Secondary Logo

Journal Logo

Analysis of the Visual Perception of Female Breast Aesthetics and Symmetry: An Eye-Tracking Study

Pietruski, Piotr M.D., Ph.D.; Paskal, Wiktor M.D.; Paskal, Adriana M. M.D.; Jaworowski, Janusz M.D.; Paluch, Łukasz M.D., Ph.D.; Noszczyk, Bartłomiej M.D., Ph.D.

Plastic and Reconstructive Surgery: December 2019 - Volume 144 - Issue 6 - p 1257-1266
doi: 10.1097/PRS.0000000000006292
Breast: Original Articles
Free
SDC
Discussion
Press Release
Watch Video

Background: Because of its subjective character, the term “breast attractiveness” is poorly defined and thus its reliable standardized assessment can be particularly challenging. Because of objective analysis of the observer’s gaze pattern, eye-tracking technology may provide a better insight into the visual perception of breast aesthetics and symmetry.

Methods: One hundred observers, 50 women and 50 men, assessed the aesthetics and symmetry of eight types of female breasts displayed as digital images on frontal, lateral, and oblique projections. The gaze pattern of each observer was recorded using eye-tracking technology, and gaze data were subsequently analyzed.

Results: Although sex and breast type exerted an effect on attention capturing by some areas of interest, key characteristics of gaze patterns in female and male observers were essentially the same. Irrespective of observers’ sex, the longest fixation duration and the highest fixation number were recorded for lower breast regions, in particular, for the nipple-areola complex. Mean fixation duration in this area corresponded to 58 and 57 percent of overall observation time for female and male observers, respectively, during the assessment of breast aesthetics; and to 56 and 52 percent of overall observation time for female and male observers, respectively, during the assessment of breast symmetry.

Conclusions: Nipple-areola complex and lower breast are key focus areas for the assessment of breast aesthetics and symmetry. Gaze data collected during this study may constitute a valuable source of reference values for future eye-tracking research on various patient groups’ visual perception of breast attractiveness and deformities.

Warsaw and Otwock, Poland

From the Department of Plastic and Reconstructive Surgery, Medical Centre of Postgraduate Education, Prof. W. Orlowski Memorial Hospital; Department of Research Methodology, Center for Preclinical Research, Medical University of Warsaw; Department of Histology and Embryology, Warsaw Medical University; Timeless Plastic Surgery Clinic; and the Department of Radiology, Medical Centre of Postgraduate Education, Gruca Orthopaedic and Trauma Teaching Hospital.

Received for publication June 20, 2018; accepted February 13, 2019.

Disclosure:The authors have no financial interest in any of the products or devices mentioned in this article. No funding was received for this article.

Related digital media are available in the full-text version of the article on www.PRSJournal.com.

A “Hot Topic Video” by Editor-in-Chief Rod J. Rohrich, M.D., accompanies this article. Go to PRSJournal.com and click on “Plastic Surgery Hot Topics” in the “Digital Media” tab to watch.

Piotr Pietruski, M.D., Ph.D., Department of Plastic and Reconstructive Surgery, Medical Centre of Postgraduate Education, Prof. W. Orlowski Memorial Hospital, Czerniakowska 231 Street, 00-416 Warsaw, Mazowieckie, Poland, pietruski.piotr@gmail.com

A primary objective of aesthetic and reconstructive breast operations is to restore normal, satisfactory for the patient, presentation of the breasts. Successful breast surgery requires preoperative evaluation, selection of an appropriate operative technique, and analysis of surgical outcomes as a basis for future clinical advancements. This puts particular emphasis on the understanding of breast beauty components. Because of its subjective character, the term “breast aesthetics” is poorly defined and thus its reliable standardized assessment can be particularly challenging.1–9

Subjective evaluation is based mainly on visual inspection of the breasts during a physical examination and on the analysis of photographs and three-dimensional models, using various scales and guidelines.3,8,10,11 In recent years, patient-reported outcome instruments have also been gaining in popularity.3,12,13 Widely applied objective evaluation methods include direct and indirect anthropometry, with the latter involving measurements taken on breast photographs and three-dimensional models.3,14–17 These analyses are suitable for accurate quantitative assessment of important characteristics of the breasts, such as size, shape, and symmetry.3,18–20 Although many methods for aesthetic evaluation of the breasts exist, in particular for the analysis of postoperative outcomes, none of them is considered a gold standard.8 Subjective methods provide low intraobserver and interobserver agreement.3 The anthropometric analysis is time-consuming, complex, and expensive in the case of three-dimensional scanning. Moreover, their quantitative results cannot be easily translated on a qualitative perception of the aesthetic outcome.3,8 This seems to be a strong argument supporting subjective assessment as a direction of future research on breast aesthetics.

Equally important as the technique for subjective assessment of breast aesthetics is the evaluator. A discrepancy in the perception of body attractiveness and surgical outcomes by patients and plastic surgeons is a widely recognized issue, and this problem also extends to the concept of breast beauty.7,13,21,22 For example, it was shown that the opinions of plastic surgeons and patients regarding the shape and attractiveness of augmented breasts may differ considerably.22 Without doubt, patient satisfaction is an ultimate measure of surgical outcome. However, subjective perception of the outcome is influenced by other factors, such as cultural and ethnic background or medical history. Thus, development of algorithms used in plastic and reconstructive surgery of the breasts that are based on patient opinion does not seem to be justified. An optimal solution seems to be a universal scale for the evaluation of breast aesthetics that provides good interobserver agreement between medical professionals and patients.

Eye-tracking technology is suitable for noninvasive objective assessment of the visual perception process based on the analysis of ocular movements. This provides better insight into gaze patterns, including the stimuli that draw an observer’s attention, fixation duration, and order. Although more popular in marketing, eye tracking has also found application in various medical sciences.23–29 Surprisingly, however, there are only single reports on the use of this technology in plastic and reconstructive surgery.30–34

In this study, we used eye-tracking technology to objectively analyze visual processes taking place during the assessment of female breast aesthetics and symmetry. In our opinion, understanding of women’s and men’s visual perception with regard to breast symmetry and aesthetics may facilitate the development of new reliable methods for subjective evaluation of surgical outcomes.

Back to Top | Article Outline

MATERIALS AND METHODS

This study was reviewed and approved by our institutional review board. Nine three-dimensional high-poly models of the trunk and neck with the skin corresponding to type II according to the Fitzpatrick skin type scale were created for this study, based on archival photographic documentation and three-dimensional images.35 Each model represented a different type of female breasts (Fig. 1). The models were used to render highly standardized breast images on frontal, lateral (right and left), and oblique 45-degree (right and left) projection. [See Video 1 (online), which demonstrates the visual data acquisition.] This enabled us to eliminate potential bias related to lighting conditions, the position of the breasts, type of the skin envelope, and the presence of skin lesions. The images from one model were used to familiarize observers with the experimental procedures, and others constituted the material for the study proper.

Fig. 1.

Fig. 1.

{"href":"Single Video Player","role":"media-player-id","content-type":"play-in-place","position":"float","orientation":"portrait","label":"Video 1.","caption":"This video demonstrates the visual data acquisition.","object-id":[{"pub-id-type":"doi","id":""},{"pub-id-type":"other","content-type":"media-stream-id","id":"1_xl6uvs48"},{"pub-id-type":"other","content-type":"media-source","id":"Kaltura"}]}

The panel of observers included 100 medical laypersons (persons who neither had a medical background nor had been receiving medical education at the time of the study): 50 women (mean age, 34.2 ± 11.5 years; range, 20 to 60 years) and 50 men (mean age, 34.8 ± 10.6 years; range, 20 to 58 years) with Caucasian heritage and the same cultural background. Exclusion criteria were eye motility disorders, cataract, astigmatism and other eye diseases, schizophrenia, autism spectrum disorders, and history of oncologic or plastic operations involving the chest/breasts. The observers were informed that they would assess breast images using their eyes solely, without the involvement of a computer mouse or keyboard. After providing consent, each observer was familiarized with the experimental setup.

The observer sat on a chair at a standardized distance of 60 cm in front of a 15-inch computer screen calibrated with EyeTribe eye-tracker device (TheEyeTribe, Copenhagen, Denmark). After stabilization of the head, the observer was instructed to keep it stable until the end of the study. Visual gaze was recorded and analyzed with Ogama 5.0 software (Adrian Voßkühler, Freie Universität, Berlin, Germany). The eye-tracking system was calibrated by means of a 12-point calibration procedure. The next stage started after the quality of the calibration was graded as “perfect.” Then, a sample breast image was displayed. The observers were informed that they would have 30 seconds to assess each image, and a screen with circles containing scores from 1 to 10 would be displayed after that. To proceed to another image, the observer needed to gaze at a selected circle for 2 seconds. Placement of the circles on the right side of the monitor prevented accidental fixations during the early assessment of another image. After the observer confirmed their readiness to start the experiment, the main screen with the instruction to assess breast aesthetics from 1 to 10 (1 = very poor and 10 = excellent) was displayed. The procedure started after the observer gazed at the white “start” rectangle. First, eight frontal images were assessed. After the last image in the series was assessed, the start screen was displayed again, and the same procedure was repeated for eight lateral images, and then for eight images in oblique projection. After an obligatory 40-minute break, the second session was conducted according to the same rules, but instead of breast aesthetics, the observer was asked to assess the symmetry of breast images, using a scale from 1 to 10. Whereas in the case of frontal images the same photographs as during the first session were displayed, each right-side lateral and oblique image was accompanied by its left-side mirror copy. After completing the second session, the observers were informed about the aim of the study and again signed the consent form. Such approach enabled us to maximize the likelihood of spontaneous, unbiased assessment of breast images.

Based on our 2-year experience with eye-tracking technology and published evidence, the fixation threshold was set at greater than or equal to 100 msec.33,36 All visual fixations shorter than this value were defined as irrelevant saccades. Each breast image was divided into several areas of interest (Fig. 2). Visual gaze data collected during the assessment of breast aesthetics and symmetry were objectively analyzed in terms of fixation duration and number per each area of interest. Moreover, the sequence of three initial fixations (excluding repeated fixations in the same area of interest) was established. [See Video 2 (online), which demonstrates the visual data analysis.]

Fig. 2.

Fig. 2.

{"href":"Single Video Player","role":"media-player-id","content-type":"play-in-place","position":"float","orientation":"portrait","label":"Video 2.","caption":"This video demonstrates the visual data analysis.","object-id":[{"pub-id-type":"doi","id":""},{"pub-id-type":"other","content-type":"media-stream-id","id":"1_hfuqasw0"},{"pub-id-type":"other","content-type":"media-source","id":"Kaltura"}]}

Normal distribution of the study variables was verified with the Shapiro-Wilk test. Statistical characteristics of the results were presented as arithmetic means and standard deviations. Intergroup comparisons were carried out with the Mann-Whitney U test, the chi-square test, and Fisher’s exact test. The influence of sex and breast type on the eye-tracking parameters was examined with factorial analysis of variance; both main effects and interaction effects of the explanatory variables were considered. All calculations were carried out with the Statistica 10 package (StatSoft, Inc., Tulsa, Okla.), with the threshold of statistical significance set at p < 0.05.

Back to Top | Article Outline

RESULTS

Gaze Patterns during Breast Aesthetics Assessment

Mean fixation duration in female observers calculated for all eight breast types was 21.5 ± 2.95 seconds for images on frontal projection and 21.21 ± 3.5 seconds for images on lateral projection. In male observers, mean fixation duration for images on frontal and lateral projection was 21.86 ± 2.42 and 22.08 ± 3.43 seconds, respectively. The rest of the 30-second observation time was spent on blinking, saccade movements, and fixation-to-fixation transitions. Mean fixation number in female observers, averaged for all analyzed breast types, was 60.07 ± 12.22 for images on frontal projection and 46.1 ± 12.21 for images on lateral projection. In male observers, mean fixation numbers for images on frontal and lateral projection were 63.34 ± 10.23 and 46.84 ± 10.26, respectively (Tables 1 and 2). (See Table, Supplemental Digital Content 1, which shows the results of analysis of variance for gaze pattern in the assessment of breast aesthetics and symmetry, http://links.lww.com/PRS/D785.)

Table 1. - Summarized Results of Gaze Pattern Analysis in the Assessment of Breast Aesthetics on Frontal View
Area of Interest Visual Fixation Duration (%) Visual Fixation Number (%)
Female Male p Female Male p
Mean SD Mean SD Mean SD Mean SD
Frontal projection view
 Neck 0.20 0.59 0.19 0.51 0.90 0.30 0.99 0.26 0.70 0.88
 Clavicular region, R 0.20 0.50 0.24 0.54 0.29 0.27 0.69 0.33 0.75 0.25
 Clavicular region, L 0.20 0.51 0.20 0.53 0.88 0.27 0.74 0.27 0.77 0.94
 Upper sternum 2.08 3.62 1.55 2.34 0.45 2.05 2.86 1.60 1.93 0.38
 Lower sternum 6.92 6.97 5.13 4.66 <0.01* 6.56 5.44 5.00 3.69 <0.01*
 Upper outer quadrant, R 2.32 2.74 3.77 3.26 <0.01* 2.58 2.82 3.98 3.18 <0.01*
 Upper inner quadrant, R 2.59 3.26 2.83 2.51 <0.01* 2.82 3.10 3.14 2.56 <0.01*
 Lower outer quadrant, R 28.99 13.18 30.61 8.76 <0.01* 27.44 9.04 29.23 7.61 <0.01*
 Lower inner quadrant, R 10.13 6.57 8.58 4.26 0.02* 11.42 6.24 9.77 4.64 <0.01*
 Nipple-areola complex, R 27.73 13.22 28.34 9.29 0.07 25.94 9.41 26.61 7.85 0.07
 Inframammary fold, R 7.88 8.50 6.99 7.18 0.73 8.23 7.49 7.59 7.23 0.27
 Upper inner quadrant, L 2.86 4.02 2.63 2.69 0.08 3.05 3.69 2.85 2.77 0.24
 Upper outer quadrant, L 2.52 2.65 4.17 3.69 <0.01* 2.87 2.76 4.40 3.63 <0.01*
 Lower inner quadrant, L 11.29 7.93 8.27 4.02 <0.01* 11.71 6.33 9.21 4.14 <0.01*
 Lower outer quadrant, L 29.68 12.07 31.83 8.73 <0.01* 28.66 9.23 29.95 7.40 0.01*
 Nipple-areola complex, L 28.07 12.83 28.98 9.21 0.03* 26.39 9.97 26.46 7.84 0.48
 Inframammary fold, L 9.00 8.16 8.41 7.40 0.53 9.80 11.64 8.88 6.63 0.46
R
, right; L, left.
*
Statistically significant.

Table 2. - Results of Gaze Pattern Analysis in the Assessment of Breast Aesthetics and Symmetry on Lateral View
Area of Interest Visual Fixation Duration (%) Visual Fixation Number (%)
Female Male p Female Male p
Mean SD Mean SD Mean SD Mean SD
Aesthetics assessment: lateral projection view
 Neck 1.66 2.1 2.64 4.20 <0.01* 2.46 3.56 3.36 4.62 0.01*
 Upper breast 18.97 15.90 21.56 14.30 <0.01* 23.19 14.53 25.81 13.00 <0.01*
 Upper pole 3.80 4.50 5.35 5.71 <0.01* 4.97 5.26 6.84 6.31 <0.01*
 Lower breast 79.37 16.92 75.80 15.51 <0.01* 74.36 15.88 70.83 14.67 <0.01*
 Nipple-areola complex 62.76 37.76 56.75 16.31 <0.01* 51.11 14.07 47.20 12.81 <0.01*
 Inframammary fold 13.38 23.01 12.16 8.39 0.21 14.74 10.15 14.58 7.62 0.37
 Lower pole 10.42 12.61 10.36 10.38 <0.01* 10.95 10.57 11.15 8.46 0.01*
Symmetry assessment: right lateral projection view
 Neck 0.65 1.77 0.60 1.23 0.35 0.76 1.80 0.83 1.83 0.14
 Upper breast 6.68 5.35 7.69 5.99 0.02* 7.30 5.40 8.13 5.88 0.04*
 Upper pole 2.41 3.11 2.78 3.11 0.03* 2.53 2.88 2.84 2.91 0.05
 Lower breast 41.89 11.64 41.58 12.48 0.66 39.95 8.95 40.27 9.40 0.45
 Nipple-areola complex 33.44 11.74 33.66 13.37 0.90 30.28 8.88 30.34 9.83 0.70
 Inframammary fold 6.31 5.35 6.22 5.55 0.48 6.50 5.35 6.64 5.24 0.52
 Lower pole 3.38 3.57 3.40 3.77 0.72 3.27 3.30 3.40 3.38 0.61
Symmetry assessment: left lateral projection view
 Neck 1.05 1.81 0.96 1.70 0.47 1.34 2.28 1.10 1.82 0.33
 Upper breast 8.00 5.61 8.55 5.86 0.14 9.05 5.70 9.14 5.54 0.42
 Upper pole 3.11 2.91 3.33 3.32 0.53 3.46 3.08 3.49 3.06 0.69
 Lower breast 41.72 11.17 40.62 9.28 0.13 41.61 8.48 40.52 7.64 0.05
 Nipple-areola complex 30.58 11.00 31.61 9.66 0.03* 29.48 8.00 29.32 6.82 0.98
 Inframammary fold 7.91 6.08 6.88 4.85 0.05* 7.86 5.41 7.58 4.80 0.86
 Lower pole 3.85 3.31 3.83 3.28 0.85 3.64 2.83 3.81 3.00 0.48
*
Statistically significant.

The data illustrating the effect of sex and breast type on fixation duration and number per area of interest are presented. (See Table, Supplemental Digital Content 2, which shows the summarized results of gaze pattern analysis in the assessment of breast aesthetics, http://links.lww.com/PRS/D786.) A significant effect of sex on fixation duration was observed in 11 of 17 areas of interest from frontal images (in two areas of interest only for some breast types), eight of 13 areas of interest from oblique images, and two of seven areas of interest from lateral images. In the case of frontal and oblique images, the effect of sex on fixation duration depended on breast type. Sex also exerted a significant effect on fixation number in 11 of 17 areas of interest from frontal images; however, in three areas of interest, the effect of sex was observed only for some breast types; and in another three areas of interest, its power and direction turned out to be breast type–specific. In eight areas of interest from oblique images, the effect of sex on fixation number was observed regardless of breast type; and in one area of interest, only for some specific types. On lateral images, the effect of sex on fixation number was documented in five of seven areas of interest, including one area of interest in the case of which power and direction of the effect were breast type–specific.

The sequences of the initial three fixations for images on various projections are presented in Table 3. In the case of frontal images, the most common initial fixation patterns in female and male observers were right lower sternum, right nipple-areola complex, left nipple-areola complex (4.8 percent gaze patterns) and left lower inner quadrant, right nipple-areola complex, left nipple-areola complex (3.8 percent gaze patterns), respectively. In the case of the remaining two projections, the most common initial fixation patterns in female and male observers were the same: nipple-areola complex, lower breast, upper breast (11.3 percent women versus 14 percent men; p = 0.24) and right inframammary fold, right nipple-areola complex, left nipple-areola complex (5 percent women versus 4.5 percent men; p = 0.74) for lateral and oblique images, respectively.

Table 3. - Summarized Data on Three Initial Fixations during the Assessment of Breast Aesthetics and Symmetry
Breast Aesthetics Assessment Breast Symmetry Assessment
Area of Interest Female (%) Male (%) p Area of Interest Female (%) Male (%) p
Frontal projection view
 First fixation
  NAC, L 24.5 23.3 0.68 NAC, L 28.8 33.3 0.17
  NAC, R 15.3 26.8 <0.01* NAC, R 18.8 25.3 0.03*
  Lower inner quadrant, L 13.8 11.0 0.24 IMF, L 16.3 12.8 0.16
 Second fixation
  NAC, R 42 35.5 0.06 NAC, R 35.8 43.3 0.03*
  NAC, L 23.5 22.8 0.80 NAC, L 25.8 25.0 0.81
  Lower inner quadrant, R 8.0 11.0 0.15 Lower inner quadrant, R 9.0 5.0 0.03*
 Third fixation
  NAC, L 25.0 24.3 0.81 NAC, L 22.0 25.0 0.32
  NAC, R 20.5 18.5 0.48 NAC, R 20.3 14.3 0.03*
  Lower sternum 11.3 6.8 0.03* Lower sternum 11.0 9.5 0.48
Lateral projection view
 First fixation
  NAC 47.0 49.0 0.57 NAC, L 30.8 27.3 0.28
  Lower breast 26.3 25.8 0.87 NAC, R 21.5 25.5 0.18
  Upper breast 13.8 9.5 0.06 Lower breast, L 14.3 15.3 0.69
 Second fixation
  NAC 39.8 37.8 0.56 NAC, R 34.5 29.5 0.13
  Lower breast 27.5 27.8 0.94 NAC, L 27.3 29.0 0.58
  Upper breast 12.5 13.5 0.67 Lower breast, R 14.5 11.8 0.25
 Third fixation
  Upper breast 30.5 33.8 0.33 Lower breast, R 18.0 19.5 0.59
  Lower breast 18.0 17.8 0.93 NAC, R 18.8 17.0 0.52
  Lower pole 17.0 16.0 0.70 Lower breast, L 17.8 15.3 0.34
Oblique projection view
 First fixation
  NAC, R 37.8 37.5 0.94 RP NAC, R 15.5 20.8 0.05
  Lower sternum 11.5 11.3 0.91 LP NAC, L 13.5 11.8 0.46
  IMF, R 11.3 10.3 0.65 LP IMF, L 11.5 6.8 0.02*
 Second fixation
  NAC, R 31.5 33.8 0.50 LP NAC, L 15.3 16.0 0.78
  NAC, L 26.5 24.5 0.52 RP NAC, P 13.8 16.3 0.32
  Lower breast, R 10.3 10.5 0.91 RP Lower breast, R 8.8 10.5 0.40
 Third fixation
  NAC, L 24.5 24.8 0.94 LP NAC, L 10.5 16.3 0.01*
  Lower breast, L 14.8 11.8 0.21 RP NAC, R 10.3 15.0 0.04*
  NAC, R 11.3 11.5 0.91 RP NAC, L 9.0 11.3 0.29
R
, right; L, left; NAC, nipple-areola complex; IMF, inframammary fold; RP, right projection; LP, left projection.
*
Statistically significant.

Back to Top | Article Outline

Gaze Patterns during Breast Symmetry Assessment

Mean fixation duration in female observers calculated for all breast types was 21.17 ± 3.4 seconds for frontal images and 19.79 ± 4.04 seconds for lateral images. In male observers, mean fixation duration for frontal and lateral images was 22.09 ± 2.66 and 21.27 ± 3.18 seconds, respectively. Mean fixation number per area of interest from a single frontal image was 61.12 ± 13.76 and 63.02 ± 12.23 for female and male observers, respectively (Table 4). In women, mean fixation number per area of interest from two lateral images was 60.02 ± 13.37, as compared with 54.70 ± 13.13 in men (Tables 2 and 4). (See Table, Supplemental Digital Content 3, which shows the summarized results of gaze pattern analysis in the assessment of breast symmetry, http://links.lww.com/PRS/D787.)

Table 4. - Summarized Results of Gaze Pattern Analysis in the Assessment of Breast Symmetry on Frontal View
Area of Interest Visual Fixation Duration (%) Visual Fixation Number (%)
Female Male p Female Male p
Mean SD Mean SD Mean SD Mean SD
Neck 0.16 0.50 0.11 0.45 0.04* 0.21 0.67 0.15 0.62 0.04*
Clavicular region, R 0.18 0.50 0.20 0.50 0.44 0.22 0.63 0.25 0.64 0.45
Clavicular region, L 0.13 0.43 0.15 0.46 0.44 0.18 0.63 0.21 0.61 0.40
Upper sternum 1.52 2.72 1.21 2.70 0.08 1.51 2.31 1.19 2.06 0.06
Lower sternum 7.41 7.17 4.39 3.64 <0.01* 6.50 5.54 4.45 3.13 <0.01*
Upper outer quadrant, R 2.45 3.47 3.28 3.25 <0.01* 2.55 3.42 3.42 3.14 <0.01*
Upper inner quadrant, R 1.45 2.14 1.54 1.92 0.10 1.67 2.29 1.75 2.00 0.23
Lower outer quadrant, R 26.45 10.65 32.61 10.77 <0.01* 26.43 8.34 31.44 9.03 <0.01*
Lower inner quadrant, R 10.82 5.92 8.89 5.54 <0.01* 11.92 5.64 9.56 5.57 <0.01*
Nipple-areola complex, R 22.72 10.32 28.43 10.66 <0.01* 22.04 7.91 26.67 8.83 <0.01*
Inframammary fold, R 9.74 7.44 8.71 7.82 <0.01* 10.16 7.03 9.08 7.33 <0.01*
Upper inner quadrant, L 1.87 2.78 1.73 2.22 0.57 2.06 2.80 1.93 2.29 0.92
Upper outer quadrant, L 2.94 4.03 3.28 3.21 <0.01* 3.05 3.68 3.52 3.24 <0.01*
Lower inner quadrant,, L 12.43 7.81 9.71 6.20 <0.01* 12.80 6.81 10.13 5.69 <0.01*
Lower outer quadrant, L 32.18 12.57 32.91 8.91 0.02* 30.89 9.21 32.01 7.32 0.01*
Nipple-areola complex, L 27.71 12.35 28.69 9.37 0.01* 26.08 9.40 27.05 7.34 0.02*
Inframammary fold, L 12.63 10.18 9.03 6.96 <0.01* 12.62 9.27 9.80 8.22 <0.01*
R
, right; L, left.
*
Statistically significant.

The results of analysis of variance for the effect of sex and breast type on fixation duration and number are presented in Table, Supplemental Digital Content 2, http://links.lww.com/PRS/D786. Sex exerted a significant effect on fixation duration and number in 10 of 17 and 13 of 17 areas of interest from frontal images, respectively, and in seven of 14 and three of 14 areas of interest from lateral images, respectively. However, in two areas of interest from the lateral images, the effect of sex was observed only for some breast types. In the case of oblique images, a significant effect of sex on fixation duration and number was documented in 17 of 26 and 12 of 26 areas of interest, respectively.

The most common sequences of the initial three fixations during breast symmetry assessment are shown in Table 3. In both female and male observers, the most common sequence of initial fixations on frontal images was left inframammary fold, right nipple-areola complex, left nipple-areola complex (women, 5.3 percent; men, 4.8 percent; p = 0.75). The most common initial fixation patterns in women and men were left nipple-areola complex, right nipple-areola complex, left lower breast (7 percent gaze patterns) and left nipple-areola complex, right nipple-areola complex, right lower breast (8.3 percent gaze patterns), respectively, for lateral images; and left breast right nipple-areola complex, right breast right nipple-areola complex, left lower breast (1.5 percent gaze patterns) and right breast nipple-areola complex, right breast left nipple-areola complex, left breast right nipple-areola complex (2.5 percent gaze patterns), respectively, for oblique images. The latter sequence was significantly less common in women than in men (0.3 percent; p = 0.01).

Back to Top | Article Outline

DISCUSSION

Until recently, our knowledge about the perception of breast attractiveness originated primarily from the studies with various questionnaires and subjective scales. Eye-tracking technology could be a new objective method to understand this process; providing appropriate research methodology, eye tracking is less prone to observer-related bias. Only a few previous studies used eye tracking for visual assessment of female breasts, and our experiment is the first attempt to understand this process in a meticulous, standardized, complex way.31–34

Our primary objective was to analyze visual patterns of women and men during evaluation of the aesthetics and symmetry of female breasts. Using eye-tracking data, we created descriptive perception patterns for both sexes (Table 5). Moreover, we obtained the data on fixation duration for various breast types, presented in the form of heat maps. (See Figure, Supplemental Digital Content 4, which shows the heat map of the visual assessment of breast aesthetics on frontal view, http://links.lww.com/PRS/D788. See Figure, Supplemental Digital Content 5, which shows the heat map of the visual assessment of breast aesthetics on lateral view, http://links.lww.com/PRS/D789. See Figure, Supplemental Digital Content 6, which shows the heat map of the visual assessment of breast symmetry on frontal view, http://links.lww.com/PRS/D790. See Figure, Supplemental Digital Content 7, which shows the heat map of the visual assessment of breast symmetry on lateral view, http://links.lww.com/PRS/D791.) Although sex and breast type exerted an effect on attention capturing in some areas of interest, the key characteristics of gaze patterns in women and men were essentially the same. Assessing either the aesthetics or the symmetry, both female and male observers focused primarily on lower regions of the breasts. Regardless of sex, minimum duration of fixations in these regions (75 percent and 76 percent in women and men, respectively) was documented during the assessment of aesthetics on oblique images, and maximum duration during the assessment of symmetry, on lateral images in the case of women (84 percent) and on frontal images in the case of men (83 percent). However, it needs to be stressed that most fixations in this area corresponded to the nipple-areola complex, which played a key role in the assessment of breast symmetry and aesthetics. This finding is consistent with other similar studies.31,34 Mean fixation duration for the nipple-areola complex during the assessment of breast aesthetics on frontal, lateral, and oblique images was 55, 63, and 55 percent, respectively, in female observers; and 57, 57, and 56 percent, respectively, in male observers. During the assessment of breast symmetry, mean fixation duration for nipple-areola complex in female and male observers corresponded to 50 percent and 48 percent, respectively, on frontal images; 64 percent and 60 percent, respectively, on lateral images; and 54 percent and 49 percent, respectively, on oblique images. Another argument for the importance of the nipple-areola complex in visual perception of breast stems from the fact that regardless of observers’ sex, this area of interest was the most common point of initial fixation.

Table 5. - Summary of Sex-Specific Differences for Visual Preferences in the Assessment of Breast Aesthetics and Symmetry
Projection View Longer Fixation More Fixations
Breast aesthetics assessment
 Frontal
  Female Lower sternum, right LIQ, left LIQ Lower sternum, right LIQ, left LIQ
  Male Upper right breast, right LOQ, lateral left breast, left NAC Upper right breast, right LOQ, lateral left breast
 Lateral
  Female Lower breast, NAC Lower breast, NAC
  Male Neck, upper breast, upper pole, lower pole Neck, upper breast, upper pole, lower pole
 Oblique
  Female Neck, both clavicular regions, lower left breast and NAC Neck, left clavicular region, lower left breast and NAC
  Male Upper and lower right breast, right NAC and IMF Upper and lower right breast, right NAC
Breast symmetry assessment
 Frontal
  Female Neck, lower sternum, right LIQ, left LIQ, both IMFs Neck, lower sternum, right LIQ, left LIQ, both IMFs
  Male Lateral right breast, lateral left breast, both NACs Lateral right breast, lateral left breast, both NACs
 Lateral
  Female Left IMF
  Male Right upper breast with upper pole, left NAC Right upper breast
 Right oblique
  Female
  Male Both clavicular regions, right upper and lower breast, right NAC and IMF Both clavicular regions, right upper and lower breast, right NAC
 Left oblique
  Female Right upper and lower breast, right NAC Right upper and lower breast, right NAC
  Male Both clavicular regions, upper left breast, left NAC Both clavicular regions, upper left breast, left NAC
L
IQ, lower inner quadrant; LOQ, lower outer quadrant; NAC, nipple-areola complex; IMF, inframammary fold.

Surprisingly, upper regions of the breasts generally did not attract the observers’ attention during the assessment of aesthetics and symmetry. Although our study demonstrated that these regions gained significantly more attention from men than from women, compared with the lower breasts, the values of eye-tracking parameters were substantially lower. A subjective analysis of visual data showed that both in the upper regions and in the lower regions other than nipple-areola complex, fixations centered primarily at the breast periphery, especially in outer regions and the inframammary fold. This implies that the observers focused primarily on the assessment of lower breast shape.

Another interesting finding is the observation that both fixation duration and number for the lower sternum, inframammary fold, and lower inner quadrants increased with breast size and the degree of ptosis. Attention capturing by the lower sternum was particularly evident among female observers and during the assessment of breast symmetry. Furthermore, our findings clearly show that the neck, shoulders, and clavicular regions had a lesser impact on breast visual assessment and thus could be potentially omitted in similar future eye-tracking studies.

It needs to be stressed that we still do not know how the visual pattern parameters translate on a subjective assessment of breast attractiveness. The results of some eye-tracking studies imply, albeit not unequivocally, that observers may focus their visual attention on some unique features, either attractive or unattractive ones.31–33,37–39 Considering this assumption and the results of our study, one may hypothesize that visual analysis of nipple-areola complex and lower breast shape, rather than the assessment of the upper breast’s fullness, might exert a profound effect on both positive and negative perception of breast symmetry and aesthetics. If verified empirically, this notion might play a fundamental role in planning aesthetic and reconstructive breast operations. However, ultimate verification of this hypothesis requires further research on correlations between visual perception and subjective preferences regarding breast attractiveness.

In our opinion, eye tracking, especially the analysis of attention-capturing time, might be used for the identification of parameters that are critical for the perception of breast attractiveness by various groups of patients or plastic surgeons. These data may be helpful for understanding and subsequent standardization of subjective breast assessment procedure, and perhaps might even facilitate surgeon-patient communication regarding priorities of the operation. Our research, providing the reference data for healthy persons, constitutes a basis for further eye-tracking studies targeting specific groups (e.g., patients who seek breast augmentation or break reconstruction) or analyzing perception of some specific breast deformations. Furthermore, eye-tracking technologies might improve our understanding of body dysmorphic disorders and perhaps even facilitate detection thereof based on comparison of gaze patterns for patients and healthy controls. However, it should be remembered that visual perception of breast symmetry and aesthetics may be influenced by other factors, such as cultural background or socioeconomic status.9,10,22,39–42 The results of our study were obtained in the selected population; thus, similar experiments should be repeated in other groups of observers.

Back to Top | Article Outline

CONCLUSIONS

This study provided objective data on visual patterns of women and men during the assessment of breast aesthetics and symmetry. The study clearly showed that, irrespective of the observer’s sex, lower regions of the breasts and nipple-areola complex were the key focus areas of gaze pattern. The large set of eye-tracking data collected during this study may constitute a valuable reference for similar experiments analyzing the visual perception of female breast attractiveness and deformities.

Back to Top | Article Outline

REFERENCES

1. Van Limbergen E, van der Schueren E, Van Tongelen K. Cosmetic evaluation of breast conserving treatment for mammary cancer: 1. Proposal of a quantitative scoring system. Radiother Oncol. 1989;16:159–167.
2. Lowery JC, Wilkins EG, Kuzon WM, Davis JA. Evaluations of aesthetic results in breast reconstruction: An analysis of reliability. Ann Plast Surg. 1996;36:601–606; discussion 607.
3. Kim MS, Sbalchiero JC, Reece GP, Miller MJ, Beahm EK, Markey MK. Assessment of breast aesthetics. Plast Reconstr Surg. 2008;121:186e–194e.
4. Hall-Findlay EJ. The three breast dimensions: Analysis and effecting change. Plast Reconstr Surg. 2010;125:1632–1642.
5. Brown N, Scurr J. The need for a standardised anthropometric protocol for objective assessment of pre- and postoperative breast surgery. Gland Surg. 2012;1:142–145.
6. Swanson E. A measurement system for evaluation of shape changes and proportions after cosmetic breast surgery. Plast Reconstr Surg. 2012;129:982–992.
7. Alderman A, Chung KC. Measuring outcomes in aesthetic surgery. Clin Plast Surg. 2013;40:297–304.
8. Begic A, Stark B. The Telemark Breast Score: A reliable method for the evaluation of results after breast surgery. Plast Reconstr Surg. 2016;138:390e–400e.
9. Mallucci P, Branford OA. Population analysis of the perfect breast: A morphometric analysis. Plast Reconstr Surg. 2014;134:436–447.
10. Mallucci P, Branford OA. Concepts in aesthetic breast dimensions: Analysis of the ideal breast. J Plast Reconstr Aesthet Surg. 2012;65:8–16.
11. Martinovic ME, Blanchet NP. BFACE: A framework for evaluating breast aesthetics. Plast Reconstr Surg. 2017;140:287e–295e.
12. Andrade WN, Semple JL. Patient self-assessment of the cosmetic results of breast reconstruction. Plast Reconstr Surg. 2006;117:44–47; discussion 48–49.
13. Pusic AL, Lemaine V, Klassen AF, Scott AM, Cano SJ. Patient-reported outcome measures in plastic surgery: Use and interpretation in evidence-based medicine. Plast Reconstr Surg. 2011;127:1361–1367.
14. Galdino GM, Nahabedian M, Chiaramonte M, Geng JZ, Klatsky S, Manson P. Clinical applications of three-dimensional photography in breast surgery. Plast Reconstr Surg. 2002;110:58–70.
15. Liu C, Luan J, Mu L, Ji K. The role of three-dimensional scanning technique in evaluation of breast asymmetry in breast augmentation: A 100-case study. Plast Reconstr Surg. 2010;126:2125–2132.
16. Becker H. The role of three-dimensional scanning technique in evaluation of breast asymmetry. Plast Reconstr Surg. 2012;130:893e–894e; author reply 894e–896e.
17. Yang J, Zhang R, Shen J, Hu Y, Lv Q. The three-dimensional techniques in the objective measurement of breast aesthetics. Aesthetic Plast Surg. 2015;39:910–915.
18. Hudson DA. Factors determining shape and symmetry in immediate breast reconstruction. Ann Plast Surg. 2004;52:15–21.
19. Kovacs L, Zimmermann A, Papadopulos NA, Biemer E. Re: Factors determining shape and symmetry in immediate breast reconstruction. Ann Plast Surg. 2004;53:192–194.
20. Eriksen K, Nordstrand Lindgren E, Olivecrona H, Frisell J, Stark B. Evaluation of volume and shape of breasts: Comparison between traditional and three-dimensional techniques. J Plast Surg Hand Surg. 2011;45:14–22.
21. Ching S, Thoma A, McCabe RE, Antony MM. Measuring outcomes in aesthetic surgery: A comprehensive review of the literature. Plast Reconstr Surg. 2003;111:469–480; discussion 481–482.
22. Hsia HC, Thomson JG. Differences in breast shape preferences between plastic surgeons and patients seeking breast augmentation. Plast Reconstr Surg. 2003;112:312–320; discussion 321–322.
23. Sweeney JA, Luna B, Srinivasagam NM, et al. Eye tracking abnormalities in schizophrenia: Evidence for dysfunction in the frontal eye fields. Biol Psychiatry 1998;44:698–708.
24. Yee E, Blumstein SE, Sedivy JC. Lexical-semantic activation in Broca’s and Wernicke’s aphasia: Evidence from eye movements. J Cogn Neurosci. 2008;20:592–612.
25. Farzin F, Rivera SM, Hessl D. Brief report: Visual processing of faces in individuals with fragile X syndrome: An eye tracking study. J Autism Dev Disord. 2009;39:946–952.
26. Tourassi G, Voisin S, Paquit V, Krupinski E. Investigating the link between radiologists’ gaze, diagnostic decision, and image content. J Am Med Inform Assoc. 2013;20:1067–1075.
27. Bianchi R, Laurent E. Emotional information processing in depression and burnout: An eye-tracking study. Eur Arch Psychiatry Clin Neurosci. 2015;265:27–34.
28. Duque A, Vázquez C. Double attention bias for positive and negative emotional faces in clinical depression: Evidence from an eye-tracking study. J Behav Ther Exp Psychiatry 2015;46:107–114.
29. Alamudun F, Yoon HJ, Hudson KB, Morin-Ducote G, Hammond T, Tourassi GD. Fractal analysis of visual search activity for mass detection during mammographic screening. Med Phys. 2017;44:832–846.
30. Meyer-Marcotty P, Gerdes AB, Stellzig-Eisenhauer A, Alpers GW. Visual face perception of adults with unilateral cleft lip and palate in comparison to controls: An eye-tracking study. Cleft Palate Craniofac J. 2011;48:210–216.
31. Kim MS, Burgess A, Waters AJ, et al. A pilot study on using eye tracking to understand assessment of surgical outcomes from clinical photography. J Digit Imaging 2011;24:778–786.
32. Dixson BJ, Grimshaw GM, Linklater WL, Dixson AF. Eye tracking of men’s preferences for female breast size and areola pigmentation. Arch Sex Behav. 2011;40:51–58.
33. van Schijndel O, Litschel R, Maal TJ, Bergé SJ, Tasman AJ. Eye tracker based study: Perception of faces with a cleft lip and nose deformity. J Craniomaxillofac Surg. 2015;43:1620–1625.
34. Cai LZ, Paro JAM, Lee GK, Nazerali RS. Where do we look? Assessing gaze patterns in breast reconstructive surgery with eye-tracking technology. Plast Reconstr Surg. 2018;141:331e–340e.
35. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol. 1988;124:869–871.
36. Manor BR, Gordon E. Defining the temporal threshold for ocular fixation in free-viewing visuocognitive tasks. J Neurosci Methods 2003;128:85–93.
37. Sütterlin B, Brunner TA, Opwis K. Eye-tracking the cancellation and focus model for preference judgments. J Exp Soc Psychol. 2008;44:904–911.
38. Godoy A, Ishii M, Byrne PJ, Boahene KD, Encarnacion CO, Ishii LE. The straight truth: Measuring observer attention to the crooked nose. Laryngoscope 2011;121:937–941.
39. Litschel R, Majoor J, Tasman AJ. Effect of protruding ears on visual fixation time and perception of personality. JAMA Facial Plast Surg. 2015;17:183–189.
40. Thakerar JN, Iwawaki S. Cross-cultural comparisons in interpersonal attraction of females toward males. J Soc Psychol. 1979;108:121–122.
41. Furnham A, Baguma P. Cross-cultural differences in the evaluation of male and female body shapes. Int J Eat Disord. 1994;15:81–89.
42. Sarwer DB, Grossbart TA, Didie ER. Beauty and society. Semin Cutan Med Surg. 2003;22:79–92.

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2019 by the American Society of Plastic Surgeons