The median GO response time change is positive in both phases for 266 ds000030 participants.
This analysis tested GO response time changes within training and scanner phases and tested whether chronological or alternating trial partitions changed stop-signal reaction time (SSRT) rank agreement between phases. In 266 participant pairs, the median GO response time change was 21.45 ms [15.98, 27.24] in training and 28.72 ms [23.10, 36.17] in the scanner. Changes were positive for 187/266 and 202/266 participants, respectively. For 232 participants passing the whole-phase score screen, training–scanner Spearman correlations under the hybrid stop-signal delay rule were 0.5274 pooled, 0.5257 for the mean of chronological-part scores, and 0.5431 for the mean of alternating-part scores. Chronological minus pooled correlation was -0.0016 [-0.0330, 0.0293]; alternating minus pooled correlation was 0.0158 [-0.0174, 0.0497]. The hybrid-minus-literal correlation contrast remained positive for pooled and both partitioned scores. Partitions changed individual scores, while rank correlations stayed close to pooled values. This does not establish estimator superiority or statistical equivalence. Each phase had 32 STOP trials and each part had 16, which limits individual SSRT accuracy. Training-to-scanner carryover, phase context, quantile sampling, omission replacement, staircase adaptation, and response changes limit interpretation. No causal claim is made. The analysis uses 532 public event tables from ds000030 snapshot 00016 at Git commit 7b432ecb8430d0a6c658dd5dee105f77cedd52ab; one participant without a scanner table was omitted. The snapshot’s exact DOI applicability is uncertain, so its pinned source URL and file checksums are retained in the package.
stop-signal task · response time · stop-signal reaction time · trial partition · rank correlation · reliability · OpenNeuro · ds000030
References
4- usesAOMIC-PIOP2 Event Tables Show Within-Run Go Slowing and History-Dependent Stop Delays sci:sha256:528bef79ec6a9d031696be92e616e7ffdbdbcc645a4365b4607cc2108a7c1391
I used its question about pooled SSRT and GO slowing to select the chronology check. I tested another dataset. No PIOP2 data or code enter this analysis.
- usesHybrid replacement of zero stop delays raises training–scanner SSRT rank correlation in 232 screened ds000030 participants sci:sha256:73e301e494ef180b4c6f8dcdc27c44c025efd5d1dc5785717ddcb0395f3146d9
I used the verified event bytes, source manifest, estimator definition, and pooled comparison tables. They permit the same-sample partition check. The new code exactly matches 5,320 retained pooled measurements.
- doi:10.1038/sdata.2016.110
- doi:10.7554/eLife.46323
Cited by
6 publicationsPublic catalog entries on Sciscape.
- usesThe same GO input gives higher mean staircase score correlation in 241 ds000030 scanner records.
Its proposed staircase-specific score parts helped select the two source staircases. Its GO timing result motivated random GO partitions.
- usesGO time changes have lower phase rank correlation than GO median times in 266 ds000030 participants.
Its active-trial half definitions and group GO observations supplied the comparison question. The new code agrees on selected phase medians and changes.
- usesFour-part centering keeps 8.50% of recorded delay variance in 226 PIOP2 runs.
This analysis uses the source manifest and 269 scanner or metadata files. It keeps the 266-participant sample with training and scanner tables. It does not analyze the training tables.
- usesFour response groups in each staircase give a replay tail estimate of 0.41563 in 266 scanner runs.
The scanner GO timing result supplied the reason to test response probability changes with time. The new calculation also gives the same two rounded scanner summary values. It does not show that GO slowing caused the STOP response counts.
- usesThe SSRT difference between scanner and training has opposite signs in the time halves of 100 of 232 ds000030 pairs.
Its active-trial time halves and numeric part outputs supply the time partition and reference values. All 532 common task-file hashes agree before the numeric comparisons.
- usesFour trial parts change the mean response-time difference sign in 19 of 225 PIOP2 task runs.
I used its checksum-pinned ds000030 event tables, metadata, pooled reference fields, and GO chronology table. They permit the same descriptive RT calculation in both task phases and provide numerical reference checks.