From Fossils To Mind Workshop 2021: Nicole Labra Avila

Comparison of the morphological evolution of hominin brains by the alignment of the cortical sulci in endocast

Nicole Labra Avila


(1) UMR 7194, CNRS, PaleoFED team, Département Homme et Environnement, Muséum national d’Histoire naturelle, Paris, France
(2) Department of African Zoology, Royal Museum for Central Africa, Tervuren, Belgium
(3) NeuroSpin, CEA, Université Paris-Saclay, F-91191, Gif-sur-Yvette, France


The morphological comparison of the brain cortex of different extant species is a difficult task and it is even more challenging if we speak about extinct hominin species. The only clues to a possible configuration of their cerebral cortex are the traces left by the cerebral grooves in the endocranium of the fossilized skull. The endocasts constitute one of the main elements in the study of the evolutionary processes among different hominin species, which led to the current brain configuration. In this context, we present the adaptation of a method previously used in neuroimaging with MRI images to quantify anatomical differences between human, chimpanzee and gorilla brains but to be used in endocasts. This method, known as DISCO, is based on a non-linear registration technique that uses sulci as explicit landmarks to estimate a diffeomorphic deformation field for each species by the coregistration of their sulci.

Then, the pairwise deformation fields between the species are extracted, encoding the total deformation that must be applied to the brain of a source species in order to map its sulci onto the corresponding sulci of a target species. The determinant of the Jacobian of the deformation fields quantifies the local amount of expansion (>1) or contraction (<1) in each point of the brain, between two given species. This method might have the potential to give quantitative results as the evolution of the volume of a given area across species. In principle, we have considered only four primary sulci which are deemed to be present in 3 endocasts of different species (Homo erectus, Homo sapiens and Homo neanderthalensis) but in the future, we would like to be able to use average endocasts of each species with semi-automatically delimited sulci to have a better accuracy of the results.