The P300 neural communication prosthesis allows an individual to type words from an on-screen menu by recording visually evoked potentials related to the user’s intention. Typing rates are affected in part by two sources of uncertainty: (a) noise in the P300 signal, and (b) statistical language structure. Early system designs focused on classification to address P300 noise. More recently, word completion has been applied to exploit statistical language structure. In general, the time to specify a menu option is proportional to menu size. However, previous approaches employ fixed menu size, independent of P300 noise or statistical language structure. Here we propose developing statistical models of the two sources of uncertainty and applying dynamic programming/stochastic control theory along with feedback information theory for the principled design of variable-length menus based upon these models. In treating the case of no noise in the P300, we illustrate how the optimal policy of this dynamic programming problem satisfies properties reminiscent of the Huffman code in data compression and information theory. By exploiting this optimality property, we illustrate for certain action spaces how the state space of the stochastic control problem can be greatly diminished (from exponential growth to linear growth), and we exhibit optimal policies for a probabilistic models pertaining to the statistical structure of the English language. Through this methodology, we plan to explore a new principled, mathematical formulation of the P300 communication prosthesis problem, that couples a quantitative characterization of the brain with a novel application of stochastic control theory and feedback information theory that ultimately demonstrates faster, more accurate communication prostheses for individuals with related deficits in neural or muscular function.

BIOGRAPHY

Todd P. Coleman is an Assistant Professor at UIUC in the Coordinated Science Laboratory in the Department of Electrical and Computer Engineering. He's also affiliated with the Neuroscience Program and the Beckman Institute. His research interests are in network information theory, wireless communication, operations research, and computational neuroscience. Todd received B.S. degrees in electrical engineering as well as computer engineering from the University of Michigan in 2000. He completed the M.S. degree in 2002 and Ph.D. degree in November 2005, both in electrical engineering from MIT, under the supervision of Prof. Muriel Medard. Todd's PhD thesis was titled "Low-Complexity Approaches to Distributed Data Dissemination". For the remaining 2005-2006 academic year, Todd was a postdoctoral scholar in the Neuroscience Statistics Research Laboratory at MIT's Department of Brain and Coginitive Sciences and Massachusetts General Hospital, under the supervision of Prof. Emery Brown, M.D.,Ph.D.