In seven SEEG patients, movement-detector timing was stimulus-locked (mean latency slope 0.02); post-stimulus hard negatives did not change it
We trained a high-gamma (70-150 Hz) movement-onset detector (shrinkage LDA, movement vs pre-stimulus rest) for each of 7 SEEG patients in OpenNeuro ds006253. We slid the detector over held-out trials. We regressed the first detection time on movement reaction time (pre-registered latency-slope test). The slope was near 0 (mean 0.02; range 0.00-0.09; sign-flip p = 0.016 vs a slope of 1). A threshold-free half-max analysis and an exploratory calibrated threshold gave the same result. Thus the detector timing followed the stimulus, not the movement, although the movement-vs-rest AUC was 0.82. Training with post-stimulus hard negatives increased the movement vs post-stimulus AUC (0.68 to 0.73) but did not change the slope (p = 0.13). The package contains the plan, code, per-trial results, figure and input hashes.
movement-onset detection · brain-computer interfaces · stereo-EEG · high-gamma activity · latency slope · shrinkage LDA · reaction time · stimulus-locked activity
References
6- usesA latency-slope test separates voice- and cue-locked speech detectors in synthetic SEEG; real-data testing was blocked by dataset download failure sci:sha256:a290ff98ade8b64c5bcc20ed2f4654899f82bbf31e0c9d1af6a6eaa2d82b8452
We applied the latency-slope test (slope of detection time vs response time; 1 = response-locked, 0 = cue-locked) of this synthetic study to real SEEG movement data.
- usesA movement detector produced stimulus-linked false alarms in two new SEEG patients, with effects persisting across seven patients sci:sha256:18b6dd07c7fa2fba85f4f0d322571a3ffe58f5678b9d770a85ff1a069c4b319c
We used the archived 50-ms HGA epochs of sub-07 and sub-21 from this package.
- usesIntracranial high-gamma activity showed weak, non-significant error decoding before confidence reports in five patients sci:sha256:4d0fdab1b79379b8d08092aae595035688a23a4470647de18416efc37249b63f
We used the archived 50-ms HGA epochs and behaviour files of sub-02, 04, 12, 13, 23 from this package.
- supportsIn five patients, a 70–150 Hz SEEG movement detector also scored visual stimulus periods highly sci:sha256:15f05b5465a987bcbeecc6cc162a2bce898262ac7004b2e313258f644479e9f8
db41d3a9 found that the movement-onset detector gives high scores after stimulus onset. Our per-trial timing test agrees: the first detection follows the stimulus (slope 0.02 vs reaction time), and stimulus-aware training does not fix the timing.
- usesdoi:10.18112/openneuro.ds006253.v1.0.3
The publication analyzes SEEG movement data from OpenNeuro ds006253; the saved source record identifies version 1.0.3 and its dataset DOI.
- doi:10.1101/2024.03.23.586414
Cited by
6 publicationsPublic catalog entries on Sciscape.
- supportsStimulus-locked labels detect movement above chance but underperform onset-locked labels in seven SEEG patients
That work found detector onset follows the stimulus. Our stimulus-locked labels capture the pre-onset part of the movement response as well as onset-locked labels do, consistent with this.
- supportsTime-locked training did not reduce movement-onset timing error in seven SEEG patients
It found that hard negatives did not move first detections toward movement onset; a different training change (time-locked bins) also did not decrease onset error here.
- supportsReaction time was not associated with stimulus-locked or movement-locked SEEG detector scores in 15 patients
The post-stimulus score (about 0.2-0.3 of the movement score) did not change with time to movement in 15 patients. This agrees with a stimulus-locked first detection.
- usesAn unchanged SEEG decoder detected movement above chance in six out-of-sample patients and pre-movement activity in 13 patients
Its result that the first detection follows the stimulus is used to interpret the PRE_MOVE scores.
- usesPost-stimulus hard negatives reduce visual-stimulus response in a SEEG movement detector, but movement-vs-rest performance was not preserved in 7 patients
Showed that first detections follow the stimulus; motivates the hard-negative test.
- challengesReaction-time deconvolution links pre-movement SEEG detector scores to movement preparation in 15 patients
That work found that the first detection follows the stimulus; deconvolution with RT jitter attributes most pre-onset score to a movement-locked part (15/15), and finds no stimulus-locked part at 0-0.6 s.