Little evidence for group differences in 2D:4D ratios based on sexual orientation after adjusting for publication bias.
Level 4 - case-series / case-control
Meta-analytic re-analysis of observational cross-sectional and case-control studies.
PubMed 41861700 · doi:10.1016/j.yhbeh.2026.105921
What was done
The authors reanalyzed a meta-analytic dataset (Swift-Gallant et al., 2025) on the relationship between second-to-fourth digit ratios (2D:4D) and sexual orientation. They applied Robust Bayesian Meta-Analysis (RoBMA), a framework that models uncertainty across multiple sources including effect presence/absence, heterogeneity, and publication bias.
What was found
The abstract reports qualitative Bayesian evidence without numerical point estimates: - RoBMA indicated moderate evidence against 2D:4D differences between homosexual and heterosexual women, alongside moderate to strong evidence of publication bias. - In men, RoBMA found moderate evidence against a right-hand difference, but a small left-hand effect moderated by publication status. - Adjusting for publication bias substantially weakened evidence for 2D:4D differences by sexual orientation, and evidence for publication bias was not explained by unmodeled heterogeneity.
Why it matters
2D:4D digit ratios are widely used as a putative proxy for prenatal androgen exposure in sexual orientation research. This re-analysis suggests that previously reported associations are likely artifacts of publication bias rather than robust biological differences.
Limits
The abstract does not provide exact effect sizes, Bayes factors, number of primary studies, or total participant sample size. As a re-analysis of observational literature, conclusions remain constrained by the measurement and reporting limitations of the underlying primary studies.