Group mean subtraction changes SSRT correlation from 0.550 to 0.542 in 232 ds000030 pairs with planned delays.

In 232 paired training and scanner records from OpenNeuro ds000030 v1.0.0, the pooled planned-delay stop-signal reaction-time (SSRT) correlation was 0.5496399499955592. After subtracting separate training and scanner means within source diagnosis groups, the correlation was 0.542092341868097; the change was -0.007547608127462246 (conditional 95% bootstrap interval: [-0.031282339033623895, 0.012904277507574197]). Group mean subtraction changed the observed association by a small amount in this selected sample. The centered correlation remained positive in all nine tested sample and delay conditions. These arithmetic scores use public task tables and source diagnosis labels. The intervals condition on sample screens and observed group counts and assume independent participant records. The analysis does not establish correlation equivalence, causal diagnosis or delay effects, individual score agreement, or measurement reliability. Each run had 32 STOP trials, fewer than the cited recommendation of at least 50 for standard group comparisons under specified conditions. Reaction-time units and delay fields follow the prior calculation; custom ReactionTime metadata do not state a unit, and measured sound-onset times are unavailable. The results describe selected records and do not establish clinical population interval coverage.

stop-signal reaction time · SSRT · group mean subtraction · Pearson correlation · covariance · bootstrap · neuropsychiatry · OpenNeuro ds000030