The mean patient median energy change beyond baseline was -0.60 percent of the baseline scale in 176 SEEG blocks from five patients

We tested whether repeated electrical pulses changed signed waveform energy in public SEEG data. We used OpenNeuro ds004457 version 1.0.2. The analysis compared the first and second pulse halves. It subtracted the matching change before stimulation. The sample had five patients, 19 receiving contacts, 176 consecutive blocks, and 590 contact-block links. Each block had 8 to 18 retained good biphasic 6.0 mA pulses. The equal mean of patient median changes was -0.60 percent of the fixed patient baseline energy scale. The descriptive patient bootstrap 95 percent interval was -3.56 to +2.36 percent. The patient sign test gave p = 0.75. The conditional order test gave p = 0.8365, under joint pulse-label invariance. We proved an exact linear formula for balanced half energy differences. We checked all half scores with direct pulse products. The package preserves source epochs, raw retrieval code, source hashes, code, all link results, and null labels. Five patients, shared noise, and uncertain recording reference limit interpretation. The result does not prove stable responses or a neural cause.

SEEG · electrical stimulation · pulse order · waveform energy · signed energy · intracranial EEG · balanced-label test