- Email: [email protected]
More than just skin-deep? A pilot study integrating physical and non-physical factors in the perception of physical attractiveness
Personality and Individual Differences 42 (2007) 563–572 www.elsevier.com/locate/paid
More than just skin-deep? A pilot study integrating physical and non-physical factors in the perception of physical attractivene Viren Swami a
a,*
, Corina Greven b, Adrian Furnham
b
Division of Public Health, University of Liverpool, Whelan Building, Qudrangle, Brownlow Hill, Liverpool L69 3GB, UK b Department of Psychology, University College London, London, UK Received 24 March 2006; received in revised form 28 June 2006 Available online 5 October 2006
Abstract The perceived beauty of a potential partner has traditionally been explained in terms of physical characteristics such as body weight and the waist-to-hip ratio (WHR), while non-physical characteristics have been neglected. This study looked at the contribution of body weight, WHR and personality to ratings of overall female attractiveness. Seventy-six participants rated 30 line drawings that varied 2 levels of personality (introverted, extraverted), 3 levels of body weight and 5 of WHR. Effect sizes were largest for body weight (0.58), followed by personality (0.25) and finally WHR (0.10). There were significant interactions between all three variables, with effect sizes ranging between (0.3 and 0.4). These results highlight the importance of examining non-physical characteristics when studying perceptions of human beauty, a finding consistent with the notion that beauty is more than just skin-deep. 2006 Elsevier Ltd. All rights reserved. Keywords: Physical attractiveness; Non-physical attractiveness; Body weight; Waist-to-hip ratio; Personality
*
Corresponding author. Tel.: +44 (0)151 794 5270; fax: +44 (0)151 794 5588. E-mail address: [email protected] (V. Swami).
0191-8869/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2006.08.004
564
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
1. Introduction The topic of human beauty continues to be one that attracts considerable debate and controversy both within psychological and popular circles (e.g., Swami, in press; Swami & Furnham, 2006). In the past decade, a productive field of research has examined perceptions of male and female bodily attractiveness (e.g., Singh, 1993; Tove´e, Maisey, Emery, & Cornelissen, 1998; Tove´e, Hancock, Mahmoodi, Singleton, & Cornelissen, 2002; Wilson, Tripp, & Boland, 2005), with some recent studies taking a cross-cultural perspective (e.g., Sugiyama, 2004; Swami & Tove´e, 2005a; Swami, Antonakopoulos, Tove´e, & Furnham, 2006; Swami, Caprario, Tove´e, & Furnham, 2006). These studies have typically examined the relative contribution of body shape and weight measures to ratings of female physical attractiveness. In one of the earliest studies of physical attractiveness, Singh (1993) argued that the waist-tohip ratio (WHR), a measure of body shape, is a better predictor of female attractiveness judgements than overall body weight. He presented evidence showing that a low WHR is associated with higher levels of fertility, and went on to argue that the WHR acts a first-pass filter, excluding women who are unhealthy or have low reproductive capability. Evidence for this has been reported in a number of countries using line-drawn stimuli developed by Singh (e.g., Furnham, Tan, & McManus, 1997). However, some investigators have argued that such findings are an artefact of the experimental design. The line drawings and photographic stimuli used in these studies appeared to confound body weight and shape measures. When researchers have investigated the relative importance of body shape and weight (the latter quantified as the body mass index, or BMI) using photographs of real women (e.g., Puhl & Boland, 2001; Tove´e, Maisey, et al., 1998, 2002; Tove´e & Cornelissen, 2001) and three-dimensional images (Fan, Liu, Wu, & Dai, 2004), variation in BMI appears to be the greater predictor of female attractiveness than WHR. These effects cannot simply be explained by the stimuli not adequately capturing shape cues, as when pictures of men are used in the same format, attractiveness is typically determined by shape cues rather than BMI (Maisey, Vale, Cornelissen, & Tove´e, 1999; Swami & Tove´e, 2005b), thus demonstrating that shape cues are salient in this format. These results also hold when cross-cultural data sets are examined (e.g., Swami & Tove´e, 2005a; Swami, Antonakopoulos, et al., 2006; Swami, Caprario, et al., 2006), and psychosocial explanations for these have been postulated (e.g., Swami & Tove´e, 2006). Of course, the relative ranges of BMI and WHR in these studies are unequal, and the WHR seems to be a more important predictor of attractiveness within a normal or average BMI range (cf. Streeter & McBurney, 2003; Wilson et al., 2005). More recent conceptualisations of the WHR hypothesis of attractiveness have also shifted away from the notion of it being a first-pass filter (e.g., Marlowe & Wetsman, 2001; Marlowe, Apicella, & Reed, 2005). Sugiyama (2004), for example, has presented evidence of a male preference for WHRs lower-than-the-local-female-average WHR, which takes into account the effect of parasitic loads and nutrition in increasing or decreasing local WHRs ranges. Within this paradigm, the WHR is one of several factors influencing physical attractiveness, with such factors as body weight, skin colour and facial attractiveness all also playing major roles. The theoretical background behind most such studies stems from the evolutionary psychological idea that human physical attractiveness should be seen as an assessment of fitness value, indexed by such factors as fecundity or health (Buss, 1994, 1999). Within this view, the most
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
565
fitness-enhancing trait (for example, a curvaceous female body) is perceived as the most attractive and individuals are drawn to what is regarded as most attractive by definition. Typically, however, researchers have assumed that the assessment of physical attractiveness will be based on purely physical traits, to the exclusion of non-physical factors. When non-physical factors are considered within the evolutionary psychological paradigm, it is usually as an epiphenomenon of physical characteristics. That is, individuals may be attracted to others because of physical attributes (a slender body weight or a curvaceous body shape), with non-physical virtues merely enhancing desirability as a secondary consideration. Kniffin and Wilson (2004), however, have argued that a more rounded evolutionary psychology should allow for assessments of physical attractiveness to be based on overall fitness value, including non-physical traits. Indeed, a large body of social psychological research has emphasised the influence of non-physical characteristics on judgements of physical attractiveness (e.g., Feingold, 1992; Felson & Bohrenstedt, 1979; Gross & Crofton, 1977; Nisbett & Wilson, 1977). More recent studies inspired by evolutionary psychology show that social status (Townsend & Levy, 1990) and prosocial orientation (Jensen-Campbell, Graziano, & West, 1995) enhance perceptions of physical attractiveness. In their study, Kniffin and Wilson (2004) also demonstrated that non-physical factors have a very potent effect in the perception of physical attractiveness. In general, women were more strongly influenced by non-physical factors than men and there were also large individual differences within each gender. These studies highlight the importance of examining non-physical factors in research questions (Kniffin & Wilson, 2004). To this end, we designed a simple experimental study to compare the relative effects of body weight, WHR and personality, the latter acting as a basic non-physical attribute. The study used a selection of modified line drawings reported in Furnham, Swami, and Shah (2006), with the final set combining three body weight categories, five WHR levels, and two levels of personality. This manipulation allowed us to investigate the contribution of a target’s personality on judgements of physical attractiveness.
2. Methods 2.1. Participants A total of 76 British undergraduate students, all of whom were males, participated in the study. The mean age of the sample was 23.28 (SD = 7.93). Only participants who self-reported as heterosexual were invited to participate in the study. Just over 80% of the sample were British Caucasians, while the rest of the sample were of either Asian (11.8%), Afro-Carribean (3.9%) or other (3.9%) descent. Participant ethnicity did not have a significant effect in subsequent ratings. The socio-economic and educational backgrounds of the participants were fairly homogenous. All participants were naı¨ve to the aims of the study and participated on a voluntary basis. 2.2. Stimuli and procedure The stimuli used in this study were selected and modified from Furnham et al. (2006). To begin with, we selected 15 images that varied by three levels of body weight (underweight, average
566
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
weight and overweight) and five levels of WHR (0.6, 0.7, 0.8, 0.9, and 1.0). The ranges of these variables are in line with those used in previous studies. The arms and legs were narrowed or thickened within each weight category, and all other facial and bodily features (e.g., breast size) were kept constant. In addition, the stimuli had originally been designed to be ethnically ambiguous. These stimuli were then given the designatory symbol ‘A,’ indicated at the bottom-right corner of the image. A second set of identical stimuli were given the designatory symbol ‘B.’ Participants were told that they would be shown different images of a woman named Rozie and that the images depicted Rozie with different physical characteristics. In addition, Rozie could have two different types of personality, indicated by the letter (A or B) at the bottom of the image. The characteristics associated with each personality were Personality A (Extravert): Rozie is friendly and likes to be around others. She is actionoriented, is a ‘people person,’ and often feels energetic; she is open and shows her emotions easily, but usually acts before thinking. Overall, Rozie has a cheerful outlook, reacts to things spontaneously and has an entertaining curiosity about most things. Personality B (Introvert): Rozie is shy and quiet, and likes to spend time alone. She likes ideas and concepts, and prefers solitude for concentration; she is thoughtful and keeps her emotions private, but usually thinks before acting. Overall, Rozie is highly analytical and precise in her thinking and tends to shy away from social interactions. These descriptors were designed based on characteristics differentiating individuals on an introversion–extraversion scale (e.g., Costa & McCrae, 1992). While the use of symbols (A or B) to indicate personality type may have been overly simple, manipulating other aspects of the stimuli (e.g., clothing or hairstyle) to indicate personality type would have introduced possible confounds. This would also have been true had we used two different individuals with different personalities. In a pilot study (n = 18), we found that participants were able to make judgements based on the use of symbols to designate personality. Further, both descriptors were in reading view of participants throughout the experiment, and participants were told to consider both physical and nonphysical factors in making their judgements. The task was conducted by a female experimenter trained in the methods of the study. However, for a subset of the participants (n = 16), the task was conducted by a male experimenter. There were no significant differences in the responses to male and female experimenters (p > 0.10). For each of the 30 images, participants were asked to provide a rating for overall attractiveness, that is, how attractive they thought the image of Rozie was in physical and nonphysical terms. Each image was presented on sheets of card measuring 210 · 297 mm and remained in view for 45 s. Participants were provided with a booklet which included brief instructions and on which they made their ratings according to a 9-point Likert scale (9 = very attractive). All participants completed the task individually, and were later debriefed.
3. Results A 2 · 5 · 3 repeated measures analysis of variance (ANOVA) with 76 participants was computed. Body weight, WHR and personality were treated as within subject factors. Where a factor had more than two levels, and the Mauchly’s Test of Sphericity was shown to be significant, the
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
567
Greenhouse–Geisser correction was applied. A summary of the ANOVA results and the main effects of WHR, body weight and personality, and their interactions are shown in Table 1. The ANOVA revealed that body weight, WHR and personality all had significant main effects on the ratings of the figures. In general, extraverts were preferred over introverts, under- and normal weight figures were preferred over overweight figures, and lower WHRs were preferred over higher WHRs (see Fig. 1). However, the effects sizes revealed body weight accounted for more of Table 1 ANOVA results with main effects of WHR, body weight and personality, and their interactions Source
Df
Body weight Personality Waist-to-hip ratio Personality · body weight Personality · WHR Waist-to-hip ratio · body weight Personality · WHR · body weight
102.52 25.40** 8.34** 3.85* 2.69* 2.47* 3.07*
0.58 0.25 0.10 0.04 0.03 0.03 0.04
Greenhouse–Geisser corrected. p < 0.05. p < 0.001.
5.4
6.0
5.2
5.5
5.0 Mean ratings
**
Effect size **
1.66, 124.26 1, 75 2.56, 191.62a 1.77, 132.69a 3.33, 250.03a 6.28, 471.30a 6.10, 457.33
5.0
4.5
4.8
4.6 4.0
4.4
3.5 Under weight
Average weight
4.2 Extroverted
Over weight
Introverted
5.1 5.0 4.9 Mean ratings
*
Mean ratings
a
F a
4.8 4.7 4.6 4.5 4.4 0.6
0.7
0.8
0.9
1.0
Fig. 1. Graphs of the main effects for body weight, personality and waist-to-hip ratio.
568
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
the variance in the data ðg2p ¼ 0:58Þ than did personality ðg2p ¼ 0:25Þ or WHR ðg2p ¼ 0:10Þ. However, there were also significant interactions which would suggest caution in interpreting the main effects. There were significant two-way interactions of personality · body weight, personality · WHR, and body weight · WHR, as well as a significant three-way interaction between personality · WHR · body weight. The two-way interactions are illustrated in Fig. 2, and the three-way interaction in Fig. 3. From Fig. 3, it can be seen that for extraverted figures, the preferred WHRs for the underweight and overweight figures was 0.6. The preferred WHR for the average weight category was either 0.7 and 0.9. For the introverted figures, the preferred WHR for the underweight category was 0.7, whereas for the average weight category it was 0.8 and for the overweight category it was 0.6. In general, lower WHRs were preferred over higher ratios across categories; moreover, underweight and average weight figures were rated fairly similarly, but overweight figures were
6.5
6.5
Personality
Personality 6.0
6.0
Extraverted
Extraverted 5.5
Introverted
Mean ratings
Mean ratings
5.5
5.0 4.5
5.0
4.5
4.0
4.0
3.5
3.5
3.0
Introverted
3.0 1
2
3
1
2
Body weight
3
4
5
Waist-to-hip ratio
6.5
6.0
Mean ratings
5.5
Body weight
5.0
Under weight 4.5
Average weight 4.0
Over weight 3.5
3.0 1
2
3
4
5
Waist-to-hip ratio
Fig. 2. Graphs of the two-way interactions between personality · WHR, personality · body weight and body weight · WHR.
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572 6.5
6.5
6.0
6.0
Body weight
5.5
Mean ratings
Mean ratings
5.5
Under weight
5.0
Average weight
4.5
569
Over weight
5.0
Body weight 4.5
Under weight
4.0
4.0
3.5
3.5
Average weight
3.0 1
2
3
Waist-to-hip ratio
4
5
3.0 0.6
Over weight
0.7
0.8
0.9
1.0
Waist-to-hip ratio
Fig. 3. Graphs of the three-way interaction between body weight, personality and waist-to-hip ratio.
rated more negatively. The ANOVA also revealed that the body weight and WHR interaction had a significant effect on the ratings of all attributes. However, it is important to note that the effect sizes for these interactions were fairly small, and therefore the significance of these interactions may have been caused by the large number of participants in the study.
4. Discussion This study was designed to address a relative paucity of research examining the effects of non-physical characteristics on judgements of physical attractiveness. Our findings highlight this neglected area of research: non-physical factors, in this case the personality of the target, can have an important influence on judgements of attractiveness both independently and in interactions with body shape and weight. Although our results are not directly comparable with those of Kniffin and Wilson (2004) because of the different methodologies used, they are in general agreement in showing that non-physical factors can affect perceptions of attractiveness. In particular, extraverted or sociable individuals are perceived as more attractive than introverted or quiet individuals. On its own, this is perhaps not a surprising finding; what is more striking, however, is the role of the interaction between personality, body shape and body weight. Kniffin and Wilson (2004) have argued that any assessment of an individual’s physical attractiveness should also include perceptions of non-physical characteristics, which lend themselves to a judgement of overall attractiveness. Although it is conceivable that assessments of attractiveness begin with physical attributes before weighting non-physical characteristics (cf. Singh, 1993), a more likely scenario is that both sets of factors are assessed simultaneously when such information is available. Kniffin and Wilson (2004) argue that in ancestral social environments, interactions probably took place in small groups of people whose physical attributes were roughly average and whose non-physical attributes were intimately known to each other. Although we think that conjecture about supposed past evolutionary environments should be avoided (cf. Buller, 2005),
570
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
such a scenario suggests that non-physical traits may have been important components of interpersonal relations in human evolutionary history. An obvious objection to the methodology used in the present study is that participants were asked to make ratings of relative strangers, of which some information about personality is known. In real life situations, perceptions of strangers do not often include information about extraversion or introversion. Rather, perceptions of strangers are often predicated on first impressions, which primarily involve judgements of non-verbal, physical cues (see Baron, Byrne, & Branscombe, 2006). In such scenarios, it is not inconceivable that judgements of attractiveness should proceed with perceptions of physical rather than non-physical attractiveness. However, at a basic level, all inter-personal relationships must involve some level of social interaction, and here non-physical characteristics may play an important role in determining an individual’s overall attractiveness (Swami, in press). It is this interaction between physical and non-physical factors that may underscore the popular belief that beauty is more than just skin-deep. It should also be noted that the personality descriptions used in the present study were extremes of behaviour, and in reality many individuals combine aspects of introversion and extraversion in their overall personality types (e.g., cheerful and emotionally private). In this respect, we would support Kniffin and Wilson’s (2004) call for studies of physical attractiveness to include both physical and non-physical factors within naturalistic designs. An experimental design such as this is able to manipulate various factors simultaneously, but naturalistic designs are more able to approximate the way perceptions of attractiveness are formed in real life situations. Finally, it is possible that participants in the present study tended to be extraverts, which would explain the overall preference for extraverted others. Studies in social psychology, for example, have shown that an important factor that moderates attractiveness is similarity of attitudes or values (e.g., Newcomb, 1961). At a basic level, people who are evenly matched in their physical appearance, social background and personality are more likely to be attracted to one another. Future studies would therefore do well to include simple personality measures of the participants, to examine the possible effect of similarity of personality on attraction. In spite of these limitations, the results of the present study suggest that the personality of a target may be an important component of physical attractiveness judgements. Indeed, in the present study, personality was found to explain more of the variance in the data than WHR. We believe that these findings are an important starting point for future research. In particular, studies should investigate in greater detail gender differences in the importance of non-physical characteristics, which Kniffin and Wilson (2004) found to be important. Studies could also examine the way in which the personality of the observer influences preferences for personality and physical characteristics in a target individual. Better technology would also be useful: the use of photographic and three-dimensional stimuli would certainly be an improvement on the use of line drawings (Treleaven, Furnham, & Swami, 2006). In conclusion, this study suggests that non-physical attributes have an important effect on the perception of physical attractiveness, which is consistent with more rounded evolutionary psychological theories of mate selection. Like Kniffin and Wilson (2004), our results point to extraverts being more valued social partners than introverts. Of course, the value of introversion in some case should not be disregarded (Storr, 1989), but in general being sociable and outgoing may be just as important as having a particular body weight or shape.
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
571
References Baron, R., Byrne, D., & Branscombe, N. (2006). Social psychology. Boston: Pearson. Buller, D. J. (2005). Adapting minds: evolutionary psychology and the persistent quest for human nature. London: MIT Press. Buss, D. (1994). The evolution of desire. New York: Basic Books. Buss, D. (1999). Evolutionary psychology: the new science of the mind. Boston: Allyn & Bacon. Costa, P. T., & McCrae, R. (1992). Four ways five factors are basic. Personality and Individual Differences, 135, 653–665. Fan, J., Liu, F., Wu, J., & Dai, W. (2004). Visual perception of female physical attractiveness. Proceedings of the Royal Society of London B, 271, 347–352. Feingold, A. (1992). Good-looking people are not what we think. Psychological Bulletin, 111, 304–341. Felson, R. B., & Bohrenstedt, G. W. (1979). ‘Are the good beautiful or the beautiful good?’ The relationship between children’s perceptions of ability and perceptions of physical attractiveness. Social Psychology Quarterly, 42, 386–392. Furnham, A., Swami, V., & Shah, K. (2006). Female body correlates of attractiveness and other ratings. Personality and Individual Differences, 41, 443–454. Furnham, A., Tan, T., & McManus, C. (1997). Waist-to-hip ratio and preferences for body shape: a replication and extension. Personality and Individual Differences, 22, 539–549. Gross, A. E., & Crofton, C. (1977). What is good is beautiful. Sociometry, 40, 85–90. Jensen-Campbell, L. A., Graziano, W. G., & West, S. G. (1995). Dominance, prosocial orientation and female preferences: do nice guys really finish last? Journal of Personality and Social Psychology, 68, 427–440. Kniffin, K. M., & Wilson, D. S. (2004). The effect of non-physical traits on the perception of physical attractiveness: three naturalistic studies. Evolution and Human Behavior, 25, 88–101. Maisey, D. M., Vale, E. L. E., Cornelissen, P. L., & Tove´e, M. J. (1999). Characteristics of male attractiveness for women. Lancet, 353, 1500. Marlowe, F. W., Apicella, C. L., & Reed, D. (2005). Men’s preferences for women’s profile waist-hip-ratio in two societies. Evolution and Human Behavior, 26, 458–468. Marlowe, F., & Wetsman, A. (2001). Preferred waist-to-hip ratio and ecology. Personality and Individual Differences, 30, 481–489. Newcomb, T. M. (1961). The acquaintance process. New York: Holt, Rinehart, and Winston. Nisbett, R. E., & Wilson, T. D. (1977). The halo effect: evidence for unconscious alteration of judgements. Journal of Personality and Social Psychology, 35, 250–256. Puhl, R., & Boland, F. (2001). Predicting female physical attractiveness. Psychology, Evolution and Gender, 3, 27–46. Singh, D. (1993). Adaptive significance of female physical attractiveness: role of the waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293–307. Storr, A. (1989). Solitude: a return to the self. London: Ballantine Books. Streeter, S. A., & McBurney, D. H. (2003). Waist-to-hip ratio and attractiveness: new evidence and a critique of ‘a critical test’. Evolution and Human Behaviour, 24, 88–98. Sugiyama, L. S. (2004). Is beauty in the context-sensitive adaptations of the beholder? Shiwiar use of waist-to-hip ratio in assessments of female mate value. Evolution and Human Behaviour, 25, 51–62. Swami, V. (in press). The missing arms of ve´nus de milo: reflections on the science of physical attractiveness. Lewes: The Book Guild. Swami, V., Antonakopoulos, N., Tove´e, M. J., & Furnham, A. (2006). A critical test of the waist-to-hip ratio hypothesis of female physical attractiveness in Britain and Greece. Sex Roles, 54, 201–211. Swami, V., Caprario, C., Tove´e, M. J., & Furnham, A. (2006). Female physical attractiveness in Britain and Japan: a cross-cultural study. European Journal of Personality, 20, 69–81. Swami, V., & Furnham, A. (2006). The science of attraction. The Psychologist, 19, 362–365. Swami, V., & Tove´e, M. J. (2005a). Female physical attractiveness in Britain and Malaysia: a cross-cultural study. Body Image, 2, 115–128. Swami, V., & Tove´e, M. J. (2005b). Male physical attractiveness in Britain and Malaysia: a cross-cultural study. Body Image, 2, 383–393.
572
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
Swami, V., & Tove´e, M. J. (2006). Does hunger influence judgements of female physical attractiveness? British Journal of Psychology, 97, 353–363. Tove´e, M. J., & Cornelissen, P. L. (2001). Female and male perceptions of female physical attractiveness in front-view and profile. British Journal of Psychology, 92, 391–402. Tove´e, M. J., Hancock, P., Mahmoodi, S., Singleton, B. R. R., & Cornelissen, P. L. (2002). Human female attractiveness: waveform analysis of body shape. Proceedings of the Royal Society of London B, 269, 2205–2213. Tove´e, M. J., Maisey, D. S., Emery, J. L., & Cornelissen, P. L. (1998). Visual cues to physical attractiveness. Proceedings of the Royal Society of London B, 266, 211–218. Townsend, J. M., & Levy, G. D. (1990). Effects of potential partners’ costume and physical attractiveness on sexuality and partner selection. Journal of Psychology, 124, 371–389. Treleaven, P., Furnham, A., & Swami, V. (2006). The science of body metrics. The Psychologist, 19, 416–419. Wilson, J. M. B., Tripp, D. A., & Boland, F. J. (2005). The relative contributions of subjective and objective measures of body shape and size to body image and disordered eating in women. Body Image, 2, 233–247.
More than just skin-deep? A pilot study integrating physical and non-physical factors in the perception of physical attractivene Viren Swami a
a,*
, Corina Greven b, Adrian Furnham
b
Division of Public Health, University of Liverpool, Whelan Building, Qudrangle, Brownlow Hill, Liverpool L69 3GB, UK b Department of Psychology, University College London, London, UK Received 24 March 2006; received in revised form 28 June 2006 Available online 5 October 2006
Abstract The perceived beauty of a potential partner has traditionally been explained in terms of physical characteristics such as body weight and the waist-to-hip ratio (WHR), while non-physical characteristics have been neglected. This study looked at the contribution of body weight, WHR and personality to ratings of overall female attractiveness. Seventy-six participants rated 30 line drawings that varied 2 levels of personality (introverted, extraverted), 3 levels of body weight and 5 of WHR. Effect sizes were largest for body weight (0.58), followed by personality (0.25) and finally WHR (0.10). There were significant interactions between all three variables, with effect sizes ranging between (0.3 and 0.4). These results highlight the importance of examining non-physical characteristics when studying perceptions of human beauty, a finding consistent with the notion that beauty is more than just skin-deep. 2006 Elsevier Ltd. All rights reserved. Keywords: Physical attractiveness; Non-physical attractiveness; Body weight; Waist-to-hip ratio; Personality
*
Corresponding author. Tel.: +44 (0)151 794 5270; fax: +44 (0)151 794 5588. E-mail address: [email protected] (V. Swami).
0191-8869/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2006.08.004
564
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
1. Introduction The topic of human beauty continues to be one that attracts considerable debate and controversy both within psychological and popular circles (e.g., Swami, in press; Swami & Furnham, 2006). In the past decade, a productive field of research has examined perceptions of male and female bodily attractiveness (e.g., Singh, 1993; Tove´e, Maisey, Emery, & Cornelissen, 1998; Tove´e, Hancock, Mahmoodi, Singleton, & Cornelissen, 2002; Wilson, Tripp, & Boland, 2005), with some recent studies taking a cross-cultural perspective (e.g., Sugiyama, 2004; Swami & Tove´e, 2005a; Swami, Antonakopoulos, Tove´e, & Furnham, 2006; Swami, Caprario, Tove´e, & Furnham, 2006). These studies have typically examined the relative contribution of body shape and weight measures to ratings of female physical attractiveness. In one of the earliest studies of physical attractiveness, Singh (1993) argued that the waist-tohip ratio (WHR), a measure of body shape, is a better predictor of female attractiveness judgements than overall body weight. He presented evidence showing that a low WHR is associated with higher levels of fertility, and went on to argue that the WHR acts a first-pass filter, excluding women who are unhealthy or have low reproductive capability. Evidence for this has been reported in a number of countries using line-drawn stimuli developed by Singh (e.g., Furnham, Tan, & McManus, 1997). However, some investigators have argued that such findings are an artefact of the experimental design. The line drawings and photographic stimuli used in these studies appeared to confound body weight and shape measures. When researchers have investigated the relative importance of body shape and weight (the latter quantified as the body mass index, or BMI) using photographs of real women (e.g., Puhl & Boland, 2001; Tove´e, Maisey, et al., 1998, 2002; Tove´e & Cornelissen, 2001) and three-dimensional images (Fan, Liu, Wu, & Dai, 2004), variation in BMI appears to be the greater predictor of female attractiveness than WHR. These effects cannot simply be explained by the stimuli not adequately capturing shape cues, as when pictures of men are used in the same format, attractiveness is typically determined by shape cues rather than BMI (Maisey, Vale, Cornelissen, & Tove´e, 1999; Swami & Tove´e, 2005b), thus demonstrating that shape cues are salient in this format. These results also hold when cross-cultural data sets are examined (e.g., Swami & Tove´e, 2005a; Swami, Antonakopoulos, et al., 2006; Swami, Caprario, et al., 2006), and psychosocial explanations for these have been postulated (e.g., Swami & Tove´e, 2006). Of course, the relative ranges of BMI and WHR in these studies are unequal, and the WHR seems to be a more important predictor of attractiveness within a normal or average BMI range (cf. Streeter & McBurney, 2003; Wilson et al., 2005). More recent conceptualisations of the WHR hypothesis of attractiveness have also shifted away from the notion of it being a first-pass filter (e.g., Marlowe & Wetsman, 2001; Marlowe, Apicella, & Reed, 2005). Sugiyama (2004), for example, has presented evidence of a male preference for WHRs lower-than-the-local-female-average WHR, which takes into account the effect of parasitic loads and nutrition in increasing or decreasing local WHRs ranges. Within this paradigm, the WHR is one of several factors influencing physical attractiveness, with such factors as body weight, skin colour and facial attractiveness all also playing major roles. The theoretical background behind most such studies stems from the evolutionary psychological idea that human physical attractiveness should be seen as an assessment of fitness value, indexed by such factors as fecundity or health (Buss, 1994, 1999). Within this view, the most
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
565
fitness-enhancing trait (for example, a curvaceous female body) is perceived as the most attractive and individuals are drawn to what is regarded as most attractive by definition. Typically, however, researchers have assumed that the assessment of physical attractiveness will be based on purely physical traits, to the exclusion of non-physical factors. When non-physical factors are considered within the evolutionary psychological paradigm, it is usually as an epiphenomenon of physical characteristics. That is, individuals may be attracted to others because of physical attributes (a slender body weight or a curvaceous body shape), with non-physical virtues merely enhancing desirability as a secondary consideration. Kniffin and Wilson (2004), however, have argued that a more rounded evolutionary psychology should allow for assessments of physical attractiveness to be based on overall fitness value, including non-physical traits. Indeed, a large body of social psychological research has emphasised the influence of non-physical characteristics on judgements of physical attractiveness (e.g., Feingold, 1992; Felson & Bohrenstedt, 1979; Gross & Crofton, 1977; Nisbett & Wilson, 1977). More recent studies inspired by evolutionary psychology show that social status (Townsend & Levy, 1990) and prosocial orientation (Jensen-Campbell, Graziano, & West, 1995) enhance perceptions of physical attractiveness. In their study, Kniffin and Wilson (2004) also demonstrated that non-physical factors have a very potent effect in the perception of physical attractiveness. In general, women were more strongly influenced by non-physical factors than men and there were also large individual differences within each gender. These studies highlight the importance of examining non-physical factors in research questions (Kniffin & Wilson, 2004). To this end, we designed a simple experimental study to compare the relative effects of body weight, WHR and personality, the latter acting as a basic non-physical attribute. The study used a selection of modified line drawings reported in Furnham, Swami, and Shah (2006), with the final set combining three body weight categories, five WHR levels, and two levels of personality. This manipulation allowed us to investigate the contribution of a target’s personality on judgements of physical attractiveness.
2. Methods 2.1. Participants A total of 76 British undergraduate students, all of whom were males, participated in the study. The mean age of the sample was 23.28 (SD = 7.93). Only participants who self-reported as heterosexual were invited to participate in the study. Just over 80% of the sample were British Caucasians, while the rest of the sample were of either Asian (11.8%), Afro-Carribean (3.9%) or other (3.9%) descent. Participant ethnicity did not have a significant effect in subsequent ratings. The socio-economic and educational backgrounds of the participants were fairly homogenous. All participants were naı¨ve to the aims of the study and participated on a voluntary basis. 2.2. Stimuli and procedure The stimuli used in this study were selected and modified from Furnham et al. (2006). To begin with, we selected 15 images that varied by three levels of body weight (underweight, average
566
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
weight and overweight) and five levels of WHR (0.6, 0.7, 0.8, 0.9, and 1.0). The ranges of these variables are in line with those used in previous studies. The arms and legs were narrowed or thickened within each weight category, and all other facial and bodily features (e.g., breast size) were kept constant. In addition, the stimuli had originally been designed to be ethnically ambiguous. These stimuli were then given the designatory symbol ‘A,’ indicated at the bottom-right corner of the image. A second set of identical stimuli were given the designatory symbol ‘B.’ Participants were told that they would be shown different images of a woman named Rozie and that the images depicted Rozie with different physical characteristics. In addition, Rozie could have two different types of personality, indicated by the letter (A or B) at the bottom of the image. The characteristics associated with each personality were Personality A (Extravert): Rozie is friendly and likes to be around others. She is actionoriented, is a ‘people person,’ and often feels energetic; she is open and shows her emotions easily, but usually acts before thinking. Overall, Rozie has a cheerful outlook, reacts to things spontaneously and has an entertaining curiosity about most things. Personality B (Introvert): Rozie is shy and quiet, and likes to spend time alone. She likes ideas and concepts, and prefers solitude for concentration; she is thoughtful and keeps her emotions private, but usually thinks before acting. Overall, Rozie is highly analytical and precise in her thinking and tends to shy away from social interactions. These descriptors were designed based on characteristics differentiating individuals on an introversion–extraversion scale (e.g., Costa & McCrae, 1992). While the use of symbols (A or B) to indicate personality type may have been overly simple, manipulating other aspects of the stimuli (e.g., clothing or hairstyle) to indicate personality type would have introduced possible confounds. This would also have been true had we used two different individuals with different personalities. In a pilot study (n = 18), we found that participants were able to make judgements based on the use of symbols to designate personality. Further, both descriptors were in reading view of participants throughout the experiment, and participants were told to consider both physical and nonphysical factors in making their judgements. The task was conducted by a female experimenter trained in the methods of the study. However, for a subset of the participants (n = 16), the task was conducted by a male experimenter. There were no significant differences in the responses to male and female experimenters (p > 0.10). For each of the 30 images, participants were asked to provide a rating for overall attractiveness, that is, how attractive they thought the image of Rozie was in physical and nonphysical terms. Each image was presented on sheets of card measuring 210 · 297 mm and remained in view for 45 s. Participants were provided with a booklet which included brief instructions and on which they made their ratings according to a 9-point Likert scale (9 = very attractive). All participants completed the task individually, and were later debriefed.
3. Results A 2 · 5 · 3 repeated measures analysis of variance (ANOVA) with 76 participants was computed. Body weight, WHR and personality were treated as within subject factors. Where a factor had more than two levels, and the Mauchly’s Test of Sphericity was shown to be significant, the
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
567
Greenhouse–Geisser correction was applied. A summary of the ANOVA results and the main effects of WHR, body weight and personality, and their interactions are shown in Table 1. The ANOVA revealed that body weight, WHR and personality all had significant main effects on the ratings of the figures. In general, extraverts were preferred over introverts, under- and normal weight figures were preferred over overweight figures, and lower WHRs were preferred over higher WHRs (see Fig. 1). However, the effects sizes revealed body weight accounted for more of Table 1 ANOVA results with main effects of WHR, body weight and personality, and their interactions Source
Df
Body weight Personality Waist-to-hip ratio Personality · body weight Personality · WHR Waist-to-hip ratio · body weight Personality · WHR · body weight
102.52 25.40** 8.34** 3.85* 2.69* 2.47* 3.07*
0.58 0.25 0.10 0.04 0.03 0.03 0.04
Greenhouse–Geisser corrected. p < 0.05. p < 0.001.
5.4
6.0
5.2
5.5
5.0 Mean ratings
**
Effect size **
1.66, 124.26 1, 75 2.56, 191.62a 1.77, 132.69a 3.33, 250.03a 6.28, 471.30a 6.10, 457.33
5.0
4.5
4.8
4.6 4.0
4.4
3.5 Under weight
Average weight
4.2 Extroverted
Over weight
Introverted
5.1 5.0 4.9 Mean ratings
*
Mean ratings
a
F a
4.8 4.7 4.6 4.5 4.4 0.6
0.7
0.8
0.9
1.0
Fig. 1. Graphs of the main effects for body weight, personality and waist-to-hip ratio.
568
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
the variance in the data ðg2p ¼ 0:58Þ than did personality ðg2p ¼ 0:25Þ or WHR ðg2p ¼ 0:10Þ. However, there were also significant interactions which would suggest caution in interpreting the main effects. There were significant two-way interactions of personality · body weight, personality · WHR, and body weight · WHR, as well as a significant three-way interaction between personality · WHR · body weight. The two-way interactions are illustrated in Fig. 2, and the three-way interaction in Fig. 3. From Fig. 3, it can be seen that for extraverted figures, the preferred WHRs for the underweight and overweight figures was 0.6. The preferred WHR for the average weight category was either 0.7 and 0.9. For the introverted figures, the preferred WHR for the underweight category was 0.7, whereas for the average weight category it was 0.8 and for the overweight category it was 0.6. In general, lower WHRs were preferred over higher ratios across categories; moreover, underweight and average weight figures were rated fairly similarly, but overweight figures were
6.5
6.5
Personality
Personality 6.0
6.0
Extraverted
Extraverted 5.5
Introverted
Mean ratings
Mean ratings
5.5
5.0 4.5
5.0
4.5
4.0
4.0
3.5
3.5
3.0
Introverted
3.0 1
2
3
1
2
Body weight
3
4
5
Waist-to-hip ratio
6.5
6.0
Mean ratings
5.5
Body weight
5.0
Under weight 4.5
Average weight 4.0
Over weight 3.5
3.0 1
2
3
4
5
Waist-to-hip ratio
Fig. 2. Graphs of the two-way interactions between personality · WHR, personality · body weight and body weight · WHR.
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572 6.5
6.5
6.0
6.0
Body weight
5.5
Mean ratings
Mean ratings
5.5
Under weight
5.0
Average weight
4.5
569
Over weight
5.0
Body weight 4.5
Under weight
4.0
4.0
3.5
3.5
Average weight
3.0 1
2
3
Waist-to-hip ratio
4
5
3.0 0.6
Over weight
0.7
0.8
0.9
1.0
Waist-to-hip ratio
Fig. 3. Graphs of the three-way interaction between body weight, personality and waist-to-hip ratio.
rated more negatively. The ANOVA also revealed that the body weight and WHR interaction had a significant effect on the ratings of all attributes. However, it is important to note that the effect sizes for these interactions were fairly small, and therefore the significance of these interactions may have been caused by the large number of participants in the study.
4. Discussion This study was designed to address a relative paucity of research examining the effects of non-physical characteristics on judgements of physical attractiveness. Our findings highlight this neglected area of research: non-physical factors, in this case the personality of the target, can have an important influence on judgements of attractiveness both independently and in interactions with body shape and weight. Although our results are not directly comparable with those of Kniffin and Wilson (2004) because of the different methodologies used, they are in general agreement in showing that non-physical factors can affect perceptions of attractiveness. In particular, extraverted or sociable individuals are perceived as more attractive than introverted or quiet individuals. On its own, this is perhaps not a surprising finding; what is more striking, however, is the role of the interaction between personality, body shape and body weight. Kniffin and Wilson (2004) have argued that any assessment of an individual’s physical attractiveness should also include perceptions of non-physical characteristics, which lend themselves to a judgement of overall attractiveness. Although it is conceivable that assessments of attractiveness begin with physical attributes before weighting non-physical characteristics (cf. Singh, 1993), a more likely scenario is that both sets of factors are assessed simultaneously when such information is available. Kniffin and Wilson (2004) argue that in ancestral social environments, interactions probably took place in small groups of people whose physical attributes were roughly average and whose non-physical attributes were intimately known to each other. Although we think that conjecture about supposed past evolutionary environments should be avoided (cf. Buller, 2005),
570
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
such a scenario suggests that non-physical traits may have been important components of interpersonal relations in human evolutionary history. An obvious objection to the methodology used in the present study is that participants were asked to make ratings of relative strangers, of which some information about personality is known. In real life situations, perceptions of strangers do not often include information about extraversion or introversion. Rather, perceptions of strangers are often predicated on first impressions, which primarily involve judgements of non-verbal, physical cues (see Baron, Byrne, & Branscombe, 2006). In such scenarios, it is not inconceivable that judgements of attractiveness should proceed with perceptions of physical rather than non-physical attractiveness. However, at a basic level, all inter-personal relationships must involve some level of social interaction, and here non-physical characteristics may play an important role in determining an individual’s overall attractiveness (Swami, in press). It is this interaction between physical and non-physical factors that may underscore the popular belief that beauty is more than just skin-deep. It should also be noted that the personality descriptions used in the present study were extremes of behaviour, and in reality many individuals combine aspects of introversion and extraversion in their overall personality types (e.g., cheerful and emotionally private). In this respect, we would support Kniffin and Wilson’s (2004) call for studies of physical attractiveness to include both physical and non-physical factors within naturalistic designs. An experimental design such as this is able to manipulate various factors simultaneously, but naturalistic designs are more able to approximate the way perceptions of attractiveness are formed in real life situations. Finally, it is possible that participants in the present study tended to be extraverts, which would explain the overall preference for extraverted others. Studies in social psychology, for example, have shown that an important factor that moderates attractiveness is similarity of attitudes or values (e.g., Newcomb, 1961). At a basic level, people who are evenly matched in their physical appearance, social background and personality are more likely to be attracted to one another. Future studies would therefore do well to include simple personality measures of the participants, to examine the possible effect of similarity of personality on attraction. In spite of these limitations, the results of the present study suggest that the personality of a target may be an important component of physical attractiveness judgements. Indeed, in the present study, personality was found to explain more of the variance in the data than WHR. We believe that these findings are an important starting point for future research. In particular, studies should investigate in greater detail gender differences in the importance of non-physical characteristics, which Kniffin and Wilson (2004) found to be important. Studies could also examine the way in which the personality of the observer influences preferences for personality and physical characteristics in a target individual. Better technology would also be useful: the use of photographic and three-dimensional stimuli would certainly be an improvement on the use of line drawings (Treleaven, Furnham, & Swami, 2006). In conclusion, this study suggests that non-physical attributes have an important effect on the perception of physical attractiveness, which is consistent with more rounded evolutionary psychological theories of mate selection. Like Kniffin and Wilson (2004), our results point to extraverts being more valued social partners than introverts. Of course, the value of introversion in some case should not be disregarded (Storr, 1989), but in general being sociable and outgoing may be just as important as having a particular body weight or shape.
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
571
References Baron, R., Byrne, D., & Branscombe, N. (2006). Social psychology. Boston: Pearson. Buller, D. J. (2005). Adapting minds: evolutionary psychology and the persistent quest for human nature. London: MIT Press. Buss, D. (1994). The evolution of desire. New York: Basic Books. Buss, D. (1999). Evolutionary psychology: the new science of the mind. Boston: Allyn & Bacon. Costa, P. T., & McCrae, R. (1992). Four ways five factors are basic. Personality and Individual Differences, 135, 653–665. Fan, J., Liu, F., Wu, J., & Dai, W. (2004). Visual perception of female physical attractiveness. Proceedings of the Royal Society of London B, 271, 347–352. Feingold, A. (1992). Good-looking people are not what we think. Psychological Bulletin, 111, 304–341. Felson, R. B., & Bohrenstedt, G. W. (1979). ‘Are the good beautiful or the beautiful good?’ The relationship between children’s perceptions of ability and perceptions of physical attractiveness. Social Psychology Quarterly, 42, 386–392. Furnham, A., Swami, V., & Shah, K. (2006). Female body correlates of attractiveness and other ratings. Personality and Individual Differences, 41, 443–454. Furnham, A., Tan, T., & McManus, C. (1997). Waist-to-hip ratio and preferences for body shape: a replication and extension. Personality and Individual Differences, 22, 539–549. Gross, A. E., & Crofton, C. (1977). What is good is beautiful. Sociometry, 40, 85–90. Jensen-Campbell, L. A., Graziano, W. G., & West, S. G. (1995). Dominance, prosocial orientation and female preferences: do nice guys really finish last? Journal of Personality and Social Psychology, 68, 427–440. Kniffin, K. M., & Wilson, D. S. (2004). The effect of non-physical traits on the perception of physical attractiveness: three naturalistic studies. Evolution and Human Behavior, 25, 88–101. Maisey, D. M., Vale, E. L. E., Cornelissen, P. L., & Tove´e, M. J. (1999). Characteristics of male attractiveness for women. Lancet, 353, 1500. Marlowe, F. W., Apicella, C. L., & Reed, D. (2005). Men’s preferences for women’s profile waist-hip-ratio in two societies. Evolution and Human Behavior, 26, 458–468. Marlowe, F., & Wetsman, A. (2001). Preferred waist-to-hip ratio and ecology. Personality and Individual Differences, 30, 481–489. Newcomb, T. M. (1961). The acquaintance process. New York: Holt, Rinehart, and Winston. Nisbett, R. E., & Wilson, T. D. (1977). The halo effect: evidence for unconscious alteration of judgements. Journal of Personality and Social Psychology, 35, 250–256. Puhl, R., & Boland, F. (2001). Predicting female physical attractiveness. Psychology, Evolution and Gender, 3, 27–46. Singh, D. (1993). Adaptive significance of female physical attractiveness: role of the waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293–307. Storr, A. (1989). Solitude: a return to the self. London: Ballantine Books. Streeter, S. A., & McBurney, D. H. (2003). Waist-to-hip ratio and attractiveness: new evidence and a critique of ‘a critical test’. Evolution and Human Behaviour, 24, 88–98. Sugiyama, L. S. (2004). Is beauty in the context-sensitive adaptations of the beholder? Shiwiar use of waist-to-hip ratio in assessments of female mate value. Evolution and Human Behaviour, 25, 51–62. Swami, V. (in press). The missing arms of ve´nus de milo: reflections on the science of physical attractiveness. Lewes: The Book Guild. Swami, V., Antonakopoulos, N., Tove´e, M. J., & Furnham, A. (2006). A critical test of the waist-to-hip ratio hypothesis of female physical attractiveness in Britain and Greece. Sex Roles, 54, 201–211. Swami, V., Caprario, C., Tove´e, M. J., & Furnham, A. (2006). Female physical attractiveness in Britain and Japan: a cross-cultural study. European Journal of Personality, 20, 69–81. Swami, V., & Furnham, A. (2006). The science of attraction. The Psychologist, 19, 362–365. Swami, V., & Tove´e, M. J. (2005a). Female physical attractiveness in Britain and Malaysia: a cross-cultural study. Body Image, 2, 115–128. Swami, V., & Tove´e, M. J. (2005b). Male physical attractiveness in Britain and Malaysia: a cross-cultural study. Body Image, 2, 383–393.
572
V. Swami et al. / Personality and Individual Differences 42 (2007) 563–572
Swami, V., & Tove´e, M. J. (2006). Does hunger influence judgements of female physical attractiveness? British Journal of Psychology, 97, 353–363. Tove´e, M. J., & Cornelissen, P. L. (2001). Female and male perceptions of female physical attractiveness in front-view and profile. British Journal of Psychology, 92, 391–402. Tove´e, M. J., Hancock, P., Mahmoodi, S., Singleton, B. R. R., & Cornelissen, P. L. (2002). Human female attractiveness: waveform analysis of body shape. Proceedings of the Royal Society of London B, 269, 2205–2213. Tove´e, M. J., Maisey, D. S., Emery, J. L., & Cornelissen, P. L. (1998). Visual cues to physical attractiveness. Proceedings of the Royal Society of London B, 266, 211–218. Townsend, J. M., & Levy, G. D. (1990). Effects of potential partners’ costume and physical attractiveness on sexuality and partner selection. Journal of Psychology, 124, 371–389. Treleaven, P., Furnham, A., & Swami, V. (2006). The science of body metrics. The Psychologist, 19, 416–419. Wilson, J. M. B., Tripp, D. A., & Boland, F. J. (2005). The relative contributions of subjective and objective measures of body shape and size to body image and disordered eating in women. Body Image, 2, 233–247.