نسبت دور کمر به باسن و جذابیت صورت: یک مطالعه مقدماتی

The aim of this study was to evaluate the influence of facial attractiveness and waist to hip ratio (WHR) on five judgements of attractiveness (attractiveness, sexiness, healthiness, fertility and pregnancy). It was hypothesised that facial attractiveness would be more important than WHR for all measures except for pregnancy. Four levels of WHR were matched with two levels of facial attractiveness to produce eight individual stimuli based on the same figure. These were rated on a seven point bi-polar scale for each measure by 100 male subjects. A correlation matrix revealed that attractiveness was most highly correlated with sexiness; healthiness was most highly correlated with fertility, and pregnancy was independent of all other measures. Results revealed that both facial attractiveness and WHR were highly significant in influencing the five judgement measures. Facial attractiveness seems more important than WHR for all measures except pregnancy. Following overall linear trend analysis for all measures it was suggested that WHR is the best predictor of perceived pregnancy. The results are discussed in the context of evolutionary theory. Limitations of the study, particularly reliance on limited stimuli, are acknowledged.

In evolutionary terms selection of an appropriate mate is largely dependent on that mate’s perceived ability to reproduce and take good care of its young. Owing to the cryptic nature of human ovulation, it had been suggested that the male uses more observable body cues such as physical attractiveness and health in order to deduce a female’s capability for reproduction (Buss, 1989, Kenrick, 1989 and Symons, 1979). Researchers have recently questioned whether there is a cross-cultural standard for attractiveness which is universally understood by potential mates as a major cue for fertility. This has been proposed in the form of body fat distribution as measured by the waist to hip ratio (Singh, 1993a, Singh, 1993b, Singh, 1994 and Singh and Luis, 1995). The waist to hip ratio is dependent on sex hormones which become active at puberty. A high waist to hip ratio which is typical of males is maintained by testosterone which stimulates fat deposits in the abdominal region and inhibits fat deposits in the gluteofemoral region. By contrast the lower waist to hip ratio of the female is maintained by oestrogen which inhibits fat deposits in the abdominal region and stimulates deposits in the gluteofemoral region (Bjorntorp, 1987, Rebuffe-Scrive, 1987, Rebuffe-Scrive, 1988 and Rebuffe-Scrive, 1991). The influence of these sex hormones give rise to an android or gynoid body fat distribution which can be accurately measured by the waist to hip ratio (WHR). The WHRs of males and females are similar before puberty and in old age but it is during the reproductive period that a woman’s ratio lowers to between 0.67 and 0.80 (Lanska, Lanska, Hartz & Rimm, 1985) whilst a man’s is maintained between 0.85 and 0.95 (Jones, Hunt, Brown & Norgan, 1986). It has been reported that married women with high waist to hip ratios and low body mass index (BMI) (weight in proportion to squared height) report more difficulty in becoming pregnant and have their first live birth at a later age than married women with a low WHR (Kaye, Folsom, Prineas, Potter & Gapstur, 1990). Another study in Holland found that there is a negative correlation between WHR and probability of conception in an artificial insemination programme (Zaadstra et al., 1993). With regard to general health it has been found that particular fat distribution in the body (i.e. belly fat), rather than total fat is a major risk factor in diabetes, heart attack and stroke (Bjorntorp, 1988 and Bjorntorp, 1991). It would thus seem that as an indicator of good health and fertility it would be advantageous for the male to select a female mate with a low WHR. If such a selection process were to be present in males it would be logical to assume that perceived attractiveness is not an arbitrary concept but one which is related to fertility and health so that females who are perceived as most attractive are also perceived as most healthy and fertile. A recent Swedish study found WHR was a good predictor of pregnancy rate in in-vitro fertilisation and embryo transfer (Wass, Waldenstrom, Rössner & Hellberg, 1997). Thus women with a WHR of between 0.70 and 0.79 had a pregnancy rate of 29.9% compared to 15.9% for those with a WHR of >0.80. There was no relationship between BMI and pregnancy rate. The authors were able to show that the relationship between WHR and pregnancy is not due to hormonal imbalance or egg fertilisation. Indeed few clinical studies show any relationship between BMI and fertility. Indeed, several experiments by Singh, 1993a, Singh, 1993b and Singh, 1994) have shown a marked preference for females presenting a low WHR (0.7). In both Singh experiments, 12 line drawings were created to represent females of four different WHRs (0.7, 0.8, 0.9, 1.0) and three different weight categories: normal, underweight, overweight. When these figures were presented to subjects of both sexes who were asked to rank them on the following attributes: attractiveness, youthfulness, healthiness, sexiness, and capability of and desire for reproduction, it became apparent that stimuli with higher WHRs were attributed lower ranks whilst stimuli with WHR 0.7 were ranked most highly. Further, females of normal weight were ranked most highly. Weight also had a bigger effect (determined by effect size) on rankings than WHR. There are two types of limitations related to this study; stimuli and method. With regards to the format of the stimuli it can be argued that the presentation of line drawings is detrimental to the ecological validity of such experiments in that the 2D format is used to give an impression of something which is 3D and as such does not allow for differences in the depth of body which could elicit different WHRs whilst having the same 2D outline. It can also be argued that these line drawings are not sufficiently lifelike in order to precipitate a satisfactory response with regards to attractiveness and other factors. As to the methods involved, a repeated measures design was chosen but subjects were shown all the stimuli at the same time, allowing for the possibility that the subjects were aware of experimental manipulations. In order to overcome some of these flaws Henss (1995a) undertook a replication of Singh’s experiments with some changes. He too presented male and female subjects with twelve line drawings but decided to use a between subjects design so that each subject assessed one female figure on a scale of 1–9. This modification prevented subjects from becoming aware of the experimental manipulations. The results from Henss’s experiment showed that females with a WHR of 0.8 (rather than 0.7) were rated most highly on the attractiveness factor though the differences between these WHR conditions was minimal. Females who were underweight were consistently rated most highly, although the underweight female was judged to be 55.7 kg which Henss (1995a) argued was not ‘particularly meagre’. He accounted for the discrepancies between his and Singh’s findings by referring to differences in methodology such as the rating scales chosen by the experimenters. In another attempt to rectify previous shortcomings in WHR research, Furnham, Tan and McManus (1997) conducted a repeated measures study again using the twelve line drawings which were presented in random order to 90 undergraduate students. The line drawings were rated on a 7 point bipolar scale for factors relating to reproductive ability. This experiment found that subjects gave females with a high WHR (1.0) the lowest scores across all three ratings factors (attractiveness, healthiness and youthfulness) but in contrast to Singh’s findings (and in accordance with Henss), a WHR of 0.8 was most preferred for all factors. With regards to weight, overweight stimuli were regarded as the most unattractive but not the least healthy. Contrary to Henss’s findings, Furnham et al. (1997) found that the normal weight category (as opposed to the ‘thin’) was the most highly rated on all attributes except youthfulness. In an attempt to discover how the effect of WHR might be mediated by an ecological variable other than weight, a study by Furnham, Dias and McClelland (1998) used the same line drawings as in previous studies on WHR, but introduced a breast size variable. It was found that breast size was a less powerful factor in determining the ratings of attractiveness than WHR, but its effect was dependent on both overall body fat and the size of WHR. However, there have been criticisms of some of the stimulus material used in these studies. Tassinary and Hansen (1998) showed that judgments of attractiveness and fecundity can be either unrelated or related, positively or negatively, to WHR depending on waist size, hip size and weight. They argue that the stimulus material used by Singh, Henss and Furnham confounded weight with hip size, as well as WHR with relative waist size. Hence they developed new unconfounded material which was rated on attractiveness and fecundity. They found weight of figure a more powerful predictor of attractiveness than WHR, a finding which has been consistently obtained by all WHR researchers and is an important objection to Singh’s ‘first pass filter’ hypothesis. Specifically, the results indicated that relative hip size and weight were positively associated with fecundity and negatively associated with attractiveness, regardless of WHR. Further, there was no correlation between ratings of attractiveness and fecundity. Relative hip size is a more potent influence on attractiveness than is relative weight, while the opposite is true of fecundity. They concluded that “weight and hip size are important and independent co-determinants of both relative attractiveness and fecundity, and that the WHR is of marginal importance for predicting relative attractiveness. This pattern of results is inconsistent with either the good sense or the good taste model of human female attractiveness, and thus constitutes a clear and unambiguous disconfirmation of the WHR hypothesis.” (pp. 154–155). Tovee et al., 1997 and Tovee et al., 1998 have also been critical of the WHR research. They hypothesised and found that the body-mass index is more closely related to fertility and health than waist-to-hip ratio and therefore also of ratings of sexual attractiveness. “Body-mass index accounted for 73.5% of variance, whereas waist-to-hip ratio accounted for only 1.8%. Waist-to-hip ratio correlated poorly with attractiveness, whereas even small changes in body-mass index radically altered the attractiveness rating.” (p. 548). Despite the fact that Singh argues that WHR plays a critical role in judgments of female attributes, especially attractiveness, it is apparent in all his studies that weight plays a more significant role than WHR. Singh’s argument is that WHR could act as a wide ‘first-pass filter’, which would exclude those women who are unhealthy or have a low reproductive capability (Singh, 1993b). It could be argued that the WHR is a distal cue that can be used to make probalistic judgments of age and fertility where other cues (i.e. facial) are not available or clear. However it is implied in the work of some WHR researchers that the WHR may dominate in instances where there are more than one cue — i.e. weight, breast size, facial attractiveness, ethnicity. WHR interacts with other features to influence ratings, especially attractiveness. In biological systems, filters are assumed to be sequential devices followed by narrower filters, each requiring greater specificity. WHR is a distal cue which enables the male seeking a mate to get information about the reproductive age and fertility of a female when, for some reason, the face is not observable. However this is not necessarily an evolutionary argument and faces are frequently clearly observable. Inevitably, importance is attached to the face when judging attractiveness, and it is for this reason that the present study involves the presentation of female figures with WHRs of 0.7, 0.8, 0.9 and 1.0, but uses facial attractiveness (two levels) as the other independent variable. However, there is a great deal of cross cultural consensus as to what people agree is facially attractive. Also a meta analysis of over 100 studies showed considerable consensus in the judgments of facial attractiveness (Henss, 1995b). Cunningham, Roberts, Wu, Barbee and Druen (1995) argue that specific facial features such as prominent cheekbones, small chin, large wide set eyes, and a small nose can give an indication of perceived youth, health and thus fertility. In addition, shiny hair and clear skin also contribute to ratings of overall attractiveness. In order to present the four levels of WHRs and the two levels of attractiveness, it was decided to use computer manipulated photographs which would reduce the problems associated with line drawings. Facial attractiveness was varied using a photograph of a model, which was enhanced by making the nose smaller (for the attractive face). For the unattractive stimulus the nose was made larger whilst the eyes and mouth were made smaller. This is not to suggest that there is a simple monotonic relationship between nose length and attractiveness. The changes were made according to Cunningham et al.’s four criteria (see above) and a pilot study using ten males showed that the ‘attractive’ face was always perceived as more attractive than the unattractive face. It was also decided to collect data from male subjects only. The design of this experiment is a repeated measures design with five dependent measures: attractiveness, sexiness, fertility, healthiness and pregnancy. These particular ratings were chosen based on Singh’s original measures (attractive, sexy, healthy, desire for children, capability for having children, youthfulness). Singh found desire, and capability for having children, part of the same factor, hence it was decided to use the measure of fertility to encompass both. A measure for pregnancy was added to ascertain whether WHR was in fact a measure of pregnancy above all other things. The aim of this study was to investigate the nature of the interaction between facial attractiveness and WHR. It set out to investigate whether facial attractiveness is more important than WHR in ratings of overall attractiveness. If Singh is right and WHR acts as a ‘first pass filter’ then females with low WHRs should always be judged as more attractive than females with high WHRs — regardless of facial attractiveness. It was also hypothesised that the cues for fertility and health are sufficiently observable from the face as to make facial attractiveness a more important factor in determining ratings of attractiveness, healthiness, fertility and sexiness than WHR. It was further hypothesised that the WHR is primarily a measure of the likelihood of pregnancy and is used to screen out females only after they have been screened for facial attractiveness.