Applying multivariate methods to dental development sequences for subadults from archaeological samples

BRACHETTA-APORTA, NATALIA; D'ADDONA, LUCAS ANDRÉS

International Journal of Osteoarchaeology

John Wiley and Sons Ltd

Lugar: Nueva York; Año: 2020

1047-482X

Dentition has been traditionally used to estimate the age of subadults from archaeological samples. However, age estimations face some difficulties related to the factors that affect tooth preservation and the lack of standards with adequate references for past populations. Moreover, the differences within and among populations associated with the origin of the population, the sex of the individuals, and other factors make it difficult to develop standards in samples with no adequate references. The aim of this work was to explore a procedure that allows sequencing individuals according to their degree of dental development by employing simultaneously the information available of the target individual and the rest of the sample. The information of dental development as well as post-mortem tooth loss was recorded in computed tomography images and by macroscopic examination in subadults from two archaeological samples from South America. Then, we performed a multiple imputation to estimate the degree of development of missing teeth and carried out a principal component analysis for summarizing the information of the formation of permanent teeth. The first principal component was used to obtain a dental development score and to establish a continuous sequence of the individuals on the basis of their degree of dental maturation. Our results show that multivariate analysis can be applied for ordering individuals according a sequence of maturation in archaeological studies, even in the presence of post-mortem tooth loss. Furthermore, multivariate methods allow to evaluate the contribution of each tooth to maturation sequences. As a whole, the procedure proposed here is a promissory tool for cases where the use of chronological age sequences built from nonpopulation-specific standard can lead to misrepresentations of data.