Applying a geometric morphometric surface semilandmark-based approach for assessing sexual dimorphism in cranial bones

Traditionally, sex estimation in human skulls has been carried out applying non-metric and inter-landmark distance approaches. Such estimation has substantially improved with the introduction of geometric morphometric semilandmark-based registration, representing a useful tool for  capturing the complex morphology of 3D  surfaces, particularly in human skulls. Taking in account these improvements, the main purpose of this study is to show the benefits of these  techniques for assessing by means of a quantitative, hypothesis testing approach the  differential phenotypic expression of sexual  dimorphism in human cranial traits, using the  frontal bone, the mastoid process of the temporal  bone, and the zygomatic bone as proxies.  According to our main results: i) accuracy and  precision of sex estimation is higher when the  shape variable is included in the analysis than  when it is not, ii) numerical classification tends to  be more accurate in males than in females for  both shape and size components of sexual dimorphism, and iii) a single trait or two traits in  combination would be more successful in sex  classification than the three traits taken as a  whole. The introduction in the present study of a  surface semilandmark-based approach for  quantitatively assessing sexual dimorphism in human skull considerably improves the capture of  sexual dimorphic signatures, corroborating the  importance of the information carried by the surface of the frontal bone, mastoid process, and zygomatic bone. Our results suggest that shape  and size components should be considered as  relevant factors in making a differential, bone- dependent sex assessment in human crania.