A mechanistic understanding of human brain function is a central goal of modern science. Progress in this area could answer fundamental questions about the mystery of consciousness, and potentially provide breakthrough medical treatments for a variety of neurological and psychiatric disorders. We take an information processing view of the brain and apply techniques from electrical engineering in our investigation.
Research in our laboratory is organized into two interwoven streams:
- Decoding human brain activity. We record neural activity using electrodes placed on the scalp (EEG) and with functional magnetic resonance imaging (fMRI) while human participants perform a variety of behavioral tasks or are exposed to natural stimuli. Non-invasive measures of neural activity are both high-dimensional and noisy, and therefore we are developing new machine learning techniques to map spatiotemporal patterns of neural activity to behavior in a statistically optimal fashion. This research has the potential to increase our understanding of how the brain represents its environment and directs action.
- Non-invasive brain stimulation. We are investigating techniques that employ electric currents, near-infrared lasers, and ultrasonic waves to perturb neural activity in a specific and reversible manner. These interventions are coupled with recordings of brain activity using EEG and fMRI in order to understand the effects of the stimulation on the brain. A central challenge in brain stimulation research is that there is an infinite set of possible configurations with which to stimulate. Therefore, deriving general principles to guide rational use of brain stimulation is necessary to advance neuromodulation for clinical interventions and to increase our ability to interrogate brain circuits.