An important comparison of software for Scale Sensitive Fractal Analysis : are ancient and new results compatible?
Surface texture analysis in Toothfrax and MountainsMap® SSFA module: Different software packages, different results?
Recommendation: posted 24 November 2022, validated 01 December 2022
The community of archaeologists, bioanthropologist and paleontologists relying on tools use-wear and dental microwear has grown in the recent years, mainly driven by the spread of confocal microscopes in the laboratories. If the diversity of microscopes is quite high, the main software used for 3D surface texture data analysis are mostly different versions of the same Mountains Map core. In addition to this software, since the beginning of 3D surface texture analysis in dental microwear, surface sensitive fractal analysis (SSFA) initially developed for industrial research (Brown & Savary, 1991) have been performed in our disciplines with the Sfrax/Toothfrax software for two decades (Ungar et al., 2003). This software being discontinued, these calculations have been integrated to the new versions of Mountains Map, with multi-core computing, full integration in the software and an update of the calculation itself.
New research based on these standard parameters of surface texture analysis will be, from now on, mainly calculated with this new add-on of Mountains Map, and will be directly compared with the important literature based on the previous software. The question addressed by Calandra et al. (2022), gathering several prominent researchers in this domain including the Mountains Map developer F. Blateyron, is key for the future research: can we directly compare SSFA results from both software?
Thanks to a Bayesian approach to this question, and comparing results calculated with both software on three different datasets (two on dental microwear, one on lithic raw materials), the authors show that the two software gives statistically different results for all surface texture parameters tested in the paper. Nevertheless, applying the new calculation to the datasets, they also show that the results published in original studies with these datasets would have been similar. Authors also claim that in the future, researchers will need to re-calculate the fractal parameters of previously published 3D surfaces and cannot simply integrate ancient and new data together.
We also want to emphasize the openness of the work published here. All datasets have been published online and will be probably very useful for future methodological works. Authors also published their code for statistical comparison of datasets, and proposed a fully reproducible article that allowed the reviewers to check the content of the paper, which can also make this article of high interest for student training.
This article is therefore a very important methodological work for the community, as noted by all three reviewers. It will certainly support the current transition between the two software packages and it is necessary that all surface texture specialists take these results and the recommendation of authors into account: calculate again data from ancient measurements, and share the 3D surface measurements on open access repositories to secure their access in the future.
Brown CA, and Savary G (1991) Describing ground surface texture using contact profilometry and fractal analysis. Wear, 141, 211–226. https://doi.org/10.1016/0043-1648(91)90269-Z
Calandra I, Bob K, Merceron G, Blateyron F, Hildebrandt A, Schulz-Kornas E, Souron A, and Winkler DE (2022) Surface texture analysis in Toothfrax and MountainsMap® SSFA module: Different software packages, different results? Zenodo, 7219877, ver. 4 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7219877
Ungar PS, Brown CA, Bergstrom TS, and Walker A (2003) Quantification of dental microwear by tandem scanning confocal microscopy and scale-sensitive fractal analyses. Scanning: The Journal of Scanning Microscopies, 25, 185–193. https://doi.org/10.1002/sca.4950250405
Alain Queffelec and Florent Rivals (2022) An important comparison of software for Scale Sensitive Fractal Analysis : are ancient and new results compatible?. Peer Community in Archaeology, 100024. https://doi.org/10.24072/pci.archaeo.100024
The recommender in charge of the evaluation of the article and the reviewers declared that they have no conflict of interest (as defined in the code of conduct of PCI) with the authors or with the content of the article. The authors declared that they comply with the PCI rule of having no financial conflicts of interest in relation to the content of the article.
Reviewed by Antony Borel, 16 Nov 2022
Reviewed by anonymous reviewer, 15 Nov 2022
Evaluation round #1
DOI or URL of the preprint: https://doi.org/10.5281/zenodo.6669276
Author's Reply, 11 Nov 2022
Decision by Alain Queffelec and Florent Rivals, posted 02 Sep 2022
We have received feedback from three reviewers who have reviewed your work in depth. They all agree that your work is important to the field and requires only minor changes before it can be recommended. They also all recognize the very open manner in which you provided the data and code, as well as the quality of your supplementary materials.
We would therefore be very pleased to receive the responses you can provide to all the comments from the three reviewers, and we will be most willing to arrange a quick round of peer-review of this new version, as some of the reviewers' comments are very technical and would require them to check your responses themselves.
Alain Queffelec and Florent Rivals