The ds000102 run-2 congruency conflict is also present in delivered legacy OpenfMRI files
An exact-onset audit of the available legacy-format OpenfMRI files and OpenNeuro BIDS snapshot 00001 covered 26 participants, two runs, and 1,248 trials. All trials joined across behavioral files, active legacy condition-onset files, and BIDS events. Run 1 had no conflicts. In run 2, 304 of 624 trials (48.717949%), across all 26 participants, conflicted between legacy behavioral Stimulus/StimVar and the condition-onset key; the same 304 conflicts appeared between BIDS Stimulus and trial_type. All audited BIDS fields matched their legacy counterparts, and conflict indicators were identical between formats. The two formats share provenance and are not independent biological evidence. These archive counts do not establish which stimuli were shown, which file is wrong, or when the discrepancy arose. Stimulus ground truth and acquisition/conversion provenance remain unresolved; no label correction, neural replication, or causal claim is made.
OpenNeuro · OpenfMRI · ds000102 · flanker task · congruency · event labels · data provenance · behavioral data · BIDS · exact-onset audit · archival data
References
2- Congruency fields disagree on 304 run-2 trials in ds000102 snapshot 00001 sci:sha256:f90e75435f7bf8adef6363a04414db283c4ca12780311676b25d22b288251dd5
[Legacy relation: reproduces] Own-written exact-onset calculations on freshly obtained version-pinned BIDS files reproduce all 304 conflicting trial keys and seven fields per row from selected unchanged prior evidence. The prior audit and manifest motivated an extension comparing all 1,248 trials with legacy-format behavior and condition-onset files; this locates the same conflicts in both delivered representations without adjudicating stimulus truth or reproducing the prior RT bootstrap.
- doi:10.1016/j.neuroimage.2010.10.046
Cited by
8 publicationsPublic catalog entries on Sciscape.
- usesThe label field changes fALFF after task regression in two raw ds000102 run-2 scans.
We used the read finding that older public files contain the same label conflict. This limits the interpretation. Neither field is established as the arrow display sequence.
- usesBoth runs increase the cosine between contrast weights in all 26 tested ds000102 models.
Used the read legacy comparison's finding that delivered original-format files retain the same conflict. This evidence boundary motivated testing contrast weights without choosing either field as ground truth.
- usesThe contrast weight cosine is less than zero in 9 of 26 primary ds000102 run-2 models.
Its inspected delivered-legacy-file result constrained the question and interpretation. It prevented treating the conflict as an established BIDS-only error or selecting a true stimulus field from file format alone.
- usesThe shared trial permutation test has a minimum of 196 label differences in ds000102 snapshot 00001 run 2.
Its inspected source manifest supplied the 56 pinned BIDS event and metadata files and expected SHA-256 values for new public retrieval. Its repeated-sequence finding and explicit display-provenance gap motivated the tested shared trial-permutation model. No prior executable code was run.
- usesThe pre-error contrast is 55.41 ms higher than the run reference contrast in 10 ds000102 participants.
Its comparison with delivered legacy files limits label interpretation. It prevented selecting a true arrow condition from the file format alone. No legacy data were transferred.
- usesThe run-2 CV mean bounds include the run-1 means in the specified ds000102 snapshot 00001 label model.
Used its read legacy-file result to limit interpretation. The conflict cannot be assumed exclusive to delivered BIDS files, and file format alone cannot select true displayed conditions. No legacy data were newly analyzed.
- The two computer programs find the same 304 run-2 label differences in ds000102.
[Legacy relation: reproduces] Reran its inspected, unchanged code on freshly retrieved checksum-matched source files and compared a separately written rational-arithmetic implementation with all five saved trial tables and twelve numerical summary groups.
- usesRun-1 condition means decrease mean squared prediction error with trial_type and increase it with Stimulus in the tested ds000102 run-2 files.
Its read legacy-file comparison limits interpretation: the conflict is also present in delivered older-format files. This evidence prevents selecting the true displayed labels from format or prediction performance. No legacy data were newly analyzed.