No effects of androgen receptor gene CAG and GGC repeat polymorphisms on digit ratio (2D:4D): a comprehensive meta-analysis and critical evaluation of research

Abstract

A series of meta-analyses assessed whether differentially efficacious variants (CAG and GGC repeat-length polymorphisms) of the human androgen receptor gene are associated with digit ratio (2D:4D), a widely investigated putative pointer to prenatal androgen action. Extensive literature search strategies identified a maximum of 18 samples (total N=2909) vs. 5 samples (N=1497) for the CAG-related vs. GGC-related meta-analyses, respectively. In contrast to a small-sample (N=50) initial report, widely cited affirmatively in the literature, meta-analysis of the entire retrievable evidence base did not support any associations between CAG variants and right-hand, left-hand, or right-minus-left-hand 2D:4D. Effects of GGC variants on digit ratios likewise were almost exactly null. For the CAG literature, time trend analysis indicated shrinking effects among more recent studies. Both quantitative and qualitative citation analyses documented that citation bias exists in the research literature: CAG-related studies yielding larger effects were cited more frequently within the same time unit, and the initial, unreplicated report continued to be cited frequently and mostly solely as well as confirmatively, while non-replications were cited much less often. The meta-analytical null findings, along with several further strands of evidence consistent with these, undermine one validity claim for 2D:4D as a retrospective pointer to prenatal testosterone action. Discussed are alternative interpretations of the evidence and avenues for future research.