11 Domesticated animals display suites of morphological, behavioural and physiological 12 differences compared to their wild ancestors, a phenomenon known as the 13 domestication syndrome (DS). Domestication experiments, and the convergent 14 patterns seen across domesticated species, have been adduced to support a singular 15 developmental source for the DS. Specifically, the suite of DS traits are hypothesized 16 to arise via selection solely upon tameness [1] resulting in neural crest deficit, which 17 as a developmental by-product gives rise to morphological changes such as white 18 pigmentation and floppy ears [2]. Consistent with this, genomic studies highlight 19 evidence of selection upon genes associated with neural crest development, in e.g. 20 domesticated foxes [3], horses [4] and dogs [5]. However, genes associated with 21 neural crest development were only a subset of many showing selective signatures, 22Results). Then, to test for a moderating effect of breed morphology on behaviour, we 59 designed a Bayesian phylogenetic multilevel meta-analytic model. This model uses 60 the estimated effect sizes from the correlations as its response variable and compares 61 the average effect size between morphological categories. The level of support and 62 effect of morphology were allowed to vary between the different predicted 63 associations through the inclusion of group level effects. We also accounted for 64 repeated measures from the same breeds, as well as the non-independence of breeds 65 due to shared ancestry. We performed two versions of this nested meta-analysis. To 66 test whether the presence of floppy ears, curly tails or white pigmentation predicts the 67 strength of any of the behavioural correlations, we evaluated these traits as binary 68
Supplemental information 103Supplemental information including methods, statistical analyses, two figures and 104 supplemental results can be found with this article online at [link] 105 106 Acknowledgements 107We thank Rasmus Erlandsson for designing the dog illustrations in Figure 1a. 108