Models with temporal derivatives have more error after cue removal than after cue time changes in 84 ds000030 schedules.
Across 84 stop-signal schedules from 126 controls in OpenNeuro ds000030 snapshot 1.0.0, cue removal produced more relative projection error than changing cue times in all flagged schedules and all 54 forward comparison groups. With an isolated failed-stop predictor and its exact temporal derivative, median errors were 37.675812% for cue removal and 4.934010% for cue time change; adding the derivative reduced the median cue time change error by 7.792628% within schedules, while cue removal error fell by 0.203759%. The smallest cue-removal minus cue-time-change difference across forward groups was 17.521141 percentage points. These are calculations on constructed predictors, not observed brain responses or evidence that a tone was delivered. The 54 groups reuse the same schedules and are not independent samples. Results depend on the stated hypothetical cue timing, HRF, sampling grid, drift removal, and model columns; no image noise model, biological tests, significance tests, confidence intervals, or independent verification were used.
stop-signal task · temporal derivative · hemodynamic response function · fMRI · event schedule · projection error · OpenNeuro · ds000030
References
5- usesTwo methods agree on cue predictor changes to 1e-9 in 126 ds000030 control runs. sci:sha256:389d2890e7a3886f007771f9bfb71758ffd5cd5f39b21c32d34c4658cbe584dc
I used its gamma cumulative distribution integration method in new code. I compared the outputs with quadrature before the derivative comparison.
- usesCue removal changes the tested failed-stop predictors more than a cue time change in 84 control runs. sci:sha256:ec38dcb3ea2d8063a8f950782a778d6d4d3791c20a5b2e558f3c9481d67b3e35
I used its cue models, nine settings, source manifest, selection, and primary raw metrics. These inputs made the derivative comparison and numerical check possible.
- doi:10.18112/openneuro.ds000030.v1.0.0
- doi:10.1038/sdata.2016.110
- doi:10.7554/eLife.88652
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.