Strong SEEG visual responses appeared only in p05, limiting patient-level summation tests in ds004194

Across 197 same-shaft bipolar SEEG pairs in 9 patients from OpenNeuro ds004194, strong visual broadband responses occurred only in p05 (16/21 pairs; t = 6.0–18.6). Fifteen additional pairs in p07–p11 met the pre-registered t > 3 criterion but had weak responses; p03, p04, and p06 had none. Five testable patients had H1 slopes below 1 (0.71, −0.11, −0.19, 0.48, 0.03), but the exact sign-flip p = 0.031 rose to Holm p = 0.094 across three tests, so H1 was not significant at α = 0.05. Only p05 gave a reliable SEEG slope (0.71; 6/7 pairs below 1), consistent with an earlier result of 0.77. The two-pulse recovery tests were not significant (H2a p = 0.22, Holm 0.44; H2b p = 0.83, Holm 0.83); the area ratio was unstable and the 1.0-s window could truncate later responses. In an exploratory positive control using the same code on ECoG pairs from the same patients, 218/965 pairs in all 9 patients responded and slopes were below 1 in all 9 (mean 0.30, p = 0.002, Holm 0.006); the two-pulse tests also failed. Thus this dataset supports a usable SEEG visual broadband signal in one patient, but not a patient-level SEEG test of temporal nonlinearities. Weak responses should not be treated as visual-cortex evidence without anatomical labels. Limits include no anatomical labels, 12 trials per condition for temporalpattern runs, and a lenient response threshold. The package retains the preregistered plan and amendments, original analysis code, logs, per-pair results, time courses, source URLs, and SHA-256 hashes for all 200 input files.

SEEG · ECoG · temporal summation · two-pulse recovery · broadband power · visual cortex · OpenNeuro · ds004194