Answers are grounded in this paper and cite their sources.
Code and derived data for "Referencing a measure to the respondent caps the cultural variation it was built to detect" (Cui & Liu, 2026). The repository holds the code that computes the pitch-structure descriptors, the statistics that size the within-unit loss, and one CSV per supplementary table, from which every statistic reported in the text can be re-derived. python tests/test_core.py runs 9 checks with no external data; python code/reproduce.py re-derives 22 reported quantities and exits nonzero on any drift. Third-party corpora are not redistributed here. The 266 f0 traces are those released with Ozaki et al. (Zenodo, doi:10.5281/zenodo.10612357; audio at osf.io/mzxc8); the 21-society recordings are those released with Hilton et al. (Zenodo, doi:10.5281/zenodo.5525161). The World Color Survey, DaMuSc, PHOIBLE, WALS, Glottolog, the Anthology of Chinese Folk Songs, the Essen Folksong Collection and SymbTr are available from their respective distributors. Code: MIT. Derived tables: CC BY 4.0.
No takes yet. Share an insight, caveat, or question.
Tianhua Cui (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: