Visible skin condition and perception of human facial appearance

Authors


Dr Bernhard Fink, Department of Sociobiology/Anthropology, Institute of Zoology & Anthropology, University of Goettingen, Kellnerweg 6, D-37077 Goettingen, Germany. Tel.: +49 551 39 934; fax: +49 551 39 7299; e-mail: bernhard.fink@ieee.org

Synopsis

Evolutionary psychology suggests that certain human beauty standards have evolved to provide reliable cues of fertility and health. Hence, preferences for some physical characteristics of the face and body are thought to reflect adaptations for the promotion of mate choice. Studies that have investigated facial attractiveness have concentrated mainly on features such as symmetry, averageness and sex-typical traits, which are developed under the influence of sex steroids. Few studies, however, have addressed the effect of human skin condition on perception of facial appearance in this context, and possible implications for sexual selection. There is now accumulating evidence that skin pigmentation and skin surface topography cues, particularly in women, have a significant influence on attractiveness judgements, as they seem primarily to signal aspects of age and health. This article (i) reviews briefly some of the main determinants of visible skin condition, (ii) presents recent evidence on its signalling value in face perception and (iii) suggests areas for future research with reference to an evolutionary psychology framework.

Résumé

La psychologie évolutive suggère que certaines normes de beauté de l’homme ont évolué pour fournir des indices fiables de la fécondité et la santé. Par conséquent, les préférences pour certaines caractéristiques physiques du visage et du corps sont censées refléter les adaptations pour la promotion de ce choix. Les études portant sur l’attractivité du visage se sont essentiellement concentrées sur des fonctionnalités telles que la symétrie, l’aspect d’ensemble et le sexe, éléments typiques développés dans le cadre de l’influence des stéroïdes sexuels. Mais peu d’études ont porté sur l’effet de l’état de la peau humaine sur la perception de l’apparence du visage, et ses implications possibles sur la sélection sexuelle. Il y a maintenant un faisceau de preuves montrant que la pigmentation de la peau et l’aspect de surface comme repères topographiques, notamment des femmes, ont une influence significative sur l’attractivité, car ils apparaissent comme des signaux de l’âge et de la santé. Cet article (i) passe brièvement en revue certains des déterminants principaux de l’état cutané visible, (ii) présente les dernières preuves de leur valeur de signalisation dans la perception du visage, et (iii) propose des domaines de recherche future comme référence-cadre de la psychologie évolutive.

Introduction

Variability in visible skin appearance is an important signal and moderator in human social interaction. For example, flushing because of increased blood flow when embarrassed [1] or from sexual arousal [2] facilitates non-verbal communication, as it reflects an individual’s emotional state. Moreover, deliberate modifications of the skin, such as scarification, tattoos and piercing, are used for personal expression, rite of passage or fashion trends (e.g., [3–5]). The significance of the appearance of the skin is also evident in dermatological disorders, which can have a major impact on patients’ daily activities, self-esteem, mental well being and social relationships because of their conspicuous visibility [6, 7]. Interestingly, a study of patients from different social and ethnic groups in South Africa revealed that women were more likely than men to report the effects of skin disease on self-esteem, clothing choice, treatment problems and anxiety [8]. Because of the existence of an attractiveness stereotype [9], public response to skin diseases such as psoriasis, atopic dermatitis and acne often results in prejudice and stigmatization. The origins of such response are thought to be rooted firmly in history and culture because skin disease has long been associated with disgrace and danger in several cultures [10]. Hence, it has been reported that the prevalence of psychiatric and psychosocial co-morbidity, including depressive illness, obsessive-compulsive disorder, social phobia and body dysmorphic disorder in dermatological patients ranges from 25% to 43% [11]. Moreover, in a study by Gupta et al. [12], 6% of patients suffering from psoriasis reported suicidal thoughts.

The cosmetic industry has often conveyed the image that youthful and healthy looking skin, free of blemishes and hair, is most desirable, primarily for women [13]. As the face is of particular importance in human social communication [14, 15] and body image [16], even minor imperfections can have an often disproportionately major impact on mental health and quality of life (QoL). Hence, it is now known that women with facial skin complaints may be particularly prone to depression and an increased risk of suicide [16]. In accordance with previous studies, make-up was found to improve significantly the QoL in women with disfiguring skin disorders [17, 18]. For example, Hayashi et al. [18] reported a significant decrease in the score of the ‘General Health Questionnaire 30’ (GHQ 30) in female acne sufferers after receiving advice on the appropriate use of make-up. The GHQ 30 was originally designed to screen people for psychological or mental health care services, and patients who scored higher than seven were recommended for psychological counselling. Women scored on average >9 on the GHQ 30 before the makeup lessons, whereas afterwards, the score improved to a normal level (<5).

Importantly, unblemished facial skin also drives the positive perception of attractiveness in healthy women without dermatological disease. In a study by Mulhern et al. [19], 10 female volunteers were made up by a beautician to produce five possible scenarios: (1) no make up, (2) foundation only, (3) lip make-up only, (4) eye make-up only, (5) and full facial make-up. The results revealed that faces with full make-up were judged higher on attractiveness than without make-up. Moreover, it was found that attractiveness ratings of certain facial regions were also influenced by the sex of the rater. Whereas women judged eye-make up as having the greatest impact on attractiveness, in men both eye make-up and skin foundation influenced their attractiveness ratings. In accordance with the effect of make-up on mental well being in patients with skin disorders, cosmetic usage also enhances self-esteem in healthy women, as it positively influences both self-perception and perception by others [20–22]. Moreover, Cash et al. [22] reported a systematic relationship between the use of cosmetics and locus of control, suggesting that cosmetic products are used to achieve an attractive appearance. Given that men seem to prefer women with facial make-up [19], it is likely that women may, consciously or unconsciously, use this improvement of visible skin condition to ultimately increase their mating success through the enhancement of facial attractiveness. A more recent study by Nash et al. [23] supports the hypothesis that women can successfully employ cosmetic products to manipulate their appearance. They found that, in contrast to bare skin, Caucasian women who wore makeup were perceived as healthier and more confident, as well as having greater earning potential and more prestigious jobs. Most studies of this nature are conducted with college students or middle-aged women, although the positive effect on self-perception because of the use of facial makeup has also been validated with elderly women aged 60–96 years [24].

Given the importance that humans attach to skin appearance, evolutionary psychologists have proposed that skin condition, particularly that of women, may signal aspects of mate value (e.g. age and health) [25]. Findings that disfiguring skin disorders, such as acne vulgaris and hirsutism (male pattern of body hair in women), reflect elevated androgen levels in blood plasma [26, 27] and may thus have consequences on an individual’s reproductive potential [28], support this hypothesis. Of note, Bunker et al. [29] also showed that 83% of women with acne had polycystic ovaries. The polycystic ovary syndrome is the most common endocrine disorder among women of reproductive age and is regarded as a leading cause of infertility [30, 31].

Evolutionary psychologists are now concerned with visible skin condition not only in an anecdotal, but also in an empirical way. Recent studies provide evidence that facial skin colour distribution and skin surface topography cues provide information about a woman’s age and health [32, 33], both of which are linked to female fecundity [34, 35]. However, not all of the respective concepts in evolutionary psychology have been accessible, or of primary importance to other disciplines, including cosmetic science. Given the common interest in the signalling quality of visible human skin of biology, psychology, and other social and health related sciences, it seems timely to review some of the key components by which skin appearance is determined and discuss recent attempts linking the dermatological knowledge on visible skin condition with evolutionary psychology theories on facial appearance and its perception.

The human skin

Human skin accounts for one-sixth of total body weight [36], and is the largest, independent endocrine organ [37] in the body. It consists of numerous components that give the skin a complex, multi-layered structure. Skin appearance is determined primarily by its colouration and surface topography.

Skin colouration

Human skin colouration shows remarkable variation both within and among human populations. The German anthropologist Johann Friedrich Blumenbach (1752–1840) was the first to propose a human classification based on skin colour variation. He distinguished five ‘races’: (1) pale-skinned Caucasians, living in Europe, in Western Asia up to the Ganges river, and in Northern Africa; (2) yellow-brown-skinned Mongolians, living in Asian regions, which are not occupied by Europeans, and including the Finns, Laplanders, and Inuit; (3) black-skinned Ethiopians, living in all parts of Africa except from the very North; (4) copper-coloured Americans, living in South- and North-America; and (5) dark-brown skinned Malaysians, living on Pacific islands (for full review, see [38]). This classification of the geography of human skin colour, however, has changed over the years [39]. Moreover, the conception of race is often controversial for scientific as well as for social and political reasons, as it leads some societies not only to create myths about people with different skin and/or hair pigmentation [40], but also to the development of racism and ‘colourism’ [41, 42]. In 1950, UNESCO issued a statement on ‘the race question’, which suggested the use of the term ‘ethnic groups’ rather than ‘races’ [43].

Skin pigmentation, leading to skin colour differences among different ethnic groups, is a highly heritable trait [44, 45]. So far, six genes have been identified in the expression of normal pigmentation, and there is evidence that several additional genes may play a role in skin, hair, and iris pigmentation [45–48]. The colouration of human skin is determined by only four major pigments: carotenoids of exogenous origin (yellow) and endogenously produced melanin (brown) in the epidermis, oxygenated haemoglobin (red) and reduced haemoglobin (red-blue) in the capillaries and the venules of the dermis [49–51]. Of these, the chromophores melanin and haemoglobin are the major determinants of differences in skin colour between individuals, with melanin playing the primary role [48, 52, 53]. Interestingly, it has been reported that women differ from men in these two main components of skin colour. That is, the skin of women is generally poorer in melanin and (oxygenated) haemoglobin than that of men [49]. This appears to be a genuine biological sexual dimorphism. In women, lightness of skin colour correlates with the ratio of second finger length to fourth finger length (2D : 4D), a possible biomarker for prenatal sex steroids [54]. Furthermore, women were also reported to have greater concentrations of carotene within their skin [49]. Besides the four chromophores, the structural dermal protein collagen, additionally, contributes to skin colouration by scattering light in the visible spectrum [55].

In humans, melanin can be found in two forms: eumelanin and phaeomelanin. The former chromophore is a black-brown pigment and, because of its location within the skin, hair, and eyes, it is also called ‘cutaneous’ melanin. Phaeomelanin, however, varies in colour from yellow to reddish brown and is also found in the skin and eyes, as well as in hair, while in red hair, it is found in high concentrations [40]. In hair, varying proportions of the two pigments produce a wide range of colours, from the original black to brown, flaxen, golden, and red. Phaeomelanin, however, is less photostable than eumelanin and also occurs in less density [40, 44]. Both pigments are synthesized by melanocytes, which are dendritic cells located in the basal layer of the epidermis. This synthesis process is called melanogenesis and is catalysed by the enzyme tyrosinase [46]. Melanocyte activity is influenced by a variety of stimuli, such as UV-radiation [56, 57], melanocyte stimulating hormones, adrenocorticotropic hormone [58] and corticosterone [53]. After melanin synthesis, melanin chromophores are transferred in discrete packages (melanosomes) along the dendritic processes of the melanocytes and delivered into surrounding keratinocytes by exo- and endocytosis. Once within keratinocytes, melanin species co-localize within membranous structures (reviewed in [40, 50]. Interestingly, the number of melanocytes is actually equal in human ethnic groups, so that the melanin component of skin colour is determined solely by the size and pattern of distribution of melanosomes, the eumelanin and phaeomelanin content within the melanosomes, and metabolic and tyrosinase activity within the melanocytes [40, 44].

East Asians, for instance, have a yellowish complexion because their skin has a higher proportion of phaeomelanin to eumelanin. Moreover, the clustered pigments are organized spherically rather than in an ellipsoidal fashion [59]. Further examples of ethnic differences in skin pigmentation are provided by Alaluf et al. 2002 [60]. These authors revealed that the most light-skinned individuals (e.g. European, Chinese and Mexican), have approximately half the epidermal melanin as the most darkly pigmented (e.g., sub-Saharan African and Indian), but also that the size of melanosomes apparently varies progressively with ethnicity, that is, sub-Saharan African skin having the largest melanosomes followed in turn by Indian, Mexican, Chinese and European. Regardless of ethnicity, however, it should be noted that epidermal melanin fraction volume correlates positively with the average dose of surface solar UV-radiation received at the geographical location of the ethnic group in question (itself largely a function of latitude). Jablonski and Chaplin [61] demonstrated that skin reflectance was most strongly correlated with the quantity of UV-radiation required to produce a barely perceptible reddening of lightly pigmented skin (UVMED), but that some notable variations of UVMED relative to latitude existed, especially in extremely arid and high-altitude environments.

Standardized methods to measure skin pigmentation have been used since the early 20th century. The first attempts were based on colour matching techniques in which an individual’s skin pigmentation was compared with a chromatic scale (e.g., von Luschan’s chromatic scale [62]). These methods were then replaced by reflectance spectrophotometry when portable instruments became available. Early models of such reflectance spectrometry measured the percentage of light reflected from the skin by using up to nine filters corresponding to the different wavelengths of the visible spectrum [48, 49]. The most commonly used instrument was the E. E. L. reflectance spectrophotometer (Evans Electroselenium Ltd., currently distributed by Diffusion Systems, U.K.) (e.g., [63, 64]), which is still used in some studies (e.g., [61, 65]).

Modern, objective in vivo measures of skin colour utilize spectrophotometric or colorimetric techniques and the use of derived colour coordinates such as L*a*b*, and various digital imaging/image analysis methods (for a full review of these approaches, the reader is directed to the review by Pierard [66]). While these measures certainly bring objectivity to the measurement of skin colour, they still are not able to separate the individual contributions of the chromophores responsible for either the measured, integrated remittance spectrum or the final photographic image (no matter how high a quality it may be). Consequently, recent years have seen the development of the so-called ‘Melanin Index’ and ‘Erythema Index’ (MI and EI, respectively), to try and provide a linear, interval data scale for these chromophores [67, 68]. Instruments that derive MI and EI [for example, the Mexameter™ (Courage & Khazaka GmbH, Cologne, Germany), the DermaSpectrometer™ (Cortex Technology, Hadsund, Denmark) and the Erythema/Melanin Meter™ (DiaStron Ltd, Andover, U.K.)] utilize the same basic approach, taking the log of ratios of reflectance within two to three selected wavebands in the visible and infrared. These approaches represent a significant step forward in the quantification of the chromophores responsible for skin colour, but are limited by (a) their limited measurement area (a maximum of approximately 10 mm diameter), (b) their integration over the measured area, with no resolution of spatial distribution, (c) their direct contact with the skin surface (which can lead to artefacts such as blanching through excessive applied probe pressure, etc.) and (d) the inability of the log-ratio method to separate completely, contributions from the two chromophores. As a general comment, these methods fall short because they do not take into account the complex interaction of visible light with, and transport within, human skin, summarized in Fig. 1a, b.

Figure 1.

 (a) Schematic diagram of light transport in human skin; (b) schematic representation of skin surface topography, and its interaction with light, as a `continuum' across a human lifetime.

To address this need, therefore, a new, non-invasive measurement technique has been developed by Cotton and Claridge [69] and later modified by Astron Clinica (Cambridge, U.K.): spectrophotometry intracutaneous analysis (SIA), also called ‘SIAscope’. The SIAscope was developed originally for the early diagnosis of malignant melanoma [70, 71], but has since demonstrated great utility in the measurement of normal skin [72]. This new measurement takes into account light transport within skin, operating on the principle of ‘chromophore mapping’ and determines not only the concentration and distribution of (eu)melanin and (oxy)haemoglobin in vivo (see Fig. 2), but also dermal collagen. Currently, two different SIA measures are available: a hand-held contact probe and a non-contact measure enabling larger acquisition areas (for more details, please see [72, 73]).

Figure 2.

 Non-Contact SIAscopy: (a) original cross-polarised image; (b) corresponding eumelanin greyscale concentration map; (c) corresponding greyscale oxyhaemoglobin map.

Skin surface topography

The surface topography of the skin is basically determined by the structure of the dermis and the mechanical forces imposed on this tissue. It is characterized by regular patterns of intersecting lines and irregular dispersed pilosebaceous follicles and eccrine pores [74]. The main mechanical roles of the human skin are (1) to instantly and/or permanently match dynamic changes in shape and volume of the viscera and adipose tissue, (2) to protect them against external mechanical stress and (3) to facilitate, via palmar and plantar skin, the gripping of objects and negotiation of the surrounding environment [75].

The skin is composed of three primary layers, each with varying mechanical properties: the superficial layer [stratum corneum (SC)], the epidermis and the dermis. These three layers cover the hypodermis (also referred to as subcutaneous adipose layer or subcutis).

The SC is the outermost layer of the skin and constitutes the body’s waterproof, protective integument, varying in thickness from <10 μm on the face, to 10–20 μm on the trunk and limbs, to 200–300 μm on palmar and plantar skin (see the review by Rawlings and Matts [76] for a more detailed discussion of skin structure and function). The SC is composed of corneocytes, tightly stacked flat, polygonal cells approximately 30–50 μm in diameter, containing densely packed keratin protein filaments, surrounded by a tough involucrin-rich protein cell ‘envelope’. The mechanics of native, dry SC are characterized by stiffness and brittleness. Dry SC, therefore, even with an abundance of excess surface area in the form of micro-topography, tends to crack and split so that its barrier function is compromised physically and chemically. Something is needed, therefore, to ‘plasticize’ the SC, to confer suppleness and fluidity of movement. Water is the only endogenous ‘plasticizer’ of the SC in vivo, which is why ‘moisturization’ (both endogenous and supplemented) of this outer layer is of such importance.

The less rigid 50- to 100-μm-thick epidermis is able to conform and flow with overlying SC [75]. As it contains no blood vessels, it is nourished by diffusion of oxygen and nutrients from the dermal vasculature. Keratinocytes and melanocytes form the majority of the cell population of the epidermis. The function of melanocytes has already been described above. Keratinocytes originate in the basal layer (stratum basale) of the epidermis and are the predominant cell type in this compartment (forming approximately 90% of the epidermal cells). Keratinocytes migrate towards the surface in a process of progressive keratinization (differentiation), eventually undergoing a dramatic transformation into the squamous corneocyte cells of the SC. At the SC surface, ‘desquamation’ takes place, a rolling system to replace ‘spent’ surface corneocytes with fresh cells from beneath, equivalent approximately to a layer of surface corneocytes being exfoliated and replaced approximately every day. The exfoliation of corneocytes from the surface of the skin is facilitated by the action of specific hydrolytic enzymes in the SC.

The major mechanical component of the skin is the dermis, which can be several millimetres thick, consisting of connective tissue embedded in an amorphous extracellular matrix. This layer provides cushioning and shock protection to a variety of delicate structures, including blood and lymphatic vessels, hair follicles, sweat glands, sebaceous and apocrine glands, and a variety of mechanoreceptor/nerve endings, which endow the sense of touch and heat. The two fibrous proteins collagen and elastin provide skin with tensile and elastic strength, respectively. With ageing (and particularly photoageing), the quantity and quality of both of these proteins deteriorate, resulting in a loss of tensile strength and elasticity, and, thus, formation of fine lines and wrinkles (particularly in areas of continuous or repeated flexure). The thickness of the hypodermis can range from 1 mm to more than 5 cm, depending on the amount of adipocytes contained within it. Its purpose is to attach the skin to the underlying bone and muscle as well as supplying it with vasculature and innervations. The ageing process causes certain areas of the face to undergo fat atrophy, while in others, adipocytes undergo increased mitosis, resulting in the characteristic fat dysmorphism of senescence [77].

In their review, Piérard et al. [74] describe four types of wrinkles, which result from structural changes in the skin layers elucidated above. Depending on their histological aspects, pathogenesis, orientation and depth they are classified as (1) atrophic, (2) elastotic, (3) expressional and (4) gravitational wrinkles. Atrophic wrinkles are fine, almost parallel lines, which vanish when the skin is put under transversal tension. They occur because of collagen degradation both in the dermis and the hypodermis. Elastotic wrinkles, however, become progressively permanent lines, which do not lessen upon stretching. As a result of an accumulation of abnormal, thickened, tangled, and non-functional fibres of elastin in and around these wrinkles (histologically known as solar elastosis) [78, 79], the skin takes on a characteristic ‘cobblestone’ appearance and becomes significantly more rigid. Expressional wrinkles gradually become permanent furrows, caused by the repeated contraction of the muscles of facial expression (the frown lines, glabellar lines and the ‘crow’s-feet’ being typical examples). Expressional wrinkles are of a lower frequency and higher amplitude than those described previously. Finally, gravitational wrinkles result from folding and sagging of the skin, which has lost its turgidity, under its own weight. The structural changes responsible for these changes are found in the hypodermis.

Numerous objective and non-invasive methods are available to quantify the severity of wrinkles. These methods range from classification according to their visual representation by the use of rating scales, to the accurate replication of skin surface topography with subsequent mechanical/laser profilometry, through to the current state of the art non-contact 3D techniques such as fringe-projection (e.g., [74, 75, 78, 80]). Currently, there are three grading scales that attempt to classify the type and severity of wrinkles: Fitzpatrick’s scale, Glogau’s scale and Hamilton’s scale (for details, see [75]). There is, however, no consensus about the definition of such terms as wrinkles, lines, and furrows, and choice of specific scale or measure is often left to the researcher and his/her particular branch of research.

Skin ageing

The ageing process represents a steady accumulation of cell and tissue change as the result of progressive disorder of regulatory mechanisms and an associated reduction in systemic reserves to counter stress and disease [81]. The human skin undergoes an ageing process in a similar manner as the viscera (at least in non sun-exposed sites) and, thus, represents a unique visible indicator of systemic age [36]. Wrinkling and changes in pigmentation are obvious signs of cutaneous ageing, influenced by both intrinsic and extrinsic factors [82]. However, as both factors interact, it is often difficult to assign phenotypic consequences of the ageing process to one or the other.

Intrinsic ageing

Intrinsic ageing processes are structural changes that occur as natural consequences of ageing and are determined genetically [36]. Consequently, these processes are also seen in most internal organs [83]. Intrinsic ageing refers, in particular, to a decrease in the gonadal production of oestrogen in females (menopause) and testosterone in males (andropause), the adrenal production of the androgen dehydroepiandrosterone (DHEA) and its metabolite DHEA sulphate (DHEAS), and the activity of growth hormone, as well as of the insulin-like growth factor (somatopause) (reviewed in [81]). Lowered secretion of these hormones is thought to be linked to decreased proliferation capacity leading to senescence and altered biosynthetic activity of the skin [83].

Although SC thickness is apparently unaffected by the menopause, deficiencies in neutral lipid neosynthesis have been found which, in turn, affects the barrier/interface function of the skin. Consequently, more time is needed to reconstitute an effective SC barrier following injuries to the skin, superficial or otherwise [84, 85]. Furthermore, lowered oestrogen levels lead to an overall decrease in skin strength and elasticity, because of degenerative changes in collagen and elastin and a progressive atrophy in cutaneous blood supply. In association with this, hyperpigmentation, wrinkling and pallor of the skin may be seen, although at lower intensities compared with skin alterations caused by cumulative photodamage. Alongside these findings in women, lowered testosterone levels result in a decrease of elasticity, extensibility and turgor in male skin. Moreover, appendages such as hair follicles, and apocrine and endocrine glands are apparently decreased in number [81, 86]. Somatopause leads to an uneven distribution of adipose tissue, where certain areas of the face undergo fat atrophy, while others experience a hypertrophy of fat. A decrease in the amount of adipocytes is found primarily in the periorbital, buccal, temporal and perioral areas, as well as on the forehead, whereas an increase in adipocytes is seen in the jowl, lateral nasolabial fold, lateral labiomental crease and lateral malar areas [77, 85].

Extrinsic ageing

Extrinsic ageing is driven by environmental factors, including exposure to solar UV-radiation (e.g., [87–89]), IR-radiation [90]), smoking (e.g., [88, 91, 92]), ozone [89, 93], and dust [89], and involves changes in cellular biosynthetic activity and a progressive disorganization of the dermal matrix [83, 89, 94]. Among these environmental factors, casual exposure to solar UV-radiation is the most potent and prominent driver of so-called ‘premature skin ageing’ [83], causing the production of free radicals or ‘reactive oxygen species’ (ROS) which, by nature, damage virtually every class of cell component, including protein, lipid and nucleic acid [95]. For instance, it is now believed that free radicals can damage the guanine residues that make up 50% of the telomere overhang structure, accelerating telomere shortening and, thus, speeding up premature ageing (e.g., [95, 96]). These reactive species have also been shown to stimulate the degeneration of dermal matrix components, e.g., collagen and elastin, leading to an accumulation of compromised, so-called ‘elastotic’ tissue [89] and associated mechanical failure. ROS are also known to drive a variety of pigmentation disorders, resulting in selective over- (e.g., lentigos, diffuse hyperpigmentation) and under- (e.g. guttate idiopathic hypomelanosis) expression of melanin, causing a progressive visible heterogeneity in melanin distribution (reviewed in [97]).

Because variation in epidermal melanin content and melanosome distribution, pigmentary alterations vary in their severity and manifestation among different ethnic groups (reviewed in [95, 98]). Consequently, the lifetime protection from solar UVR afforded by melanin accounts for smaller differences between sun-exposed sites and sun-protected sites in sub-Saharan African vs. Caucasian skin. Additionally, these darker skin types appear to express certain wrinkle types with less severity and at a greater age than fairer skin [36, 98, 99]. Elastotic wrinkles, for instance, are apparently less prominent in people with darker skin than in Caucasians [74]. Somewhat in contrast to Caucasian skin, it has been proposed that East Asian facial skin tends to age with more emphasis on weaker skeletal support, heavier soft tissue, larger amounts of malar fat, thicker skin, and a weaker chin, driving downward gravitational migration of facial skin tissue [99].

In general, it is thought that the other environmental noxa such as IR-radiation, smoking, ozone, and airborne pollution,act via mechanisms similar to those identified for UV-radiation (see [89]). Interestingly, environmental factors, such as pollution and smoking, seem to produce skin wrinkling, but not pigmentary abnormalities [95].

The effect of human skin condition on face perception

Evolutionary and socialization theory suggests that human facial appearance and attractiveness in particular influence the perception of others in social interactions, as well as development of certain behaviours (e.g. social skills, dating and sexual experience) and traits (e.g. mental and physical health) (reviewed in [100, 101]). Studies have shown that people assign more positive qualities to attractive children and adults than to unattractive ones [101, 102]. In addition, facial attractiveness correlates positively with mating success and, thus, supports the hypothesis that the attractiveness of the face is important in human mate selection [103]. Moreover, people’s view of facial attractiveness seems to be remarkably consistent, regardless of race, nationality, or age (reviewed in [101, 104]). It is, therefore, hardly surprising that humans attach great importance to a beautiful, healthy, and youthful-looking skin. An improvement in facial skin appearance can be accomplished relatively easily through the use of cosmetic products [17, 18, 24]. The visible signs of ageing can also be considerably reduced via injections of Botulinum toxin, the use of dermal fillers, and chemical peels [105]. Despite the highly significant role which humans assign to their facial skin condition, the influence of the biology and associated appearance of skin on mate choice has so far received only little attention in evolutionary psychological research.

Skin colour, natural and sexual selection

The influence of feather and skin patch colouration on sexual attractiveness is known in a wide variety of non-human animals [106]. Studies in avian species particularly have shown that colouration can serve as a secondary sexual ornament, advertising individual quality in terms of physical condition and reproductive potential (e.g., [107, 108]). However, there are also examples in mammals, notably in non-human primates, indicating an honest signal function of skin (e.g., [109–114]). Following these findings, evolutionary psychology suggests that visible skin condition, particularly those of women, may also signal aspects of an individual’s mate quality in humans [115, 116]). Hence, skin pigmentation has been shown to influence judgments of attractiveness in a profound manner.

Geographic variation in skin colour has been attributed to adaptation via natural selection, at least in part [61, 117, 118]. Given that melanin has both UV absorption and ROS scavenging capacity, dark skin is thought to likely be an adaptation to the intensity of ground-level UV-radiation at lower latitudes. Additionally, varying degrees of depigmentation are thought to evolve to permit UV-induced synthesis of previtamin D3 [61, 119]. Contradictory to the natural selection pressure for darker skin, cross-cultural studies indicate male preference for lighter-than-average skin colour in sexual partners [120–122]. In addition to female menstrual cycle-dependent preferences for darker skin complexions in males [121, 123], both preferences lead to a sexual dichromatism in humans [49, 61]. The observed latitudinal gradient in skin colour seems to result from a balance between natural and sexual selection [119]. Sexual selection theory, however, does not presume that sex-differences in skin pigmentation arise exclusively because of sexual selection but, rather, assigns sexual selection a secondary, facultative role. Hence, lighter skin in females is thought to have evolved originally for other reasons and only later became a mate selection criterion for men. Van den Berghe and Frost [120] argued that lighter skin in females arose first by coincidence, e.g. the differing effects of male and female sex hormones on melanin production, causing female skin colouration to fluctuate slightly with the menstrual cycle and being lightest, smoothest and most free of blemishes near ovulation [124, 125]. These authors further suggested that men might then have used this visible signal, subconsciously, to assess a female’s hormonal status and thus, her reproductive potential. Other researchers have argued that lighter skin in women acts as infantile mimicry to lessen aggressiveness in men and to stimulate their provisioning instincts [126], while others assume that women first acquired a lighter skin to facilitate vitamin D synthesis and, thereby, ensure higher calcium reserves for pregnancy and lactation [61]. The different explanations for a preference for lighter skin in women, however, are still controversial [62, 65].

There are also arguments for lighter skin being the most beautiful and desirable because of its association with power, wealth, and privilege due to the persistent effect of European colonization and slavery [42, 63, 127]. Bond and Cash [127], for instance, found a distinct idealization of lightness among female African Americans. They found that, although women were generally satisfied with their skin tone, those who desired a different skin tone favoured being lighter over being darker. They also found that, in contrast to light- and dark-skinned Black women, the ideal skin tone for medium-toned Black women was significantly lighter than their self-perceived colour. Moreover, the majority of respondents believed that Black men consider light skin most attractive. A more recent study by Hunter [42] provides additional support for the effects of skin colour on women’s perceived attractiveness. As hypothesized, skin tone was found to predict educational and income status in both African-American and Mexican-American women. Moreover, lightness of the skin correlated positively with spousal status in African-American women (but not in Mexican-American women).

This hypothesis is challenged by van den Bergh and Frost [120], who stated that there seems to have been a general preference for lighter-than-average female skin in human societies, even before the area of European colonialism in regions that had never been dominated by a lighter-skinned ethnic group. Furthermore, they argued that this preference is also demonstrated in societies where higher-status individuals tend to be darker than the mass of the population. Finally, preference for skin lightness often coincides with rejection of other European physical features, as noted by Wagatsuma [41].

Skin condition as a signal in mate choice

Studies on facial attractiveness have predominantly investigated the influence of symmetry, averageness and sex hormones on facial appearance and its perception (see for review [128–134]), by suggesting that these physical characteristics indicate the underlying genetic quality of an individual (e.g., [128, 135]). However, only a few studies in evolutionary psychology have been concerned with facial skin as an indicator of mate value, even though findings indicate that the signal value of the skin is not solely restricted to sexual dichromatism. When presenting small skin patches (i.e. isolated fields of skin images) extracted from the left and right cheeks of male facial images, Jones et al. [136] found a positive correlation between women’s rating of apparent health and those of attractiveness, this being independent of facial shape information. Further support for an association between skin condition, genetic quality, and perceived attractiveness is provided by Roberts et al. [137]. These authors demonstrated that facial skin patches of men who where heterozygous at all three loci of the major histocompatibility complex were judged by women to be more attractive and healthier than those of men who were homozygous at one or more of these loci.

Neither study, however, investigated the relative contributions of skin colour distribution and surface topography cues to judgments of attractiveness and health. Recent research by Fink et al. [25, 33, 138] aimed to disentangle both determinants of visible skin condition. Using images of 170 British women aged from 11 to 76 years, they found that facial skin colour distribution alone, independent of skin topography and facial shape, significantly influenced the perception of age, with evenness in skin colouration accounting for up to 20 years of perceived age. Furthermore, the authors found a remarkably high correlation between estimated age and judgments of facial attributes, such as attractiveness, health, and youth [138]. Further studies showed that, although both skin colour distribution and skin surface topography cues significantly influence the perception of age and health, they convey differential information with regard to the strength of these effects. That is, skin surface topography is a greater visual cue for an individual’s age, whereas skin colour distribution seems to be a stronger visual cue for health [32] (see Fig. 3). The processes of intrinsic and extrinsic ageing produce changes in both skin colouration and surface topography and, thus, drive perception of both age and health and provide information about an individual’s mate value.

Figure 3.

 Examples of stimulus images from the study of Fink and Matts (2007): (1) original face; (2) topography smoothed; (3) colour smoothed; (4) topography and colour smoothed.

As the face provides important non-verbal information through facial expression and the display of facial form and skin condition, face perception and visual attention towards faces are critical in human social communication [15]. Studies by Langlois et al. [139, 140] and Maner et al. [141] have reported a positive correlation between visual attention and perceived facial attractiveness. In accordance with these studies, Fink et al. [33] were able to link visual attention to perceived age and attractiveness judgements. This study not only revealed that the number of eye fixations and dwell time were positively correlated with skin colour homogeneity, but also that they were negatively correlated with perceived age and positively so with perceived attractiveness [33]. The detrimental effects of facial ageing on attractiveness ratings in women are consistent with other studies reporting that old faces are generally perceived to be less attractive than young faces [142, 143]. The signs of cutaneous ageing, however, do not negatively influence people’s judgments on facial attractiveness in an equal fashion. For instance, Perrett et al. [144] showed that the evaluation of facial attractiveness reflects the learning of parental characteristics. When considering the attractiveness of female images for possible long-term relationships, men whose mothers were older than 30 years of age at their birth were less impressed by visual youth cues than men with younger mothers. However, maternal age did not impact upon judgements of similar images for possible short-term relationships. In women, and in contrast to the findings in men, the age of both parents influenced preferences for men for both possible short-term and long-term relationships. Judgements on same-sex faces (i.e. the judge was the same gender as the subject) revealed decreased attractiveness with increasing age cues [144]. Furthermore, the age of the person surveyed appears to influence attractiveness rating of facial stimuli, that is, older respondents evaluated faces more positively for attractiveness than younger participants [143].

It is reported that female mate value is determined principally by fertility and health, which are both correlated with age [34, 35]. Furthermore, variance in reproductive success is lower for females than for males and they also have a narrow reproductive window compared to males [145]. Selection pressure on males to choose females on the basis of age should be much stronger than in females [34], whereas females should compete with each other for high quality spouses by advertising reproductive potential and exaggerating morphological indicators of youthfulness [115]. With reference to the skin, a more unblemished, relative hairless and smooth skin indicates low androgen and high oestrogen concentrations (reviewed in [36]) and, thus, may signal fertility [25, 125]. In this view, skin wrinkling and changes in skin colour distribution are the most obvious cues that indicate a person’s age. Schneider et al. [146] propose that the quality of skin condition is important in attracting mates during the reproductive phase of human life whereas, after menopause, it is no longer a factor in selection pressure. For instance, the loss of ovarian oestrogen causes a decline in skin collagen content [147] as well as alterations in skin pigmentation [148].

As the face is the most frequently exposed part of the body and always visible to others, facial skin condition might be used as an honest indicator of female mate value. This hypothesis is supported by the findings of Furnham et al. [149], who found facial age to have a more significant influence on the assessment of attributes such as youthfulness, attractiveness, fertility, healthiness, and fecundity, than other oestrogen-dependent body features such as waist-to-hip ratio (WHR).

In contrast, many male characteristics, which seem to be chosen by females, are correlated positively with age (reviewed in [150]). On this basis, women should prefer men of a greater age [151]. However, females do not only choose their mates on the basis of characteristics which advertise adequate parental investment in terms of resource and social status [152], but also on physical attributes [153, 154], which indicate individual quality in terms of body condition and reproductive quality [155, 156]. The study by Jones et al. [136] provides the first evidence that women may also use the condition of skin in male faces to evaluate their health and attractiveness. Both are particularly important when considering an appropriate mate to sire offspring. Although many desired male characteristics are correlated positively with age, previous findings by Adams and Huston [157] as well as Cross and Cross [158] did not support the hypothesis that middle-aged men are considered to be more attractive than middle-aged women (35–55 years). Hence, cutaneous ageing also has detrimental effects on attractiveness ratings in men. Recent studies indicate that the older the putative father is, the higher the chance of passing on genetic defects to his offspring (reviewed in [159]). Moreover, studies have shown that fertility significantly declines with age in men older than 45 (e.g., [160–162]). Given these findings, women should also derive benefit from correctly perceiving the age of a putative mate and almost certainly use signs of cutaneous ageing in the male face to assess his inherent mate value.

Conclusions

Studies in evolutionary psychology investigating facial attractiveness have focused mainly on facial form. Yet, the significance of visible skin condition on the perception of human facial appearance needs to be determined. There is now evidence accumulating from recent research that visible facial skin condition influences perception and attractiveness judgements, and could thus also signal aspects of mate value. Preliminary results indicate that humans are highly sensitive to signs of age such as (even subtle) alterations in skin colour distribution and surface topography, which occur because of intrinsic and extrinsic ageing processes. This sensitivity towards facial age cues is expressed not only through the judgment of facial attributes, such as attractiveness, health and youth but also through altered visual attention [32, 33, 138]. As fertility and health are both negatively correlated with age, facial age may thus provide reliable information on an individual’s mate quality. Furthermore, as female skin colouration changes slightly across the menstrual cycle [124], more detailed research on alterations in both melanin and haemoglobin concentration and distribution in the course of this cycle would be interesting.

As studies investigating human skin condition were conducted primarily with Caucasian women from the United Kingdom, Austria, or Germany, cross-cultural studies are needed to investigate if the putative signal value of skin condition is universal or, rather, depends on socio-ecological conditions. In consideration of ethnic differences in the cutaneous ageing process, attractiveness ratings of non-Caucasians with varying skin surface topography are also needed. Furthermore, there are hardly any published data on the signal value of male skin condition. Indeed, as it has been argued that the appearance of facial skin might be more important in women than in men (because of a relatively narrow reproductive window), most of the studies investigating human skin condition are restricted to female stimuli. However, recent findings also suggest a decrease in fertility with age in men (e.g. reviewed in [159]) and, thus, investigations into the putative impact of age signals in male faces might also be worthwhile.

So far, Jablonski and Chaplin [61] have made significant progress in investigating the adaptive value of skin colour across the world. They conclude that the degree of melanin pigmentation in human skin is an adaptation to solar UV-radiation (see also [163]) but did not, however, quantify cutaneous melanin concentration (or relative concentrations of eu- vs. phaeomelanin). Furthermore, possible differences in the information content of the major chromophores haemoglobin and melanin, or even the two forms of melanin, have not been investigated yet. Studies in barn owls (Tyto alba) have shown that eumelanin-based and phaeomelanin-based traits signal different information relevant for mate choice [164]. Additionally, chronically elevated corticosterone levels during embryonic development result in a decrease in melanin production [53]. Similar changes (i.e., heterogeneous skin colour distribution from birth onwards) in melanin concentrations because of prenatal stress might also be possible in humans.

Finally, given that the ageing process causes an uneven distribution in facial adipose tissue and that facial features such as symmetry and sexual dimorphism strongly influence attractiveness judgements, the interaction of skin topography and facial form deserves special attention in future studies. In closing, we believe that the determinants of skin appearance and perception identified and discussed in this review are of particular interest to the cosmetic scientist, as they are at the heart of the continuing consumer demand for better technologies to address these issues. We believe that a more complete understanding of the mechanisms of skin appearance and the psychology of perception will inevitably improve our ability to identify new and more relevant cutaneous targets and concurrent innovation of better technology.

Acknowledgements

We express our thanks to Nick Neave, who helped with improving an earlier version of the manuscript. Preparation of this article was supported by the Procter & Gamble Company and the German Research Foundation [Deutsche Forschungsgemeinschaft (DFG)] through the Institutional Strategy of the University of Göttingen. B.F. is currently funded by an Emmy-Noether Fellowship of the DFG (FI 1450/4–1).

Ancillary