A meta-analysis of correlated behaviours with implications for behavioural syndromes: mean effect size, publication bias, phylogenetic effects and the role of mediator variables

By L. Garamszegi, G. Markó, G. Herczeg. Evolutionary Ecology, 2012

143 citations recorded by the source corpus.

In evolutionary and behavioural ecology, increasing attention is being paid to the fact that functionally distinct behaviours are often not independent from each other. Such phenomenon is labelled as behavioural syndrome and is usually demonstrated by phenotypic correlations between behaviours like activity, exploration, aggression and risk-taking across individuals in a population. However, published studies disagree on the strength, and even on the existence of such relationships. To make general inferences from this mixed evidence, we quantitatively reviewed the literature using modern meta-analytic approaches. Based on a large dataset, we investigated the overall relationship between behaviours that are expected to form a syndrome and tested which factors can mediate heterogeneities in study outcomes. The average strength of the phenotypic correlation between behaviours was weak; we found no effect of the phylogeny of species but did observe significant publication bias. However, even accounting for this bias, the mean effect size was positive and statistically different from zero (r = 0.198). Effect sizes showed considerable heterogeneity within species, implying a role for population-specific adaptation to environmental factors and/or between-study differences in research design. There was a significant positive association between absolute effect size and repeatability of behaviours, suggesting that within-individual variation of behavioural traits can set up an upper limit for the strength of the detected phenotypic correlations. Moreover, spatial overlap between the contexts in which different behaviours were assayed increased the magnitude of the association. The small effect size for the focal relationship implies that a huge sample size would be required to demonstrate a correlation between behaviours with sufficient statistical power, which is fulfilled only in very few studies. This suggests that behavioural syndromes often remain undetected and unpublished. Collectively, our meta-analysis revealed a number of points that might be worth to consider in the future study of behavioural syndromes.

Subfields: Ecology, Evolution, Behavior and Systematics

View Paper: https://www.semanticscholar.org/paper/941c64cab6dd648c5cbb1186fed6d0b40247eca7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology, Evolution, Behavior and Systematics

Citations in the corpus

Boldness–exploration behavioral syndrome: interfamily variability and repeatability of personality traits in the young of the convict cichlid (Amatitlania siquia) (2015, background): “…shown that animal personalities and behavioral syndromes covary with ecological conditions, depend on the state of the individual, and can be maintained by frequency-dependent selection (Dall et al. 2004, 2012; Sih and Bell 2008; Garamszegi et al. 2012; Wolf and Weissing 2012; Sih et al. 2015).”
An empiricist guide to animal personality variation in ecology and evolution (2014, background): “The extent to which these different reproductive strategies are state-dependent (e.g., dependent on condition when first encountering a suitable breeding habitat) is currently unclear however there are no obvious genetic or morphological predictors of the strategy that an individual will follow.”