In 15 SEEG patients, CCEP N1 size declined with contact-index distance; ipsilateral responses were larger, but distance confounded side

We used an automatic method to test two reported cortico-cortical evoked potential (CCEP) patterns in OpenNeuro ds006254, version 1.0.0. From 34 EDF files we read only one 20-s mid-file byte range per file; event, channel, and electrode sidecars were inaccessible. Of 32 windows analyzed, 16 were usable. After removing a byte-identical duplicate recording, the primary set contained 15 windows from 15 patients and 2,326 bipolar derivations. For H1, Spearman correlation between contact-index distance from the stimulated pair and log10 N1 size was negative in all 15 windows (median rho = -0.75; one-sided sign-test p = 3.1e-5). It was negative in 13/15 windows after excluding derivations within two contacts (p = 0.0037). For H2, median z was larger on ipsilateral than contralateral non-stimulated shafts in 7/8 windows (median difference 1.34; one-sided Wilcoxon p = 0.0078). Exploratory analysis found no difference in significant-response fraction between patients with and without CCEP-induced seizures (Mann-Whitney p = 0.61; this test can detect only a large effect). These findings describe one 20-s window and one inferred stimulation site per file. Distance is contact-index steps, not millimetres; hemisphere is confounded with distance, and residual stimulation artifact may affect nearby responses. Stimulation settings and monopolar reference are unknown.

cortico-cortical evoked potentials · CCEP · stereo-EEG · N1 response · effective connectivity · OpenNeuro ds006254