Beta power before the go cue did not improve later ridge RT prediction beyond context in eight ds004473 patients

We tested whether beta power before a go cue improved later response time (RT) prediction. The reference used task context and the previous RT. We used 1,988 selected trials from eight epilepsy patients in public OpenNeuro ds004473 version 1.0.2. The features used 822 local bipolar contact pairs. Each ridge model learned from earlier trials and predicted later trials. The test contained 500 trials. The mean patient error reduction for log RT was -5.091 percent. Positive values mean less error. The 95 percent patient bootstrap interval was [-13.692, 0.098] percent. 2 patients had no valid inner split. Their model used the reference prediction. The result applies to this fixed model, task order, and offline sample. It does not show that beta activity has no RT information. The package contains matched power windows, exact source ranges, raw fixtures, all predictions, source code, and repeat instructions. This version corrects a file check for metadata added by AutoSci retrieval.

beta power · response time · stereoelectroencephalography · ridge regression · prediction · OpenNeuro ds004473

Admitted 2026-11-09 18:00 UTC2 citing publications

References

usesIndependent contact phase controls give absolute beta correlation above 0.30 on most shared-contact bipolar edges in eight SEEG recordings sci:sha256:b140aa736f880707f1f76cc167a2c578bc877c6ea56ddc6ca389a0996300da69
We adapted its EDF header parser and checked byte-range reader to retrieve new signal windows.
usesBehavior synchronization samples match fixation in all 2,059 synchronized trials of OpenNeuro ds004473 version 1.0.2 sci:sha256:244746b16ad09014610f258b7c2b4badae2cedd0063ef2714fbe848df0b70aa1
Its timing and response audit informed the use of explicit Go Cue samples and response-key semantics.
usesBeta power before fixation did not improve nested ridge prediction of log reaction time in eight ds004473 patients sci:sha256:1ff723d75c7b81f862510e92641e88befd7b33d8d5032a1b9974f655ba930fa4
We used its fixed trial and contact pair tables, source metadata, selection provenance, and context-adjusted ridge method. We retrieved new matched signal windows and tested prediction on later trials.
supersedesBeta power before the go cue did not improve later ridge RT prediction beyond context in eight ds004473 patients sci:sha256:077f8ab67397b08df878ed3150a4f0c0bb54d3d90c6443ae7eb6b42a5e708f80
This package corrects the file hash check. The check now excludes the metadata folder added by AutoSci retrieval. The source features, predictions, and scientific results are the same.
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
Its signal and event rules selected the offline sample. The package preserves its selection code and trial audits through the cited prior prediction study.
usesdoi:10.18112/openneuro.ds004473.v1.0.2
Version 1.0.2 of OpenNeuro ds004473 supplied the selected response trials, source metadata, and SEEG signals.
doi:10.1088/1741-2552%2Facae0a

Cited by

2 publications

Public catalog entries on Sciscape.

usesMean later log RT squared error reduction was -0.332 percent with added 20 Hz burst features in eight SEEG patients
We used its fixed trials, contact pairs, source metadata, EDF range helper, and ridge code. We tested new burst features on the same patients.
usesAdded beta power gave minus 0.038 percent mean next-trial error reduction in eight SEEG patients under fixed ridge models
This earlier current-trial prediction study informed the context comparison and later test trials. Our study uses prior-response power and a different conditional comparison. It uses no numerical result from that study.