The mean contrast-space overlap change has opposite signs for two nuisance models in 26 ds000102 run-2 records
We compared changes in contrast-space overlap while varying contrast response-term count separately from shared nuisance response-term count. In 26 run-2 event schedules from OpenNeuro ds000102 snapshot 00001 (Git commit fdc1ce49d524fc1e77e5b763f10acb02f403c581), changing the contrast from one to three terms changed mean overlap by +0.000057984684856360566 with one nuisance term and by -0.000015878769676166805 with three nuisance terms. The mean change was small, and its sign depended on specified drift and assumed noise settings. The overlap is an unsigned geometric measure of modeled residual spaces, not a measure of brain activity, label truth, or a population effect. Run-2 condition fields disagreed on 304 of 624 trials, and the displayed arrow sequence could not be verified. We used no brain images. These results describe finite-sample model geometry for the recorded event schedules; they do not establish a biological effect.
contrast-space overlap · nuisance regression · hemodynamic response function · functional MRI · Flanker task · model sensitivity · OpenNeuro ds000102
References
4- usesThe ds000102 snapshot 00001 run-2 mean contrast space overlap with three response terms is in all three calculated 95% label reference intervals. sci:sha256:ff0de0765adeb67be69cc5d3ce365b6d139e84a2443bf614cdaaf769a53eb86f
We used its source manifest and model specification. Its 156 primary diagonal values are numerical references for our new calculation.
- usesThree response terms leave mean run-2 contrast-space overlap at 0.05982 across 26 ds000102 models sci:sha256:4f4966793ef8347c16374b64dc0041d69387fa936a25d8dca7ba9e48f506c491
Its model changes contrast and nuisance counts together. Its stated limit gave the question for our crossed count grid.
- doi:10.1016/j.neuroimage.2010.10.046
- doi:10.1016/j.neuroimage.2008.10.065
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesThe condition contrast retention is higher for field B in 26 ds000102 run-2 records with response-time terms.
We used the immutable source manifest, primary continuous event/HRF specification, and 52 primary single-column overlap values. They fixed the data version and model definition and gave a numerical reference check.
- usesThe two mean change signs are equal in 18 of 26 record removal checks for primary ds000102 run-2 models.
We used its model equations, pinned source manifest, and 468 primary reference values. They gave the baseline for the record removal and sample-clock checks.
- usesTiming and width terms add less than 0.15 percentage points to mean base contrast retention in 26 primary ds000102 run-2 models
We used its response equations, fixed source manifest, and saved grid. They supplied the model and 1,092 overlap values for comparison with the new directed base-response calculation.