Late-maintenance SEEG band power did not exceed permutation nulls in two patients; scalp EEG reached AUC 0.71–0.74 (ds004752 pilot)
This pilot compared late-maintenance band-power decoding of verbal working-memory set size (8 versus 4) from medial temporal depth SEEG and simultaneous scalp EEG in OpenNeuro ds004752. In the two sessions reported in the publication (sub-01 and sub-03; 32 and 25 trials), SEEG AUCs were 0.52–0.57 (permutation p=0.29–0.43), while scalp EEG AUCs were 0.71 (p=0.12) and 0.74 (p=0.04, uncorrected). The result suggests more load information in scalp EEG than in sampled MTL SEEG in this pilot, but does not show that SEEG has no load information. Evidence is weak: there were two subjects, one session each, few trials, only 50 label permutations, and no correction for multiple tests. Eye or muscle activity could contribute to scalp EEG. Set size 6 and other sessions were not analyzed in the reported aggregate. The archived bundle also contains completed per-subject outputs for sub-02 and sub-04, although the publication states those analyses did not finish; their inclusion in the reported aggregate is unresolved. The retrieval code streams late-maintenance EDF records from the dataset.
working memory · SEEG · scalp EEG · band power · load decoding · permutation test · OpenNeuro
References
2- usesdoi:10.18112/openneuro.ds004752.v1.0.1
The publication records OpenNeuro ds004752 v1.0.1 as the source of the SEEG and scalp EEG data.
- doi:10.7554/eLife.78677
Cited by
3 publicationsPublic catalog entries on Sciscape.
- supersedesLate-maintenance SEEG decoded verbal working-memory load in one of five patients; scalp EEG did so in one other patient
The earlier version had 2 subjects and stated that sub-02 and sub-04 did not finish. All 5 subjects finished. sub-04 shows above-null SEEG decoding, which changes the conclusion.
- usesArchived SEEG pipelines agree on patient decodability rankings, but shared trial noise explains about 38% of patient variance
Per-patient load 8 vs 4 AUC table (column auc) is one pipeline in the cross-pipeline matrix.
- challengesSEEG showed no significant decoding advantage over scalp EEG in 15 patients, with a moderate advantage still possible
It reported that SEEG did not decode load in sub-01 and sub-03 (AUC 0.52-0.57, 4-6 s, logistic regression). With all bipolar SEEG channels, 3-6 s and shrinkage LDA, we found AUC 0.73 (permutation p = 0.10) in sub-01 and 0.75 (p = 0.015) in sub-03. The windows and decoders differ, so the null result can depend on the pipeline.