{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

symmetry and odour - 5 THE ROYAL Lu SOCIETY Human body...

Info icon This preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 5 THE ROYAL Lu SOCIETY Human body odour, symmetry and attractiveness Anja Rikowski and Karl Grammer Ludwig Boltzmann Institute for Urban Ethology, c/o Institute for Human Biology, Alt/zamtrasse 14, A-1090 Vienna, Austria Several studies have found body and facial symmetry as well as attractiveness to be human mate choice criteria. These characteristics are presumed to signal developmental stability. Human body odour has been shown to influence female mate choice depending on the immune system, but the question of whether smell could signal general mate quality, as do other cues, was not addressed in previous studies. We compared ratings of body odour, attractiveness, and measurements of facial and body asymmetry of 16 male and 19 female subjects. Subjects wore a T—shirt for three consecutive nights under controlled conditions. Opposite-sex raters judged the odour of the T—shirts and another group evaluated portraits of the subjects for attractiveness. We measured seven bilateral traits of the subject’s body to assess body asymmetry. Facial asymmetry was examined by distance measurements of portrait photographs. The results showed a significant positive correlation between facial attractiveness and sexiness of body odour for female subjects. We found positive relationships between body odour and attractiveness and negative ones between smell and body asymmetry for males only if female odour raters were in the most fertile phase of their menstrual cycle. The outcomes are discussed in the light of different male and female reproductive slralegies. Keywords: humans; body odour; fluctuating asymmetry; attractiveness; mate choice 1. INTRODUCTION Recent studies on human body odour indicate the relevance of olfactory communication in various social situations. Apart from mother—child interactions and olfactory identification of relatives and partners (Hold & Schleidt 1977), the sense of smell apparently has important implications for human sexual behaviour (for a review see Schaal & Porter 1991). For example, several substances have been discovered in human excretions that act as sexual pheromones, such as male androstenes in sweat (for a review, see Gower & Ruparelia 1993) and female copulines in vaginal secretions (Michael et a]. 1975). Moreover, human female olfactory preferences seem to induce disassortative mating for components of the major histocompatibility complex (MHC) as in other mammals (Wedekind et a1. 1995). Olfactory cues may thus be able to trigger an incest avoidance mechanism by reflecting parts of an individual’s genetic equipment. In this study we focused on the question of whether human scent, apart from the above-mentioned functions, could signal mate condition like other cues in mate selection. Recent research has focused on the significance of developmental stability as such a mate-choice-relevant feature. Develop- mental stability describes an individual’s ability to cope with genetic and environmental perturbations during development (see Moller & Swaddle 1997). Fluctuating asymmetry, that is, small random devia- tions from perfect bilateral symmetry of morphological structures (Van Valen 1962), is presumed to indicate the developmental instability of an individual (e.g. Watson & Thornhill 1994). Low levels of fluctuating asymmetry are P706. R. Soc. Lond. B (1999) 266, 869*874 Received 23 November 1998 Accepted 21 January 1999 869 preferred in a potential mate in many species (see Moller & Thornhill 1998). In humans, body and facial asymmetry are negatively associated with facial attractiveness (Gangestad et a1. 1994; Grammer & Thornhill 1994), another confirmed mate choice criterion. Cross-cultural studies revealed that physical attractiveness plays an important role in human mate choice (Buss 1989). Although it appears to be a major criterion only for men, attractiveness of potential partners also rates highly among criteria used by females in choosing mates. Furthermore, Gangestad & Buss (1992) pointed out that attractiveness of the potential partner plays a major role in societies with higher parasite loads. Each condition-sensitive mate choice criterion reflects the physical state of a possible mate with a certain error. Thus, the probability of selecting a truly high-quality mate will be improved when the choosing individual examines two or more sexual traits (Moller & Pomiankowski 1993). This should also be the case for signals of different modal- ities, that is, multichannel signals may well provide a better overall indication of mate condition than single- channel information. Fluctuating asymmetry and physical attractiveness both communicate the condition of develop- mental stability via the Visual channel. We hypothesized that human body odour transmits information about an individual’s developmental stability as an additional, redundant olfactory signal. Since olfactory and visual cues have different physiological roots, the signalling errors are likely to be uncorrelated. Thus, taking the information of both signals into account reduces the error and allows much more reliable mate choice decisions. © 1999 The Royal Society 870 A. Rikowski and K. Grammer Human body odour, symmetry and attractiveness Table 1. Results of the t-test to examine deviations from zero of the mean asymmetry in the measured trait (directional asymmetry) trait mean s.d. t d.f. wrist breadth 0.687 0.833 2.96‘ 60 hand breadth 0.935 1.536 3.06‘ 64- elbow width 0.793 1.282 2.89‘ 66 ear length —0.500 1.507 1.68 66 car breadth 0.319 1.108 1.27 68 ankle width — 1.560 2.100 3.88M 50 foot breadth —0.507 2.197 1.4-1 66 *p S 0.01; ”p < 0.001; two-tailed. 2. METHODS {s ‘ We compared ratings of attractiveness and measurements of facial and body asymmetry with the evaluated quality of body odour of 19 female and 16 male subjects (mean age=23.54, s.d.=3.74). The participants were mainly students from the faculties of biology and psychology of the University of Vienna. Female subjects did not take oral contraceptives. (a) Measurement of body asymmetry To assess body asymmetry we measured seven bilateral traits of the subjects’ bodies: hand breadth, wrist breadth, elbow width, car length, ear breadth, ankle width and foot breadth. According to Livshits & Kobyliansky (1989), the summed fluctuating asymmetry of these traits shows considerable heritability. Measurement error was minimized by remeasuring each trait nine times. We used the mean of these repeated measurements as a basis for further calculations. According to Swaddle et a1. (1994) accuracy of the asymmetry measurement was estimated by a three-way mixed model ANOVA. Between- individual variation in asymmetry was significantly and sufficiently higher than the estimated measurement error for each of the seven traits (F34, 612 = between 8.60 and 45.77,p < 0.001). The subsequent data analysis revealed that most examined traits showed directional asymmetry in our sample, that is the mean signed differences between left and right side deviated significantly from zero (table 1). Only the values for ear breadth, ear length and foot breadth fulfilled the criteria for fluctuating asymmetry. However, in most other studies on human fluctu- ating asymmetry, none of these seven measured traits showed directional asymmetry (e.g. Furlow et at. 1996). On the other hand, we have no reason to reject the possibility that the amount of directional asymmetry indeed might be higher in our investigated population. We therefore made the analysis both with an index solely of the three non-directional asymmetrical traits and with all seven characters. Each asymmetry index was calculated as the mean of the relative differences between right and left side (that is |r — 1|, divided by 0.5X(r+ 1)), excluding two and seven cases of injured characters for the two different indices, respectively. Subjects indicated their body heights and weights to control for eflects of obesity on fluctuating asymmetry (Manning 1995). We used these data to calculate the corpulence index, that is (100Xbody weight in grams) divided by (body height in cm)3. According to KnuBmann (1988), this index is a more appro- priate measure for separate-sex samples than the usual body mass index. Proe. R. Soc. Lond. B (1999) (b) Measurement of facial asymmetry We recorded faces of the subjects with a video camera and digitized snapshots of these film sequences that showed the best upright position and neutral face expression of each person in 303 pixelsX303 pixels size. By means of a computer program, the black-and-white portraits of the subjects were standardized to the same size and orientation. For the standardization the faces first were coded with 52 landmarks (source coordinates). In the second step the mean coordinates (destination co- ordinates) for all faces and the respective landmarks were calculated. In the third step the centre of gravity of the source coordinates was calculated for each face. Then the face was moved on the picture plane so that the centre of gravity of the face fell on the centre of gravity of the destination coordinates. Finally, each face was resized to 150% of its original size. Then the face was scaled down in one-pixel steps until the square sum of the difference between the source and destination coordinates reached a minimum (least squares method). After scaling, the face was rotated about the centre of gravity for 45°. Then the same method as above was applied for stepwise rotation. This results in size- and orientation-optimized pictures in relation to the centre of gravity of the face. For the facial asymmetry analysis we used the method described by Grammer & Thornhill (1994). We manually set 12 landmarks on bilateral facial structures of each face that could be reliably identified. The program then connected these marks and computed the midpoints of the six resulting horizontal lines. Summing up the x-axis differences between the midpoints gives an index of horizontal asymmetry in the face. To assess reliability of this measurement, another person unfamiliar with the study’s purpose set the landmarks in the portraits again. The inter-correlation between the resulting two facial asymmetry indices was r = 0.85 at a significance level of 0.001. (c) Olfactory experiments As in previous studies (e.g. Hold & Schleidt 1977; Wedekind et a1. 1995), body odour was collected by ‘T-shirt experiments’. Each subject wore aT—shirt for three consecutive nights directly on the skin and had to follow a certain discipline during this time to avoid disturbing scents. The rules included refraining from (i) using deodorants and perfumed products, (ii) eating odour-producing food like garlic, onions, cabbage and aspar- agus, (iii) smoking cigarettes and drinking alcohol, (iv) sexual activity, and (v) avoiding pubs and discotheques because of the smoky air. To control for possible hormonally induced variation in body odour (Poran 1995), each woman had to start the T-shirt experiment between the eighth and the eleventh day of her menstrual cycle. This restraint made it necessary to conserve the odour of the worn T—shirts by deep-freezing. The white T-shirts were prepared by washing them with non- perfumed soap powder; each T-shirt was then kept in two freezing bags after drying. Each subject received one T—shirt, a package of soap powder to wash his/her bedclothes before the experiment, a perfume-free washing bar (Neutrogena®) for personal hygiene once every morning and odourless liquid soap for hair cleaning. Subjects were informed about the experi- mental procedure and received a written explanation along with a list of the behavioural restrictions. We asked them after the experiment to report honestly possible violations of the rules, their moods during these days, and if they had restless nights. None of the reported minor disturbances had a significant effect on the odour evaluations. Human boaj/ odour, symmetry and attractiveness Odour evaluations took place in four distinct rating sessions on different days at the same time of day. By presenting not more than ten T-shirts to a given group of raters we intended to avoid habituation of scent perception. Two groups of 15 and 11 male raters judged the odour of the ten and nine different T-shirts worn by female subjects, respectively. Twenty-one and 22 female raters judged the smell of the ten and six different T-shirts worn by male subjects, respectively. Raters have all been non-smokers. Female raters did not take oral contraceptives and reported the first date of their last menstrual bleeding and their mean cycle length to calculate day in cycle at the time of evaluation. Before each odour evaluation a person unfamiliar with the experiment recoded the T-shirt numbers and put each frozen piece of clothing into a glass bottle. The filled containers were then heated to 38 0C for 30 min. Raters judged the T—shirt odour on a scale from one (lowest agreement) to seven (highest agreement) with regard to three adjectives: ‘intense’ as a control variable for odour perception and quantity, and ‘pleasant’ (‘angenehm’ in German) and ‘sexy’ (‘erotisch’ in German) as parameters of quality. After the judgement, we presented photographs of the subjects to the raters and asked them to indicate persons they knew. Such ratings were excluded from the data analysis. Hence, we received between nine and 22 valid judgements per worn T-shirt (mean=15.l7, s.d.=4.l3), of which we used the mean values as evaluations of each subject’s odour. According to Grammer (1993), female perception of the male pheromone androstenone changes with cycle phase. Therefore, we divided female raters into two groups of fertility phases (see Baker & Bellis 1995, p. 161): (i) least fertile, ranging from days 1 to 4 and 17 to 32 (n=26) and (ii) most fertile, ranging from days 5 to 16 of the menstrual cycle (n=l4-). Female raters were only classified as being in the most fertile phase if their reported mean cycle length did not exceed 32 days. Hence, three raters had to be excluded from this analysis. Generally, women in the most fertile phaSe of their menstrual cycle rated male body odour as more sexy than did women in the least fertile phase (t: 2.31, d.f. =38,p<0.05, two-tailed). (d) Attractiveness ratings The standardized portraits of the facial asymmetry analysis were printed out in black-and-white on a 300 dpi greyscale laser printer. Twenty-three male and 24 female students from the Economics Department, who were presumably unfamiliar with the evolutionary theories of beauty and symmetry, rated the portraits of opposite-sex subjects for facial attractiveness. They were asked on questionnaires to judge each subject’s face with regard to the adjectives ‘attractive’ and ‘sexy’ on seven-point scales, again seven being the highest value. 3. RESULTS Since all mean ratings and measurements normally distributed, we used parametric statistical tests. Initial analysis of all collected data revealed no general sex differences for the subjects in this sample, except for sexiness of body odour. The smell of female subjects was judged significantly more sexy by men than the odour of male subjects by female raters (t: —2.06, d.f.:28.68, p < 0.05, two-tailed). Physical attractiveness and sexiness ratings of the portraits correlated highly positively both for female (r=0.99, d.f. =17, p<0.001, two-tailed) and male subjects were Prue. R. Soc. Lond, B (1999) A. Rikowski and K. Grammer 871 Table 2. Pearson correlations between the adjectives of oa’our evaluation (Correlations of female odour (11:19, rated by males) are presented on the bottom left. Correlations of male odour (n: 16) are presented on the top right separately for female raters in the least fertile phase (first value) and most fertile phase (second value) of their menstrual cycle.) odour odour intensity pleasantness sexiness intensity 7 — 036/012 006/042 pleasantness 0.04- 07570.40 sexiness 0.19 0.64" *p < 0.01; two-tailed. (r=0.95, d.f. =14, pSODOI, two-tailed). Therefore, we combined these two variables to produce a mean value of facial attractiveness for further analysis. Ratings of male and female odour pleasantness and sexi- ness did not show such a strong agreement, as shown in table 2. The association between these two variables was not even significant for women in the most fertile cycle phase j udging male body odour. Moreover, whereas female raters in the least fertile phase tended to find an intense odour more unpleasant, women in their most fertile phase rated more intense smells as rather sexier (table 2). To account for possible actual differences in the associations of pleasantness and sexiness 0f odour we used these two adjec- tives of quality separately in the further analysis. Generally, the ratings of odour pleasantness and sexiness given by women in their most and least fertile phases were nega- tively correlated, although not significantly (pleasantness: r: —0.34, n.s.; sexiness: r: —0.37, n.s.). For female subjects, the corpulence index showed a significant negative correlation with both body asymmetry indices (r: —0.59 and r: -0.63, respec- tively, d.f. :17, p < 0.01, two-tailed), attractiveness (r=—0.55, p < 0.05), and in tendency with sexiness of body odour (r: —0.40, n.s.). That means body symmetry increased with higher relative body mass, but facial attractiveness and sexiness of odour decreased. Male corpulence had only slight and non-significant influences on the variables of interest (for all variables: r < 0.3, n.s.). (a) Attractiveness, body and facial asymmetry As shown in table 3, the two different indices of body asymmetry were significantly and positively correlated with each other for both sexes. We found nearly no relationship between body asymmetry and facial attrac- tiveness for female or male subjects (table 3). Female facial asymmetry showed no significant correla- tion with facial attractiveness or body asymmetry, as presented in table 3. In tendency, the relation between facial asymmetry and attractiveness rather pointed in a theory-contradicting direction. That is, the more attractive the woman, the more asymmetrical her face. Male facial asymmetry, however, significantly decreased with increasing facial attractiveness (table 3). Moreover, the relation between facial and body asymmetry was positive, although not significant. Table 3 further shows 872 A. Rikowski and K. Grammer Human body odour, symmetry and attractiveness Table 3. Zero-order (re) and partial correlations (rp) between attractiveness, boa’y asymmetry and facial asymmetry for female and - male subjects controlled for ejects of the corpulence index (Degrees of freedom are presented in parentheses.) female subjects male subjects attractiveness BdAl BdAQ attractiveness BdAl BdA2 r0(17) rP(16) r0(l7) r1) (16) r0(17) rP(16) r0(14-) rp(l3) r0(14-) rP(13) r0(l4-) rp(l3) BdAl“ 0.27 ——0.08 —0.09 —0.07 BdAQb 0.20 —O.21 0.75"M 0.60“ — 0.04- 0.05 0.76“” 0.76"M — — FcA 0.14 0.33 —0.27 —O.l7 —0.22 —0.09 —0.60" —0.64‘ 0.33 0.33 0.30 0.31 2‘ BdAl, body asymmetry measured as index ofear length, ear breadth and foot breadth. b BdA2, body asymmetry measured as index of all measured traits. FcA, facial asymmetry. "p s 0.05; "p S 0.01; "‘1; S 0.001; two-tailed. £- ' a (II b) sexiness of body odour (mean) N A 1 2 3 4 5 6 7 facial attractiveness (mean) Figure 1. Relationship between facial attractiveness and sexiness of body odour for female subjects. The smell of women was rated by men as more sexy, the more attractive their faces. that controlling for the corpulence index by partial correlation left the results for male subjects nearly unchanged, as expected. (b) Female subjects We found remarkable positive correlations between facial attractiveness and quality of body odour for female subjects (table 4-). This effect was most pronounced and reached significance for body odour sexiness, that is, ...
View Full Document

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern