SEEG power gave no clear gain over retained task events for 100 ms keypress forecasts without Go Cue messages in eight people
We tested 100 ms keypress forecasts in eight public SEEG recordings from OpenNeuro ds004473 v1.0.2. All Go Cue messages were removed before the reader processed task events. Exact Fixation, ISI Onset, and past Response messages remained. The test had 26,879 windows and 343 positive windows. Adding 32 past SEEG power inputs to a fitted task-event reader changed mean Brier error by -0.000108. The descriptive 95% person interval was -0.000586 to +0.000343. This test did not establish an average benefit. A planned fixed-penalty test gave a small benefit in all eight people, but its mean errors were higher than those of the tune-selected readers. The package preserves source recipes, event masks, observer code, fitted models, predictions, scores, three figures, and independent checks. This is an offline test with exact retained messages and reused clinical data. It does not test uncued behavior, live control, or brain stimulation.
brain-computer interfaces · stereoelectroencephalography · SEEG · keypress forecasting · event availability · Brier score · causal history · logistic regression
References
3- usesFixed SEEG power models gave no clear gain beyond task events for continuous 100 ms keypress forecasts in eight people sci:sha256:5a550bbec827507f5125afef09901460929b3204e989fb4167f8ca343bdf5140
Used all eight continuous SEEG feature files, source events, selected contacts, target and split rules, and parent fitting and raw-retrieval code. Reproduced all 32 selected and fixed parent models before testing Go Cue message loss with a new causal observer and new fits.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
The analysis uses the eight public recordings in OpenNeuro ds004473 version 1.0.2; the exact-version dataset metadata record CC0.
- doi:10.1088/1741-2552%2Facae0a
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.