Michael Trumpis

PhD Recipient and Postdoctoral Fellow, now a Senior Computational Neuroscientist at Paradromics

I am an electrical engineer by training, with a fondness for timeseries analysis, inverse problems, machine learning, and open source science software. As a student of timeseries, I stumbled into electrophysiology over a decade ago and have never yet gotten over the wonder of the brain's capacity for perception, inference, and action. Much is known to neuroscience from very sparse (100s of neuronal units) and aggregate (hemodynamics) samples of neural activity. The frontier of neurophysiology is to sample activity from large scale networks in the behaving brain with high temporal and spatial precision. Such is the proposed capability of micro-ECoG. My graduate work involves developing methodology to validate and quantify the information content of recordings from implanted micro-ECoG arrays. Lacking strong spiking content, I assess the cortical field potential in terms of stimulus decoding and the consistency of spatial random field statistics. I believe these first steps will help establish micro-ECoG as an important new tool in neurophysiology.

When I am not in the lab, I am biking, climbing, hiking, brewing, or playing guitar to the birds on my front porch.

Contact Information

• Email Address: michael.trumpis@duke.edu