Liam Paninski, PhD
The neural coding problem is perhaps the fundamental question in systems neuroscience: given some input stimulus (or movement, or thought, etc.), what is the probability of a neural response? In other words, what is the neural code?
The roadblock is that there are too many possible stimuli and responses and not enough time to record data. So we have to approach this problem statistically, to build good models of the neural code and use these models to make predictions about mechanisms, decode spike trains, predict responses to novel stimuli, etc. I'm interested in statistical analysis at various levels of the neural code, from ensembles of simultaneously-recorded spike trains down to voltage fluctuations in individual dendritic compartments.
Lab website: www.stat.columbia.edu/~liam/
Grossman Center website: grossmancenter.columbia.edu