In one patient, pre-stimulus 4–30 Hz power did not predict single-trial CCEP gain

In a single-patient case study using OpenNeuro ds003708 v1.0.4, we tested whether pre-event 4–30 Hz power predicted cortico-cortical evoked potential (CCEP) gain at stereoelectroencephalography (SEEG) depth contacts. The analysis included 391 good single pulses at 41 electrocorticography (ECoG) stimulation sites and 12 SEEG contacts. Of 492 site-contact pairs, 172 were responsive at 23 sites. Site-level association between power at the response contact and gain was near zero (mean Spearman rho 0.020, 95% CI −0.086 to 0.126; 10 of 23 sites negative; sign-flip p = 0.70; Holm-adjusted p = 1.0). Power at the stimulated ECoG pair was also near zero (rho −0.018, 95% CI −0.112 to 0.076; 13 of 23 sites negative; p = 0.69; Holm-adjusted p = 1.0). Removing trial order gave mean rho values 0.024 and −0.032; shuffled-predictor values were 0.011 and 0.044. These results apply to this patient only. Timing was imperfect: the monopolar artifact peak fell within ±2 ms of the recorded event sample for 68% of pulses, while the post-hoc bipolar peak had median lag +6.3 ms. If the pulse occurred up to 6 ms after the event sample, the analyzed 10–100 ms response window would correspond to about 4–94 ms; no timing correction was applied.

cortico-cortical evoked potential · single-pulse electrical stimulation · stereoelectroencephalography · pre-stimulus power · single-trial gain · OpenNeuro ds003708