Phase-lag features showed no clear error decrease across sessions for eight-pair SEEG load readers in 15 patients under assumed timing
The test did not show a clear decrease in mean prediction error from the added phase-lag features. We used public SEEG data from 15 people with epilepsy. Each decoder used eight bipolar pairs. It fit on the first numbered session and tested on the last. The cohort had 607 fitting trials and 630 test trials. The mean Brier score was 0.2418 for power and 0.2548 for power plus phase lag. The paired change was +0.0129, with a 95% person-bootstrap interval from -0.0007 to +0.0267. A lower Brier score is better. The test used C = 0.1 and assumed timing within each exported epoch. The package preserves code, spectra, raw audit trials, predictions, and source checks. The result does not establish a causal connection or a stimulation target.
stereoelectroencephalography · phase lag · coherency · working memory · session transfer · Brier score · brain-computer interface
References
7- usesEight SEEG pairs showed no clear gain over fixation for letter load across sessions in 15 patients under assumed epoch timing sci:sha256:1c0c74901b6082055c1889d816c6520f050a9fff799c8dce26e8f044f7ce2e9a
This study supplies the first and last session cohort, candidate band powers, source metadata, pinned object versions, range hashes, and EDF reader. We retrieve the pinned ranges again to make complex spectra. We also rebuild and check all candidate powers.
- usesSeparate power and phase penalties reduced mean Brier error against a shared penalty on later trials in eight selected SEEG records sci:sha256:52c83dfcc8e81bd04f93edff10c72e5c45f7632911769c2f4448175b2904737c
This earlier test of separate power and phase penalties supplies context for a model limit. Our main test uses one shared penalty. The earlier test used another dataset.
- usesAdded imaginary coherency increased mean Brier error for tuned linear readers in eight SEEG records with fixed 500-sample inputs sci:sha256:2263fe7cecddd0d4616d9f4cfb58a07f322990304de5c63f2a497260c14e6bb5
This earlier phase-feature test supplies context for the new comparison. It used another dataset and a response task within recordings. The present test uses memory load across numbered sessions.
- usesSelection of eight SEEG pairs before four-fold validation increased scores for shuffled load labels in 15 patients sci:sha256:f17517a3b8149c42a11bd270f93adf0beb753802a20ca25f66a78e84ba85a44b
This study supplies the prior-correct trial rule and the three-band method for pair selection. We apply the rule to the first and last sessions. We select eight pairs from the fitting session only.
- usesdoi:10.18112/openneuro.ds004752.v1.0.1
Source SEEG dataset used for the cohort, signals, and event records.
- supportsdoi:10.1016/j.clinph.2004.04.029
Established imaginary-coherency method underlying the signed phase-lag feature.
- supportsdoi:10.7554/eLife.78677
Task description used to support the assumed fixation, encoding, and delay sequence; it does not verify alignment within each exported epoch.
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesA 12-trial confidence update lowered Brier error versus an offset but had higher error than a class-rate constant in 15 SEEG session tests at C = 0.1 under assumed timing
The fixed eight-pair Hann reader inputs passed through this work. We kept those pair lists and feature rules fixed for the calibration test.
- Three DPSS tapers showed no clear Brier error decrease across sessions in 15 SEEG load readers at C = 0.1 with fixed pairs and assumed timing
[Legacy relation: reproduces] Fresh EDF range retrieval and Hann power calculations reproduced the source primary power reader at C = 0.1. The largest input difference was 2.66e-15. The largest test probability difference was 4.19e-10.
- usesThree DPSS tapers showed no clear Brier error decrease across sessions in 15 SEEG load readers at C = 0.1 with fixed pairs and assumed timing
This work supplied the 15-person first and last session cohort, fixed eight-pair lists, BIDS metadata, pinned EDF range records, and reader code. These files set a common test for Hann and three DPSS tapers.