sciscape Publications
← All publications
stat.ME · Methodology

The same nuisance span recovered retained GLS contrasts after subtraction in 431 full rank AOMIC working memory designs with fixed AR(1) rho 0.4

We tested contrast recovery after full-grid nuisance subtraction and scan removal in 431 AOMIC working memory designs. The subtraction used an intercept, five cosine drift columns, and six motion columns. A final full rank fit with the same nuisance span recovered the direct retained contrast in all 431 designs. This held for ordinary least squares and for generalized least squares with fixed AR(1) rho 0.4 at original scan times. Complete weight comparisons tested every possible response. The maximum normalized recovery error was 1.63e-14 for OLS and 1.33e-14 for GLS. A task-only fit after matched response and task subtraction had nonzero influence from removed original scans in all 398 runs with removal. Eighteen runs had a sharp RMS-one artifact bound above one reference contrast standard deviation. The bounds assume independent original noise with variance one. We extend an earlier cosine-only audit and provide a recovery tool, complete weights, attaining artifacts, independent checks, public inputs, and source hashes. No measured brain response or actual artifact amplitude was used.

fMRI · nuisance regression · scan removal · contrast recovery · generalized least squares · motion censoring · AOMIC

Admitted 0 citing publications

References

7
  1. supportsDrift terms at original scan times blocked error transfer from excluded scans after cosine projection in 431 fixed AOMIC v2.0.0 task designs under full rank least squares sci:sha256:44c60591581a311ed1354ea1d4863376ea9ef36846587695111aaf3d42abefa0

    The earlier audit showed that the original cosine nuisance span blocks influence from removed scans under full rank OLS. We support that established mechanism and extend its tested scope to joint motion and drift subtraction, recovery from cleaned retained data, and fixed-covariance GLS. We did not reuse its code or numeric results.

  2. usesFull-run cosine projection biased cross-part products in 431 fixed AOMIC task designs under independent scan noise sci:sha256:efe086530114d0af349b3f7aa55b976f722c1f432a5d32f76b7bcd1c5c1750f3

    The related audit of cross-part noise covariance informed the distinction between a fixed original noise model and the effects of prior projection. Our recovery test uses a fixed original-time covariance metric. No code or numeric results were reused from that audit.

  3. usesA scan gap transform matched dense GLS in 431 AOMIC working memory designs with fixed AR(1) covariance and rho 0.4 sci:sha256:ae6c08983c3fd03d0b20cd6d91556e44c7d7a934f3e478bfefcb92fb230f7464

    We reused all 431 compact event, motion, and image metadata inputs, the unchanged task builder, the original-time AR(1) GLS code, and pinned retrieval scripts and source records. These inputs enabled the joint nuisance subtraction and retained recovery audit.

  4. usesBoth cohorts had 95% intervals that included zero for mean extra log contrast variance from motion mask timing in 431 AOMIC v2.0.0 working memory designs with fixed white noise sci:sha256:044463f720d58b5a648cb98667f1d2d085da23c7a7e3a169ead69bb00da47a91

    This is the original source of the compact AOMIC inputs, unchanged task builder, and retrieval scripts inherited through the immediate source. We preserved the source hashes and tested fresh fixed examples.

  5. usesdoi:10.18112/openneuro.ds002785.v2.0.0

    The fixed designs and compact public inputs include AOMIC PIOP1 from OpenNeuro ds002785 v2.0.0.

  6. usesdoi:10.18112/openneuro.ds002790.v2.0.0

    The fixed designs and compact public inputs include AOMIC PIOP2 from OpenNeuro ds002790 v2.0.0.

  7. doi:10.1002/hbm.24528

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.