What gives a face its gender?

Face configuration processing in the human brain: the role of symmetry

Symmetry is an important cue in face perception. We manipulated symmetry and other configurational variables to study their role in face processing in the human brain. We employed 2 types of symmetry: image symmetry (where one part of the image is defined as the mirrored transform of the other part about an axis) and object symmetry (where the spatial relationships among the image components are interpreted as parts of a symmetric 3-dimensional object). We compared blood oxygenation level dependent responses in healthy human observers for upright front-view faces with responses to different symmetry-controlled images. The cortical areas activated by the face images, relative to Fourier-matched scrambled images, were the fusiform (FFA) and occipital (OFA) face areas, the middle occipital gyri (MOG), and areas around the superior temporal and intraoccipital sulci (IOS). Contrasting faces and their image-symmetric scrambled versions showed a similar activation pattern except in the right OFA, suggesting an involvement in facial symmetry processing. The upright versus inverted faces (with the same image symmetry but unfamiliar object identity) showed robust differential activation in the FFA, OFA, MOG, IOS, and precuneus. The response to frontal-view versus 3/4-view faces (having the same object symmetry but disrupted image symmetry) showed little differential activation in the FFA or the OFA but strong responses in the MOG and IOS, suggesting that face processing in the FFA and the OFA is holistic and viewpoint invariant.

How accurately do young children and adults discriminate the gender of natural faces?

The effects of gender-relevant facial features and hairstyles on discrimination of sex of photographed men and women were examined. The stimuli were 36 photographs of natural faces: 9 masculine-males, 9 feminine-males, 9 masculine-females, and 9 feminine-females. Each of the nine faces had one of three hairstyles, hair concealed, short hair, or long hair. 33 children (M age=5:10) and 28 adults were asked to identify the sex of each stimulus. Children had difficulties with cross-gender faces and cross-gender hairstyles, such as a feminine-male face with long hair. Adults' discriminations were accurate except for masculine-female faces, while children's decisions depended both on hairstyle and internal facial features. Children's discriminations might be based both on their perceptual skill in detecting critical visual features and their knowledge of men and women.

Face identification in the near-absence of focal attention

In contrast to artificial geometric shapes, natural scenes and face-gender can be processed even when spatial attention is not fully available. In this study, we investigate whether a finer discrimination, at the level of the individual, is possible in the near-absence of focal attention. Using the paradigm, subjects performed face identification on faces of celebrities and relatively unfamiliar individuals, along with a task that is known to engage spatial attention. We find that face-identification performance is only modestly impaired under dual-task conditions. These results suggest that the visual system is well able to make complex judgments of natural stimuli, even when attention is not fully available.

The role of facial attractiveness and facial masculinity/femininity in sex classification of faces

We tested whether adults (experiment 1) and 4 - 5-year-old children (experiment 2) identify the sex of highly attractive faces faster and more accurately than not very attractive faces in a reaction-time task. We also assessed whether facial masculinity/femininity facilitated identification of sex. Results showed that attractiveness facilitated adults' sex classification of both female and male faces and children's sex classification of female, but not male, faces. Moreover, attractiveness affected the speed and accuracy of sex classification independently of masculinity/femininity. High masculinity in male faces, but not high femininity in female faces, also facilitated sex classification for both adults and children. These findings provide important new data on how the facial cues of attractiveness and masculinity/femininity contribute to the task of sex classification and provide evidence for developmental differences in how adults and children use these cues. Additionally, these findings provide support for Langlois and Roggman's (1990 Psychological Science 1 115 121) averageness theory of attractiveness.

Sex categorization of conspecific pictures in Japanese monkeys (Macaca fuscata)

We investigated whether monkeys discriminate the sex of individuals from their pictures. Whole-body pictures of adult and nonadult monkeys were used as stimuli. Two male Japanese monkeys were trained for a two-choice sex categorization task in which each of two choice pictures were assigned to male and female, respectively. Following the training, the monkeys were presented with novel monkey pictures, and whether they had acquired the categorization task was tested. The results suggested that while monkeys discriminate between the pictures of adult males and females, discrimination of nonadult pictures was difficult. Partial presentations of the pictures showed that conspicuous and sexually characteristic parts (i.e., underbellies including male scrotums or breasts including female nipples) played an important role in the sex categorization.

Is face distinctiveness gender based?

Two experiments were carried out to study the role of gender category in evaluations of face distinctiveness. In Experiment 1, participants had to evaluate the distinctiveness and the femininity-masculinity of real or artificial composite faces. The composite faces were created by blending either faces of the same gender (sexed composite faces, approximating the sexed prototypes) or faces of both genders (nonsexed composite faces, approximating the face prototype). The results show that the distinctiveness ratings decreased as the number of blended faces increased. Distinctiveness and gender ratings did not covary for real faces or sexed composite faces, but they did vary for nonsexed composite faces. In Experiment 2, participants were asked to state which of two composite faces, one sexed and one nonsexed, was more distinctive. Sexed composite faces were selected less often. The results are interpreted as indicating that distinctiveness is based on sexed prototypes. Implications for face recognition models are discussed.

Does inversion abolish the left chimeric face processing advantage?

Experiments using chimeric stimuli have shown that the right hemisphere is more influential in processing facial information. Here, again, we found clear evidence that study participants used the information from the left side of the face to inform their gender decisions when chimeric male/female, female/male stimuli were presented. Most interestingly though, this effect was not only present for upright faces but also for inverted (flipped) faces (although the effect was significantly reduced). We propose that the chimeric bias effects found here argue against the idea that inversion destroys the right hemisphere superiority for faces. If this was indeed the case, flipping the chimeric faces should have resulted in a loss of the left face bias. This was not the case.

Categorizing Others: The Dynamics of Person Construal

A fundamental question in social cognition is whether people categorize others on the basis of the social groups to which they belong. Integrating ideas from related work on face processing, the current research explored the emergence and boundary conditions of person categorization. Using speeded responses to facial stimuli as a marker of category activation, the authors showed in 3 experiments that person categorization: (a) occurs only under active-encoding conditions and (b) does not extend to applicable but task-irrelevant categorical dimensions, but (c) is sensitive to overlap in the perceptual features that support multiple categorical construals. The authors consider the implications of these findings for models of social-cognitive functioning and the component processes that support person perception.

The Perceptual Determinants of Person Construal: Reopening the Social-Cognitive Toolbox

Guided by a heuristic account of social-cognitive functioning, researchers have attempted to identify the cognitive benefits that derive from a categorical approach to person construal. While revealing, this work has overlooked the fact that, prior to the application of categorical thinking as an economizing mental tool, perceivers must first extract category-triggering information from available stimulus cues. It is possible, therefore, that basic perceptual processes may also contribute to people's propensity to view others in a category-based manner. This possibility was explored in 3 experiments in which the authors investigated the ease with which perceivers can extract categorical and identity-based knowledge from faces under both optimal and suboptimal (i.e., inverted faces, blurred faces, rapidly presented faces) processing conditions. The results confirmed that categorical knowledge is extracted from faces more efficiently than identity-related knowledge, a finding that underscores the importance of perceptual operations in the generation of categorical thinking.

Categorizing and Individuating Others: The Neural Substrates of Person Perception

People are remarkably adroit at understanding other social agents. Quite how these information-processing abilities are realized, however, remains open to debate and empirical scrutiny. In particular, little is known about basic aspects of person perception, such as the operations that support people's ability to categorize (i.e., assign persons to groups) and individuate (i.e., discriminate among group members) others. In an attempt to rectify this situation, the current research focused on the initial perceptual stages of person construal and considered: (i) hemispheric differences in the efficiency of categorization and individuation; and (ii) the neural activity that supports these social-cognitive operations.

Noting the greater role played by configural processing in individuation than categorization, it was expected that performance on the former task would be enhanced when stimuli (i.e., faces) were presented to the right rather than to the left cerebral hemisphere. The results of two experiments (Experiment 1—healthy individuals; Experiment 2—split-brain patient) confirmed this prediction. Extending these findings, a final neuroimaging investigation revealed that individuation is accompanied by neural activity in regions of the temporal and prefrontal cortices, especially in the right hemisphere. We consider the implications of these findings for contemporary treatments of person perception.

Sex differences in face gender recognition in humans

Human faces are ecologically-salient stimuli. Face sex is particularly relevant for human interactions and face gender recognition is an extremely efficient cognitive process that is acquired early during childhood. To measure the minimum information required for correct gender classification, we have used a pixelation filter and reduced frontal pictures (28,672 pixels) of male and female faces to 7168, 1792, 448 and 112 pixels. We then addressed the following questions: Is gender recognition of male and female faces equally efficient? Are male and female subjects equally efficient at recognising face gender? We found a striking difference in categorisation of male and female faces. Categorisation of female faces reduced to 1792 pixels is at chance level whereas categorisation of male faces is above chance even for 112 pixel images. In addition, the same difference in the efficiency of categorisation of male and female faces was detected using a Gaussian noise filter. A clear sex difference in the efficiency of face gender categorisation was detected as well. Female subject were more efficient in recognising female faces. These results indicate that recognition of male and female faces are different cognitive processes and that in general females are more efficient in this cognitive task.

Electrophysiological Correlates of Age and Gender Perception on Human Faces

In a previous experiment using scalp event-related potentials (ERPs), we have described the neuroelectric activities associated with the processing of gender information on human faces (Mouchetant-Rostaing, Giard, Bentin, Aguera, & Pernier, 2000). Here we extend this study by examining the processing of age on faces using a similar experimental paradigm, and we compare age and gender processing. In one session, faces were of the same gender (women) and of one age range (young or old), to reduce gender and age processing. In a second session, faces of young and old women were randomly intermixed but age was irrelevant for the task, hence, age discrimination, if any, was assumed to be incidental. In the third and fourth sessions, faces had to be explicitly categorized according to their age or gender, respectively (intentional discrimination). Neither age nor gender processing affected the occipito-temporal N170 component often associated with the detection of physiognomic features and global structural encoding of faces. Rather, the three age and gender discrimination conditions induced similar fronto-central activities around 145–185 msec. In our previous experiment, this ERP pattern was also found for implicit and explicit categorization of gender from faces but not in a control condition manipulating hand stimuli (Mouchetant-Rostaing, Giard, Bentin, et al., 2000). Whatever their exact nature, these 145–185 msec effects therefore suggest, first, that similar mechanisms could be engaged in age and gender perception, and second, that age and gender may be implicitly processed irrespective of their relevance to the task, through somewhat specialized mechanisms. Additional ERP effects were found at early latencies (45–90 msec) in all three discrimination conditions, and around 200–400 msec during explicit age and gender discrimination. These effects have been previously found in control conditions manipulating nonfacial stimuli and may therefore be related to more general categorization processes.

Representation of the gender of human faces by infants: a preference for female

Six experiments based on visual preference procedures were conducted to examine gender categorization of female versus male faces by infants aged 3 to 4 months. In experiment 1, infants familiarized with male faces preferred a female face over a novel male face, but infants familiarized with female faces divided their attention between a male face and a novel female face. Experiment 2 demonstrated that these asymmetrical categorization results were likely due to a spontaneous preference for females. Experiments 3 and 4 showed that the preference for females was based on processing of the internal facial features in their upright orientation, and not the result of external hair cues or higher-contrast internal facial features. While experiments 1 through 4 were conducted with infants reared with female primary caregivers, experiment 5 provided evidence that infants reared with male primary caregivers tend to show a spontaneous preference for males. Experiment 6 showed that infants reared with female primary caregivers displayed recognition memory for individual females, but not males. These results suggest that representation of information about human faces by young infants may be influenced by the gender of the primary caregiver.

Categorical perception of race

Traditionally, research demonstrating categorical perception (CP) has assumed that CP occurs only in cases where natural continua are divided categorically by long-term learning or innate perceptual programming. More recent research suggests that this may not be true, and that even novel continua between novel stimuli such as unfamiliar faces can show CP effects as well. Given this, we ask whether CP is dependent solely on the representation of individual stimuli, or whether stimulus categories themselves can also cause CP. Here, we test the hypothesis that continua between individual faces that cross the categorical boundary between races show an enhanced CP effect. We find that continua running from a black face to a white face do, indeed, show stronger CP effects than continua between two black faces or two white faces. This suggests that CP effects are enhanced when continua run between two distinctly represented individual stimuli, and are further enhanced when those individuals are, in turn, members of different stimulus categories.

Evaluating the independence of sex and expression in judgments of faces

Face recognition models suggest independent processing for functionally different types of information, such as identity, expression, sex, and facial speech. Interference between sex and expression information was tested using both a rating study and Garner's selective attention paradigm using speeded sex and expression decisions. When participants were asked to assess the masculinity of male and female angry and surprised faces, they found surprised faces to be more feminine than angry ones (Experiment 1). However, in a speeded-classification situation in the laboratory in which the sex decision was either "easy" relative to the expression decision (Experiment 2) or of more equivalent difficulty (Experiment 3), it was possible for participants to attend selectively to either dimension without interference from the other. Qualified support is offered for independent processing routes.

Infants' ability to match dynamic phonetic and gender information in the face and voice

Six experiments tested young infants' sensitivity to vowel and gender information in dynamic faces and voices. Infants were presented with side-by-side displays of two faces articulating the vowels /a/ or /i/ in synchrony. The heard voice matched the gender of one face in some studies and the vowel of one face in other studies and, in some studies, vowel and gender were placed in conflict. Infants of age 4.5 months showed no evidence of matching face and voice on the basis of gender, but were able to ignore irrelevant gender information and match on the basis of the vowel. Robust evidence of the ability to match on the basis of gender was not evident until 8 months of age. This set of findings suggests that, when identical stimuli are used, gender matching does not emerge until a later age than does phonetic matching. Results are discussed in relation to key theories of intermodal development.

Discriminating the sex of faces by 6- and 8-mo.-old infants

The goal of this study was to ascertain at what age infants could discriminate male and female faces using only the internal features of the face. The habituation-dishabituation technique was used to estimate infants' discrimination between male and female faces. Analysis showed that 8-mo.-old infants discriminated female and male faces, whereas 6-mo.-old infants did not, but showed an asymmetry in discrimination. 6-mo.-old infants who were habituated to the female face fixated consistently longer on the novel male face in test trials, so sex discrimination was complete but not observed after habituation to male faces. Data are discussed in relation to the role of experience in face discrimination.

Identification of gender in domestic-cat faces with and without training: perceptual learning of a natural categorization task

Three experiments were conducted to determine whether human observers could identify the gender of 40 domestic cats (20 female, 20 male) depicted in individual color photographs. In experiment 1a, observers performed at chance for photographs depicting whole cats, cat heads (bodies occluded), and cat bodies (heads occluded). Experiment 1b showed that chance performance was also obtained when the photographs were full-face close-ups of the cats. Experiment 2a revealed that even with gender-identification training on 30 (15 female, 15 male) of the 40 face close-ups, observers were unable to generalize their training to reliably identify the gender of the 10 remaining test faces (5 female, 5 male). However, experiment 2b showed that gender-identification training with the 14 most accurately identified faces from experiment 1b (7 female, 7 male) was successful in raising gender identification of the 10 test faces above chance. Experiments 3a and 3b extended this facilitative effect of gender-identification training to a population of animal-care workers. The findings indicate that, with appropriate training, human observers can identify the gender of cat faces at an above-chance level. A perceptual category learning account emphasizing the on-line formation of differentiated male versus female prototypes during training is offered as an explanation of the findings.

More about brows: how poses that change brow position affect perceptions of gender

The speeded categorisation of gender from photographs of men's and women's faces under conditions of vertical brow and vertical head movement was explored in two sets of experiments. These studies were guided by the suggestion that a simple cue to gender in faces, the vertical distance between the eyelid and brow, could support such decisions. In men this distance is smaller than in women, and can be further reduced by lowering the brows and also by lowering the head and raising the eyes to camera. How does the gender-classification mechanism take changes in pose into account? Male faces with lowered brows (experiment 1) were more quickly and accurately categorised (there was little corresponding 'feminisation' of raised-brow faces). Lowering gaze had a similar effect, but failed to interact with head lowering in a simple manner (experiment 2). We conclude that the initial classification of gender from the facial image may not involve normalisation of the face image to a canonical state (the 'mug-shot view') for expressive pose (brow movement and direction of gaze). For head pose (relative position of the features when the face is not viewed head-on), normalisation cannot be ruled out. Some perceptual mechanisms for these effects, and their functional implications, are discussed.

An investigation into the perception of dominance from schematic faces: a study using the World-Wide Web

The World-Wide Web (WWW) is considered to be a viable tool for scientific research, and several investigators have already made use of it in their studies. Although the WWW allows researchers to access a vast subject pool, questions of reliability and validity need to be addressed before it is incorporated into mainstream research. By replicating, on the Internet, an existing study (Keating, Mazur, & Segall, 1977) on the perception of schematic faces, we were able to conclude that experimental work carried out in this manner is not necessarily biased by the medium. One difference from previous work was the effect of a smiling versus a nonsmiling face on the perception of dominance, given an identical brow position for the two faces. This was replicated on a different occasion with different subjects, which may represent a shift in attitudes to this facial configuration since the original study was conducted. Young North American males were overrepresented in our sample, but comparison with other surveys indicates that the population sampled by the Internet is becoming more representative.

Presentation-time measures of the effects of manipulations in colour space on discrimination of famous faces

Caricaturing the distinctive shape of famous faces can produce an advantage in reaction-time paradigms but the role of distinctive colour and intensity information in recognition of facial identity has not previously been explored. A presentation-time paradigm was developed by which stimuli could be presented for a range of brief display periods. Subjects were required to identify photo-realistic colour representations of famous faces which either were veridical, were caricatured in colour space, or had enhanced colour saturation and intensity contrast (as contrast controls). Recognition accuracy was greater when viewing the caricatured stimuli than either the veridical images or the contrast controls. The removal of colour information to produce grey-scale images also decreased accuracy of face recognition. Both results indicate that colour information aids differentiation of a class of natural stimuli with similar configurations. Thus it is demonstrated that caricaturing faces can be extended to the colour domain and, as with shape caricaturing, enhancement of distinctive information can produce a recognition advantage for famous faces.

The perception of face gender: the role of stimulus structure in recognition and classification

The perception of face gender was examined in the context of extending "face space" models of human face representations to include the perceptual categories defined by male and female faces. We collected data on the recognizability, gender classifiability (reaction time to classify a face as male/female), attractiveness, and masculinity/femininity of individual male and female faces. Factor analyses applied separately to the data for male and female faces yielded the following results. First, for both male and female faces, the recognizability and gender classifiability of faces were independent--a result inconsistent with the hypothesis that both recognizability and gender classifiability depend on a face's "distance" from the subcategory gender prototype. Instead, caricatured aspects of gender (femininity/masculinity ratings) related to the gender classifiability of the faces. Second, facial attractiveness related inversely to face recognizability for male, but not for female, faces--a result that resolves inconsistencies in previous studies. Third, attractiveness and femininity for female faces were nearly equivalent, but attractiveness and masculinity for male faces were not equivalent. Finally, we applied principal component analysis to the pixel-coded face images with the aim of extracting measures related to the gender classifiability and recognizability of individual faces. We incorporated these model-derived measures into the factor analysis with the human rating and performance measures. This combined analysis indicated that face recognizability is related to the distinctiveness of a face with respect to its gender subcategory prototype. Additionally, the gender classifiability of faces related to at least one caricatured aspect of face gender.

Encoding activity and face recognition

A series of studies conducted over the past 20 years have explored the effects of various tasks on recognition memory for faces. Memory for faces appears better when the study task involves judgements about an abstract trait rather than a physical feature. The various situations in which these results were obtained raise important methodological questions regarding the learning conditions, whether incidental or intentional, and the duration of exposure to the stimulus during the study phase. We consider here two alternative explanations for the reported results. One concerns depth of processing and the other the opposition between component and holistic processing. Possible strategies for improving face recognition performance are considered.

Gender discrimination and prediction on the basis of facial metric information

Horizontal and vertical facial measurements are statistically independent. Discriminant analysis shows that five of such normalized distances explain over 95% of the gender differences of "training" samples and predict the gender of 90% novel test faces exhibiting various facial expressions. The robustness of the method and its results are assessed. It is argued that these distances (termed fiducial) are compatible with those found experimentally by psychophysical and neurophysiological studies. In consequence, partial explanations for the effects observed in these experiments can be found in the intrinsic statistical nature of the facial stimuli used.

Physical attractiveness of girls with gender identity disorder

University students, masked to group status, judged the physical attractiveness of girls with gender identity disorder and clinical and normal control girls, whose photographs were taken at the time of assessment (mean age, 6.6 years). Each student made ratings for all girls for five traits: attractive, beautiful, cute, pretty, and ugly. A multivariate analysis of variance showed a significant group effect. Multiple comparisons of the significant univariate effects showed that the girls with gender identity disorder had significantly less attractive ratings than the normal control girls for the traits attractive, beautiful, and pretty who, in turn, had less attractive ratings than the clinical control girls. Girls with gender identity disorder and the normal controls also had less attractive ratings than the clinical controls for the trait cute. Correlational analyses showed that age was substantially negatively related to the attractiveness ratings in the group of girls with gender identity disorder, but was considerably less so in the two control groups. The extent to which the group differences in attractiveness were due to objective, structural differences in facial attractiveness vs. socially created, or subjective, processes is discussed.

You can tell by the nose--judging sex from an isolated facial feature

Measurements taken from the nose are among the most important physical variables which discriminate statistically between male and female faces, yet several investigators have claimed that it is difficult to judge sex on the basis of noses presented in isolation. Previous work on the isolated nose has, however, involved the use of frontal views only, which may have obscured important physical differences between the noses of males and females. An investigation of the accuracy of judgments of the sex of isolated noses observed in frontal, profile, and three-quarter views by male and female subjects is reported. Judgment of sex was performed significantly more accurately than chance in all cases except for frontal views of female noses, where judgment was significantly less accurate than chance. Analysis of variance demonstrated a significant interaction of sex of nose and view of nose, such that male noses were identified better in frontal and in profile views, but female noses better in the three-quarter view. It is suggested that one possible reason for the seemingly contradictory role of the nose in previous studies of sex judgment is that all noses look more male in frontal views. For a nose to be perceived as female, its distinctive shape must be made available to the perceiver; this is most likely from the three-quarter view.

Judgment of gender through facial parts

Japanese male and female undergraduate students judged the gender of a variety of facial images. These images were combinations of the following facial parts: eyebrows, eyes, nose, mouth, and the face outline (cheek and chin). These parts were extracted from averaged facial images of Japanese males and females aged 18 and 19 years by means of the Facial Image Processing System. The results suggested that, in identifying gender, subjects performed identification on the basis of the eyebrows and the face outline, and both males and females were more likely to identify the faces as those of their own gender. The results are discussed in relation to previous studies, with particular attention paid to the matter of race differences.

Perception of age in adult Caucasian male faces: computer graphic manipulation of shape and colour information

This study investigated visual cues to age by using facial composites which blend shape and colour information from multiple faces. Baseline measurements showed that perceived age of adult male faces is on average an accurate index of their chronological age over the age range 20-60 years. Composite images were made from multiple images of different faces by averaging face shape and then blending red, green and blue intensity (RGB colour) across comparable pixels. The perceived age of these composite or blended images depended on the age bracket of the component faces. Blended faces were, however, rated younger than their component faces, a trend that became more marked with increased component age. The techniques used provide an empirical definition of facial changes with age that are biologically consistent across a sample population. The perceived age of a blend of old faces was increased by exaggerating the RGB colour differences of each pixel relative to a blend of young faces. This effect on perceived age was not attributable to enhanced contrast or colour saturation. Age-related visual cues defined from the differences between blends of young and old faces were applied to individual faces. These transformations increased perceived age.