Entry Date:
February 28, 2017

Close-Packed Silicon Microelectrodes for Scalable Spatially Oversampled Neural Recording

Principal Investigator Clifton Fonstad

Co-investigator Edward Boyden


A major goal of neuroscience is to understand how the activity of individual neurons yields network dynamics, and how network dynamics yields behavior (and causes disease states). Innovative neuro-technologies with orders-of-magnitude improvements over traditional methods are required to reach this goal. Nanofabrication can provide the scalable technology platform necessary to record with single-spike resolution the electrical activity from a large number of individual neurons, in parallel and across different regions of the brain. By combining innovations in fabrication, design, and system integration, we can scale the number of neural recording sites: from traditionally a small number of sparse sites, to currently over 1000 high-density sites, and in the future beyond many thousands of sites distributed through many brain regions.

We designed and implemented close-packed silicon microelectrodes, to enable the spatially oversampled recording of neural activity in a scalable fashion, using a tight continuum of recording sites along the length of the recording shank, rather than discrete arrangements of tetrode-style pads or widely spaced sites. This arrangement, thus, enables spatial oversampling continuously running down the shank so that sorting of spikes recorded by the densely packed electrodes can be facilitated for all the sites of the probe simultaneously. We use MEMS microfabrication techniques to create thin recording shanks and a hybrid lithography process that allows a dense array of recording sites which we connect to with submicron dimension wiring. We have performed neural recordings with our probes in the live mammalian brain, and illustrate the spatial oversampling potential of closely packed electrode sites.