Neural network analysis by using the Self-Organizing Maps (SOMs) applied to human fossil dental morphology: A new methodology

DOI: 10.1007/978-1-4020-5845-5_6

ABSTRACT Recent studies focusing on dental morphology of extinct and extant human populations have shown, on a global scale, the considerable
potential of dental traits as a tool to understand the phenetic relations existing between populations. The aim of this paper
is to analyze the dental morphologic relationships between archaic Homo and anatomically modern Homo sapiens by means of a new methodology derived from artificial neural networks called Self Organizing Maps (SOMs). The graph obtained
by SOMs to some extent recalls a classical Multidimensional Scaling (MDS) or a Principal Component Analysis (PCA) plot. The
most important advantages of SOMs is that they can handle vectors with missing components without interpolating missing data.
The analyzed database consisted of 1055 Lower-Middle and (Early) Late Pleistocene specimens, which were scored by using dental
morphological traits of the Arizona State University Dental Anthropology System (ASUDAS). The principal result indicates a
close relationship between the Homo erectus s.l. and Middle Pleistocene specimens and the later Neandertal groups. Furthermore, the dental models of anatomically modern
Homo sapiens are particularly different compared to the more archaic populations. Thus, SOMs can be considered a valuable tool in the
field of dental morphological studies since they enable the analysis of samples at an individual level without any need i) to interpolate missing data or ii) place individuals in predetermined groups.

Keywordsneural network analysis-Self-Organizing Maps (SOMs)-multidimensional scaling-dental morphology-Arizona State University Dental Anthropology System (ASUDAS)-Lower Pleistocene specimen-Middle Pleistocene specimen-Late Pleistocene specimen

  • [Show abstract] [Hide abstract]
    ABSTRACT: The relationships between a range of modern human samples are assessed from cladistic analyses of the published population frequencies of tooth crown characters, using new data on the Krapina Neanderthal sample as an outgroup. All of the most parsimonious trees show an early divergence of African and Australasian groups. This result is compared with an alternative dendrogram proposed by Turner (1992). Reconstruction of a hypothetical dental ancestor suggests that the similarities between the African and Australasian groups result from the retention of symplesiomorphous dental traits. Additionally, despite expectations from multiregional evolution, recent Europeans are dentally less like the Krapina Neanderthals than are Africans and Australians.
    Journal of Human Evolution 05/1997; 32(4):389-402. · 3.87 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This work contains a theoretical study and computer simulations of a new self-organizing process. The principal discovery is that in a simple network of adaptive physical elements which receives signals from a primary event space, the signal representations are automatically mapped onto a set of output responses in such a way that the responses acquire the same topological order as that of the primary events. In other words, a principle has been discovered which facilitates the automatic formation of topologically correct maps of features of observable events. The basic self-organizing system is a one- or two-dimensional array of processing units resembling a network of threshold-logic units, and characterized by short-range lateral feedback between neighbouring units. Several types of computer simulations are used to demonstrate the ordering process as well as the conditions under which it fails.
    Biological Cybernetics 12/1981; 43(1):59-69. · 1.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The winner-take-all (WTA) property is essential to competitive-learning systems. This article discusses WTA-type neural networks composed of nonlinear dynamic model neurons, characterized by a nonlinear loss term. It is shown by mathematical analyses that these networks have the WTA property even when their neurons have nonidentical characteristics and the interconnections have nonidentical strengths. The class of WTA networks is further generalized to allow explicitly modeled interneurons between the principal cells. A model of a cyclically operating WTA system, capable of handling changing inputs by automatically inactivating the winner, is then set up and demonstrated by computer simulations.
    Neural Networks 01/1994; · 2.08 Impact Factor