Two condition correlations from correct trials agree with Mennes et al. (2011) at two decimal places under trial_type in ds000102 snapshot 00001.
We compared participant-level Pearson correlations between mean reaction time (RT) and coefficient of variation (CV) in OpenNeuro ds000102 snapshot 00001 with three values in Mennes et al. (2011): congruent r=0.44, incongruent r=0.52, and overall r=0.38. With correct positive RTs pooled across two runs, the correlations using trial_type were 0.435638, 0.521161, and 0.409387. The two condition values agree with the report at two decimal places; the overall value does not. In a planned grid of 64 settings, no single setting matched all three. A post-grid calculation using all positive RTs for the overall value gave r=0.3842236698951937, a possible explanation that does not establish the historical procedure. The archive has 26 participants and 1,248 event rows. The two available condition fields disagree on 304 run-2 rows; their true labels are unknown. Bootstrap intervals are wide and include zero for the field differences. This consistency check does not identify the historical processing, true labels, or a biological cause.
reaction time · coefficient of variation · Pearson correlation · Flanker task · OpenNeuro · ds000102 · behavioral variability · reproducibility · bootstrap
References
2- usesRun-2 flanker reaction-time contrasts in ds000102 depend on the condition encoding sci:sha256:d6a63429439d411dd1ec62210b1c560cb7a14e504645d1f2caef43ead9079d1e
The inspected report gives the condition fields and source file hashes. This analysis uses them to retrieve the same public text files and calculate the source-report comparison.
- supportsdoi:10.1016/j.neuroimage.2010.10.046
Mennes et al. (2011), section 3.2, supplies the three reported behavioral correlations used for this consistency check.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesThe condition time summaries do not identify the labels of two ds000102 snapshot 00001 trials.
The report gave the correct positive response rule and two condition correlations. These values enabled the finite test of label choices with the same summaries.