White-matter stimulation did not increase early SEEG response size in five subjects from OpenNeuro ds004696

We tested whether stereoelectroencephalography (SEEG) stimulation pairs with at least one white-matter (WM) contact caused larger 15–100 ms responses than gray-matter (GM) pairs at the same distance. We analyzed OpenNeuro ds004696 (Mayo Clinic; 6 mA, 0.2 ms biphasic single pulses) in five of eight subjects; three subjects were not analyzed because of a time limit. For each subject, we selected one adjacent GM bipolar recording pair per shaft (10–16 pairs), excluded shared contacts and pairs with minimum contact distance below 10 mm, and analyzed 2,504 site-recording rows. Response size was log10 of the RMS of the mean response from 15–100 ms divided by baseline standard deviation. We fit per-subject linear models with WM site, log10 distance, and recording-pair fixed effects, and used 2,000 site-label permutations. The mean WM coefficient was −0.026 log10 (WM/GM ratio 0.94); two of five subject coefficients were positive, and the exact sign-flip p value was 0.31. One subject had a significant negative coefficient (p = 0.015). A secondary, uncalibrated peak-z detection rate was higher at WM sites in two of five subjects. With this response-size measure and protocol, the WM advantage reported in earlier ds005448 work using N1 detection counts at 8 mA did not replicate. Limits include five subjects, a subset of recording channels, atlas tissue labels that were not checked, an uncalibrated detection threshold, Euclidean dataset-coordinate distances, and possible effects at other currents. The smallest possible exact sign-flip p value with five subjects is 0.0625.

stereoelectroencephalography · SEEG · electrical stimulation · white matter · gray matter · evoked response · OpenNeuro ds004696 · response size