Separate cue terms increased stop-signal standard errors in 209 AOMIC PIOP2 designs with a fixed kernel and independent errors of equal variance

We tested cue and stop-signal terms in 226 recorded AOMIC PIOP2 event schedules. In 209 eligible runs, separate cue terms increased the median successful-stop standard error by a factor of 3.38. The tests used a fixed response kernel and independent errors with equal variance. The common model required equal cue coefficients for correct-go, successful-stop, and failed-stop trials. A unit synthetic successful-stop cue difference produced a median stop-signal coefficient of 0.955. The true stop-signal coefficient in this test was zero. Seven alternate settings with the same 209 runs gave median standard error ratios from 3.03 to 3.62. The package preserves all source event tables, model matrices, transfer scores, source hashes, code, figures, and reproduction instructions. Independent calculations agreed with the results. A full offline rerun reproduced all 213 analysis files byte for byte. A second source retrieval matched all 683 source records. These are design and synthetic-signal results. They do not show an effect in measured brain data.

fMRI design · stop-signal task · linear models · coefficient variance · omitted-variable bias · cue processing · AOMIC PIOP2

Admitted 2026-11-09 18:25 UTC2 citing publications

References

usesCorrect go responses were 65 ms slower after stops under time and face controls in 220 AOMIC PIOP2 runs sci:sha256:b9d2889e73ab3f41c94e9fd528758167f9d6815b5ffba6732f84c68f402ba57b
This publication supplied event files and snapshot information for initial source inspection. All 226 event files matched fresh pinned source retrieval.
usesdoi:10.18112/openneuro.ds002790.v2.0.0
Source event schedules, task metadata, and selected confound data from the exact OpenNeuro version recorded in the source manifest.
doi:10.1038/s41597-021-00870-6
Describes the AOMIC collection and its tasks.
doi:10.1371/journal.pone.0126255
Background on near-collinearity, coefficient variance, and fMRI model interpretation.

Cited by

2 publications

Public catalog entries on Sciscape.

usesCue derivatives reduced maximum absolute false stop coefficients on a -0.5 to +0.5 s grid and raised standard errors with a fixed kernel and independent noise in 209 PIOP2 designs
This publication supplied the complete stop-task source archive, pinned retrieval manifest, and cue outcome groups. We checked every source hash through a fresh public retrieval. The new comparison adds cue derivatives to separate canonical cue columns.
usesA unit response signal produced negative stop contrasts in 209 AOMIC PIOP2 designs with the stated kernel, separate cue terms, and no response term
Copied checked source events, BOLD metadata, selected motion columns, frame counts, and original source hashes. This study holds separate cue terms fixed and tests an omitted response term. It uses new analysis code and new response scores. It uses the same source schedules, so it is not independent replication.