Reciprocal CCEP N1 latency coupling was not significant after distance and contact effects were removed in 8 SEEG patients

We tested whether contact-level cortico-cortical evoked potential (CCEP) N1 latency in one direction predicts latency in the reverse direction after accounting for distance and sender- and receiver-contact effects. We analyzed expert-checked N1 peaks from OpenNeuro ds005448 v1.0.0 (STReEF): 8 SEEG patients as the primary cohort and 5 ECoG patients in secondary analyses. For each patient, we estimated directed contact-level latency, fitted ordinary least squares models with sender and receiver effects, log distance, squared log distance, and same-shaft status, and measured Spearman correlation between residual latencies for reciprocal dyads. Across 1,593 SEEG dyads, residual coupling was not significant (mean rho = 0.086; positive in 5/8 patients; exact sign-flip p = 0.14; Holm-adjusted p = 0.14). Raw coupling was 0.180 (7/8 positive; p = 0.023; Holm p = 0.070), and fell to 0.139 after distance adjustment. The latency-distance slope was positive in all 8 SEEG patients (0.52 ms/mm; Holm p = 0.031). Across all 13 patients, residual coupling was 0.107 (10/13 positive; p = 0.0078), driven mostly by ECoG; residual coupling was not shown at the 13 mm minimum-distance sensitivity threshold. These results do not show a significant dyad-specific shared path in SEEG. A small residual effect remains possible because the primary cohort has 8 patients. Limitations include a single center, Euclidean rather than fiber distance, mapping bipolar stimulation pairs to contacts, and non-independent dyads within patients.

cortico-cortical evoked potentials · CCEP · N1 latency · SEEG · ECoG · reciprocal connectivity · intracranial EEG · OpenNeuro