New code gives the same six next GO fraction differences for 70 ds000030 CONTROL participants.

New Python code reproduces the six archived participant-weighted differences in next-GO correct-response fractions for 70 CONTROL participants in OpenNeuro ds000030. The analysis uses 723 STOP-to-next-GO pairs (168 with stored zero delay and 555 with positive delay) from snapshot 1.0.0. The zero-delay minus positive-delay difference was +20.730107 percentage points (pp) at 500 ms (pointwise bootstrap 95% interval +11.864145 to +29.459583 pp) and -7.068543 pp at 1500 ms (−13.339531 to −1.780561 pp). The code also reports an approximate simultaneous band across the six analysis limits and a paired 500-to-1500 ms change. The reproduction matched all six exact point estimates and archived participant, trial-audit, outcome-count, interval, and unknown-choice results within the reported tolerance. Interpretation is limited by sparse zero-delay observations, adaptive stop delays, uncertain pilot inclusion, unknown outcomes, and the fact that stored zero delay does not establish tone delivery. The participant bootstrap assumes independence, and the added simultaneous band does not have calibrated finite-sample coverage. The analysis is descriptive and does not establish a causal effect.

stop-signal task · next-GO response · correct-response fraction · OpenNeuro ds000030 · bootstrap · reproduction · uncertainty