Focal Length Affects Depicted Shape and Perception of Facial Images

Vít Třebický; Jitka Fialová; Karel Kleisner; Jan Havlíček

doi:10.1371/journal.pone.0149313

Abstract

Static photographs are currently the most often employed stimuli in research on social perception. The method of photograph acquisition might affect the depicted subject’s facial appearance and thus also the impression of such stimuli. An important factor influencing the resulting photograph is focal length, as different focal lengths produce various levels of image distortion. Here we tested whether different focal lengths (50, 85, 105 mm) affect depicted shape and perception of female and male faces. We collected three portrait photographs of 45 (22 females, 23 males) participants under standardized conditions and camera setting varying only in the focal length. Subsequently, the three photographs from each individual were shown on screen in a randomized order using a 3-alternative forced-choice paradigm. The images were judged for attractiveness, dominance, and femininity/masculinity by 369 raters (193 females, 176 males). Facial width-to-height ratio (fWHR) was measured from each photograph and overall facial shape was analysed employing geometric morphometric methods (GMM). Our results showed that photographs taken with 50 mm focal length were rated as significantly less feminine/masculine, attractive, and dominant compared to the images taken with longer focal lengths. Further, shorter focal lengths produced faces with smaller fWHR. Subsequent GMM revealed focal length significantly affected overall facial shape of the photographed subjects. Thus methodology of photograph acquisition, focal length in this case, can significantly affect results of studies using photographic stimuli perhaps due to different levels of perspective distortion that influence shapes and proportions of morphological traits.

Citation: Třebický V, Fialová J, Kleisner K, Havlíček J (2016) Focal Length Affects Depicted Shape and Perception of Facial Images. PLoS ONE 11(2): e0149313. https://doi.org/10.1371/journal.pone.0149313

Editor: Pablo Brañas-Garza, Middlesex University London, UNITED KINGDOM

Received: September 16, 2015; Accepted: January 29, 2016; Published: February 19, 2016

Copyright: © 2016 Třebický et al. 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: All relevant data are within the paper and its Supporting Information files.

Funding: JF was supported by Charles University Grant Agency GAUK 918214 (http://www.cuni.cz/ukeng-33.html), VT and KK by Czech Science Foundation GAČR P407/15-05048S (http://gacr.cz/en/) and VT, JF, and JH by the project "National Institute of Mental Health (NIMH-CZ)", grant number ED2.1.00/03.0078 (and the European Regional Development Fund, http://ec.europa.eu/regional_policy/en/funding/erdf/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Human face research has received immense attention in fields ranging from social psychology to behavioural neuroscience and economics. Studies on perception of human faces involve various social contexts such as mate choice, cooperation, and parental care [1]. These studies have many practical implications; for instance, it has been shown that facial appearance affects electoral success, career progress, and child treatment [2]. Further, perceived characteristics are frequently associated with facial appearance [3]. For instance, it was recently found that facial width-to-height ratio (fWHR) is related to perceived aggressiveness [4–6], dominance [7,8], and trustworthiness [9]. Moreover, perceptions of certain psychological characteristics such as dominance [10], intelligence [11], and aggressiveness [12] can be, to some extent, predicted from overall facial morphology as analysed by geometric morphometrics.

The majority of studies on facial perception employ portrait photographs as stimuli. The facial images can be either obtained from free access sources [12–16] or taken for the purpose of the specific study [17–20]. There are several major factors affecting the resulting stimuli including exposure (e.g., under- or overexposure of images or inappropriate depth of field), optical aberrations of the lens used (e.g., radial and perspective distortions) [21,22], colour representations and lighting set up (e.g., number and type of lights and light modifier used) [23]. For instance, recent studies reported that variation in colours of facial photographs affects assessments of health, attractiveness, aggressiveness, and dominance [20,24]. Similarly, the position and posture of the target during image acquisition influences fWHR as measured from the photograph [25] or perceived body size [26]. Although several previous studies have reported carefully the standardized procedures for photograph acquisition (e.g., [27,28], employed methodology varies across individual studies. Further, detailed descriptions of photograph acquisition and standardization within a study is often missing which may impede assessment of validity and accurate replication of previous findings.

A key factor affecting the resulting photographs, which is frequently not reported in previous studies, is focal length. Focal length represents the distance between lens optics and camera sensor and provides variance in viewing angle and zoom (from wide angle fish eye lens to narrow telephoto lens) resulting in different degrees of image distortion. The most common image distortions are radial distortions, where straight lines are rendered as curved lines (i.e., barrel and pincushion distortion) [21], and perspective distortions, which are determined by the viewpoint from which the photograph is taken in relation to the target (i.e., nearby elements are rendered larger than distant ones) [22]. Due to these distortions, artefacts in size and shape representations in photographs can occur.

Here, we investigated the possible influence of focal length on perception of facial images through the assessment of selected interpersonal characteristics. Further, we tested the potential effect of focal length (50 mm, 85 mm, 105 mm) on depicted facial shape by measuring facial width-to-height ratio and employing geometric morphometrics.

Methods

Ethics statement

The study was approved by the Institutional Review Board of Charles University, Faculty of Science (approval number 2013/11). All participants gave written informed consent prior to taking part in the study.

Participants

Targets.

In total we obtained facial images of 23 target men (mean age = 23.83, SD = 4.38) and 22 target women (mean age = 22.91, SD = 4.38). The targets were recruited via social networks (e.g., Facebook) and advertisement at a part-time jobs webpage (www.jobs.cz) and were reimbursed with 100 CZK (approximately €4).

Raters.

We recruited 369 students (176 males) from Charles University in Prague to rate a set of photographs on one of the selected characteristics (i.e., attractiveness, dominance, and femininity/masculinity). Attractiveness of the male photographs was assessed by 25 male (mean age = 21.72, SD = 2.42) and 30 female raters (mean age = 21.2, SD = 1.34). Attractiveness of the female photos was assessed by 30 male (mean age = 23.5, SD = 3.45) and 34 female raters (mean age = 22.47, SD = 3.08). Dominance of the male photographs was assessed by 30 male (mean age = 23.33, SD = 2.85) and 34 female raters (mean age = 22.53, SD = 2.08). Dominance of the female photos was assessed by 27 male (mean age = 21.85, SD = 3.19) and 30 female raters (mean age = 21.53, SD = 1.96). Masculinity of the male photos was assessed by 30 male raters (mean age = 22.97, SD = 3.01) and 33 female raters (mean age = 22.61, SD = 2.38). Finally, femininity of the female photos was assessed by 34 male (mean age = 22.56, SD = 2.35) and 32 female raters (mean age = 22.31, SD = 2.07). Raters were not reimbursed for their participation.

Photograph acquisition

Three facial photographs of each target varying in focal length (50 mm, 85 mm, 105 mm) were taken using a DSLR camera (Nikon D90) equipped with APS-C sensor (crop factor 1.5×). Crop factor indicates how many times smaller the sensor is compared to the full frame (FF) sensor (the size of a full 35mm analogue film frame). Exposure was set to ISO 100, shutter speed 1/100s, aperture F8 and 2/3 of strobe power. We used ISO 100 to maintain the highest image quality and the lowest amount of digital noise, and shutter speed 1/100s to freeze potential motion while giving enough time for synchronizing the shutter curtains with the strobe lights [29,30]. Aperture F8 was selected to obtain the sharpest possible results with the lens used (see below) and give sufficient depth of field, i.e., the target’s face from nose tip to ear ridges was sharp without needing to change settings while taking each photograph from different distances (different distances produce variance in the depth of field). The focusing point was set on the left eye in AF-S mode. White balance was set manually using a grey background to ensure constant colour representation. Photographs were processed into JPEG files in the camera with the Nikon STANDARD colour scheme.

All images were taken from a tripod (Velbon Sherpa) with the height set for each photograph depending on the height of the target, keeping the target´s face in middle of the frame. Similarly, the distance between the camera and the target was individually adjusted for each shot so that the head of the target filled the same portion of the frame (using grid lines in live view mode) [31,32].

To test the effect of the focal length we selected the closest FF equivalents of 50 mm, 85 mm, and 105 mm focal lengths. The 85 mm and 105 mm focal lengths are the most frequently used for portrait photography [30] and do not exceed ordinary space requirements between the target and the camera, concurrently giving the same field of view. Moreover, 50 mm lenses are frequently used as standard or prime lenses, as they are believed to be equivalent in focal length to the human eye and are adept at creating natural-looking images (e.g., [21]). However, to our knowledge, it appears that there is no solid evidence supporting this notion. We used a zoom type lens Nikon AF-S DX Zoom Nikkor 18–135 mm f/3.5–5.6 G IF-ED, which allowed us to set all three focal lengths without needing to switch lenses. To use the closest FF equivalent focal length on an APS-C sensor camera, one needs to multiply the focal length stated on the given lens barrel by the crop factor of a given camera (e.g., 1.5× for Nikon APS-C). The closest possible focal lengths were: 32 mm as 50 mm FF equiv. 48 mm; 56 mm as 85 mm FF equiv. 84 mm; and 70 mm as 105 mm FF equiv. 105 mm, respectively.

The targets were asked to stand 1.5 m from a plain grey background (Storm Gray, BD Company). Two studio strobes (Menik MD300, 300W, GN 54) with white reflective umbrellas (ø102 cm) as light modifiers were used to illuminate the targets. Lights were arranged in 40° angles on each side and were 1.85 m from the target, 1.8 m height and at 35° angle incline giving even illumination (Fig 1). Light conditions in the room were controlled by the use of non-translucent curtains and all ambient lights were switched off to remove any additional lighting variables.

Download:

Fig 1. Lighting setup diagram.

Visualisation was generated by the online free service Sylights.com.

https://doi.org/10.1371/journal.pone.0149313.g001

The targets were instructed to remove any facial cosmetics and adornments (e.g., jewellery or glasses) and their hair was pulled back from the face and held by a headband. We instructed them to keep a “neutral” facial expression, and look directly into the camera. In case of targets´ head not facing the camera directly, they were further instructed to adjust their head position accordingly (e.g., to move their chin up or down). All photographs were subsequently post-produced regarding the position of the face in the image (same position of the eyes in vertical and horizontal axis) using image manipulation software GIMP, ver. 2.8.

Rating sessions

Rating sessions were held in a lecture room equipped with 25 identical desktop computers and 22” DELL P2210 LCD screens (set on 50% level of contrast and brightness) under standardized light conditions. Qualtrics survey building engine (www.qualtrics.com) was used to present photographs in full screen mode. Photographs were presented in a randomized order using a forced-choice test, where raters were simultaneously shown all three photographs (taken with different focal lengths) of each target. The sides of the three photographs were fully randomized (i.e., different for each target for each rater). Photographs were judged for their attractiveness, dominance, and femininity/masculinity, characteristics frequently used in facial research (for recent reviews see [33,34]). We asked raters to order the triads of the photographs with respect to given characteristic (e.g., 1 –the least attractive, 2 –medium attractive, 3 –the most attractive). Each rater assessed photographs of all 45 targets.

Facial width-to-height ratio measurements

To investigate whether the ratio between selected facial features differs depending on the focal length used, bizygomatic width (facial width), and distance between the upper lip and brow (facial height) [4,5] was measured using tpsDig2 software, ver. 2.14 [35]. The fWHR was then calculated by dividing the facial width by the facial height. Measurements of fWHR were calculated by two independent experimenters (JF and VT). Inter-rater agreement for fWHR in images varying in focal length varied between Cronbach’s α = 0.963–0.972, indicating excellent reliability [36].

Geometric morphometrics

Geometric morphometrics were applied to examine whether different focal lengths affect facial morphology in portrait photographs.

The 82 landmarks (including 42 semi-landmarks) were digitized by tpsDig2 software, ver. 2.14 [35]. Landmarks are represented as points that are anatomically (or geometrically) homologous in different individuals, while semi-landmarks serve to denote curves and outlines. The definitions of landmark and semi-landmark locations on human faces were based on previous work [11,12,37]. Semi-landmarks were slid by tpsRelw (ver. 1.49) software. All configurations of landmarks and semi-landmarks were superimposed by Generalized Procrustes Analysis (GPA), implemented in tpsRelw, ver. 1.53 [35]. This procedure standardized the size of the objects and optimized their rotation and translation so that the distances between corresponding landmarks were minimized. Subsequently, to visualize the effect of distortion produced by focal lengths, the original photographs of men and women were unwarped to consensual configuration of a particular focal length. The composite images were generated by tpsSuper 1.14 [38].

Statistical analysis

To assess differences between judgements given to the images taken with varying focal lengths, we calculated the proportion (from 0 to 1) of the images taken with given focal length selected as the most attractive/dominant/masculine-feminine. As the judgements were collected using a forced-choice paradigm, they are not independent and only data on the first choice (e.g., the most attractive) were analysed. The proportion for each judgement type was then compared to a random distribution (equal to 0.333) using a one-sample Wilcoxon Signed Rank Test.

Differences in mean proportions of the first choice between images taken with individual focal lengths were compared using Kruskal-Wallis H test. Subsequently, we used Mann-Whitney U tests as post-hoc tests to compare each pair of focal lengths (50 mm × 85 mm; 50 mm × 105 mm; 85 mm × 105 mm). Bonferroni adjustment was used to correct for the effect of multiple comparisons in the Mann-Whitney U tests, with a threshold for significance of 0.017 (i.e., 0.05/3) [39,40]. Effect sizes are reported in the form of Cohen’s d.

A repeated measures ANOVA was employed to investigate differences in fWHR between images taken with different focal lengths. Overall effect size is reported in form of partial eta squared (η²), effect sizes of subsequent Bonferroni post-hoc tests are again reported in the form of Cohen’s d.

To test for shape differences that resulted from using various focal lengths, we performed permutational multivariate analysis of variance using distance matrices with 9,999 permutations (the Adonis function in the Vegan package in R [41]); the Euclidean method was used as a distance measure and the parameter ‘strata’ was set to constraint permutation within the groups of landmark configuration of the same photographed subject. We ran a multiple multivariate regression with principal component scores as the response variable and with focal length as an explanatory variable. Effect sizes are reported as R2.

All analyses were performed using IBM SPSS ver. 22 and R software for statistical computing [42].

Results

Perceived characteristics

To test for the potentially confounding effect of rater sex on the assessed characteristics, we performed an independent sample t-test. No significant differences between men and women were found on any of the rated characteristics for female or male faces: female attractiveness t₍₁₉₀₎ = 0.023, p = 0.982; female dominance t₍₁₆₉₎ = 0.365, p = 0.715; female femininity t₍₁₉₆₎ < 0.001, p = 1; male attractiveness t₍₁₆₃₎ = 0.395, p = 0.694; male dominance t₍₁₈₇₎ = 0.039, p = 0.969; male masculinity t₍₁₈₇₎ < 0.001, p = 1.

Subsequently, we compared the proportion of the first choices for each focal length against random distribution for each type of judgement, for results see Table 1. The differences in mean proportion of the first choices (for each characteristic) between focal lengths were analysed using Mann-Whitney tests (Table 2). We found significant effect of focal length on judgements of all characteristics for both target sexes. Figs 2, 3 and 4 represent the proportion of the first choices for each focal length with respect to their attractiveness, dominance, and femininity/masculinity, respectively. More specifically, the mean proportion of the first choices given to images taken with 50 mm focal length was the lowest for all assessed characteristics and significantly differed from both 85 mm and 105 mm focal lengths. The mean proportion of the first choices given to images taken with 105 mm focal length was the highest; however, the mean proportion of the first choices given to images taken with 105 mm focal length did not significantly differ from those given to images taken with 85 mm focal length for female attractiveness, female dominance, and male attractiveness ratings.

Download:

Fig 2. Proportion of first choices for each focal length with respect to their attractiveness.

Dashed line represents the proportion of ratings expected by chance (i.e., 0.333), error bars represent 95% CI.

https://doi.org/10.1371/journal.pone.0149313.g002

Download:

Fig 3. Proportion of first choices for each focal length with respect to their dominance.

Dashed line represents the proportion of ratings expected by chance (i.e., 0.333), error bars represent 95% CI.

https://doi.org/10.1371/journal.pone.0149313.g003

Download:

Fig 4. Proportion of first choices for each focal length with respect to their femininity/masculinity.

Dashed line represents the proportion of ratings expected by chance (i.e., 0.333), error bars represent 95% CI.

https://doi.org/10.1371/journal.pone.0149313.g004

Download:

Table 1. Comparison of the proportion of first choices for each focal length against random distribution (i.e., 0.333).

https://doi.org/10.1371/journal.pone.0149313.t001

Download:

Table 2. The differences in mean proportion of first choices for attractiveness, dominance and femininity/masculinity between focal lengths.

https://doi.org/10.1371/journal.pone.0149313.t002

Facial width-to-height ratio

Repeated measure ANOVA showed that fWHR in female targets significantly varied across the individual focal lengths (F_(2,42) = 120.511, p < 0.001, η² = 0.852). Subsequent Bonferroni post-hoc tests revealed significant differences between 50 mm (mean = 1.716, SD = 0.023) and 85 mm (mean = 1.797, SD = 0.024) (p < 0.001, d = 3.446), 50 mm and 105 mm (mean = 1.1797, SD = 0.022) (p < 0.001, d = 3.599), but not between 85 mm and 105 mm (p = 1, d = 0).

In a similar fashion, fWHR in males also significantly varied across the individual focal lengths (F_(2,44) = 176.419, p < 0.001, η² = 0.889). Bonferroni post-hoc tests revealed significant differences between 50 mm (mean = 1.695, SD = 0.024) and 85 mm (mean = 1.787, SD = 0.026) (p < 0.001, d = 3.677), 50 mm and 105 mm (mean = 1.803, SD = 0.025) (p < 0.001, d = 4.407), and 85 mm and 105 mm (p = 0.016, d = 0.627).

Geometric morphometrics

Using permutational multivariate ANOVA we found statistically significant shape differences in female targets between the faces of the same individuals taken at different focal lengths (F_(1,64) = 2.539, p = 0.004, R² = 0.038). The differences between focal length pairs were significant in all three combinations: 50 mm × 85 mm (F_(1,42) = 1.814, p < 0.001, R² = 0.041); 50 mm × 105 mm (F_(1,42) = 2.463, p < 0.001, R² = 0.055); 85 mm × 105 mm (F_(1,42) = 0.41, p < 0.001, R² = 0.01).

A similar pattern was found for the male targets (F_(1,66) = 2.765, p < 0.001, R² = 0.04). Subsequent comparison of focal length pairs again showed significant differences for all three combinations: 50 mm × 85 mm (F_(1,44) = 1.739, p < 0.001, R² = 0.038); 50 mm × 105 mm (F_(1,44) = 2.721, p < 0.001, R² = 0.058); 85 mm × 105 mm (F_(1,44) = 0.23, p = 0.007, R² = 0.005) (Fig 5).

Download:

Fig 5. Consensual configurations and composites of female and male images for the 50 mm, 85 mm and 105 mm focal lengths.

https://doi.org/10.1371/journal.pone.0149313.g005

Discussion

The main aim of our study was to test for possible effects of variation in focal length on perceptual judgements and shape distortions of male and female facial photographs. Photographs were taken of the same individual at focal lengths of 50 mm, 85 mm, and 105 mm equivalents. We found that facial photographs taken with 105 mm focal length equivalent were judged by both male and female raters as the most attractive, dominant, and masculine/feminine, irrespective of the target’s sex. In contrast, photographs of both males and females taken with 50 mm focal length equivalent were perceived as the least attractive, dominant, and masculine/feminine. The differences in perception were further supported by our shape analysis. We found that fWHR, as measured from the photographs, significantly varied across focal lengths in both male and female faces; fWHR taken at 50 mm was significantly smaller than the two greater focal lengths (85 mm and 105 mm). Results of the geometric morphometrics analyses similarly showed significant differences in overall facial shape between the focal lengths used in both sexes. In both male and female faces the following pattern was observed for images taken at the 50 mm focal length as compared to the longer focal lengths: overall rounded face, larger and wider set eyes, wider set eyebrows, rounded, longer and broader nose, taller forehead, rounded chin and disappearing ears obscured with cheeks (Fig 3).

The changes in facial dimensions, shape, and facial perception found in our study appear to be a consequence of variation in perspective distortion produced by the different focal lengths. In objects captured with shorter focal lengths, perspective distortion produces an appearance expanded in its depth, which makes faces look rounded and facial traits closer to camera are perceived seemingly bigger (e.g., nose) while more distant traits look smaller (e.g., ears). The faces captured with the shortest focal length appear overall to be rounded due to the vertically oblong shape of the human head, which produces the most noticeable radial distortion at the sides of head. In contrast, objects captured with longer focal lengths look compressed in depth, which makes faces look flatter [32]. The resulting effect shows smaller facial width-to-height ratio for faces captured with shorter focal lengths [19].

For several characteristics (attractiveness in both female and male photographs and femininity in female photographs) the differences between judgements of the facial photographs taken with 85 mm and 105 mm were not significant. This effect might be a result of a smaller range between particular focal lengths and consequently less pronounced perspective distortion. Our analysis of facial shape supports this interpretation, as the observed effect sizes were rather small. Similarly, fWHR measurements of female faces did not significantly differ between photographs taken with 85 mm and 105 mm. Further, male faces have, on average, substantially more distinctive facial traits compared to females (e.g., more pronounced zygomatic arches, eye ridges, broader chin), hence their photographs might be more affected by distortions. Alternatively, the perception of some characteristics might be of greater importance in social interactions than others. Therefore, smaller differences in facial shape might not be reflected in attributed characteristics.

Here we employed a forced-choice paradigm, i.e., raters were simultaneously shown all three photographs of each target and they were asked to order the triads of photographs with respect to a given characteristic (e.g., attractiveness). The advantage of this approach is its higher sensitivity allowing for detection of subtle effects, on the other hand this setting may overestimate the actual effect. Another rating paradigm which is frequently employed in the face research is rating of each targets´ face on Likert scales one by one. However, this approach is somewhat less sensitive in detecting subtle effects. Future studies should therefore test whether a similar pattern as observed in the current study can be generalized to the other rating paradigm.

Apart from the focal length, several other important factors may affect the resulting portrait photographs; including target to camera distance [19,28,31]. As 3D objects (e.g., faces) are captured in form of 2D images on a plain (e.g., camera sensor) via perspective projection, the resulting image changes with a distance from the centre of projection, even when equated for its size [19,31]. The target to camera distance could be set by determining a field of view (or angle of view), which is a function of sensor size, focal length, and target to camera distance. Although there is no general consensus on ideal target to camera distance [28], the results of Bryan et al. [19] show considerable impact of this variable on perceived characteristics. They found that faces captured from a shorter distance (45 cm) were rated as less attractive and less trustworthy as compared to the images taken from a longer distance (135 cm). A possible explanation lays in interpersonal distance as a variable influencing social behaviour [43]; Bryan et al. [19] hypothesized that the distance-dependent perspective projection of a face might serve as a cue for social judgments and faces photographed from within personal space were judged more negatively on certain characteristics. The same principle applies to the distance between the rater and the assessed image. Positioning the rater at a distance from the image that equals the target to camera distance should give an accurate impression of the scene. However, raters are usually not positioned in the centre of projection when viewing the images which may in turn bias their impression [31]. Moreover, it appears that people tend to prefer shorter viewing distances for images captured with longer focal length and longer distances for the images captured with shorter focal lengths [31].

Previous research has further shown that colour representation [24,44,45] and head position [25] influence the resulting image. Specifically, changes in face colour affect perceived health and attractiveness with higher judgements of redder faces [24,44–46] (but see Burriss et al. [23]), while downward head tilt is perceived as more intimidating through the manipulation of fWHR [25]. However, downward tilt also increases attractiveness of female faces in mate choice context [47,48].

In our study we found that certain camera settings, in this case focal length, can considerably influence perception of resulting photographs. Distortion of the facial shape may increase the chance of both type I (falsely positive results) and/or type II (falsely negative results, e.g., floor or ceiling effect) errors. For instance, false positive results might occur by taking one set of the photographs with a certain focal length and another set with a different focal length, therefore introducing systematic error. This may occur in studies using images downloaded from the internet, where the image acquisition settings may vary in systematic fashion. For instance, studies measuring fWHR from downloaded images may reflect systematic differences in image acquisition rather than actual differences in facial proportions (e.g., successful athletes in combat sports might be depicted as tougher by adjusting camera settings compared to their less successful counterparts). In contrast, using a rather wide lens may skew the ratings of some positively perceived characteristics such as trustworthiness or attractiveness to the lower end of the scale, which might obscure the chances of finding an actual effect.

Our results thus indicate that the focal length should be adjusted according to aims of a particular study. Based on our findings and previous work it seems that longer focal lengths (e.g., 85 mm full frame equivalent) produce more positively judged images [30,49] which might be an advantageous approach in attractiveness studies as it diminishes chances of the floor effect. On the other hand, longer focal lengths entail higher demands on target to camera distance to obtain a suitable field of view and accordingly increase space requirements. Selection of an adequate focal length is thus frequently a trade-off between the lens and space available for image acquisition.

In sum, our study provides additional evidence that the methodology of photograph acquisition can influence the results of perceptual studies. We showed that facial photographs taken with various focal lengths differ in their proportions as demonstrated by the fWHR, overall facial shape as analysed by the GMM and perceived characteristics judged by independent set of raters. These results highlight the importance of adopting a standardized methodology for photograph acquisition, at least within each particular study. This is often not possible in the case of photographs downloaded from online sources. Interestingly, although this approach is methodologically questionable, it is an increasingly popular method within various branches of psychological science due to its easy availability. Further, we urge researchers capturing images in the lab to report details of image acquisition parameters and settings such as camera brand and type, sensor size (crop factor), lens and focal length, exposure parameters (F-stop, shutter speed and ISO), light source(s), modifier(s), and scene setup (e.g., distance from photographed target). Providing such details would enable a more rigorous analysis of the discrepancies between individual studies–a standard that is currently not yet achieved even in some prestigious academic journals.

Supporting Information

S1 Dataset. Data from ratings and fWHR measurements (.XLSX).

Data on GMM are available from the authors.

https://doi.org/10.1371/journal.pone.0149313.s001

(XLSX)

Acknowledgments

We are grateful to the professional photographers Ing. Luboš Bárta and Hugo Vítámvás for their insightful comments and advice regarding photograph acquisition, Amanda C. Hahn and Caroline Allen for many useful comments and proof-reading of the manuscript. We would like to thank Jakub Binter, Lydie Kubicová, Pavel Šebesta and Zuzana Štěrbová for their help with data collection and all of the volunteers for their participation.

Author Contributions

Conceived and designed the experiments: VT JF JH. Performed the experiments: VT JF. Analyzed the data: VT JF KK. Contributed reagents/materials/analysis tools: VT JF KK. Wrote the paper: VT JF KK JH.

References

1. Ambady N, Skowronski JJ, editors. First Impressions. New York: Guilford Press; 2008. 368 p.
2. Rubešová A, Havlíček J. Facial appearance and personality judgments. In: Chadee D, Kostic A, editors. Social Psychological Dynamics. Kingston: University of the West Indies Press; 2011. p. 113–44.
3. Calder AJ, Rhodes G, Johnson M, Haxby J. Oxford Handbook of Face Perception. Rhodes G, Calder A, Johnson M, Haxby J V., editors. Oxford University Press; 2011.
4. Carré JM, McCormick CM. In your face: facial metrics predict aggressive behaviour in the laboratory and in varsity and professional hockey players. Proc R Soc B Biol Sci. 2008;275(1651):2651–6.
- View Article
- Google Scholar
5. Třebický V, Fialová J, Kleisner K, Roberts SC, Little AC, Havlíček J. Further evidence for links between facial width-to-height ratio and fighting success: Commentary on Zilioli et al. (2014). Aggress Behav. 2015;41(4):331–4. pmid:25236530
- View Article
- PubMed/NCBI
- Google Scholar
6. Haselhuhn MP, Ormiston ME, Wong EM. Men’s facial width-to-height ratio predicts aggression: A meta-analysis. PLoS One. 2015;10(4):e0122637. pmid:25849992
- View Article
- PubMed/NCBI
- Google Scholar
7. Mileva VR, Cowan ML, Cobey KD, Knowles KK, Little AC. In the face of dominance: Self-perceived and other-perceived dominance are positively associated with facial-width-to-height ratio in men. Pers Individ Dif. 2014 Oct;69:115–8.
- View Article
- Google Scholar
8. Valentine KA, Li NP, Penke L, Perrett DI. Judging a man by the width of his face: The role of facial ratios and dominance in mate choice at speed-dating events. Psychol Sci. 2014;25(3):806–11. pmid:24458269
- View Article
- PubMed/NCBI
- Google Scholar
9. Stirrat MR, Perrett DI. Valid facial cues to cooperation and trust: male facial width and trustworthiness. Psychol Sci. 2010;21(3):349–54. pmid:20424067
- View Article
- PubMed/NCBI
- Google Scholar
10. Windhager S, Schaefer K, Fink B. Geometric morphometrics of male facial shape in relation to physical strength and perceived attractiveness, dominance, and masculinity. Am J Hum Biol. 2011;23(6):805–14. pmid:21957062
- View Article
- PubMed/NCBI
- Google Scholar
11. Kleisner K, Chvátalová V, Flegr J. Perceived intelligence is associated with measured intelligence in men but not women. PLoS One. 2014;9(3):e81237. pmid:24651120
- View Article
- PubMed/NCBI
- Google Scholar
12. Třebický V, Havlíček J, Roberts SC, Little AC, Kleisner K. Perceived aggressiveness predicts fighting performance in mixed-martial-arts fighters. Psychol Sci. 2013;24(9):1664–72. pmid:23818656
- View Article
- PubMed/NCBI
- Google Scholar
13. Geniole SN, McCormick CM. Facing our ancestors: judgements of aggression are consistent and related to the facial width-to-height ratio in men irrespective of beards. Evol Hum Behav. 2015;36(4):279–85.
- View Article
- Google Scholar
14. Park JH, Buunk AP, Wieling MB. Does the face reveal athletic flair? Positions in team sports and facial attractiveness. Pers Individ Dif. 2007;43(7):1960–5.
- View Article
- Google Scholar
15. Postma E. A relationship between attractiveness and performance in professional cyclists. Biol Lett. 2014 Feb 5;10(2):20130966–20130966. pmid:24501269
- View Article
- PubMed/NCBI
- Google Scholar
16. Loehr J, O’Hara RB. Facial morphology predicts male fitness and rank but not survival in Second World War Finnish soldiers. Biol Lett. 2013;9(4):20130049–20130049. pmid:23658003
- View Article
- PubMed/NCBI
- Google Scholar
17. Kleisner K, Kočnar T, Rubešová A, Flegr J. Eye color predicts but does not directly influence perceived dominance in men. Pers Individ Dif. 2010;49(1):59–64.
- View Article
- Google Scholar
18. Roberts SC, Havlicek J, Flegr J, Hruskova M, Little AC, Jones BC, et al. Female facial attractiveness increases during the fertile phase of the menstrual cycle. Proc R Soc B Biol Sci. 2004;271:S270–2.
- View Article
- Google Scholar
19. Bryan R, Perona P, Adolphs R. Perspective distortion from interpersonal distance is an implicit visual cue for social judgments of faces. PLoS One. 2012;7(9):e45301. pmid:23028918
- View Article
- PubMed/NCBI
- Google Scholar
20. Hong G, Luo MR, Rhodes PA. A study of digital camera colorimetric characterisation based on polynomial modelling. Color Res Appl. 2001;26(3):76–84.
- View Article
- Google Scholar
21. Jenkins FA, White HE. Fundamentals of Optics, 4th Edition. New York: McGraw-Hill; 2001. 766 p.
22. Kingslake R. Optics in Photography. Bellingham, WA: SPIE Press; 1992.
23. Burriss RP, Troscianko J, Lovell PG, Fulford AJC, Stevens M, Quigley R, et al. Changes in women’s facial skin color over the ovulatory cycle are not detectable by the human visual system. PLoS One. 2015;10(7):e0130093. pmid:26134671
- View Article
- PubMed/NCBI
- Google Scholar
24. Stephen ID, Smith MJL, Stirrat MR, Perrett DI. Facial skin coloration affects perceived health of human faces. Int J Primatol. 2009;30(6):845–57. pmid:19946602
- View Article
- PubMed/NCBI
- Google Scholar
25. Hehman E, Leitner JB, Gaertner SL. Enhancing static facial features increases intimidation. J Exp Soc Psychol. Elsevier Inc.; 2013;49(4):747–54.
- View Article
- Google Scholar
26. Schneider TM, Hecht H, Carbon C- C. Judging body weight from faces: The height–weight illusion. Perception. 2012;41(1):121–4. pmid:22611670
- View Article
- PubMed/NCBI
- Google Scholar
27. Peron A, Morosini I. Photometric study of divine proportion and its correlation with facial attractiveness. Dental Press J Orthod. 2012;17(2):124–31.
- View Article
- Google Scholar
28. Verhoff MA, Witzel C, Kreutz K, Ramsthaler F. The ideal subject distance for passport pictures. Forensic Sci Int. 2008;178(2–3):153–6. pmid:18450396
- View Article
- PubMed/NCBI
- Google Scholar
29. Deutschmann R, Deutschmann R. Multiple flash photography: Off-camera flash techniques for digital photography. Buffalo: Amherst Media; 2011. 126 p.
30. Kelby S. The Digital Photography Book, Vol. 3. The Digital Photography Book, volume 3. Peachpit Press; 2010.
31. Cooper EA, Piazza EA, Banks MS. The perceptual basis of common photographic practice. J Vis. 2012;12(5):8. pmid:22637709
- View Article
- PubMed/NCBI
- Google Scholar
32. Banks MS, Cooper EA, Piazza EA. Camera focal length and the perception of pictures. Ecol Psychol. 2014;26(1–2):30–46. pmid:25089080
- View Article
- PubMed/NCBI
- Google Scholar
33. Todorov A, Olivola CY, Dotsch R, Mende-Siedlecki P. Social attributions from faces: determinants, consequences, accuracy, and functional significance. Annu Rev Psychol. 2015;66(1):519–45.
- View Article
- Google Scholar
34. Little AC. Facial attractiveness. Wiley Interdiscip Rev Cogn Sci. 2014 Nov;5(6):621–34. pmid:26308869
- View Article
- PubMed/NCBI
- Google Scholar
35. Rohlf F. TpsRelw (version 1.46). Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook. 2008.
36. Bohrnstedt GW. Reliability and validity assessment in attitude measurement. In: Summers GF, editor. Attitude Measurement. Chicago: Rand McNally; 1970. p. 81–99.
37. Kleisner K, Priplatova L, Frost P, Flegr J. Trustworthy-looking face meets brown eyes. PLoS One. 2013;8(1):e53285. pmid:23326406
- View Article
- PubMed/NCBI
- Google Scholar
38. Rohlf F. TpsSuper (version 1.14). New York: Department of Ecology and Evolution, State University of New York at Stony Brook. 2004.
39. Fritz CO, Morris PE, Richler JJ. Effect size estimates: Current use, calculations, and interpretation. J Exp Psychol Gen. 2012;141(1):2–18. pmid:21823805
- View Article
- PubMed/NCBI
- Google Scholar
40. Borenstein M, Hedges L V., Higgins JPT, Rothstein HR. Converting among effect sizes. Introduction to Meta-Analysis. Chichester, UK: John Wiley & Sons, Ltd; 2009.
41. Oksanen AJ, Blanchet FG, Kindt R, Legendre P, Minchin PR, Hara RBO, et al. Package “vegan.” 2015. p. 280.
42. Team RC. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013.
43. Hayduk LA. Personal space: Where we now stand. Psychol Bull. 1983;94(2):293–335.
- View Article
- Google Scholar
44. Lefevre CE, Perrett DI. Fruit over sunbed: Carotenoid skin colouration is found more attractive than melanin colouration. Q J Exp Psychol. 2015;68(2):284–93.
- View Article
- Google Scholar
45. Pezdirc K, Hutchesson M, Whitehead RD, Ozakinci G, Perrett DI, Collins C. Fruit, vegetable and dietary carotenoid intakes explain variation in skin-color in young caucasian women: A cross-sectional study. Nutrients. 2015;7(7):5800–15. pmid:26184306
- View Article
- PubMed/NCBI
- Google Scholar
46. Whitehead RD, Re DE, Xiao D, Ozakinci G, Perrett DI. You are what you eat: Within-subject increases in fruit and vegetable consumption confer beneficial skin-color changes. PLoS One. 2012;7(3):e32988. pmid:22412966
- View Article
- PubMed/NCBI
- Google Scholar
47. Burke D, Sulikowski D. A new viewpoint on the evolution of sexually dimorphic human faces. Evol Psychol. 2010;8(4):573–85. pmid:22947821
- View Article
- PubMed/NCBI
- Google Scholar
48. Sulikowski D, Burke D, Havlíček J, Roberts SC. Head tilt and fertility contribute to different aspects of female facial attractiveness. Ethology. 2015;121.
- View Article
- Google Scholar
49. Perona P. A new perspective on portraiture. J Vis. 2010;7(9):992–992.
- View Article
- Google Scholar

[ref1] 1. Ambady N, Skowronski JJ, editors. First Impressions. New York: Guilford Press; 2008. 368 p.

[ref2] 2. Rubešová A, Havlíček J. Facial appearance and personality judgments. In: Chadee D, Kostic A, editors. Social Psychological Dynamics. Kingston: University of the West Indies Press; 2011. p. 113–44.

[ref3] 3. Calder AJ, Rhodes G, Johnson M, Haxby J. Oxford Handbook of Face Perception. Rhodes G, Calder A, Johnson M, Haxby J V., editors. Oxford University Press; 2011.

[ref4] 4. Carré JM, McCormick CM. In your face: facial metrics predict aggressive behaviour in the laboratory and in varsity and professional hockey players. Proc R Soc B Biol Sci. 2008;275(1651):2651–6.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref5] 5. Třebický V, Fialová J, Kleisner K, Roberts SC, Little AC, Havlíček J. Further evidence for links between facial width-to-height ratio and fighting success: Commentary on Zilioli et al. (2014). Aggress Behav. 2015;41(4):331–4. pmid:25236530
View Article
PubMed/NCBI
Google Scholar

[8] View Article

[9] PubMed/NCBI

[10] Google Scholar

[ref6] 6. Haselhuhn MP, Ormiston ME, Wong EM. Men’s facial width-to-height ratio predicts aggression: A meta-analysis. PLoS One. 2015;10(4):e0122637. pmid:25849992
View Article
PubMed/NCBI
Google Scholar

[12] View Article

[13] PubMed/NCBI

[14] Google Scholar

[ref7] 7. Mileva VR, Cowan ML, Cobey KD, Knowles KK, Little AC. In the face of dominance: Self-perceived and other-perceived dominance are positively associated with facial-width-to-height ratio in men. Pers Individ Dif. 2014 Oct;69:115–8.
View Article
Google Scholar

[16] View Article

[17] Google Scholar

[ref8] 8. Valentine KA, Li NP, Penke L, Perrett DI. Judging a man by the width of his face: The role of facial ratios and dominance in mate choice at speed-dating events. Psychol Sci. 2014;25(3):806–11. pmid:24458269
View Article
PubMed/NCBI
Google Scholar

[19] View Article

[20] PubMed/NCBI

[21] Google Scholar

[ref9] 9. Stirrat MR, Perrett DI. Valid facial cues to cooperation and trust: male facial width and trustworthiness. Psychol Sci. 2010;21(3):349–54. pmid:20424067
View Article
PubMed/NCBI
Google Scholar

[23] View Article

[24] PubMed/NCBI

[25] Google Scholar

[ref10] 10. Windhager S, Schaefer K, Fink B. Geometric morphometrics of male facial shape in relation to physical strength and perceived attractiveness, dominance, and masculinity. Am J Hum Biol. 2011;23(6):805–14. pmid:21957062
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref11] 11. Kleisner K, Chvátalová V, Flegr J. Perceived intelligence is associated with measured intelligence in men but not women. PLoS One. 2014;9(3):e81237. pmid:24651120
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref12] 12. Třebický V, Havlíček J, Roberts SC, Little AC, Kleisner K. Perceived aggressiveness predicts fighting performance in mixed-martial-arts fighters. Psychol Sci. 2013;24(9):1664–72. pmid:23818656
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref13] 13. Geniole SN, McCormick CM. Facing our ancestors: judgements of aggression are consistent and related to the facial width-to-height ratio in men irrespective of beards. Evol Hum Behav. 2015;36(4):279–85.
View Article
Google Scholar

[39] View Article

[40] Google Scholar

[ref14] 14. Park JH, Buunk AP, Wieling MB. Does the face reveal athletic flair? Positions in team sports and facial attractiveness. Pers Individ Dif. 2007;43(7):1960–5.
View Article
Google Scholar

[42] View Article

[43] Google Scholar

[ref15] 15. Postma E. A relationship between attractiveness and performance in professional cyclists. Biol Lett. 2014 Feb 5;10(2):20130966–20130966. pmid:24501269
View Article
PubMed/NCBI
Google Scholar

[45] View Article

[46] PubMed/NCBI

[47] Google Scholar

[ref16] 16. Loehr J, O’Hara RB. Facial morphology predicts male fitness and rank but not survival in Second World War Finnish soldiers. Biol Lett. 2013;9(4):20130049–20130049. pmid:23658003
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref17] 17. Kleisner K, Kočnar T, Rubešová A, Flegr J. Eye color predicts but does not directly influence perceived dominance in men. Pers Individ Dif. 2010;49(1):59–64.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref18] 18. Roberts SC, Havlicek J, Flegr J, Hruskova M, Little AC, Jones BC, et al. Female facial attractiveness increases during the fertile phase of the menstrual cycle. Proc R Soc B Biol Sci. 2004;271:S270–2.
View Article
Google Scholar

[56] View Article

[57] Google Scholar

[ref19] 19. Bryan R, Perona P, Adolphs R. Perspective distortion from interpersonal distance is an implicit visual cue for social judgments of faces. PLoS One. 2012;7(9):e45301. pmid:23028918
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref20] 20. Hong G, Luo MR, Rhodes PA. A study of digital camera colorimetric characterisation based on polynomial modelling. Color Res Appl. 2001;26(3):76–84.
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref21] 21. Jenkins FA, White HE. Fundamentals of Optics, 4th Edition. New York: McGraw-Hill; 2001. 766 p.

[ref22] 22. Kingslake R. Optics in Photography. Bellingham, WA: SPIE Press; 1992.

[ref23] 23. Burriss RP, Troscianko J, Lovell PG, Fulford AJC, Stevens M, Quigley R, et al. Changes in women’s facial skin color over the ovulatory cycle are not detectable by the human visual system. PLoS One. 2015;10(7):e0130093. pmid:26134671
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref24] 24. Stephen ID, Smith MJL, Stirrat MR, Perrett DI. Facial skin coloration affects perceived health of human faces. Int J Primatol. 2009;30(6):845–57. pmid:19946602
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref25] 25. Hehman E, Leitner JB, Gaertner SL. Enhancing static facial features increases intimidation. J Exp Soc Psychol. Elsevier Inc.; 2013;49(4):747–54.
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref26] 26. Schneider TM, Hecht H, Carbon C- C. Judging body weight from faces: The height–weight illusion. Perception. 2012;41(1):121–4. pmid:22611670
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref27] 27. Peron A, Morosini I. Photometric study of divine proportion and its correlation with facial attractiveness. Dental Press J Orthod. 2012;17(2):124–31.
View Article
Google Scholar

[83] View Article

[84] Google Scholar

[ref28] 28. Verhoff MA, Witzel C, Kreutz K, Ramsthaler F. The ideal subject distance for passport pictures. Forensic Sci Int. 2008;178(2–3):153–6. pmid:18450396
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref29] 29. Deutschmann R, Deutschmann R. Multiple flash photography: Off-camera flash techniques for digital photography. Buffalo: Amherst Media; 2011. 126 p.

[ref30] 30. Kelby S. The Digital Photography Book, Vol. 3. The Digital Photography Book, volume 3. Peachpit Press; 2010.

[ref31] 31. Cooper EA, Piazza EA, Banks MS. The perceptual basis of common photographic practice. J Vis. 2012;12(5):8. pmid:22637709
View Article
PubMed/NCBI
Google Scholar

[92] View Article

[93] PubMed/NCBI

[94] Google Scholar

[ref32] 32. Banks MS, Cooper EA, Piazza EA. Camera focal length and the perception of pictures. Ecol Psychol. 2014;26(1–2):30–46. pmid:25089080
View Article
PubMed/NCBI
Google Scholar

[96] View Article

[97] PubMed/NCBI

[98] Google Scholar

[ref33] 33. Todorov A, Olivola CY, Dotsch R, Mende-Siedlecki P. Social attributions from faces: determinants, consequences, accuracy, and functional significance. Annu Rev Psychol. 2015;66(1):519–45.
View Article
Google Scholar

[100] View Article

[101] Google Scholar

[ref34] 34. Little AC. Facial attractiveness. Wiley Interdiscip Rev Cogn Sci. 2014 Nov;5(6):621–34. pmid:26308869
View Article
PubMed/NCBI
Google Scholar

[103] View Article

[104] PubMed/NCBI

[105] Google Scholar

[ref35] 35. Rohlf F. TpsRelw (version 1.46). Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook. 2008.

[ref36] 36. Bohrnstedt GW. Reliability and validity assessment in attitude measurement. In: Summers GF, editor. Attitude Measurement. Chicago: Rand McNally; 1970. p. 81–99.

[ref37] 37. Kleisner K, Priplatova L, Frost P, Flegr J. Trustworthy-looking face meets brown eyes. PLoS One. 2013;8(1):e53285. pmid:23326406
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref38] 38. Rohlf F. TpsSuper (version 1.14). New York: Department of Ecology and Evolution, State University of New York at Stony Brook. 2004.

[ref39] 39. Fritz CO, Morris PE, Richler JJ. Effect size estimates: Current use, calculations, and interpretation. J Exp Psychol Gen. 2012;141(1):2–18. pmid:21823805
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref40] 40. Borenstein M, Hedges L V., Higgins JPT, Rothstein HR. Converting among effect sizes. Introduction to Meta-Analysis. Chichester, UK: John Wiley & Sons, Ltd; 2009.

[ref41] 41. Oksanen AJ, Blanchet FG, Kindt R, Legendre P, Minchin PR, Hara RBO, et al. Package “vegan.” 2015. p. 280.

[ref42] 42. Team RC. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013.

[ref43] 43. Hayduk LA. Personal space: Where we now stand. Psychol Bull. 1983;94(2):293–335.
View Article
Google Scholar

[121] View Article

[122] Google Scholar

[ref44] 44. Lefevre CE, Perrett DI. Fruit over sunbed: Carotenoid skin colouration is found more attractive than melanin colouration. Q J Exp Psychol. 2015;68(2):284–93.
View Article
Google Scholar

[124] View Article

[125] Google Scholar

[ref45] 45. Pezdirc K, Hutchesson M, Whitehead RD, Ozakinci G, Perrett DI, Collins C. Fruit, vegetable and dietary carotenoid intakes explain variation in skin-color in young caucasian women: A cross-sectional study. Nutrients. 2015;7(7):5800–15. pmid:26184306
View Article
PubMed/NCBI
Google Scholar

[127] View Article

[128] PubMed/NCBI

[129] Google Scholar

[ref46] 46. Whitehead RD, Re DE, Xiao D, Ozakinci G, Perrett DI. You are what you eat: Within-subject increases in fruit and vegetable consumption confer beneficial skin-color changes. PLoS One. 2012;7(3):e32988. pmid:22412966
View Article
PubMed/NCBI
Google Scholar

[131] View Article

[132] PubMed/NCBI

[133] Google Scholar

[ref47] 47. Burke D, Sulikowski D. A new viewpoint on the evolution of sexually dimorphic human faces. Evol Psychol. 2010;8(4):573–85. pmid:22947821
View Article
PubMed/NCBI
Google Scholar

[135] View Article

[136] PubMed/NCBI

[137] Google Scholar

[ref48] 48. Sulikowski D, Burke D, Havlíček J, Roberts SC. Head tilt and fertility contribute to different aspects of female facial attractiveness. Ethology. 2015;121.
View Article
Google Scholar

[139] View Article

[140] Google Scholar

[ref49] 49. Perona P. A new perspective on portraiture. J Vis. 2010;7(9):992–992.
View Article
Google Scholar

[142] View Article

[143] Google Scholar

Figures

Abstract

Introduction

Methods

Ethics statement

Participants

Targets.

Raters.

Photograph acquisition

Rating sessions

Facial width-to-height ratio measurements

Geometric morphometrics

Statistical analysis

Results

Perceived characteristics

Facial width-to-height ratio

Geometric morphometrics

Discussion

Supporting Information

S1 Dataset. Data from ratings and fWHR measurements (.XLSX).

Acknowledgments

Author Contributions

References