At 6 mA, pulse-split SEEG direction agreement exceeded a before-pulse control in five patients but weakened across odd and even pulses

We tested whether the stronger measured direction of a reciprocal SEEG pair remained stronger in separate pulse sets. We used public OpenNeuro ds004457 version 1.0.2. Fixed metadata rules selected 69 stimulation sites, 111 receiving contacts, and 329 reciprocal pairs in five patients. At each site, we used the same two contacts for stimulation and bipolar recording. Stimulation current was 6 mA. The late window was about 50–250 ms. The first and last five pulses gave 65.5% mean direction agreement after pulses and 54.5% in a matched control before pulses. The equal patient mean difference was 11.0 percentage points. Four of five patient differences were positive. The exact two-sided patient sign-flip p value was 0.1250. The primary difference had a descriptive patient bootstrap interval of 2.6 to 18.7 points. The planned odd and even pulse split gave a mean excess of 2.1 points, with p = 0.8750. Raw energy gave high agreement in both after and before windows. The result depends on pulse split and baseline choice. Five patients limit group inference. The package contains pulse windows, raw source hashes, code, all pair scores, and fixed checks. The result concerns recorded voltage strength order. It does not establish anatomical connection direction or clinical value.

stereo-EEG · electrical stimulation · direction agreement · corticocortical evoked potentials · pulse reproducibility · normalized voltage · OpenNeuro