Caucasian Faces Research Articles

Recognition of own-race and other-race faces by three-month-old infants.

People are better at recognizing faces of their own race than faces of another race. Such race specificity may be due to differential expertise in the two races. In order to find out whether this other-race effect develops as early as face-recognition skills or whether it is a long-term effect of acquired expertise, we tested face recognition in 3-month-old Caucasian infants by conducting two experiments using Caucasian and Asiatic faces and a visual pair-comparison task. We hypothesized that if the other race effect develops together with face processing skills during the first months of life, the ability to recognize own-race faces will be greater than the ability to recognize other-race faces: 3-month-old Caucasian infants should be better at recognizing Caucasian faces than Asiatic faces. If, on the contrary, the other-race effect is the long-term result of acquired expertise, no difference between recognizing own- and other-race faces will be observed at that age. In Experiment 1, Caucasian infants were habituated to a single face. Recognition was assessed by a novelty preference paradigm. The infants' recognition performance was better for Caucasian than for Asiatic faces. In Experiment 2, Caucasian infants were familiarized with three individual faces. Recognition was demonstrated with both Caucasian and Asiatic faces. These results suggest that (i) the representation of face information by 3-month-olds may be race-experience-dependent (Experiment 1), and (ii) short-term familiarization with exemplars of another race group is sufficient to reduce the other-race effect and to extend the power of face processing (Experiment 2).

Same-race faces are perceived more holistically than other-race faces

People are better at recognizing same-than faces but the theoretical explanation of this phenomenon is still controversial. Here we tested the hypothesis that the other-race effect'' is related to a reduced ability to encode configural information on faces. Caucasian and Asian participants had to match whole faces to isolated facial features, or onto whole faces differing by a single feature, on both Caucasian and Asian faces. Participants performed better with whole faces as compared to isolated features, demonstrating a processing'' of faces (Tanaka & Farah, 1993). For Caucasian participants, this was observed only for Caucasian faces. Asian participants who had been living for about a year among Caucasians had a comparable whole/part advantage regardless the race of the faces. These results indicate that same-race faces are processed more holistically than faces, as a result of experience. However, despite processing Caucasian faces as holistically as Asian faces in this paradigm, Asian subjects still presented a large effect. This observation suggests that holistic encoding may be a necessary step in order to be able to recognize faces efficiently, but that it is by no means sufficient to overcome the face effect.

The Impact of Prenatal Alcohol Exposure on Neurophysiological Encoding of Environmental Events at Six Months

Using methodologies from developmental studies on infant information-processing skills, early manifestations of later long-term neurocognitive effects of prenatal alcohol exposure can be explored by assessing primary cognitive processes, including attentional regulation and processing speed. One hundred eighteen 6-month-old infants (18 high risk, 100 low risk) from a longitudinal study of moderate prenatal alcohol exposure and intrauterine growth retardation were presented with both auditory (400 and 1000 Hz pure tones) and visual stimuli (chromatic Caucasian faces) in an information-processing paradigm with cardiac response as the dependent variable. The first three habituation trials were analyzed to assess neurophysiological encoding of environmental events in infants categorized as high and low risk based on status of maternal drinking. Specific indexes of the cardiac response were used to assess the infant's (1). speed of initiating attention, (2). sustained attention, and (3). shifting attention after stimulus offset. Infants identified as high risk based on a cumulative risk index for prenatal alcohol exposure responded more slowly to stimuli and were rated as significantly higher in arousal level across the three trials but did not display differences in their sustained deceleration responses or responses to stimulus-offset. Less efficient neurophysiological encoding was observed among high-risk infants, suggesting that prenatal alcohol exposure may disrupt fundamental components of the attentional system responsible for regulating the body's responses to environmental events. Slower neurophysiological responses involved in encoding environmental stimuli and initiating attention and higher levels of behavioral arousal suggest that these infants may have had difficulties with regulating the interactions between arousal level and the attentional system needed to provide optimal efficiency in processing environmental events. These outcomes suggest that prenatal alcohol exposure results in specific impairments in early attentional regulation, which may influence subsequent cognitive development and behavioral outcomes dependent on these primary cognitive processes.

An encoding advantage for own-race versus other-race faces.

Studies have shown that individuals are better able to recognise the faces of people from their own race than the faces of people from other races. Although the so-called own-race effect has been generally regarded as an advantage in recognition memory, differences in the processing of the own-race versus other-race faces might also be found at the earlier stages of perceptual encoding. In this study, the perceptual basis of the own-race effect was investigated by generating a continuum of images by morphing an East Asian parent face with a Caucasian parent face. In a same/different discrimination task, East Asian and Caucasian participants judged whether the morph faces were physically identical to, or different from, their parent faces. The results revealed a significant race-of-participant by race-of-face interaction such that East Asian participants were better able to discriminate East Asian faces, whereas Caucasian participants were better able to discriminate Caucasian faces. These results indicate that an own-race advantage occurs at the encoding stage of face processing.

Do we all look alike to computers?

People are better able to recognize faces of their own race than those of other races. One explanation is that this ‘other-race effect’ is caused by lifelong experience in which faces from some races are more common than other races. A recent article by Furl et al. tests experience-based accounts of the other-race effect by giving computer algorithms of face recognition training on preponderantly Caucasian faces. The only algorithms that reliably produced other-race effects were those that created face representations distorted to emphasize features that individuated faces.

Evaluation of Morphometric Matching for the Identification of Caucasian Faces from Blurred Images.

The reliability of the morphometric matching method for identifying the blurred images of Caucasian faces was examined experimentally using a computer-assisted facial image identification system. For evaluating the match of the 3D and 2D facial images in the same person (a match), the 3D facial image obtained from 10 subjects was compared to the 2D oblique facial image of the same subject ten times, yielding 100 superimpositions. In the case of different persons (non-matching), the 3D facial images of 10 subjects were each compared to the 2D facial images of other 9 subjects, yielding 90 superimpositions. As a matching criterion, the sum of the values of the average perpendicular difference of the facial outlines and the average point-to-point difference between the corresponding landmarks was used (abbr.: average difference). The range of the average difference was 3.6-6.3 mm for the matching person and 5.8-15.3 mm for non-matching persons, respectively. The average difference and percentage error at the FP/FN crossover point were 6.0 mm and 3.2 percent. In this experiment, it was concluded that the morphometric matching method would be applicable to the identification of blurred images of Caucasian faces.

Proportionality in Asian and North American Caucasian faces using neoclassical facial canons as criteria.

Nine projective linear measurements were taken to determine morphometric differences of the face among healthy young adult Chinese, Vietnamese, and Thais (60 in each group) and to assess the validity of six neoclassical facial canons in these populations. In addition, the findings in the Asian ethnic groups were compared to the data of 60 North American Caucasians. The canons served as criteria for determining the differences between the Asians and Caucasians. In neither Asian nor Caucasian subjects were the three sections of the facial profile equal. The validity of the five other facial canons was more frequent in Caucasians (range: 16.7-36.7%) than in Asians (range: 1.7-26.7%). Horizontal measurement results were significantly greater in the faces of the Asians (en-en, al-al, zy-zy) than in their white counterparts; as a result, the variation between the classical proportions and the actual measurements was significantly higher among Asians (range: 90-100%) than Caucasians (range: 13.3-48%). The dominant characteristics of the Asian face were a wider intercanthal distance in relation to a shorter palpebral fissure, a much wider soft nose within wide facial contours, a smaller mouth width, and a lower face smaller than the forehead height. In the absence of valid anthropometric norms of craniofacial measurements and proportion indices, our results, based on quantitative analysis of the main vertical and horizontal measurements of the face, offers surgeons guidance in judging the faces of Asian patients in preparation for corrective surgery.

An 'other-race' effect in age estimation from faces.

Previous studies have shown that, in person-recognition tasks, people perform better for faces belonging to their own race than for those belonging to another race. Recently, however, this 'other-race' effect has also been found in a sex-discrimination task (O'Toole et al, 1996 Perception 25 669-676). In the present study, we investigated whether this finding extends to age perception. Caucasian and African participants were asked to estimate the age of Caucasian and African faces. The main result of this experiment was a significant 'race of subject' x 'race of face' interaction showing that Caucasian participants performed better at evaluating Caucasian faces than African faces. However, African participants performed equally with both type of faces. This result is explained by the Africans' time of residence in Belgium. The implication of this 'other-race' effect for age estimation is discussed with respect to eyewitness reports.

Facial Morphology and Children's Categorization of Facial Expressions of Emotions: A Comparison Between Asian and Caucasian Faces

The effects of Asian and Caucasian facial morphology were examined by having Canadian children categorize pictures of facial expressions of basic emotions. The pictures were selected from the Japanese and Caucasian Facial Expressions of Emotion set developed by D. Matsumoto and P. Ekman (1989). Sixty children between the ages of 5 and 10 years were presented with short stories and an array of facial expressions, and were asked to point to the expression that best depicted the specific emotion experienced by the characters. The results indicated that expressions of fear and surprise were better categorized from Asian faces, whereas expressions of disgust were better categorized from Caucasian faces. These differences originated in some specific confusions between expressions.

Manipulating Face Gender

Previous research has shown that faces coded as pixel-based images may be constructed from an appropriately weighted combination of statistical "features" (eigenvectors) which are useful for discriminating members of a learned set of images. We have shown previously that two of the most heavily weighted features are important in predicting face gender. Using a simple computational model, we adjusted weightings of these features in more masculine and more feminine directions for both male and female adult Caucasian faces. In Experiment 1, cross-gender face image alterations (e.g., feminizing male faces) reduced both gender classification speed and accuracy for young adult Caucasian observers, whereas same-gender alterations (e.g., masculinizing male faces) had no effect as compared to unaltered controls. Effects on femininity-masculinity ratings mirrored those obtained on gender classification speed and accuracy. We controlled statistically for possible effects of image distortion incurred by our gender manipulations. In Experiment 2 we replicated the same pattern of accuracy data. Combined, these data indicate the psychological relevance of the features derived from the computational model. Despite having different effects on the ease of gender classification, neither sort of gender alteration negatively impacted face recognition (Experiment 3), yielding evidence for a model of face recognition wherein gender and familiarity processing proceed in parallel.

Automatic face recognition in a heterogeneous population

Most face recognition results in the open literature have used Caucasian faces to test their algorithms, while no special mention is made of the other population groups. The effect of a heterogeneous population on face recognition is therefore something of an unknown quantity. It is the aim of this paper to investigate the performance of a face recognition system, with emphasis on the effect of a heterogeneous population. We have found that there is no significant difference in the performance of the system on the different population groups.

African Americans' and Caucasian Americans' Recognition and Likability Responses to African American and Caucasian American Faces

Female African American and Caucasian American participants were shown female faces of their own racial group and/or another racial group. In Experiment 1, participants saw either African American or Caucasian American faces; in Experiment 2, all participants saw faces of both races. Approximately half of the participants received instructions and forms for recognition responses, whereas the other half received instructions and forms for attraction responses. The attraction scores in Experiments 1 and 2 revealed an exposure effect when participants viewed Caucasian American faces but not when they viewed African American faces. There was no evidence of cross-racial bias in Experiment 1. Experiment 2 revealed some evidence of cross-racial bias, in the finding that Caucasian Americans performed more poorly and made more errors in recognition of African American faces than of Caucasian American faces.

Exploring the Structure of Multidimensional Face-space: The Effects of Age and Gender

Four experiments investigate the status of multidimensional face-space models (Valentine, 1991a, 1991b) using Japanese faces and Japanese subjects. Experiments 1a and 1b show how commonsense assumptions about similarities between faces based on age and gender can be accounted for by this class of model. A non-intuitive prediction from the model, namely, that facedness decisions (i.e. deciding a stimulus is a face) should be slowest for adult male faces, is demonstrated in Experiment 2. Experiment 3 replicates this effect using Caucasian faces and Causasian subjects. These results, together with the unexpected discovery that face classification times are also relatively slower for male child faces, are used to explore possible structures of multidimensional face-space.

Recognition of Emotion by Chinese and Australian Children

The performance of Chinese children in Beijing and Australian children in Sydney was compared on two types of facial expression task. Children of 4, 6, and 8 years of age were presented situation discrimination and situation inference/labeling tasks with both Chinese and Caucasian faces. Some evidence was obtained for an ethnic bias effect in emotion recognition from facial expressions. There was no indication that children from a collectivist culture are poorer at recognizing certain emotions than children from an individualistic society. Overall accuracy increased with age for children from both cultures, but the Chinese children were significantly more accurate than the Australians at all ages. These results are discussed in terms of the possible effects of different socialization practices, demographic factors, and suitability of the testing procedures for the two cultures.

Similarities and Variations in Facial Expressions of Emotions: Cross‐cultural Evidence

North American (Canadian) and Indian observers were shown photographs of six facial emotions; happiness, sadness, fear, anger, surprise, and disgust, expressed by American Caucasian and Indian subjects. Observers were asked to judge each photograph, on a 7‐point scale, for the degree of (a) distinctiveness (free from blending with other emotion categories), (b) pleasantness‐unpleasantness, and (c) arousal‐nonarousal of expressed facial emotion. The results showed significant interaction of Observer × Expressor × Emotion for the distinctiveness judgement. It was found that fearful and angry expressions in Indian faces, in comparison to Caucasian faces, were judged as less distinctly identifiable by observers of both cultural origins. Indian observers rated these two emotion expressions as being more distinctive than did North Americans irrespective of the culture of the expressor. In addition, Indian observers judged fearful and angry expressions as more unpleasant than did North Americans. Caucasians, in comparison to Indians, were judged to have more arousal in most of the emotion expressions.

Computer analysis of the African-American face.

An anthropometric study was devised to examine the facial proportions of the African-American. Two hundred African-American subjects between the ages of 18 and 30 years were studied. Objective data were gathered, and standard proportions of the African-American face were obtained. Previous anthropometric studies were performed using direct measurements of facial proportions. Farkas used direct measurements of young adult North American Caucasians and compared them with those of the neoclassical canons. Xeutong evaluated facial proportions of young adults of the Han nationality by direct measurement. The data derived from these studies are compared with our results. Measurements were obtained by standard photographic technique and scanned onto computer diskette. Computer analysis of the objective data was acquired. Thirteen measurements assessing the facial dimensions were used. We observed similar vertical proportions in the African-American face and the Caucasian face. Horizontal dimensions varied significantly between racial groups. Our conclusions will assist the aesthetic surgeon caring for the African-American patient.

Attractiveness of Composite Faces: A Comparative Study

Dataare presented which give cross cultural generality to the observation by Langlois and Roggman (1990) that young southwest American college students found composite faces more attractive than the individual faces from which they were derived. These authors attributed the phenomenon to a cognitive mechanism of prototypicality originating in an evolutionary process of stabilising selection towards facial averageness. In this study New Zealand Caucasian and New Zealand Chinese students, together with indigenous students in China, Nigeria and India chose composite New Zealand Caucasian faces as more attractive than the individual faces from which they were constructed. The preference was greater for female than for male faces. Caution is expressed over attributing the phenomenon to either typicality or stabilising selection.

Structural aspects of face recognition and the other-race effect

The other-race effect was examined in a series of experiments and simulations that looked at the relationships among observer ratings of typicality, familiarity, attractiveness, memorability, and the performance variables of d' and criterion. Experiment 1 replicated the other-race effect with our Caucasian and Japanese stimuli for both Caucasian and Asian observers. In Experiment 2, we collected ratings from Caucasian observers on the faces used in the recognition task. A Varimax-rotated principal components analysis on the rating and performance data for the Caucasian faces replicated Vokey and Read's (1992) finding that typicality is composed of two orthogonal components, dissociable via their independent relationships to: (1) attractiveness and familiarity ratings and (2) memorability ratings. For Japanese faces, however, we found that typicality was related only to memorability. Where performance measures were concerned, two additional principal components dominated by criterion and by d' emerged for Caucasian faces. For the Japanese faces, however, the performance measures of d' and criterion merged into a single component that represented a second component of typicality, one orthogonal to the memorability-dominated component. A measure of face representation quality extracted from an autoassociative neural network trained with a majority of Caucasian faces and a minority of Japanese faces was incorporated into the principal components analysis. For both Caucasian and Japanese faces, the neural network measure related both to memorability ratings and to human accuracy measures. Combined, the human data and simulation results indicate that the memorability component of typicality may be related to small, local, distinctive features, whereas the attractiveness/familiarity component may be more related to the global, shape-based properties of the face.

Gaze motor asymmetries in the perception of faces during a memory task

In 33 male and female adult volunteers, eye position recording were performed by means of an infrared reflection technique. Slides of randomly shuffled black-and-white photographs (7.5 × 10°)of faces and vases were projected for 6 or 20 sec respectively in a visual memory task. In each series, 10 slides of art nouveau vases and of the “inner part” of masked Caucasian faces were used. During recording the head was fixed by a bite-board. 1. (a) For faces the preferred targets of the centre of gaze were the eyes, the mouth and nose region, for vases the contours and some prominent ornaments. 2. (b) Left-right asymmetries in the gaze-movement sampling strategy appeared with faces, but not with vases. In faces, the overall time that the centre of gaze remained in the left half of the field of gaze was significantly longer than in the right half. 3. (c) When, however, the amplitude of the gaze excursions into the left and right halves of the inspected items was taken as a measure and normalized, a preference for the right gaze field was observed. 4. (d) The relative left-right bias during face inspection was stronger with the 6 sec than with the 20 sec inspection period and significantly stronger in female than in male subjects for the 6 sec tasks. 5. (e) Left/right inversion of the face stimuli did not abolish the side bias. Thus the asymmetric sampling strategy when faces were inspected as compared to vases was due to “internal” factors on the part of the subjects. It is hypothesized that a left-right asymmetry in hemispheric visual data processing for face stimuli was the cause of a left-right asymmetry in gaze motor strategies when faces were inspected.

Face perception: Attributions, asymmetries and stereotypes

We investigated the role of ethnic stereotypes and facial asymmetries in attributions to unfamiliar faces. Stereotypicality ratings of adjectives were collected and compared with attributions of those adjectives to left and right hemiface (from the model's perspective) composites of Chinese and Caucasian faces made by Caucasian subjects in a previous study (Rhodes & Lynskey, 1989). Stereotypicality correlated significantly with attribution frequency for other group (Chinese) faces only. Face asymmetries also influenced attributions: happy, cheerful and excitable were selectively attributed to left hemiface composites (for both Chinese and Caucasian faces); cold and cruel were selectively attributed to right hemiface composites (for Caucasian faces only).