Before-stop response time baselines lower the mean after-stop outcome contrast in selected ds000030 control trials

We compared go response times before and after successful and failed stops in OpenNeuro ds000030, version 00016. We used correct go trials directly before and after each stop. Each participant had at least three selected stops per outcome. The scanner sample had 111 control participants and 1,334 selected stops. The successful-minus-failed contrast was +11.37 ms after the stop and +23.78 ms before the stop. The mean contrast in the change from before to after was -12.41 ms, with a 95% interval of -28.66 to +3.83 ms. The earlier training task had 103 participants and a before-stop contrast of +25.18 ms. Baseline subtraction lowered the mean outcome contrast in both tasks. The primary scanner interval includes zero. Stop outcome was not assigned at random. The results do not identify a causal effect of stopping. Training is a related sample from the same study. The package contains the original event files, fixed-version retrieval code, source hashes, derived trial tables, all planned checks, figures, and reproduction instructions. Nine code checks passed. A separate calculation matched all main trial and participant results. A second run reproduced all 22 numerical output files byte for byte.

stop-signal task · response time · cognitive control · baseline subtraction · behavioral data · reproducibility