The arrow-history contrast has a half-run correlation of 0.038 in 113 ds000030 CONTROL runs.
In 113 CONTROL participants from the OpenNeuro ds000030 scanner stop-signal task, the arrow-history reaction-time contrast had a first-half to second-half Pearson correlation of 0.038 (95% participant-bootstrap interval −0.123 to 0.199). The contrast compared post-STOP slowing after repeated versus changed arrows. The interval includes zero, but this does not show that the population correlation is zero. Overall post-STOP slowing correlated 0.257 (95% interval 0.057 to 0.437), and mean reaction time correlated 0.934 (95% interval 0.897 to 0.958) across halves. The contrast mean increased from 5.755 ms to 30.905 ms. These results measure linear association between halves of one run; they do not measure absolute agreement, stability between visits, or a biological cause. Changes with time, sparse STOP cells, response selection, shared schedules, and arrow/button coupling limit interpretation. The analyzed files are identified by 131 versioned S3 objects and SHA-256 checksums; their correspondence to a named DOI snapshot is uncertain.
reaction time · stop-signal task · arrow history · half-run correlation · measurement consistency · OpenNeuro ds000030
References
5- usesThe post-STOP reaction-time difference was larger with arrow repetition in 125 ds000030 CONTROL runs sci:sha256:fa4d7f19073028c29c6a1212043b960fe26c03a06c2e7330c66c916e875d0a0a
The source manifest, four-cell selection rules, and participant output supplied exact inputs and a computational check. New code calculated the half-run correlations.
- doi:10.1037/a0021800
- doi:10.3758/s13428-017-0935-1
- supportsdoi:10.1038/sdata.2016.110
Provides the task and dataset release context, including the scanner task design. It is not a replication of the half-run correlation.
- doi:10.1016/j.cortex.2024.02.008
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesThe 126 ds000030 CONTROL runs do not contain sufficient STOP-to-GO pairs for the specified half comparison.
This calculation uses the source lock, trial rules, and participant CSV from this publication. These files give the inputs and a pooled-score check.
- usesA trial-independence model gives an error estimate of 85.25% of contrast variance in 125 ds000030 CONTROL participants.
This work uses the 131-file source lock, trial selection rules, and reference participant table. The lock permits fresh retrieval of the same input bytes. The reference table permits numerical checks of the new parser and contrasts.
- usesThe arrow-history contrast has a training–scanner Pearson correlation of −0.016 in 117 ds000030 CONTROL participants.
Its near-zero half-run contrast correlation motivated testing the same measure across the public training and scanner phases. This calculation keeps its adjacency and reference-measure definitions, with new code.