Task adjustment in pair selection increased mean offline confidence forecast error in 21 MetaRDK SEEG recordings at penalty 10
We tested pair selection for offline forecasts of reported confidence. We used public SEEG data from 21 people in MetaRDK, OpenNeuro ds006253 version 1.0.3. Each rule selected eight contact pairs from earlier trials. One rule ranked confidence correlations. The adjusted rule removed six linear task effects before ranking partial correlations. Both final readers used the same task inputs and original signal means. At penalty 10, mean squared error rose from 0.03126 to 0.03485 on 2,283 later trials. Confidence used a 0–1 scale. The mean increase was 0.003587, with a descriptive 95 percent participant interval of 0.000767 to 0.007268. Fourteen people had more error. The signal window was 50 to 450 ms after movement. The increase also occurred at fixed penalties 1 and 100. An additional check, specified after the primary result, left task slopes without a penalty and also gave an increase. The cohort was reused, and the source signals were processed offline. This test does not establish a live BCI, a command-reliability measure, or a stimulation target. The package preserves inputs, trial forecasts, model states, code, three figures, source hashes, retrieval recipes, plans, and independent checks. A fresh source rebuild matched one participant exactly. All 44 small source files matched their archive. A full repeat matched all 11 result files byte for byte.
confidence forecasting · SEEG · pair selection · partial correlation · ridge regression · MetaRDK · offline brain-computer interface
References
4- usesOffline SEEG readers with eight contact pairs and penalty 10 gave mean choice AUC 0.526 across opposite cues in 19 MetaRDK participants sci:sha256:c22d122d956476b6b42bf2d4ff09d54b02692d4e2c7e7005a613581a561a487f
This publication supplies all 21 compact signal files, extraction audits, source metadata, retrieval code, and source receipts. The new study compares confidence pair-selection rules with different eligibility rules from its choice test.
- usesAdding eight SEEG signals to a task model increases offline confidence forecast error on later trials in five of six MetaRDK sessions sci:sha256:f153e0af57732bd1a666747e35778c555a9091f0c980875e17ea5b2460f566ee
This earlier confidence test supplies the six task inputs and method context. Its six-person cohort overlaps the current 21-person cohort. The new selection comparison is not an independent confirmation.
- usesdoi:10.18112/openneuro.ds006253.v1.0.3
Source dataset for the 21 participants and the supplied SEEG signals and confidence reports.
- supportsdoi:10.1038/s41467-025-61744-8
Source paper describing the acquisition and signal-processing context discussed here; this publication uses the pinned supplied derivative and does not repeat its regional or causal tests.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesA fixed SEEG confidence score gave no clear gain in later offline error forecasts in 21 MetaRDK recordings
We used its 21 compact MetaRDK recordings, trial screens, source metadata, extraction code, and retrieval records. We checked all copied input hashes and added a separate calibration block and objective choice error target.