Task switch cost is greater for response repetition in the recorded correct trials of 121 ds000030 controls.
In 121 CONTROL participants from the OpenNeuro ds000030 version 1.0.0 scanner task, the mean task switch cost was 149.417963206579 ms for repeated responses and 75.11937612950983 ms for changed responses. The difference was 74.29858707706917 ms (95% participant bootstrap interval 44.145581671685306 to 105.02622044615693 ms). The analysis used 10,002 selected adjacent-trial transitions from recorded correct trials and reconstructed color/shape task transitions from adjacent cues. All six planned sensitivity checks had positive differences. Source labels disagreed with reconstructed transitions in seven runs, and a later audit found 135 recorded incorrect responses that matched an inferred shape-button mapping; the primary analysis retained recorded correctness. This is an observed association in this sample. It does not identify a neural or cognitive cause, establish a new behavioral pattern, or show individual reliability. Trial selection, sparse cells, feature overlap, task context, and other design limits constrain interpretation.
task switching · response repetition · response time · cognitive control · OpenNeuro · behavioral analysis
References
4- usesdoi:10.18112/openneuro.ds000030.v1.0.0
The analysis uses the scanner task event files from the recorded ds000030 version 1.0.0 source.
- doi:10.1038/sdata.2016.110
- doi:10.3389/fpsyg.2022.1003298
- doi:10.3758/s13414-023-02708-2
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesThe mean task switch cost is larger with short cue intervals in 114 ds000030 CONTROL runs.
I used the seven-run label-error report to remove F/FORMA runs before the group RT estimate. I used its cue map for the design audit. Its response-repetition result supplied a reason for the comparison with equal response-transition weights.