Statistical protocol for analyzing 3D muscle attachment sites based on the “Validated Entheses‐based Reconstruction of Activity” (VERA) approach

Abstract

AbstractIn anthropological sciences, the areas of the bones where muscles or ligaments attach (“entheses” or “muscle attachments”) are routinely used for reconstructing habitual physical activity based on skeletal remains. However, several previous studies have argued against this practice, considering the low measuring repeatability of most traditional evaluation techniques and the former lack of supportive experimental evidence for an actual effect of physical activity on entheseal morphology. In 2016, I proposed a repeatable virtual approach for reconstructing habitual physical activity using entheseal morphology, which has been validated based on several experimental studies on laboratory animal species and human skeletons with uniquely detailed and longitudinal occupational documentation. This method, which was later named the “Validated Entheses‐based Reconstruction of Activity” (VERA) method, relies on a precise protocol for virtually delineating and quantifying entheseal areas on three‐dimensional bone models. The obtained area measurements are then subjected to size adjustment and multivariate statistical analyses, which have been experimentally shown to reveal group correlations among entheses that reflect habitual coordination among muscle groups. This paper presents the first critical review of all the various statistical approaches previously employed for analyzing entheses using VERA (so far), providing the first coherent step‐by‐step protocol for the complete univariate and multivariate analysis of entheseal three‐dimensional data. These include the introduction of a new, simple, and fast plotting technique for the initial screening of group entheseal patterns prior to multivariate analysis (i.e., the “VERA Bar‐Chart”), followed by the calculation of the “group entheseal patterning index” (hereafter GEPI%).