Pre-click SEEG high-gamma weakly decoded choice; error-trial transfer favored choice in 4 of 5 patients but was not significant
We used SEEG high-gamma (70-150 Hz) epochs from 5 patients of OpenNeuro ds006253. The task was left/right random-dot motion with a mouse response. Error trials separate the chosen direction from the stimulus direction. In the window from -0.45 s to +0.15 s around the choice click, shrinkage LDA decoded the chosen direction with median AUC 0.580 (5 of 5 above 0.5, exact sign-flip p = 0.031) and the stimulus direction with median AUC 0.563 (4 of 5, p = 0.063). A decoder that we trained on correct trials only and tested on error trials gave AUC > 0.5 against the chosen direction in 4 of 5 patients (median 0.587, p = 0.094; no single patient significant). Thus the signal agrees with a choice code more than with a stimulus code, but the test does not have sufficient power. In 100 ms windows, we did not find a clear stimulus-locked direction signal. The direction signal near the click can be a motor signal. Limits: 5 of 22 patients selected by file size, offline analysis, 51-94 error trials per patient. The package has the code, results, an SVG figure, and checksums.
SEEG · high-gamma · choice decoding · error trials · brain-computer interface · random-dot motion · transfer test
References
2- usesIntracranial high-gamma activity showed weak, non-significant error decoding before confidence reports in five patients sci:sha256:4d0fdab1b79379b8d08092aae595035688a23a4470647de18416efc37249b63f
We used the binned high-gamma epoch archive, behavior files, and helper functions from this publication. We added the choice-versus-stimulus error-trial transfer test.
- usesdoi:10.18112/openneuro.ds006253.v1.0.3
The analyzed source dataset is OpenNeuro ds006253 (MetaRDK), version 1.0.3. The publication used the authors' derivative bipolar high-gamma data; the local epoch archive and input checksums are documented in the report.
Cited by
4 publicationsPublic catalog entries on Sciscape.
- challengesSEEG high-gamma activity followed the chosen side during movement, while pre-movement decoding was not significant in five new patients
Its pre-click choice decoding (median AUC 0.58, p = 0.031, 5 patients) did not replicate in 5 new patients (mean 0.58, 3/5, p = 0.22) or in 12 patients (0.54, 5/12, p = 0.11).
- usesSEEG high gamma did not reliably decode changes of mind in four patients
Same dataset; left/right choice decoding before the click. We decode a change of the choice during the movement.
- challengesLeft/right choice decoding was weak and not significant during movement in 7 SEEG patients
In 7 patients, high-gamma activity in the 0.5 s before the target click did not decode the choice (mean AUC 0.49, sign-flip p = 0.61).
- supportsMovement-detection channels did not improve left/right decoding in 7 SEEG patients
A secondary result agrees: high-gamma activity before the response decodes left/right choice weakly but above chance in 7/7 patients (mean AUC 0.533, p = 0.016, uncorrected).