|
Article Excerpt
One's physical attractiveness has important implications for many aspects of social life. Attractive individuals are perceived in a more positive light (Dion, Berscheid, & Walster, 1972), often receive preferential treatment (Snyder, Tanke, & Berscheid, 1977), and attract the romantic interest of the opposite sex more than their less attractive peers. Past research into the determinants of attractiveness has focused on cues to attractiveness that are inherent in an individual's physical morphology, and that are relatively stable, and unchanging, such as body shape, body size (Singh, 1993), body symmetry (Tovee, Tasker, & Benson, 2000), facial sexual dimorphism (Perrett et al., 1998), and facial symmetry (Grammer & Thornhill, 1994; Rhodes, Proffitt, Grady, & Sumich, 1998). However, recent research on facial attractiveness suggests that the perception of attractiveness can also be influenced by the more transient factor of experience (Rhodes, Jeffery, Watson, Clifford, & Nakayama, 2003). Here, we review that evidence and consider whether body attractiveness preferences are also shaped by our experience.
One of the more counter-intuitive predictors of facial attractiveness is mathematical averageness. Averaged composites of faces are more attractive than their component faces (Langlois & Roggman, 1990; Langlois, Roggman, & Musselman, 1994), and faces that are distorted towards the average of a set of faces are more attractive than undistorted faces, even when skin complexion and symmetry (which influence attractiveness) are controlled (Rhodes, Sumich, & Byatt, 1999; Rhodes & Tremewan, 1996). Although facial averageness is generally attractive, there are a few dimensions on which average values are not optimally attractive. For example, feminized female faces (and in some cases male faces) are more attractive than average ones (Perrett et al., 1998; Rhodes, Hickford, & Jeffery, 2000).
Recent research shows that what we find attractive in a face can be influenced by relatively small amounts of experience, with a few minutes of exposure to consistently distorted faces shifting the optimally attractive facial shape towards that distortion (Rhodes et al., 2003). Such exposure also changes what is perceived as normal or average in a face (Leopold, O'Toole, Vetter, & Blanz, 2001; O'Leary & McMahon, 1991; Rhodes et al., 2005; Watson & Clifford, 2003; Webster & MacLin, 1999), and the shift in what looks attractive is associated with a congruent shift in what looks normal or average (Rhodes et al., 2003). These results suggest that norms or average faces play an important role in the perception of facial attractiveness.
Possible mechanisms that contribute to the appeal of average faces include prototype (1) abstraction (e.g. Posner & Keele, 1968), and a preference for familiar stimuli (Bornstein, 1989) because prototypes look familiar (e.g. Franks & Bransford, 1971). From this perspective, average exemplars should be attractive in many categories, and this appears to be the case. Prototypes are preferred over their individual exemplars in object categories (Whitfield & Slater, 1979), colour categories (Martindale & Moore, 1988), and music categories (Smith & Melara, 1990). Furthermore, perceived averageness correlates positively with attractiveness in all animal (dogs, fish, birds) and object (cars and watches) categories tested so far (Halberstadt & Rhodes, 2000, 2003).
In addition to these proximate mechanisms, an understanding of why average faces are attractive may also require consideration of ultimate or evolutionary mechanisms. (2) Evolutionary psychologists have proposed that a preference for average faces may identify high quality mates and thus enhance reproductive success (Thornhill & Gangestad, 1993; Thornhill & Grammer, 1999). Consistent with this view, facial averageness is associated with good health and intelligence during development (Rhodes et al., 2000; Zebrowitz, Hall, Murphy, & Rhodes, 2002), and deviations from averageness are associated with some chromosomal abnormalities (Hoyme, 1994; Thornhill & Moller, 1997).
On a general information-processing bias account such as the prototype account described above, one might expect averageness to be attractive in human bodies, in which case, experience should influence what is perceived as attractive because experience defines what is average. It is unclear, however, whether average female body shapes are optimally attractive. Research on female body attractiveness has not directly assessed the attractiveness of average body shapes. Work on waist-to-hip ratio (WHR) and body mass index (BMI) is potentially relevant, but preferences for these traits have not been explicitly related to average population values. Furthermore, results are conflicting, with preferences for the low end (0.70) of the normal range (0.68-0.80; Singh, 1993) reported in some studies (Benson, Emery, Cohen-Tovee, & Tovee, 1999; Furnham, Tan, & McManus, 1997; Henss, 1995; Singh, 1993, 1994, 1995), but not others (Marlowe & Wetsman, 2001; Tassinary & Hansen, 1998; Yu & Shepherd, 1998). The differences may be due to differences in the populations tested, with higher WHRs preferred in more physically challenging environments (Marlowe & Wetsman, 2001; Yu & Shepherd, 1998), and confounds between WHR and weight in many stimulus sets (see Tovee, Maisey, Emery, & Cornelissen, 1999 for discussion). Body weight (or weight scaled for height; BMI), which may be a more important predictor of attractiveness than WHR (Tassinary & Hansen, 1998; Tovee et al., 1999; Wetsman & Marlowe, 1999; Yu & Shepard, 1998), also yields inconsistent results, with reported preferences ranging from light-moderate (Tassinary & Hansen, 1998; Tovee et al., 1999) to heavy (Wetsman & Marlowe, 1999; Yu & Shepard, 1998).
The primary aim of the present study was to investigate whether experience can change what is attractive in human bodies. We used an adaptation methodology where perceptions of attractiveness were assessed before and after adaptation to particular body types. A set of test bodies was created by distorting 10 original bodies to differing degrees along a single dimension of...
|
