Evolution of the mammalian brain: using phylogenetic comparative methods to study brain evolution in three mammalian clades

The research on brain evolution has almost 50-year long history. Historical and contemporary studies have attempted to relate different physiological, ecological, behavioural and life-history traits to brain structure and function variation. The scale of such inquiries ranges from whole brain size, relative brain size, sizes of different regions to neuronal numbers and sizes, genetic organisation and neurochemistry. Additionally, the methodological frameworks employed have also varied, with methods including different forms of least squares regressions, to generalized linear mixed modelling, lasso regressions, and others.

In this thesis, I address the evolution of brain size variation in three clades of mammals – primates, marsupials and lagomorphs – and at different scale of inquiry – whole brains, brain lobes, and sub-regions of brain areas. In addition, as the choice of analytical methods I apply both a classical pGLS (phylogenetic generalized least squares regression) and a contemporary GLMM (generalized linear mixed model) approach. As an additional innovation I address one of the pervasive methodological issues in PCM studies by dealing with missing data. Instead of using the traditional method of list-wise deletion, in this thesis I explore the utility of a novel phylogenetic method for multiple imputation.

Alongside the three research papers, I provide a chapter in which I explore the largely understudied problem of marsupial cognition, reviewing the literature in this field so as to assist future inquiries attempting to relate the evolution of behaviour and cognition to variation in brain structure.

In the first chapter called “Primate hippocampus size and organization are predicted by sociality but not diet, I investigated the evolution of the primate hippocampus in terms of its size and organization in relation to the evolution of social and ecological variables (home range size, diet and different measures of group size). I found that the volumes within the whole cornu ammonis coevolve with group size, while only the volume of CA1 and subiculum are also predicted by home range size. On the other hand, diet, expressed as a shift from folivory towards frugivory, was not related to hippocampal volume. Interestingly, CA2 was shown to exhibit phylogenetic signal only in certain measures of group size, but not with ecological factors. I also found that sex differences in the hippocampus are related to body size sex dimorphism. These findings support the notion that in primates, the hippocampus is a mosaic structure evolving in line with social selective pressures, whereas certain subsections evolve in line with spatial ability.

The paper is published in the Proceedings of the Royal Society B and can be found online here: https://royalsocietypublishing.org/doi/10.1098/rspb.2019.1712

In the second chapter called “Marsupial Cognition” and published in Encyclopedia of Animal Cognition and Behavior, I provide a broad review of the literature on the study of cognition in marsupials. I provide a comprehensive overview of previous investigations on perception, learning, memory, and other aspects of cognition. The chapter can be found online here: https://link.springer.com/referenceworkentry/10.1007%2F978-3-319-47829-6_1167-1

This review is followed by the third chapter “Testing hypotheses of marsupial brain size variation using phylogenetic multiple imputations and a Bayesian comparative framework”, where I use a new, high-coverage dataset of marsupial brain and body sizes, and the first phylogenetically imputed full datasets of 16 predictor variables, to model the prevalent hypotheses explaining brain size evolution using phylogenetically corrected Bayesian generalized linear mixed-effects modelling. Litter size emerged as the only significant predictor of brain size variation suggesting that studies of relative brain size evolution in placental mammals may require targeted co-analysis or adjustment of reproductive parameters like litter size, weaning age or gestation length. Moreover, in this chapter I propose a rigorous pipeline for similar studies, using phylogenetic multiple imputation and a Bayesian generalized mixed modelling approach.

The paper is published in the Proceedings of the Royal Society B and can be found online here: https://royalsocietypublishing.org/doi/10.1098/rspb.2021.0394

In my last chapter, “Down a rabbit hole: Burrowing behaviour and larger home ranges are related to larger brains in Leporids”, I investigate the evolution of variation in total brain volume and olfactory bulbs in a sample of 18 leporid species in relation to seasonality in precipitation and temperature and test four other hypotheses explaining brain evolution in mammals. Leporid brain size was shown to not be evolving under any seasonality pressures, but burrowing leporids and species with larger home ranges were shown to have larger brains. This finding suggests that burrowing behaviour might buffer seasonality conditions and facilitate increase in relative brain size in this order. Additionally, the study emphasizes the importance of examining the analyses outcomes due to small sample sizes, which in our case resulted in unreasonable lambda estimations, despite the use of phylogenetic multiple imputations to conserve the whole sample size.

The thesis demonstrates that different levels of inquiry yield different results, that are nonetheless informative, and thus the scale of brain size variation should be carefully scrutinized in respect to the actual questions being asked. Different classes of explanations reveal selection pressures or constraints on brain size evolution in different taxonomic levels. Additionally, different methodological frameworks provide different advantages and disadvantages for data analysis, and handling missing data is one crucial and often overlooked aspect of performing rigorous analysis using PCMs.