Data from: Born to aggressive mothers – are certain chicks more susceptible to early life exposure to aggression?
Description
Early development in vertebrates is a period of heightened neural plasticity, making it particularly sensitive to environmental influences. Adverse experiences, such as poor nutrition, disease or social adversity, can have lasting effects on behaviour and cognition. Yet, the extent to which these outcomes are shaped by the interplay between early environmental conditions and genetic predispositions remains poorly understood. We investigated these mechanisms in domestic canaries (Serinus canaria), where mothers often show maternal aggression (MA) towards their nestlings. We hypothesised that early MA exposure would increase impulsivity, promote resource acquisition through reduced neophobia, and enhance aggression, potentially leading to higher dominance ranks at the juvenile stage. Furthermore, we expected that these effects would be amplified under socially challenging conditions (i.e., high social densities), and in individuals genetically predisposed to aggression. To test this, we cross-fostered full clutches within and between MA and noMA mothers, creating four experimental groups to disentangle genetic and environmental influences. Once independent, juveniles underwent a series of standardised behavioural tests. Our results showed trait-specific patterns: aggression-related behaviours and response inhibition were primarily influenced by genetic background, while food motivation was mainly affected by the early-life environment, and neophobia was explained by a gene–environment interaction. These findings support the idea that MA may itself be a heritable trait (the “cycle of violence” hypothesis), perhaps mediated by shared predispositions for impulsivity or poor behavioural regulation.
Files
Steps to reproduce
All statistical analyses were performed in R 4.1.3 (R Core Team, 2013). For all statistical tests, the significance level was set at α = 0.05. To inspect violations of the model assumptions, lmm plots were generated using the packages performance (v0.10.3; Lüdecke et al., 2021)and DHARMa (v0.4.6; Hartig, 2016). To detect possible outliers in our data, we used Tukey’s method to calculate IQR and upper and lower bounds, and we used scatter plots for a visual assessment. When necessary, the data was winsorized to avoid the possible undue influence of extreme values (here, winsorization was only required for the variable time pecking at the closed door from the ET). Backward elimination for non-significant interactions (α = 0.05) based on the Akaike Information Criterium (AIC) value, was used to build the minimum models and p-values were calculated using lmerTest (v3.1.3, Kuznetsova et al., 2017) (by default this package provides type-III Anova). For further information on the specific analysis conducted for each of the behavioural tests see the manuscript: Born to aggressive mothers – are certain chicks more susceptible to early life exposure to aggression? Garcia-Co, C.; van Riel, I., Verbruggen, F.; Morales, J.; Müller, W. (2025). Manuscript submitted for publication.
Institutions
- Universiteit Antwerpen Departement Biologie