Distance explains much of the limbic CCEP similarity in ventral temporal cortex, with residual effects in 2 of 5 patients

We tested whether cortico-cortical evoked potential (CCEP) waveforms recorded in human ventral temporal cortex (VTC) are more similar for hippocampal or amygdalar stimulation sites than for limbic and non-limbic sites, and whether this difference remains after controlling for distance. Using OpenNeuro ds004457 v1.0.1, we analyzed nine bipolar VTC recording pairs from five SEEG subjects. We averaged each site's trials, correlated mean waveforms, and compared mean limbic-limbic correlation with mean limbic-non-limbic correlation using 10,000 label permutations. For the five primary pairs, the raw difference was positive in all five (Fisher p = 2.6e-7). After regressing correlation on linear and log Euclidean distance, the difference was positive in three of five, and significant in two (subjects 1 and 3; Fisher p = 5.7e-4). Across all nine pairs, raw p < 0.05 in four and distance-controlled p < 0.05 in two. Similarity decreased with log distance in every pair, and limbic-limbic pairs were about three times nearer than limbic-non-limbic pairs. These results indicate that spatial proximity accounts for much of the raw limbic-shape effect; a distance-independent effect was observed in some medial VTC sites, but not all. The study is limited by five subjects, 3–11 limbic sites per pair, automatic atlas labels not inspected by eye, one distance model that may leave residual spatial structure, possible volume conduction, and Fisher combination's sensitivity to very small p values. This correlation-based test did not use the original paper's basis-profile-curve method.

cortico-cortical evoked potentials · CCEP · ventral temporal cortex · limbic stimulation · hippocampus · amygdala · SEEG · distance control · waveform similarity · OpenNeuro ds004457