The mean stop time has no finite upper limit in 226 independent AOMIC PIOP2 race models with finite go support and positive hidden tail mass
We tested known finite go laws and fixed delay supports from all 226 AOMIC PIOP2 stop-signal records in OpenNeuro ds002790 version 2.0.0. In each independent race model, two chosen finite stop laws had exactly the same full recorded response law. Their mean stop times differed by 1000 ms. A third law used a 10% trigger failure atom and had the same recorded law. A proof shows that positive mass beyond the last observable comparison leaves the finite mean without a finite upper limit. The full law identifies cell masses, which give sharp limits for a capped mean. At a 500 ms cap, the median example range width was 1.620 ms. These are ideal model limits, not human estimates or confidence intervals. The package preserves exact proofs, a general audit tool, all source bytes and hashes, response laws, endpoint witnesses, independent checks, and a complete repeat calculation.
stop-signal task · race model · identifiability · trigger failure · capped mean · structural ambiguity · response law · AOMIC-PIOP2
References
8- usesA count rule with an offset of one reproduces all recorded stop delays in AOMIC PIOP2 version 2.0.0 sci:sha256:dadcaf2a4a3f57d9efcba907d4c3968f6343046788d1f26a523b8eaa6c40aef1
Its description of the observed delay controller informed the fixed recorded-delay scope and the separate proof for adaptive delays based on past records.
- usesIntegration estimates of stopping time correlate less across halves than go response times in 132 selected AOMIC PIOP2 runs sci:sha256:f8c60202fcda29057738eb637e45397b0734a14048aaf77d7483b78722c0a330
We retrieved its 233 unchanged ds002790 source files and pinned source manifest, read its retrieval code, and checked all bytes with a fresh original-source download.
- supersedesThe mean stop time has no finite upper limit in 226 independent AOMIC PIOP2 race models with finite go support and positive hidden tail mass sci:sha256:bad0985b213715387d85be1dc3ce8a9eea4aef11b90fde8ac1fc4ae74d0cbb65
This version corrects the package hash checker to ignore the metadata folder added by AutoSci during retrieval. The source files and scientific results are unchanged.
- usesdoi:10.18112/openneuro.ds002790.v2.0.0
The analysis uses all stop-signal event files in this exact OpenNeuro snapshot; the preserved dataset description records the matching version-specific DOI and CC0 license.
- supportsdoi:10.1038/s41597-021-00870-6
The AOMIC data paper describes the task and reports a 4000 ms trial length; this does not establish the response recording deadline.
- doi:10.7554/eLife.46323
- doi:10.1037/0033-295X.97.2.309
- doi:10.3758/s13428-015-0695-8
Cited by
3 publicationsPublic catalog entries on Sciscape.
- uses100 ms response time bins reduced the largest 500 ms capped mean stop time ranges in 223 of 226 finite independent AOMIC PIOP2 race models with exact go and summary probabilities
Used the fixed source manifest, finite independent race model, full-law cell identification result, and chosen stop-law example. New programs calculate sharp limits after response information is removed or put in bins and the largest range over all compatible summary laws.
- usesAn audit with fixed bins and trial counts limits false rejection of independent stop-signal races with a stable go distribution
This finite-design proof separated hidden-tail ambiguity from sampling error. It informed the scope of our count audit and its sampling bounds. Our audit does not identify hidden stopping-time tails.
- usesFour delay groups narrowed capped median stop time bounds in 347 fixed independent race models with known lognormal go laws
This proof separates uncapped stop means from capped targets under finite go support. It helped define our different median target under known continuous go laws.