Shared time for four people was within the independent circular shift reference range in 68 NNDb adults at a 0.2 mm FD limit

We measured common retained scan time after removal of samples with large estimated head motion in OpenNeuro ds002837 version 2.0.0. The fixed sample contained 68 adults from nine film groups. Each person supplied 1,740 one-second samples. The primary rule retained framewise displacement (FD) at most 0.2 mm, without extension. Mean shared time for all distinct four-person subsets was 64.6085%. The exact expectation under independent uniform circular shifts was 64.4911%. The one-sided shift test gave p = 0.116 from 9,999 draws. The score was within the central 95% shift reference range. This result does not establish equivalence or independence of motion. The two largest film groups retained only 187 and 168 samples for all people. With one sample before and two after each failure also removed, these totals fell to three and zero. A reusable tool calculates exact observed and expected retention for every subset size without listing the subsets. Independent checks covered 394,396 subsets. A complete rerun from 175 freshly retrieved source files gave identical numerical results. The package preserves source bytes, masks, code, offsets, results, and instructions. The statistical test assumes exchangeable independent circular phases. It does not estimate a population effect or neural signal quality.

framewise displacement · head motion · fMRI · shared scan time · circular shifts · motion censoring · naturalistic neuroimaging