Skip to Main content Skip to Navigation

Variabilité de la sénescence du squelette humain. Réflexions sur les indicateurs de l'âge au décès : à la recherche d'un outil performant.

Abstract : Age at death assessment of the adult skeletal material, while fundamental, is one of the most difficult problems in Anthropology. Prior to establishing a reliable methodology, we analysed the various biological aspects of the problem : variation in the process of biological ageing and the value of current methods of skeletal indicators of age at death. This study enabled us to establish the fundamental sources of error. There is no linear relationships between chronological age and indicators of age ; there is a great complex variability in the process of skeletal ageing and this is frequently ignored by those who are in search of precision and not reliability.

Considering these issues, a new methodology is proposed. We developed a new scoring system for some skeletal indicators (the pubic symphysis, the auricular surface of the ilium and the sternal end of the fourth rib) in order to improve repeatability and to take into account morphological changes variation in the indicators. We examined reference samples from six different geographical areas in order to include the widest variability of patterns of ageing. Data were then processed by bayesian prediction in order to classify specimens in age range categories. We also tested the potential of Artificial Neural Networks, which is a computation system appropriate for non linear relationships between variables.

The results reported here indicate that the use of multiple indicators does not improve assessment reliability. However, the auricular surface of the ilium is a very reliable age-at-death indicator. The scoring system and the data processing proposed here, permits the reliable classification of specimens as well as the identification of individuals who died past 60 years of age. Indeed, this age category is always underestimate in studies of human skeletal material. Artificial Neural Networks turn out to be a relevant data processing system and deserve further investigation.
Document type :
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download
Contributor : Aurore Schmitt <>
Submitted on : Thursday, February 14, 2008 - 9:32:37 AM
Last modification on : Saturday, October 17, 2020 - 3:08:50 AM
Long-term archiving on: : Thursday, May 20, 2010 - 10:00:12 PM


  • HAL Id : tel-00255753, version 1



Aurore Schmitt. Variabilité de la sénescence du squelette humain. Réflexions sur les indicateurs de l'âge au décès : à la recherche d'un outil performant.. Anthropologie biologique. Université Sciences et Technologies - Bordeaux I, 2001. Français. ⟨tel-00255753⟩



Record views


Files downloads