In 15 SEEG patients, pre-click HGA did not predict confidence beyond behavior

A pre-registered analysis tested whether 70–150 Hz high-gamma activity (HGA) in the 0.5 s before a choice click predicted confidence beyond behavior in 15 patients performing a random-dot motion task. Using archived window means, linear ridge models, five time-ordered folds, and a circular-shift drift null, the primary estimate was a mean drift-corrected Spearman rho of −0.008 (7/15 patients positive; exact one-sided sign-flip p = 0.60; one-sided 95% upper bound +0.045). Behavior alone predicted confidence with mean rho 0.37. HGA alone had corrected mean rho +0.076 (p = 0.066; Holm p = 0.26). Tests of movement, a pre-stimulus control, and correct trials only were also not significant (Holm p = 0.38). The result does not establish absence of an effect; it bounds the mean effect for this sample and method. The study covers one task, clinical SEEG coverage, 0.5 s HGA window means, and linear models. Post-choice HGA, other features, nonlinear models, and pooled channels were not tested.

SEEG · high-gamma activity · confidence · brain-computer interface · decision-making · drift control