N1-map overlap decreased with stimulation-site separation in four patients at a second SEEG center

This replication tested whether a reported relationship between separation of SEEG stimulation sites and overlap of their N1 response maps also appeared at a second center. It used archived N1 tables for four patients from OpenNeuro ds004696 (Mayo Clinic; 6 mA single pulses; 141 stimulation sites) and electrode coordinates downloaded from the OpenNeuro S3 mirror. Jaccard overlap decreased with separation: median J was 0.29 at ≤7.5 mm, 0.20 at 7.5–15 mm, 0.12 at 15–30 mm, and 0.00 above 50 mm. A distance-matched null explained 64–73% of mean overlap at 0–15 mm. In the primary 7.5–15 mm bin, mean excess J was +0.050 and was positive in three of four patients; the group sign-flip test was not significant (p = 0.25). With four patients, the smallest possible two-sided sign-flip p value is 0.125. Of 131 active sites, 82 (63%) had mutually distinct maps under the reported J < 0.5 criterion. At 30–50 mm, median J was higher than in the comparison study (0.09 versus 0.026), but the fixed recording set contained only 24–50 contacts per patient, so sampling may explain part of this difference. OpenNeuro ds004080 was not eligible because no site pair had at least 10 common recording contacts. Limits include four patients at one additional center, a different N1 rule, a coordinate-based separation measure, and 50 null permutations. The exact OpenNeuro release corresponding to the live-mirror files was not established.

SEEG · electrical stimulation · N1 response · Jaccard overlap · site similarity · distance-matched null · epilepsy · replication · intracranial EEG