Department of Anthropology, University at Albany
Friday Morning, 301D
Recent years have seen the increasing integration of geometric morphometrics (GMM) and finite element analysis (FEA), an engineering method that can be used to address questions in evolutionary biomechanics. These efforts represent important methodological advances, but a full integration lies before us. The fundamental question needing to be answered is when given a sample of finite element models (FEMs), how can we determine whether or not the stress and strain patterns preserved within them are similar or different? More specifically, how is it possible to move beyond either qualitative assessments of FEMs or simple quantitative assessments based only on a miniscule fraction of the stress and strain data contained within them? A statistical framework for doing so will enable analysis of vast amounts of stress and strain data while incorporating spatial information, yet serious methodological and conceptual problem confront us. This problem is illustrated through FEA of chimpanzees and fossil hominins. GMM methods were used to identify six Pan troglodytes crania falling at the extreme ends of variation within the species. FEMs were constructed of these individuals, and of crania of Australopithecus africanus and Paranthropus boisei following virtual reconstruction using GMM. When loaded to simulate molar and premolar bites under conditions that control for size, the chimpanzees exhibit strains that vary in magnitude but that appear broadly similar in distribution. Strains in A. africanus are broadly chimpanzee-like, while those in P. boisei are not. However, these assessments are not based on a statistical analysis. Could they be?
This project was funded by a grant from the National Science Foundation Physical Anthropology HOMINID program (NSF BCS 0725126).