into the brain, usually of animals, a technique that has been tremendously important. But mostly we can only record from one or several of the billions of neurons in the brain. Lots of information is encoded in the timing of the signals between neurons, which you can’t see unless you record from lots of neurons at once.
We’re also limited by the duration which we can record. If you want to study a process or behavior that takes weeks to acquire, you need to be able to look at the brain over long periods of time. That capability would open up many research questions: the processes that underlie habit formation, long-term degeneration, such as Alzheimer’s, or psychiatric diseases, which often develop over years.
Long-term recording is also important clinically for brain-stimulation treatments for Parkinson’s disease and depression. [In this procedure, an electrode is surgically implanted into part of the brain involved in the disease. Delivering electrical pulses via the implant blocks the electrical signals causing tremors and other symptoms of Parkinson’s disease, and, more recently, it has shown some promise in treating severe depression.] And it’s important in prosthetic devices for paralysis victims, in which a device records from the part of the brain involved in planning and then translates that activity into movement of a computer cursor or artificial limb. The challenge is to create something you can implant in the brain that will behave consistently over long periods of time.