Toggle Main Menu Toggle Search

Open Access padlockePrints

A pre-registered meta-analysis based on three empirical studies reveals no association between prenatal (amniotic) cortisol exposure and fluctuating asymmetry in human infants

Lookup NU author(s): Dr Gareth RichardsORCiD



This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).


Developmental instability (DI) reflects an organism’s inability to develop an ideal phenotype when challenged by genetic and environmental insults. DI can be estimated via the proxy measure of fluctuating asymmetry (FA), i.e., the small random deviations from perfect bilateral symmetry observed in the morphology of paired traits. The mechanisms involved in the genesis of FA in human populations are relatively unknown, though animal research indicates that hormonal processes may be involved. As maternal stress during pregnancy is detrimental to various developmental processes, elevated prenatal cortisol may represent a causal factor in the subsequent emergence of an asymmetrical phenotype. The main purpose of this pre-registered meta-analysis based on three empirical studies was to investigate whether mid-trimester amniotic cortisol levels predict subsequent FA in finger lengths of infants from Germany, Portugal, and the UK. No statistically significant relationships were observed, and meta-analytic combination of the effect size estimates yielded a null result. We did, however, detect significant positive correlations between the cortisol present in the amniotic fluid and maternal plasma in the Portuguese cohort, and observed that FA in the German cohort was significantly lower at 70-months than at either 9- or 20-months. Taken together, the current findings run contrary to animal research showing that elevated prenatal corticosterone exposure leads to increased FA. However, this may be because a single cortisol assay obtained via amniocentesis is an inadequate proxy for average gestational exposure, and/or that prenatal cortisol levels at an earlier (i.e., first rather than second trimester) stage of pregnancy is what explains variance in subsequent FA.

Publication metadata

Author(s): Bushell W, Heil M, Ventura T, Gomes MC, Körner LM, Lawrenz J, Schaal NK, Richards G

Publication type: Article

Publication status: Published

Journal: Evolutionary Biology

Year: 2021

Volume: 48

Pages: 54-66

Print publication date: 01/03/2021

Online publication date: 15/01/2021

Acceptance date: 27/11/2020

Date deposited: 26/11/2020

ISSN (print): 0071-3260

ISSN (electronic): 1934-2845

Publisher: Springer


DOI: 10.1007/s11692-020-09523-9


Altmetrics provided by Altmetric