Co-evolutionary dynamics of mammalian brain and body size

Despite decades of comparative studies, fundamental questions remain about how brain and body size co-evolved. Divergent explanations exist concurrently in the literature for phenomena such as variation in brain relative to body size, variability in the scaling relationship across taxonomic levels and between taxonomic groups, and purported evolutionary trends. Here we resolve these issues using a comprehensive dataset of brain and body masses across the mammal radiation, and a method enabling us to study brain relative to body mass evolution while estimating their evolutionary rates along the branches of a phylogeny. Contrary to the rarely questioned assumption of a log-linear relationship, we find that a curvilinear model best describes the evolutionary relationship between log brain mass and log body mass. This model greatly simplifies our understanding of mammalian brain-body co-evolution: it can simultaneously explain both the much-discussed taxon-level effect and variation in slopes and intercepts previously attributed to complex scaling patterns. We also document substantial variation in rates of relative brain mass evolution, with bursts of change scattered through the tree. General trends for increasing relative brain size over time are found in only three mammalian orders, with the most extreme case being primates, setting the stage for the uniquely rapid directional increase that ultimately produced the computational powers of the human brain.