Phone: (412) 268-8723
Fax: (412) 268-7828
Email: kass@stat.cmu.edu

Individual Website: http://www.stat.cmu.edu/~kass

Ph.D., University of Chicago

Research Interests

In college I was a math major, but worked summers in medical laboratories. There, all around me, people were applying statistical techniques in analyzing their data. I realized this to be an activity fundamental to the scientific process and became sufficiently intrigued that I went to graduate school to study Statistics. In my Ph.D. dissertation, and in most of my subsequent work, I focused mainly on theoretical issues but hoped someday to return to experimental life sciences. Now, in mid-life, I have found cognitive neuroscience, a marvelous area rich in data analytic challenges.

My expertise is in Bayesian statistics, meaning the use of Bayes' Theorem in reasoning from data. Over the past decade Bayesian inference has been among the fastest-growing areas in Statistics, due mainly to advances in computing technology and the advent of computational methods for the many high-dimensional integrals that must be evaluated in doing Bayesian calculations. I have helped develop some of these computational techniques, while continuing to pay attention to fundamental issues in hypothesis testing (under the rubric of "Bayes factors") and the formalization of inference when relatively little is known a priori (under the rubric of "prior distributions").

Currently I am focusing on statistical analysis of neuronal data, i.e., the output of single-electrode and multiple-electrode neurophysiological experiments. The projects my collaborators and I have carried out involve formulation of probability models and study of statistical inference procedures in the following settings: estimation of instantaneous firing rate of single neurons (i.e., smoothing the PSTH); analysis of variation in firing rate across many individually-recorded neurons; non-Poisson models for within-trial variability; trial-to-trial variability; time-evolution of correlated spiking across pairs of neurons; and real-time decoding of motor cortical signals for neural prostheses.

Recent Publications

• Brockwell AE, Kass RE, Schwartz AB: Statistical signal processing and the motor cortex. Proc IEEE 95: 881-898, 2007.

• Kass RE, Ventura V, Brown EN. Statistical issues in the analysis of neuronal data. J Neurophysiol 94: 8-25, 2005.

• Ventura V, Cai C, Kass RE: Trial-to-trial variability and its effect on time-varying dependence between two neurons. J Neurophysiol 94: 2928-2939, 2005.

• Brown EN, Kass RE, Mitra PN: Multiple neural spike trains analysis: State-of-the-art and future challenges. Nature Neurosci 7: 456-461, 2004.