Slide Slam L15
Conserved and divergent connectivity principles across the primate phylogenetic tree
Guilherme Freches1,2, Katheryne Bryant3, Joanna Sierpowska1, Christian Beckmann1,2,3, Rogier Mars1,3; 1Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands;, 2Department of Cognitive Neuroscience, Radboud University Medical Centre, Nijmegen, The Netherlands, 3Wellcome Centre for Integrative Neuroimaging, Centre for Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, University of Oxford, Oxford, United Kingdom
The primate brain has undergone massive shifts in shape, size, microstructure or connectivity across evolution. While comparative neuroscience typically aims to uncover these changes and associate them with differences in cognition, it is equally useful to look at common components across different species. These conserved elements can serve as a baseline on which to compare species with a common ancestor and offer valuable insights on the principles of organization of a taxonomic group. Recently, non-negative matrix factorization has been proposed as a way to map white matter bundles and their respective grey matter networks from diffusion tractography data. Here, we adapt the technique to use connectivity blueprints instead, leveraging on their common space approach to map common and divergent connectivity principles across the primate phylogenetic tree. Publicly available connectivity blueprints for human , chimpanzee and macaque were used. Connectivity blueprints are obtained by computing the probabilistic tractography matrices seeding from the cortex to the rest of the brain and multiplying them with individual tract maps for 22 common tracts. Thus, at the vertex level, they represent the probability of streamlines from a given vertex to connect to each of the 22 chosen tracts. Five sets of bilateral connectivity blueprints were created: 3 with the average blueprint of the 3 species in question, a primate ancestor common blueprint by concatenating all 3 species’ blueprints and a great ape common ancestor blueprint by concatenating only human and chimpanzee blueprints. Non-negative matrix factorization was applied to each of the 5 sets under consideration and the resulting white matter tract components were compared in the following way: the macaque was compared with the primate common ancestor, the great ape common ancestor with the primate common ancestor, and finally, the chimpanzee and human components with the great ape common ancestor. Our analysis gives several important insights. First, it is clear that the only major shift from the primate common ancestor to the macaque is at the level of the vertical occipital tract (VOF), confirming its influence on the Macaque-unique visual fields. Second, we see that despite the arcuate fasciculus (AF) being one of the most iconic 'human-unique' fiber bundles, some aspects of AF connectivity are shared between humans and chimpanzees, suggesting that the foundation of our language system may already have been in place as far as 6 million years ago. Also, at the great ape common ancestor level, a strong motor component can be seen in the form of novel connections from the frontal aslant and superior thalamic radiation. Finally, it appears that the greater changes that happened to the chimpanzee since its divergence from the great ape common ancestor are mostly related to inferior longitudinal fasciculus reweighting and some modifications to the temporal portion of the cingulum bundle. In summary, we found conserved and species- specific principles of connectivity suggesting that the white matter basic infrastructure for language may have been present already at the level of our last common ancestor with the chimpanzee.