AGMT3-D: A software for 3-D landmarks-based geometric morphometric shape analysis of archaeological artifacts

RESEARCH ARTICLE AGMT3-D: A software for 3-D landmarks-based geometric morphometric shape analysis of archaeological artifacts Gadi Herzlinger ID1,2*, Leore Grosman1 1 Institute of Archaeology, Mount Scopus, The Hebrew University of Jerusalem, Jerusalem, Israel, 2 The Jack, Joseph and Morton Mandel School for Advanced Studies in the Humanities, Mount Scopus, The Hebrew University of Jerusalem, Jerusalem, Israel a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Herzlinger G, Grosman L (2018) AGMT3D: A software for 3-D landmarks-based geometric morphometric shape analysis of archaeological artifacts. PLoS ONE 13(11): e0207890. https://doi. org/10.1371/journal.pone.0207890 Editor: Michael D. Petraglia, Max Planck Institute for the Science of Human History, GERMANY Received: August 26, 2018 Accepted: November 7, 2018 Published: November 20, 2018 Copyright: © 2018 Herzlinger, Grosman. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All 3-D landmarks data files are available from the Open Science Framework repository: https://osf.io/sdbjk/, DOI 10.17605/OSF.IO/SDBJK. Funding: This study was funded by the Israel Science Foundation grant no. 27/12 (https://isf.org. il/#/) given to Naama Goren-Inbar; Wilson Foundation grant (http://wilsonfdn.org/) given to Leore Grosman; and Yad Hanadiv Foundation grant (http://www.yadhanadiv.org.il/) given to Leore Grosman. The funders had no role in study design, * Abstract We present here a newly developed software package named Artifact GeoMorph Toolbox 3D (AGMT3-D). It is intended to provide archaeologists with a simple and easy-to-use tool for performing 3-D landmarks-based geometric morphometric shape analysis on 3-D digital models of archaeological artifacts. It requires no prior knowledge of programming or proficiency in statistics. AGMT3-D consists of a data-acquisition procedure for automatically positioning 3-D models in space and fitting them with grids of 3-D semi-landmarks. It also provides a number of analytical tools and procedures that allow the processing and statistical analysis of the data, including generalized Procrustes analysis, principal component analysis, a warp tool, automatic calculation of shape variabilities and statistical tests. It provides an output of quantitative, objective and reproducible results in numerical, textual and graphic formats. These can be used to answer archaeologically significant questions relating to morphologies and morphological variabilities in artifact assemblages. Following the presentation of the software and its functions, we apply it to a case study addressing the effects of different types of raw material on the morphologies and morphological variabilities present in an experimentally produced Acheulian handaxe assemblage. The results show that there are statistically significant differences between the mean shapes and shape variabilities of handaxes produced on flint and those produced on basalt. With AGMT3-D, users can analyze artifact assemblages and address questions that are deducible from the morphologies and morphological variabilities of material culture assemblages. These questions can relate to issues of, among others, relative chronology, cultural affinities, tool function and production technology. AGMT3-D is aimed at making 3-D landmarks-based geometric morphometric shape analysis more accessible to archaeologists, in the hope that this method will become a tool commonly used by archaeologists. Introduction Landmarks-based geometric morphometric shape analysis is a powerful tool for the quantitative description of shape variability within and between groups. More than a decade ago, PLOS ONE | https://doi.org/10.1371/journal.pone.0207890 November 20, 2018 1 / 17 AGMT3-D: A software for 3-D geometric morphometric shape analysis data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Lycett and colleagues [1] published a seminal paper that initiated a continuous rise in the application of geometric morphometric (GM) shape analysis methods to archaeological objects of material culture. Although some authors had previously attempted to apply morphometric methods, these pioneering attempts had little impact due to the limited computing power and 3-D scanning possibilities available at that time (e.g. [2, 3]). Since the work by Lycett and colleagues, several calls have been made to encourage wider adoption of these methods for the shape analysis of lithic artifacts and other objects of material culture [4,5]. These calls, which were indeed answered by an ever-growing volume of works [6–18], outlined some of the problems and difficulties entailed in the application of landmarks-based GM methods to material culture objects. The main problem was the lack of readily identifiable homologous landmarks on such artifacts, among others [5]. While solutions have been suggested for some of the problems, such as landmark homology, another substantial problem has been overlooked: the actual process of positioning landmarks and recording their coordinates [19]. In their original work, Lycett and colleagues [1] presented a protocol and an instrument, which they called a crossbeam co-ordinate caliper (CCC), for the positioning and recording of 3-D homologous semi-landmarks on lithic artifacts. While the instrument and protocol were the first to enable the application of the landmarks-based 3-D geometric morphometric (3-DGM) method to lithic artifacts, the procedure has two main disadvantages. The first is that the instrument is manually operated, and as such its operation is extremely costly in time and resources, raises serious concerns about accuracy and inter-analyst bias, and practically limits the resolution (i.e. number of landmarks) at which analysis can be conducted. The second is the physical nature of the instrument and protocol; a researcher wishing to use this method must acquire the instrument and have physical access to the studied material. These problems had a substantial effect on the number of researchers and studies [1, 7, 14, 15, 18] applying the landmarks-based 3-DGM method. Although there are a few alternatives to the CCC for positioning and recording of landmarks, none of them provides a complete and comprehensive solution to the problems embedded in the process. One of these uses a 3-D digitizer like those produced by Microscribe [6]. However, as this too is a physical and manually operated instrument, it basically has the same disadvantages as the CCC. There are also several freely available computer programs designed for landmarks based GM analysis, such as MorphoJ or tpsDig. However, these lack 3D (...truncated)