ECoG N1 latency rises with distance, while stored SEEG detections show no reliable slope

In an independent cohort of 16 hybrid ECoG and SEEG patients, the CCEP N1 latency-distance relation reported in a prior study was reproduced at ECoG recording contacts: the median slope was 0.27 ms/mm (95% CI 0.21–0.43), positive in all 16 patients (p = 0.002), while median R² was 0.074. Similar slopes occurred at 6 SD and 15 SD detection thresholds. Distance did not improve leave-one-patient-out latency prediction over a constant model. At SEEG depth recording contacts, the primary stored 2.6 SD detections showed no reliable slope (median 0.11 ms/mm, 95% CI −0.22–0.21; p = 0.23; median R² 0.018). These depth detections had broad latency spread and cannot establish whether true depth N1 latency depends on distance. The analysis used archived rows derived from OpenNeuro ds004080; the exact dataset release is not recorded. The study was conducted at one center, used mostly ECoG stimulation, automatic detections without visual checks, one run per patient, and Euclidean distance.

cortico-cortical evoked potentials · N1 latency · ECoG · SEEG · distance · replication · intracranial EEG · brain-computer interface