SEARCH

SEARCH BY CITATION

Keywords:

  • Facial paralysis;
  • facial reanimation surgery;
  • smiling;
  • attractiveness;
  • DIBS score;
  • psychosocial outcomes

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY

Objectives/Hypothesis

Determine the effect of facial reanimation surgery on observer-graded attractiveness and negative facial perception of patients with facial paralysis.

Study Design

Randomized controlled experiment.

Methods

Ninety observers viewed images of paralyzed faces, smiling and in repose, before and after reanimation surgery, as well as normal comparison faces. Observers rated the attractiveness of each face and characterized the paralyzed faces by rating severity, disfigured/bothersome, and importance to repair. Iterated factor analysis indicated these highly correlated variables measure a common domain, so they were combined to create the disfigured, important to repair, bothersome, severity (DIBS) factor score. Mixed effects linear regression determined the effect of facial reanimation surgery on attractiveness and DIBS score.

Results

Facial paralysis induces an attractiveness penalty of 2.51 on a 10-point scale for faces in repose and 3.38 for smiling faces. Mixed effects linear regression showed that reanimation surgery improved attractiveness for faces both in repose and smiling by 0.84 (95% confidence interval [CI]: 0.67, 1.01) and 1.24 (95% CI: 1.07, 1.42) respectively. Planned hypothesis tests confirmed statistically significant differences in attractiveness ratings between postoperative and normal faces, indicating attractiveness was not completely normalized. Regression analysis also showed that reanimation surgery decreased DIBS by 0.807 (95% CI: 0.704, 0.911) for faces in repose and 0.989 (95% CI: 0.886, 1.093), an entire standard deviation, for smiling faces.

Conclusions

Facial reanimation surgery increases attractiveness and decreases negative facial perception of patients with facial paralysis. These data emphasize the need to optimize reanimation surgery to restore not only function, but also symmetry and cosmesis to improve facial perception and patient quality of life. Laryngoscope, 124:84–90, 2014


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY

The face is the focal point of most social interactions, and plays a crucial role in communication.[1-4] Among the many facial attributes that influence complex social interactions is facial attractiveness. In this article, we study the observable, physical aspects of attractiveness, something that humankind has attempted to define for ages. Although the defining aspects of facial attractiveness remain under debate, four core components are identified throughout the literature: symmetry, averageness, youthfulness, and sexual dimorphism.[5-9]

In our study, the importance of facial attractiveness lies in its role as an influential component of social interaction and quality of life.[10-18] In general, attractive individuals are more likely to have higher self-esteem, achieve greater academic and occupational satisfaction, have more fulfilling relationships, and have a higher quality of life.[19] Given the social value of facial attractiveness and the amount of attention focused on the human face, any disfigurement of the face can have grave consequences.

Patients with facial paralysis experience diminished facial movements and facial asymmetry, which penalize two key components of facial attractiveness: symmetry and averageness. A 2012 study by Ishii et al. found that observers rate paralyzed faces 1 standard deviation less attractive than normal faces.[20] Although the disfigurement is physical, it is often the psychosocial sequelae that are most detrimental.[10-15]

Facial reanimation surgery is the main treatment option for chronic facial paralysis. The umbrella term facial reanimation surgery encompasses a wide array of reconstructive procedures united by a common goal of restoring function, symmetry, and cosmesis. Despite the established importance of facial attractiveness, data are lacking on the impact of facial reanimation surgery on facial attractiveness. The objective of the current study was to assess the efficacy of reanimation surgery to improve observer-graded attractiveness and decrease negative facial perception of patients with facial paralysis. Negative facial perception was measured by our disfigured, important to repair, bothersome, severity (DIBS) score to more broadly characterize the social impact of facial paralysis.

Data from a previous study established that faces with deformity are considered more disfigured and bothersome, properties highly correlated with observer-rated severity of deformity and importance to repair the deformity.[21] Observer ratings of disfigured and bothersome were absent in normal faces, indicating that faces with deformity are characterized by different perceptive domains, a finding supported by objective data showing that observers look differently on faces with deformity.[22-24] Considering these findings, we first hypothesized that reanimation surgery will significantly lower observer-graded negative facial perceptive measures, represented by the DIBS score. Our second main hypothesis was that reanimation surgery improves observer-rated attractiveness. We subhypothesized that despite this improvement, reanimation surgery is unable to restore attractiveness to normal levels.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY

Participants

Our Johns Hopkins Medicine Institutional Review Board-approved study recruited 122 naïve observers; 32 were lost to follow-up. We excluded from recruitment: 1) those individuals younger than 18 years, or 2) those reporting to have an affective psychiatric condition (schizophrenia, autism, or related spectrum disorders) due to established differences in the way individuals with those disorders direct attention toward a face.[25] The survey was administered online to participants from around the United States to capture a heterogeneous and representative demographic.

Instrument

We created four surveys using images of paralyzed faces before and after reanimation surgery as well as “normal” faces without paralysis or other obvious deformity. Images were selected from our clinical archive if they met the following inclusion criteria: 1) chronic unilateral facial paralysis, 2) preoperative House-Brackmann grade between IV and VI, 3) undergone facial reanimation surgery, and 4) consented to have their pictures used in research. Table 1 lists the etiologies of facial paralysis in our study population.

Table 1. Patient Facial Paralysis Etiologya
 % (No.)
  1. a

    Characterization of the primary etiology of facial paralysis for each of the 20 paralysis patients whose photographs were used in the study.

Vestibular schwannoma60 (12)
Facial schwannoma5 (1)
Meningioma5 (1)
Cavernous angioma5 (1)
Bell's palsy5 (1)
Parotid carcinoma10 (2)
Submandibular adenoid cystic carcinoma5 (1)
Cutaneous squamous cell carcinoma5 (1)

In this pilot study, we did not analyze the efficacy of individual facial reconstructive techniques. Given the numerous techniques, it was not practical to select individual procedures for investigation until we determined if reanimation surgery in general impacted the outcome measures. All but two of the patients in our study underwent procedures to reanimate the upper and lower divisions of the facial nerve; more specifically, seven patients had static slings, 10 patients underwent temporalis tendon transfer, seven patients XII-VII cross-face nerve grafting, and 12 endoscopic brow lift.

After compiling our image database, 20 facial paralysis patients were randomly selected. Eight normal images were selected to demographically match the paralysis patients for age, gender, and race (Table 2). As the primary aim was to assess the efficacy of reanimation surgery, each paralysis patient served as his/her own control. Normal faces were used as a metric for comparison to postoperative faces to analyze the ability of reanimation surgery to restore attractiveness to the degree seen in normal faces.

Table 2. Patient Demographicsa
 Facial Paralysis Patients (n =20)Normal Patients (n = 8)
  1. a

    Demographic information for the 20 facial paralysis patients and eight normal patients whose photographs were used in the study.

Female55%62.5%
Male45%37.5%
Age, yr53 ± 1343 ± 16
Caucasian80%62.5%
Asian15%25%
African American5%12.5%

For each paralysis patient, four images were included: preoperative repose, preoperative smile, postoperative repose, and postoperative smile. For each normal patient, two images were selected: repose and smile. To limit repeated measures effects, these images were evenly and randomly distributed into four mutually exclusive surveys such that: 1) no survey contained more than one image of the same patient, 2) each survey contained one image of every facial paralysis patient, and 3) each survey contained the same number of images in each of the four paralysis groups. To limit observer fatigue, these four mutually exclusive surveys were split in half to create eight paired surveys (1a/1b, 2a/2b, 3a/3b, 4a/4b) designed such that observers would complete half of a survey (i.e., 1a), and then after 2 weeks, they received a link to the second half of the survey (i.e., 1b).

For each picture in the survey, observers used a normalized slider bar to rate attractiveness from 0 to 10 (0 = least attractive and 10 = most attractive). For the facial paralysis images, observers rated paralysis severity on a scale from 0 to 10 (0 = no paralysis and 10 = the most severe paralysis), how disfigured and bothersome they found the face on a scale from 0 to 10 (0 = not disfigured and bothersome and 10 = the most disfigured and bothersome), and how important it is to repair the paralysis on a scale from 0 to 10 (0 = no need for repair and 10 = greatest need for repair).

Procedure

Observers were randomized to complete one of the surveys using a Latin square randomization algorithm. After obtaining consent, observers completed a brief demographic questionnaire to control for potential covariant factors. Two weeks after completing the first survey, observers were sent the second half of their survey pair, thereby creating a mutually exclusive dataset.

Data Analysis

Data were collected and stored using the survey and database functions of Research Electronic Data Capture (REDCap; http://www.project-redcap.org) and were analyzed using Stata 12 SE (Stata Corp., College Station, TX).[26]

Observer-graded scores: attractiveness, severity, disfigured/bothersome, important to repair

The means and standard deviations of the four observer-graded variables (attractiveness, paralysis severity, disfigured/bothersome, and important to repair) were tabulated for each of the six study groups (preoperative repose, postoperative repose, normal repose, preoperative smiling, postoperative smiling, and normal smiling). The four variables were checked for correlation. The severity, disfigured/bothersome, and important to repair variables were highly correlated to each other, and negatively correlated with the attractiveness score; all data with P < .001. Principal factor analysis confirmed this, and the three were combined into a unified DIBS (disfigured, important to repair, bothersome, severity) factor for the remainder of the analysis. We measured the intraclass correlation coefficient for the DIBS score using a test-retest experiment and found it equal to 0.904, suggesting it is highly repeatable. We also tested the validity of our survey and the DIBS score. First, we applied the Delphi method to our survey to develop a standard DIBS score for each image. We then regressed the DIBS scores from 24 lay observers who completed the same survey against the standard DIBS scores and found the slope and intercept equal to 0.752 (95% confidence interval [CI]: 0.682, 0.822) and <0.001 (95% CI: −0.185,1.185), respectively. The standard deviation of the intercept was 0.431, confirming that there is variability in observer baselines; however, because the intercept was not significant, lay observers are not biased from experts. We also found that as paralysis was judged to be more severe, disfiguring/bothersome, and important to repair, the observer-graded DIBS score increased, suggesting we are able to measure this domain.

Mixed effects linear regression: impact of reanimation surgery and smiling

Two multilevel mixed effects linear regression models were created to determine the impact of reanimation surgery and smiling on attractiveness and DIBS score, respectively. Mixed effects linear regression was selected to determine the statistical significance of the findings while accounting for multiple levels of random effects in the model. Given that the images and observers were randomly selected, there were two main sources of variation in attractiveness rating. The random intercept term in the model accounted for intrinsic differences in the way observers rate attractiveness, and the random error term accounted for variance in attractiveness naturally occurring in a random population as well as sources of error inherent in regression modeling. Details about constructing the models can be found in the online Appendix (see Supporting Information, Appendix, in the online version of this article). For both attractiveness and DIBS, the best-fit model included two fixed covariates (surgery and smiling) and an interaction term (surgery interacting with smiling).

Attractiveness: difference between postoperative and normal

Planned analysis was conducted to address the subhypothesis that reanimation surgery is unable to restore attractiveness to normal levels. This hypothesis testing was performed using the bootstrap method, with 4,000 samples to provide the mean difference between the attractiveness rating for each postoperative to normal comparison. The experiment-wide α was set at .05. We corrected for multiple comparisons using the Bonferroni method.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY

Demographics

The observer population included 39 men and 51 women, with ages that ranged from 19 to 70 years; the mean age was 31 ± 13. This population was also heterogeneous with regard to race, level of education, and marital status. Of the 122 observers initially enrolled in the study, 32 were removed as they failed to complete their paired surveys, leaving 90 participants whose data were analyzed. Demographic heterogeneity did not significantly change with this exclusion.

Observer-graded scores: attractiveness, severity, disfigured/bothersome, important to repair

The mean values and standard deviations for each observer-rated variable are shown in Table 3. There was a highly positive correlation between three variables (with correlation coefficients): paralysis severity (0.924), disfigured/bothersome (0.901), and important to repair (0.877), indicating they measure a similar domain (P < .001). The three variables were negatively correlated with attractiveness.

Table 3. Observer Graded Scores for Each Study Groupa
GroupAttractivenessSeverityDisfigured/BothersomeImportant to RepairDIBS
  1. a

    Mean values and standard deviations for each observer-rated variable as well as the DIBS factor score for each of the six study groups.

  2. A negative DIBS score means the observers rated the face as less severe, disfigured/bothersome, and important to repair; a positive DIBS score means the observers rated the faces as more severe, disfigured/bothersome, and important to repair. These findings provide an overview of the effects of reanimation surgery and smiling on each observer-graded variable. The statistical significance of these data is supported by mixed effects linear regression and statistical hypothesis testing using the bootstrap method.

  3. Standard deviations are in parentheses.

  4. DIBS = disfigured, important to repair, bothersome, severity factor score; N/A = not applicable.

Repose     
Preoperative3.93 (1.66)4.61 (2.49)4.05(2.45)4.48 (2.77)0.106 (0.915)
Postoperative4.77 (1.57)2.43 (2.34)2.00 (2.09)2.58 (2.79)−0.701 (0.863)
Normal6.44 (1.42)N/AN/AN/AN/A
Smile     
Preoperative3.50 (1.80)6.39 (2.36)5.66 (2.55)5.98 (2.64)0.753 (0.890)
Postoperative4.74 (1.64)3.72 (2.50)3.13 (2.31)3.66 (2.71)−0.236 (0.905)
Normal6.88 (1.49)N/AN/AN/AN/A

Subsequent iterated principal factor analysis supported the positive correlation between them, and combined the three variables into a single normalized factor that accounted for nearly 100% of total variance. The factor loadings showed that each variable was well represented by the single factor with a P < .001, a finding supported by scree testing. This factor score was defined as the DIBS score, a unified measure representing the domain measured by the correlated variables.

The data in Table 3 compare preoperative, postoperative, and normal faces, smiling and in repose. There is an attractiveness penalty of 2.51 on a 10-point scale (1.5 standard deviations) for paralyzed faces in repose, as supported by a previous study.[20] The penalty for paralyzed smiling faces was 3.38, over 2 standard deviations. For faces in repose, reanimation surgery improved attractiveness by 0.84; however, the postoperative faces in repose remained 1.67 or 1 standard deviation less attractive than comparable normal faces. For paralyzed smiling faces, reanimation surgery improved attractiveness by 1.24. However, yet again, comparing postoperative smiling to normal smiling faces, there was a residual 2.14 difference. Figure 1 portrays the relationship between these groups.

image

Figure 1. Boxplot of attractiveness ratings for each of the six study groups. These data help support the conclusion that on average, for faces smiling and in repose, reanimation surgery improves attractiveness but does not restore it to a normal level. PostOp = postoperative; PreOp = preoperative; R = repose; S = smile. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

As indicated by Table 3, reanimation surgery decreased severity ratings for both repose (2.18, 1 standard deviation) and smile (2.67, more than 1 standard deviation). The disfigured/bothersome and important to repair scores followed suit, a finding supported by the factor analysis. By convention, the DIBS score has a mean of 0 and a standard deviation of 1. A negative DIBS score means the observers rated the face as less severe, disfigured/bothersome, and important to repair; a positive DIBS score means the observers rated the faces as more severe, disfigured/bothersome, and important to repair. Reanimation surgery decreased DIBS by 0.807 for faces in repose and by 0.986, a standard deviation, for smiling faces. These data are graphically displayed in Figure 2.

image

Figure 2. The DIBS factor score has a mean of 0 and a standard deviation of 1. Data from our mixed effects linear regression DIBS model are used to show that reanimation surgery decreases the DIBS score, our proxy for negative facial perception. Error bars represent the 95% confidence intervals. These data indicate that reanimation surgery decreased DIBS by about one standard deviation for paralyzed faces smiling and in repose. These data also show that for a paralyzed face, smiling increases negative perception as measured by DIBS. DIBS = disfigured, important to repair, bothersome, severity factor score; PostOp = postoperative; PreOp = preoperative. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Mixed effects linear regression: impact of reanimation surgery and smiling

Data from the mixed effects linear regression support the data in Table 3. The entire data output from the regression models can be found in the online Appendix (see Supporting Information, Appendix, in the online version of this article). The regression showed that all data in the model were statistically significant as defined by a P value < .05 and 95% CIs that did not include 0. To summarize the key findings from the attractiveness regression: reanimation surgery increased attractiveness of paralyzed faces by 0.84 (95% CI: 0.67, 1.01), smiling decreased attractiveness by 0.43 (95% CI: 0.26, 0.61), and the interaction term (surgery and smiling) increased attractiveness 0.40 (95% CI: 0.16, 0.65). Figure 3 illustrates this impact of reanimation surgery on attractiveness of paralyzed faces with comparison to normal faces.

image

Figure 3. Data from the mixed effects linear regression attractiveness model illustrate the impact of reanimation surgery on attractiveness of paralyzed faces with comparison to normal faces. Error bars represent the 95% confidence intervals. These data corroborate the conclusion that despite improvement, reanimation surgery does not normalize facial attractiveness, a conclusion supported by statistical hypothesis testing. PostOp = postoperative; PreOp = preoperative. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

A mixed effects linear regression model was also created to test the effects of surgery and smiling on the DIBS score. A complete output from this regression can be found in the appendix, but we summarize that reanimation surgery decreased DIBS by 0.807 (95% CI: 0.704, 0.911), smiling increased DIBS 0.647 (P < .001) (95% CI: 0.544, 0.751), and the interaction (surgery and smiling) decreased DIBS 0.182 (P = .015) (95% CI: 0.036, 0.329).

Attractiveness: difference between postoperative and normal

Planned hypothesis testing was performed to determine if reanimation surgery was able to statistically significantly improve facial attractiveness to a normal level. The bootstrap data are presented in Table 4, and show for both faces in repose and smiling there are statistically significant differences between the attractiveness rating of postoperative versus normal faces.

Table 4. Hypothesis Test of Attractiveness Ratings of Postoperative Versus Normal Facesa
ComparisonMean AttractivenessMean Difference95% Confidence Interval for Difference
PostoperativeNormal
  1. a

    Mean attractiveness ratings of postoperative and normal faces both smiling and in repose.

  2. To test the ability of reanimation surgery to normalize facial attractiveness, we performed statistical hypothesis testing, using the bootstrap method, to provide the mean difference between the attractiveness ratings for each postoperative to normal comparison with the adjusted confidence intervals displayed. The confidence intervals were corrected for multiple comparisons and were considered statistically significant (†) if they did not contain 0. These data demonstrate that for both faces in repose and smiling, there are statistically significant differences in attractiveness ratings between postoperative and normal faces.

Repose4.776.441.671.46-1.88†
Smile4.746.882.141.92-2.36†

DISCUSSION

In 2012, Ishii et al. showed that facial paralysis induces a significant attractiveness penalty.[20] The present study sought to assess the effect of reanimation surgery on observer-graded attractiveness as well as negative facial perceptive measures. The data herein provide evidence that reanimation surgery both improves attractiveness and markedly decreases negative facial perception.

Reanimation surgery decreased DIBS, a facial perception measure, by 0.807 for faces in repose and 0.986 (1 standard deviation) for smiling faces. These data represent an important decrease in negative facial perception, which almost completely eliminates negative perception in postoperative faces. The data further support the second main hypothesis and demonstrate that reanimation surgery improves attractiveness for paralyzed faces both smiling and in repose by 1.24 and 0.84 points, respectively. The data also support the subhypothesis illustrating a persistent difference in postoperative and normal attractiveness ratings.

Although reanimation surgery improved attractiveness, in this study it was unable to restore attractiveness to the degree seen in normal faces. To explain this imperfect improvement, we return to the four core components of attractiveness. Reanimation attempts to restore symmetry and make the face more normal or “average,” which may be the reason for the increase in attractiveness ratings. However, as appreciated in Figure 4, although reanimation surgery improves facial symmetry in both repose and smile, there is incomplete resolution, which may account for the residual postoperative attractiveness deficit.

image

Figure 4. Representative image set from a facial paralysis patient used in the study. This set represents the four images used for each paralysis patient: (A) preoperative repose, (B) postoperative repose, (C) preoperative smile, (D) postoperative smile. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Other data from our study generated additional questions for reflection. Our regression data show that when normal faces smile, it increases attractiveness by 0.44, but when paralyzed faces smile, attractiveness decreases 0.43. What can explain this phenomenon? Several studies support the finding that for normal faces, smiling increases attractiveness.[20, 27-29] However, studies show that even normal smiles are asymmetric and therefore may decrease attractiveness.[30] This smiling paradox may be answered by studies that indicate the degree of asymmetry in the average normal smile is not detected by casual observers. Thus, smiling may increase perceived facial symmetry and attractiveness ratings.[31] Another idea is that there is an underlying mechanism by which smiling faces are interpreted as more attractive despite their asymmetries. Smiling is associated with positive affect, and attractiveness is a positive attribute. Therefore, there may be a latent connection between positive affect and observer-graded attractiveness. In essence, positivity is attractive. Despite the lack of definitive answers, we conclude that the smile is a powerful modulator of facial attractiveness. Depending on its quality, a smile can increase or decrease attractiveness. These findings emphasize the importance of optimizing smile reconstruction in patients with facial paralysis.

The implications of these data are profound. For patients with facial paralysis, reanimation surgery may improve attractiveness and decrease negative facial perception. Attractiveness is an influential component of social interaction and quality of life, so these improvements have the potential to decrease the negative psychological and social outcomes associated with facial paralysis. These benefits may extend beyond the patient, as depressive disorders may be associated with higher medical costs and great societal costs due to loss of productivity. These data should inform clinicians of the importance of restoring aesthetics and function for facial paralysis.

A major limitation of this study was the inability to determine the efficacy of individual reanimation procedures. Given the myriad reconstructive methods, it was impractical in this pilot study to investigate the efficacy of individual procedures. We first wanted to see if reanimation surgery in general had a statistically significant impact on our measures. Future research will be focused on determining which procedures are most efficacious.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY

Facial reanimation surgery increased attractiveness and decreased negative facial perception of patients with facial paralysis. These findings emphasize the importance of optimizing facial aesthetics as a component of a successful reanimation surgery protocol.

BIBLIOGRAPHY

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. CONCLUSION
  7. BIBLIOGRAPHY
  • 1
    Jones RM, Victor JD, Conte MM. Detecting symmetry and faces: separating the tasks and identifying their interactions. Atten Percept Psychophys 2012;74:9881000.
  • 2
    Solbakken OA, Hansen RS, Havik OE, Monsen JT. Assessment of affect integration: validation of the affect consciousness construct. J Pers Assess 2011;93:257265.
  • 3
    Falkenberg I, Bartels M, Wild B. Keep smiling! Facial reactions to emotional stimuli and their relationship to emotional contagion in patients with schizophrenia. Eur Arch Psychiatry Clin Neurosci 2008;258:245253.
  • 4
    Hatfield E, Cacioppo JT, Rapson RL. Emotional Contagion. Cambridge, UK: Cambridge University Press; 1994.
  • 5
    Etcoff N. Survival of the Prettiest: The Science of Beauty. New York, NY: Doubleday; 2002.
  • 6
    Rhodes G, Yoshikawa S, Clark A, Lee K, McKay R, Akamatsu S. Attractiveness of facial averageness and symmetry in non-western cultures: in search of biologically based standards of beauty. Perception 2001;30:611625.
  • 7
    Perrett DI, Lee KJ, Penton-Voak I, et al. Effects of sexual dimorphism on facial attractiveness. Nature 1998;394:884887.
  • 8
    Perrett DI, Burt DM, Penton-Voak IS, Lee KJ, Rowland DA, Edwards R. Symmetry and human facial attractiveness. Evol Hum Behav 1999:295307.
  • 9
    Little AC, Jones BC, DeBruine LM. Facial attractiveness: evolutionary based research. Philos Trans R Soc Lond B Biol Sci 2011;366:16381659.
  • 10
    Valente SM. Visual disfigurement and depression. Plast Surg Nurs 2009;29:1016; quiz 17–18.
  • 11
    Noles SW, Cash TF, Winstead BA. Body image, physical attractiveness, and depression. J Consult Clin Psychol 1985;53:8894.
  • 12
    Levine E, Degutis L, Pruzinsky T, Shin J, Persing JA. Quality of life and facial trauma: psychological and body image effects. Ann Plast Surg 2005;54:502510.
  • 13
    Orr DA, Reznikoff M, Smith GM. Body image, self-esteem, and depression in burn-injured adolescents and young adults. J Burn Care Rehabil 1989;10:454461.
  • 14
    McGrouther DA. Facial disfigurement. BMJ 1997;314:991.
  • 15
    Topolski TD, Edwards TC, Patrick DL. Quality of life: how do adolescents with facial differences compare with other adolescents? Cleft Palate Craniofac J 2005;42:2532.
  • 16
    Connors J, Casey P. Sex, body-esteem and self-esteem. Psychol Rep 2006;98:699704.
  • 17
    Bull R, Rumsey N. The Social Psychology of Facial Appearance. New York, NY: Springer-Verlag; 1988.
  • 18
    Patrick H, Neighbors C, Knee CR. Appearance-related social comparisons: the role of contingent self-esteem and self-perceptions of attractiveness. Pers Soc Psychol Bull 2004;30:501514.
  • 19
    Jackson LA. Physical attractiveness: a sociocultural perspective. In: Cash TF, Pruzinsky T, eds. Body Image: A Handbook of Theory, Research, and Clinical Practice. New York, NY: Guilford Press, 2002:1321.
  • 20
    Ishii L, Godoy A, Encarnacion CO, Byrne PJ, Boahene KD, Ishii M. Not just another face in the crowd: society's perceptions of facial paralysis. Laryngoscope 2012;122:533538.
  • 21
    Godoy A, Ishii M, Byrne PJ, Boahene KD, Encarnacion CO, Ishii LE. How facial lesions impact attractiveness and perception: differential effects of size and location. Laryngoscope 2011;121:25422547.
  • 22
    Ishii L, Carey J, Byrne P, Zee DS, Ishii M. Measuring attentional bias to peripheral facial deformities. Laryngoscope 2009;119:459465.
  • 23
    Chu EA, Farrag TY, Ishii LE, Byrne PJ. Threshold of visual perception of facial asymmetry in a facial paralysis model. Arch Facial Plast Surg 2011;13:1419.
  • 24
    Godoy A, Ishii M, Byrne PJ, Boahene KD, Encarnacion CO, Ishii LE. The straight truth: measuring observer attention to the crooked nose. Laryngoscope 2011;121:937941.
  • 25
    Nakano T, Tanaka K, Endo Y, et al. Atypical gaze patterns in children and adults with autism spectrum disorders dissociated from developmental changes in gaze behaviour. Proc Biol Sci. 2010;277:29352943.
  • 26
    Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377381.
  • 27
    Magda LA, Goodwin KA. Consequences of negative information on perceptions of facial attractiveness. Percept Mot Skills 2008;106:508516.
  • 28
    Reis HT, Wilson IM, Monestere C, et al. What is smiling is beautiful and good. Eur J Soc Psychol 2012;20:259267.
  • 29
    Beall AE. Can a new smile make you look more intelligent and successful? Dental clinics of North America 2007;51:289297, vii.
  • 30
    Nagy E. From symmetry to asymmetry? The development of smile. Cortex 2012;48:10641067.
  • 31
    Pinho S, Ciriaco C, Faber J, Lenza MA. Impact of dental asymmetries on the perception of smile esthetics. Am J Orthod Dentofacial Orthop 2007;132:748753.