Morphological Uniqueness: The Concept and Its Relationship to Indicators of Biological Quality of Human Faces from Equatorial Africa
Round 1
Reviewer 1 Report
Morphological uniqueness: the Concept and its relationship to indicators of biological quality of human faces from Equatorial Africa.
Karel Kleiser
Symmetry
Mansucript ID: symmetry-1399214
The aim of the manuscript it to scrutinize Morphological uniqueness, a morphological trait that could be useful to understand human evolution and adaptation.
Morphological uniqueness is computed by the estimation of the different metrics, among which the level of symmetry of the face or the distance of specimens from populations average (uniqueness).
The manuscript is very well written, perfectly organized, and really fits the aims and scopes of the journal.
I think it can be published after some minor revisions. I really insist on the fact that these comments are not made to criticize the work but really to add some improvements.
Here are some comments.
Introduction:
This part is clear but I am not fully convinced that the theoretical elements given by the authoress in the introduction are clear, or true. For example, the first paragraph is very theoretical but however is not supported by citations. Same thing for the third paragraph. Some citations would help to confirm the statement.
The end of this part is clear and the questions are well expressed.
Material and Methods.
I have more comments on this part.
- The first comment is that this part is composed of text only. Tables and figures that are absolutely necessary for a good understanding of the analytic pipeline are in supplementary informations. I think that Symmetry is a journal that allows the authors to insert these informations in the body of the manuscript. I really advise the authoress the add a table presenting the sample and a figure illustrating the dataset in this part.
- Second point, I can’t see any mention of the measurement error. To me, every study based on measurement MUST have an estimation of measurement error. This is especially the case here because the morphological variations and the covariable effects seem weak. How can the reader know what is measured, true effect or noise/ error? Here, the authoress uses Geometric Morphometrics and multivariate analyses, that is to say a great number of variables for which we don’t have any idea of the precision. A paragraph about measurement error is necessary to understand what is measured and what is analyses.
- Third comment, very similar to the previous one. The authoress analyzed Symmetry/ Asymmetry in a very clever way. However, we know that asymmetry is, here again, a very low effect on such dataset, with a level of variation very small. Again, people working on symmetry use to estimate this parameter on large dataset (when possible) and use replicates in order to distinguish measurement error from this small part of variation coming from symmetry. Here, the authoress never mentioned the presence of replicates in her dataset this element should be added in the revised version.
- Another small comment for this part, please cite the packages used for analyses. They are mentioned in the text but don’t appear in the references list.
Results:
This part is brief and mainly relies on figures and tables.
I have some questions:
- I am okay with the results given on females but I am less conviced by the results for men. For instance, the authoress says that distinctiveness is negatively associated with morphological uniqueness, but that asymmetry is not. I have some doubts about these results as values for distinctiveness and for asymmetry are very similar looking at figure 1 and at table 1. How can the authoress justify this result?
- Association of MU with the other variables is made variable by variable. Would it make sense to investigate the interactions between variables?
- On figure 2 and figure 3, the relationship between MU and distinctiveness is described by a linear model but seem non-linear. It seems that two patterns of relationship are distinguishable. A pattern for positive values of MU and a pattern for negative values. This point should be commented in the text and an interpretation is necessary.
Discussion:
This part needs to be reworked. First, a 2 pages discussion seem very small in comparison with the extended analyses performed for this study. Second, the authoress cited 9 references only, 7 in the same paragraph, the 2 others in a second one. To me, it is a bad signal. It indicates that all the other paragraphs are not interpreted in the light of the previous knowledge of this topic. Actually, the evolutionary understanding of MU, well presented in the introduction, is not really discussed.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The paper as it is is not supported by a robust statistics. The authors must provide significance for all the test presented, really supporting they statements and conclusions. Without that the paper is not suitable for publication.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
symmetry-1399214-peer-review-v1
The article presents and interesting take on studying facial morphology. In addition to presenting analysis of previously frequently studied facial sexual dimorphism and symmetry, author adds two new concepts: uniqueness and distinctiveness. The article is well written. I struggled in some places of the Methods section (I list my detailed comments in the attached pdf). It can also, however, stem from my lack of experience with Geomorphic software. I think the topic presented is very interesting and can aid to shed new light on the incongruent results of the up-to-date studies on correlates of facial morphology.
Being in a possession of this impressive data base of facial photographs, why did the author decide to pick Cameroonian sample as the target population? It is, of course, not an error. Some of the conclusions about adaptiveness could have been better grounded if the author analysed MU in different samples as well.
I struggle a bit with the idea that MU is adaptive. For it to be adaptive, it would mean that individuals high in MU would sire more children, and possibly also live longer (be healthier). This study does not measure the later directly. Author points out that sex typicality and symmetry can both be interpreted as biomarkers of health, but simultaneously we know of multiple studies (including those cited by the author in the discussion section) that do not support this claim. The former is measured (i.e. the putative adaptiveness of MU via sexual selection), however, solely the sex typicality without attractiveness, in theory, wouldn’t lead to increased offspring number. All this is purely hypothetical, as we would need to measure the actually number of children of post-reproductive individuals and correlate it with MU to have a more concrete answer on adaptiveness of MU. I do not want to undermine the entire rationale of the manuscript. I would just suggest to soften the phrasing of possible adaptiveness of the MU throughout the entire text.
Author introduces two rather new facial characteristics in the introduction (distinctiveness and uniqueness). I am aware that abstracts should be brief, but would it be possible to include a very brief definition of those two characteristics and maybe state how are they different from the computational averageness.
I find it interesting, that “Distinctiveness had no association with MU”. Maybe it would be worth it to write more about where such interesting results could stem from?
In the methods section there should be also a subsection on judgements of attractiveness.
In Table S1 author mentions information on restricted sample in brackets, and no brackets are visible in the pdf.
To make the review process easier I have made all my minor comments into the pdf file that I attach to my submission.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
It is suitable to performe the S-value test as recommended in the current literature https://doi.org/10.1186/s12874-020-01105-9
Reviewer 3 Report
All of my comments and suggestions were addressed by the author. I do not have any further issues.
