The 126 ds000030 CONTROL runs do not contain sufficient STOP-to-GO pairs for the specified half comparison.

In 126 selected control scanner runs from OpenNeuro ds000030, every run had only 18–22 raw STOP-to-GO pairs, fewer than the 24 pairs necessary for a planned half-run comparison requiring at least three trials in each of six cells per half. No run completed both halves under this rule, so the primary half-run correlation cannot be estimated. At the full-run level, 90 participants met the three-trial rule: the pooled mean contrast was 14.2556 ms and the fixed 50:50 STOP-outcome contrast was 14.3601 ms; their difference was −0.1045 ms (95% bootstrap interval −2.9437 to 2.6706 ms). At the one-trial threshold, 79 participants had complete halves; fixed-mixture half agreement was r = −0.0730 (95% interval −0.2884 to 0.1391), compared with pooled agreement r = 0.0421 in the same participants. These lower-threshold results do not answer the planned question or establish an outcome-mixture cause. The analyses use only task event files, rely on participant independence, and do not estimate corrected reliability or biological causality.

stop-signal task · reaction time · half-run agreement · outcome mixture · fixed-mixture contrast · Pearson correlation · bootstrap · OpenNeuro ds000030 · trial coverage