A 3D morphometric approach between the skull and the endocast integration in Pan troglodytes, Gorilla gorilla and Homo sapiens

Cranial morphology in fossil hominins is largely variable with different observed patterns of integration related to intraspecific and interspecific factors [1]. One of the key factors to understanding the evolution of Homo is the identification and quantification of patterns of integration in the skull and endocast [2]. By studying these patterns, we can obtain more information to understand how the brain affects or is affected by the morphology of the cranial structure and then apply this data to reconstruct partial fossil specimens [3]. During the last decades, numerous studies on cranial and endocranial integration and modularity have been carried out separately. We used new computational techniques and virtual anthropology methods to investigate the interactions between cranial and endocast regions. In detail, we focus the investigation on the patterns of covariation in Homo sapiens (N=20), Pan troglodytes (N=20) and Gorilla gorilla (N=20). The main goal is to detect shared or species-specific integration patterns that can later be applied to extinct hominin species. We defined 6 regions on the skull –viscerocranium, frontal, parietals, temporals, occipital and sphenoid bones– and 8 regions on the endocast –frontal, parietal, temporal, cerebellum and occipital lobes, as well as brainstem, endo-sphenoid area and olfactory bulb– using 105 landmarks, 492 curve points and 1448 surface sliding semilandmarks. We performed a separate Procrustes registration on the skull and on the endocast. To establish the shared and species-specific pattern of integration among the three genera we have used the two-block partial least squares (PLS) method pooling the dataset by species and sex [4]. The PLS analyses has been performed by defining a network taking into account all the combinations of the 14 skull and endocast regions. To test whether the observed patterns of covariation were species-specific or not, we compared the slopes of the first axis of covariation through Bootstrap permutations [5]. Our results show that (i) facial complex covaries with the posterior part of the endocast base, mainly with the cerebellar region and, to a lesser extent, the brainstem. Occipital bone (ii) is integrated with the brainstem region, (iii) frontal bone with temporal lobes and endo-sphenoid area, and (iv) temporal bones with temporal lobes and endo-sphenoid area. Finally, (v) parietal and occipital lobes do not covary with any of the cranial regions, possibly due to the morphological differences in males and females. All the significant covariations patterns between skull and endocast –except for the covariation patterns of parietal bones with frontal lobes and frontal bone with endo-sphenoid that are common for the 3 taxa– are not fully shared but, in all cases, they are at least shared between 2 of the 3 groups. In conclusion, the investigation of the shared and species-specific pattern of covariations between skull and endocast can help us to understand in more depth the mechanisms of cranial and brain shape evolution in fossil human species.