The mean beta power time course contrast after stimulation was 0.072 dB in 13 PAL2 SEEG participants
We tested whether recent electrical stimulation changes the beta power time course of a word pair with no stimulation. We used OpenNeuro ds005494 version 1.0.1. The test included 13 participants, 62 SEEG channels, 269 exposed pairs, and 570 control pairs. Exposed pairs started 1.398 to 1.680 seconds after the previous train ended. We compared beta power in the first and third seconds of each pair. We adjusted for pair position and list order. The mean exposed-minus-control contrast was +0.072 dB. The 95% participant bootstrap interval was -0.223 to +0.446 dB. The two-sided sign-flip test gave p=0.770 under its symmetry assumption. This sample showed no clear group difference. Participant estimates ranged from -0.805 to +1.882 dB. This result does not prove that a decoder is safe or that stimulation has no effect. The package contains exact voltage epochs, source versions, event files, code, tables, four figures, and instructions. A timing audit found that direct conversion of onset time does not match the event sample indices. One preserved stimulation trace supports use of the sample index.
SEEG · beta power · electrical stimulation · brain-computer interface · OpenNeuro · PAL2
References
1- usesdoi:10.18112/openneuro.ds005494.v1.0.1
Uses the released SEEG recordings, events, and channel metadata from OpenNeuro ds005494 version 1.0.1.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesA SEEG model fitted on lists with no train gave higher mean recall loss after stimulation than a rate model in 13 people at C = 0.01
This ancestor supplied the native SEEG epoch extraction, source object versions, remote-channel selection, and timing audit inherited through the direct parent.
- usesSEEG power reduced mean recall loss by 0.005 nats versus task features on excluded PAL2 lists in 13 people
Uses native SEEG epochs, cohort selection, source metadata, pinned EDF source versions, extraction code, and timing audit. The new outcome is later paired-word recall.
- usesAdding SEEG power increased mean test loss by 0.035 nats for recent stimulation prediction on matched PAL2 lists in 13 people at C = 0.1
This study reused the native PAL2 epoch archives, channel and trial selection, source metadata, pinned source ranges, retrieval code, timing audit, and spectral power definition. It added metadata-based list matching, complete-list test folds, channel-level classifiers, and preserved predictions. It is an extension of this cohort, not an independent replication.