Zero-time choices permitted both sex-adjusted age slope signs for two-back minus zero-back d-prime in ds005270 with positive-time choices fixed
We tested the age slope of corrected two-back minus zero-back d-prime in 158 adults from OpenNeuro ds005270 version 1.0.0. We fixed positive-time choices from accuracy under a valid two-choice task assumption. We permitted either binary choice at each eligible zero-time trial. These choices are sensitivity scenarios, not recovered responses. The exact sample bounds for the slope, adjusted for recorded sex, were -0.436 to +0.241 d-prime units per decade. Complete trial assignments attained both endpoints. Thus both signs were possible under the stated assumptions. A person bootstrap gave an approximate outer 95% set of [-0.569, +0.353]. All planned bound checks also permitted both signs. A later check split the recorded error age slope of +2.784 percentage points per decade into a zero-time contribution of +3.033 and a positive-time contribution of -0.249. This split does not establish a cause. The package includes the endpoint proof, source files and hashes, code, all assignments, all resampling records, figures, and run instructions. An independent check verified 460000 bootstrap values. A fresh source copy reproduced all 18 result and figure files exactly. The result does not show that age has no effect on working memory.
n-back task · working memory · d-prime · zero response time · sensitivity analysis · partial identification · exact bounds · age association · cognitive control · OpenNeuro
References
6- usesTwo recent-repeat trials had a higher mean error rate than other non-targets in 158 ds005270 two-back records sci:sha256:6e355f0bb4051c8efac61c59e0483d4e4c99c3828c388cce9a686cd6c3f03617
We used the archived source copy, fixed source manifest, retrieval code, repository tag record, and target-history audit. We checked all 43125 letter inclusion flags against its ledger. A fresh download then checked all source hashes. Our outcome is a new discrimination contrast; we did not repeat the earlier main lure-error estimate.
- usesdoi:10.18112/openneuro.ds005270.v1.0.0
The analysis used the n-back event tables and participant metadata from this exact dataset release.
- doi:10.1016/j.neurobiolaging.2021.03.019
- doi:10.1162/jocn_a_01685
- doi:10.3758/BF03203619
- doi:10.3389/fpsyg.2017.00352
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.