A fixed beta power model gave 48.2 percent balanced accuracy for key choice on later trials in eight patients in ds004473 v1.0.2
This study tested whether beta power helped predict the recorded key on later trials. It used public SEEG data from OpenNeuro ds004473, version 1.0.2. The sample had eight patients, 1,988 key trials, and 822 adjacent contact pairs. A fixed ridge model used earlier trials for training and later trials for testing within each task condition. Response beta features gave 48.19 percent balanced accuracy. Trial order alone gave 50.00 percent. The mean patient gain was -1.81 percentage points. The exact two-sided sign-flip p value was 0.296875. A descriptive patient bootstrap interval was -4.98 to +0.71 points. This result applies to the fixed model and split. It does not show that SEEG lacks choice information. The package preserves the plan, source arrays, code, all predictions, fitted models, raw checks, and reproduction instructions. An independent solver checked all 128 fits. A second run reproduced all numerical outputs. Fresh raw checks passed for one test trial per patient.
stereoelectroencephalography · SEEG · beta power · key choice · balanced accuracy · ridge regression · temporal validation · OpenNeuro
References
4- usesBehavior synchronization samples match fixation in all 2,059 synchronized trials of OpenNeuro ds004473 version 1.0.2 sci:sha256:244746b16ad09014610f258b7c2b4badae2cedd0063ef2714fbe848df0b70aa1
This metadata audit reported a disagreement with the declared correctness rule in sub-1. It informed our use of recorded key codes without correctness-based trial selection.
- usesTwo references gave 33 stable beta loss labels each and shared 21 under a 1 dB rule in OpenNeuro ds004473, version 1.0.2 sci:sha256:c079974997528aff1b762b9f6cf35144d660cbb33285ade3a4de59cb6ee4f33b
This publication supplied the eight trial power archives, source metadata, raw range hashes, and retrieval code. The new study used these fixed inputs to test key choice on later trials.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
This is the exact OpenNeuro dataset version used for the SEEG recordings, task metadata, and source archive inputs.
- doi:10.1088/1741-2552%2Facae0a
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesFixed high gamma and beta models gave 48.1 and 48.5 percent balanced accuracy for response keys on later trials in eight SEEG recordings
This work supplied the earlier beta key prediction study, the split within task condition, the fixed ridge method, and raw test fixtures. We compared high gamma and beta with matched power calculation.