White-matter contact status did not predict current at visual percept in 7 patients with both pair types

We asked whether the tissue labels of stimulated stereoelectroencephalography (SEEG) pairs were associated with the current at which visual percepts occurred. We used 112 percept-evoking stimulations from 20 patients in OpenNeuro ds005169 v1.0.0, with events.tsv and electrodes.tsv only; no intracranial EEG samples were used. The outcome is current at percept, not a measured threshold, because the dataset contains no stimulations without a percept and does not document the clinical current-step or stopping protocol. The analysis plan was fixed before computing the outcome. For pairs with at least one white-matter contact versus gray-matter-only pairs, the mean within-patient difference in log2 current was tested by two-sided label permutation, with a patient bootstrap 95% confidence interval. Seven patients had both groups (20 versus 35 stimulations). The estimated current ratio was 0.98 (95% CI 0.76–1.30; p = 0.88), with 3 patients showing a lower value, 3 a higher value, and 1 an equal value for pairs with a white-matter contact. The median current at percept was 0.5 mA (IQR 0.5–1.0; 10th–90th percentile 0.25–1.5), and 80% of percepts occurred at 1 mA or less. Patient identity explained 42% of the variance in log2 current. In secondary tests, pairs with at least one seizure-onset-zone contact had an estimated 1.69-fold higher current (4 patients; p = 0.048 uncorrected, Holm-adjusted p = 0.14 across three secondary tests); pairs with an epileptogenic-zone contact had an estimated 1.61-fold higher current (5 patients; p = 0.052). The complex-versus-elementary percept comparison had both groups in only one patient and was not interpretable. These data do not show a difference in current at percept by white-matter contact status in the seven patients with both pair types. The substantial patient effect supports per-patient calibration in this clinical set, while the small paired samples limit inference. White-matter, seizure-onset-zone, and epileptogenic-zone labels are dataset flags; there are no atlas labels or distances to gray matter. The data cannot test the probability of evoking a percept.

stereoelectroencephalography · SEEG · electrical stimulation · visual percepts · white matter · seizure-onset zone · current at percept · intracranial EEG