Figure - available from: Frontiers in Psychology
This content is subject to copyright.
Illustration of the familiarization/visual-paired comparison (VPC) procedure used is the infant experiment.
Source publication
Previous infant studies on the other-race effect have favored the perceptual narrowing view, or declined sensitivities to rarely exposed other-race faces. Here we wish to provide an alternative possibility, perceptual learning, manifested by improved sensitivity for frequently exposed own-race faces in the first year of life. Using the familiarizat...
Similar publications
Significance
Visual sensitivity is markedly reduced during saccades, and peri-saccadic stimuli are misperceived as being closer to the saccade target. We show here that peri-saccadic shape discrimination improves significantly with practice. Unlike classical training paradigms that lead to better performance despite external masking, this perceptua...
Perceptual learning is typically highly specific to the stimuli and task used during training. However, recently it has been shown that training on global motion can transfer to untrained tasks, reflecting the generalising properties of mechanisms at this level of processing. We investigated if a) feedback was required for learning when using an eq...
Visual perception can be affected by training mental representations. However, it remains unclear if training procedures can also affect the quality of mental representations. To investigate if training enhances the fidelity of mental representations retrieved from visual long-term memory (VLTM), we used a task including object-color associations w...
Peripheral vision is susceptible to crowding; difficulty in segregating a target object from adjacent objects. Crowding limits visual function in individuals with central vision loss who are forced to rely on their peripheral vision. Perceptual learning can reduce crowding in peripheral vision, however intensive training is required. To assess whet...
The predictive coding model of perception proposes that successful representation of the perceptual world depends upon canceling out the discrepancy between prediction and sensory input (i.e., prediction error). Recent studies further suggest a distinction to be made between prediction error triggered by non-predicted stimuli of different prior pre...
Citations
... The discrepancy in the recognition accuracy across races can be seen as the machine version of the well-known crossrace effect in emotion recognition-people are quicker and better at recognizing and interpreting emotional facial expressions in members of their own race than in members of other races (Elfenbein and Ambady, 2002). In humans, such a cross-race effect is not innate but developed through learning (Chien et al., 2016). Specifically, through development, one will learn to pay attention to the statistical regularities of facial features useful for judging the facial expressions of others in daily life. ...
Emotion has been a subject undergoing intensive research in psychology and cognitive neuroscience over several decades. Recently, more and more studies of emotion have adopted automatic rather than manual methods of facial emotion recognition to analyze images or videos of human faces. Compared to manual methods, these computer-vision-based, automatic methods can help objectively and rapidly analyze a large amount of data. These automatic methods have also been validated and believed to be accurate in their judgments. However, these automatic methods often rely on statistical learning models (e.g., deep neural networks), which are intrinsically inductive and thus suffer from problems of induction. Specifically, the models that were trained primarily on Western faces may not generalize well to accurately judge Eastern faces, which can then jeopardize the measurement invariance of emotions in cross-cultural studies. To demonstrate such a possibility, the present study carries out a cross-racial validation of two popular facial emotion recognition systems—FaceReader and DeepFace—using two Western and two Eastern face datasets. Although both systems could achieve overall high accuracies in the judgments of emotion category on the Western datasets, they performed relatively poorly on the Eastern datasets, especially in recognition of negative emotions. While these results caution the use of these automatic methods of emotion recognition on non-Western faces, the results also suggest that the measurements of happiness outputted by these automatic methods are accurate and invariant across races and hence can still be utilized for cross-cultural studies of positive psychology.
... This bias for White faces is consistent with previous studies that have shown that face recognition algorithms have a bias toward faces that are used during training (Cavazos et al., 2020;Tian et al., 2021). This also fits with developmental studies in which the bias toward the recognition of own-race faces increases with experience (Kelly et al., 2005;Chien et al., 2016) and by the fact that the ORE can be reversed or reduced if a person is exposed to another racial group during development (Sangrigoli et al., 2005;Sangrigoli and De Schonen, 2004). We also found that similarity between images was correlated with perceptual judgements of identity, particularly in the later convolutional and fully connected layers. ...
... This greater discrimination of own-race compared to other-race faces is similar to the pattern of results from the DCNN analysis in which there was a greater ability to discriminate race from White faces compared to Asian and Black faces, particularly in the later convolutional layers. This fits with the importance of the role of experience in the representation of faces in humans (Kelly et al., 2005;Chien et al., 2016;Sangrigoli et al., 2005;Sangrigoli and De Schonen, 2004) and DCNNs (Cavazos et al., 2020;Tian et al., 2021). We also found that a DCNN was able to discriminate identity more efficiently for White faces compared to Asian and Black faces, consistent with behavioural studies in White participants (Malpass and Kravitz, 1969;Meissner and Brigham, 2001). ...
Social categories such as the race or ethnicity of an individual are typically conveyed by the visual appearance of the face. The aim of this study was to explore how these differences in facial appearance are represented in human and artificial neural networks. First, we compared the similarity of faces from different races using a neural network trained to discriminate identity. We found that the differences between races were most evident in the fully-connected layers of the network. Although these layers were also able to predict behavioural judgements of face identity from human participants, performance was biased toward White faces. Next, we measured the neural response in face-selective regions of the human brain to faces from different races in Asian and White participants. We found distinct patterns of response to faces from different races in face-selective regions. We also found that the spatial pattern of response was more consistent across participants for own-race compared to other-race faces. Together, these findings show that faces from different races elicit different patterns of response in human and artificial neural networks. These differences may underlie the ability to make categorical judgements and explain the behavioural advantage for the recognition of own-race faces.
... , as indexed by longer response times and lower accuracy for other-race than own-race faces. A corresponding effect has been observed in infants (Anzures et al., 2010;Chien et al., 2016;Ferguson et al., 2009;Hayden et al., 2007;Kelly et al., 2007Kelly et al., , 2009Minar & Lewkowicz, 2018;Sangrigoli & De Schonen, 2004b;Tham et al., 2015), as indexed by less novelty preference (i.e., infants' visual preference for novel over previously seen items) for other-race relative to own-race faces. ...
... The effect could index greater familiarity and salience of own-race faces, in line with the P400 sensitivity to familiar face categories (human vs. monkeys), found at 12 months but not at 6 months of age (de Haan et al., 2002;Halit et al., 2003) and to newly acquired familiarity with faces Scott, Shannon, & Nelson, 2006). A neural ORE at 9 months of age is consistent with a behavioral ORE in 6-to 9-month-olds (Anzures et al., 2010(Anzures et al., , 2011Kelly et al., 2007Kelly et al., , 2009; the absence of a neural ORE at 5 months of age is however inconsistent with evidence for a behavioral ORE as early as 3 months of age (Chien et al., 2016;Hayden et al., 2007;Sangrigoli & De Schonen, 2004b;Tham et al., 2015). This neural pattern supports a perceptual narrowing hypothesis of development of the ORE (Nelson, 2001;Scott et al., 2007), since a neural discrimination of own-and other-race faces was only evident from 9 months of age. ...
Face race influences the way we process faces, so that faces of a different ethnic group are processed for identity less efficiently than faces of one's ethnic group - a phenomenon known as the Other-Race Effect (ORE). Although widely replicated, the ORE is still poorly characterized in terms of its development and the underlying mechanisms. In the last two decades, the Event-Related Potential (ERP) technique has brought insight into the mechanisms underlying the ORE and has demonstrated potential to clarify its development. Here, we review the ERP evidence for a differential neural processing of own-race and other-race faces throughout the lifespan. In infants, race-related processing differences emerged at the N290 and P400 (structural encoding) stages. In children, race affected the P100 (early processing, attention) perceptual stage and was implicitly encoded at the N400 (semantic processing) stage. In adults, processing difficulties for other-race faces emerged at the N170 (structural encoding), P200 (configuration processing) and N250 (accessing individual representations) perceptual stages. Early in processing, race was implicitly encoded from other-race faces (N100, P200 attentional biases) and in-depth processing preferentially applied to own-race faces (N200 attentional bias). Encoding appeared less efficient (Dm effects) and retrieval less recollection-based (old/new effects) for other-race faces. Evidence admits the contribution of perceptual, attentional, and motivational processes to the development and functioning of the ORE, offering no conclusive support for perceptual or socio-cognitive accounts. Cross-racial and non-cross-racial studies provided convergent evidence. Future research would need to include less represented ethnic populations and the developmental population.
... The ORE has its origins in perceptual narrowing in infancy, a process whereby our broad perceptual system is tuned by experience to recognize faces from familiar categories at a cost of recognizing faces from less experienced face categories 45,46 . The emerging ORE might reflect a decline in the ability to discriminate among otherrace faces 7 , a decline that can be postponed by experience [47][48][49] , or a failure to refine learning for other-race faces as learning of own-race faces improves 6 . Our finding that the other-race effect is diminished or absent in White and East Asian adults born in Toronto fits well with both accounts. ...
People often find it more difficult to recognize other- than own-race faces. This other-race effect is robust across numerous ethnic groups. Yet, it remains unclear how this effect changes in people who live in a multiracial environment, and in immigrants whose lifetime perceptual experience changes over time. In the present study, we developed a novel face recognition test that approximates face recognition in the real world. We tested five groups of White and East Asian adults (n=120) living in racially homogeneous vs. heterogeneous cities and East Asians who immigrated to a multiracial city between infancy and adulthood. Multiracial cities reduce the other-race effect. The magnitude of the other-race effect changes as a function of experience, mirroring the racial diversity in perceivers’ living environment. Our study highlights the challenge of forming reliable face representations across naturalistic facial variability and suggests a facilitative role of multiracial environments in eliminating the other-race effect.
... As the social contact questionnaire did not separate past experience from current experience participants had with other-race people, it remains possible that the malleability of ORB is determined by the age at which experience with another racial group begins. This argument is supported by evidence from developmental studies showing that infancy (Liu et al., 2015;Chien et al., 2016;Singarajah et al., 2017) and childhood (Sangrigoli et al., 2005;de Heering et al., 2010;Su et al., 2017;Mckone et al., 2019) are sensitive periods beyond which the effect of experience on face recognition is markedly reduced. Future studies using a different contact questionnaire that distinguishes past from current interracial experience may offer a greater potential for revealing the link between perceptual expertise across development and plasticity of other-race face recognition. ...
The own-race bias (ORB) is a reliable phenomenon across cultural and racial groups where unfamiliar faces from other races are usually remembered more poorly than own-race faces (Meissner and Brigham, 2001). By adopting a yes–no recognition paradigm, we found that ORB was pronounced across race groups (Malaysian–Malay, Malaysian–Chinese, Malaysian–Indian, and Western–Caucasian) when faces were presented with only internal features (Experiment 1), implying that growing up in a profoundly multiracial society does not necessarily eliminate ORB. Using a procedure identical to Experiment 1, we observed a significantly greater increment in recognition performance for other-race faces than for own-race faces when the external features (e.g. facial contour and hairline) were presented along with the internal features (Experiment 2)—this abolished ORB. Contrary to assumptions based on the contact hypothesis, participants’ self-reported amount of interracial contact on a social contact questionnaire did not significantly predict the magnitude of ORB. Overall, our findings suggest that the level of exposure to other-race faces accounts for only a small part of ORB. In addition, the present results also support the notion that different neural mechanisms may be involved in processing own- and other-race faces, with internal features of own-race faces being processed more effectively, whereas external features dominate representations of other-race faces.
... As the social contact questionnaire did not separate past experience from current experience participants had with other-race people, it remains possible that the malleability of ORB is determined by the age at which experience with another racial group begins. This argument is supported by evidence from developmental studies showing that infancy (Liu et al., 2015;Chien et al., 2016;Singarajah et al., 2017) and childhood (Sangrigoli et al., 2005;de Heering et al., 2010;Su et al., 2017;Mckone et al., 2019) are sensitive periods beyond which the effect of experience on face recognition is markedly reduced. Future studies using a different contact questionnaire that distinguishes past from current interracial experience may offer a greater potential for revealing the link between perceptual expertise across development and plasticity of other-race face recognition. ...
The own-race bias (ORB) is a reliable phenomenon across cultural and racial groups where unfamiliar faces from other races are usually remembered more poorly than own-race faces (Meissner and Brigham, 2001). By adopting a yes–no recognition paradigm, we found that ORB was pronounced across race groups (Malaysian–Malay, Malaysian–Chinese, Malaysian–Indian, and Western–Caucasian) when faces were presented with only internal features (Experiment 1), implying that growing up in a profoundly multiracial society does not necessarily eliminate ORB. Using a procedure identical to Experiment 1, we observed a significantly greater increment in recognition performance for other-race faces than for own-race faces when the external features (e.g. facial contour and hairline) were presented along with the internal features (Experiment 2)—this abolished ORB. Contrary to assumptions based on the contact hypothesis, participants’ self-reported amount of interracial contact on a social contact questionnaire did not significantly predict the magnitude of ORB. Overall, our findings suggest that the level of exposure to other-race faces accounts for only a small part of ORB. In addition, the present results also support the notion that different neural mechanisms may be involved in processing own- and other-race faces, with internal features of own-race faces being processed more effectively, whereas external features dominate representations of other-race faces.
... The ORE has its origins in perceptual narrowing in infancy, a process whereby our broad perceptual system is tuned by experience to recognize faces from familiar categories at a cost of recognizing faces from less experienced face categories 45,46 . The emerging ORE might reflect a decline in the ability to discriminate among otherrace faces 7 , a decline that can be postponed by experience [47][48][49] , or a failure to refine learning for other-race faces as learning of own-race faces improves 6 . Our finding that the other-race effect is diminished or absent in White and East Asian adults born in Toronto fits well with both accounts. ...
Adults show an other-race effect (ORE): impaired recognition of other- versus own-race faces1. The ORE manifests as a difficulty recognizing other-race faces across variability in appearance, likely due to asymmetric experience with own- and other-race faces2,3. No study has systematically investigated how the ability to form stable facial representations changes as a function of differential exposure to own- and other-race faces. In Experiment 1, using the Cambridge Face Memory Task (CFMT)4,5, we tested 3 groups of adults (n=132): White and East Asian (EA) adults born and raised in Toronto (highly diverse), and EAs who were born in China and immigrated to Toronto. Whereas White adults and EA immigrants demonstrated a reliable ORE, ps <.005, EAs born in Toronto showed no ORE, p > .080. Notably, age of arrival predicted the magnitude of the ORE (B = .006, p=.001, R2 = 0.23), suggesting that early experience is more effective in reducing the ORE. In Experiment 2, we developed a new version of the CFMT, with face images that capture extensive natural variability in appearance. We tested five adult groups (projected n=120; currently n=71): the three aforementioned groups plus White adults raised in a majority White city in Canada and EAs raised in a majority EA city in Taiwan. We predict that the ORE might be greater in Exp.2 than in Exp.1, highlighting the challenge of forming stable representations of other-race faces. Additionally, the magnitude of the ORE might present a ‘U’ shape: Largest in groups who sit on the end of experience distribution (high own-race and low other-race contact), and smaller in groups in diverse environments or with early exposure to other-race faces. These experiments provide evidence about how perceptual experience shapes adults’ ability to build stable face representations, and highlight the importance of early experience in shaping the ORE.
... Note that infants also need time to acquire the ability to distinguish the identity of faces (e.g. Chien et al. (2016)). Still, what matters here is just that, with more exposure to own-race faces and little exposure to other-race faces, infants cease to have the ability to recognize other-race faces while they retain their ability to recognize own-race faces. ...
Forms of racial cognition begin early: from about 3 months onwards, many human infants prefer to look at own-race faces over other-race faces. What is not yet fully clear is what the psychological mechanisms are that underlie racial thoughts at this early age, and why these mechanisms evolved. In this paper, we propose answers to these questions. Specifically, we use recent experimental data and evolutionary biological insights to argue that early racial cognition is simply the result of a “facial familiarity mechanism”: a mental structure that leads infants to attend to faces that look similar to familiar faces, and which probably has evolved to track potential caregivers. We further argue that this account can be combined with the major existing treatments of the evolution of racial cognition, which apply to (near-) adult humans. The result is a heterogeneous picture of racial thought, according to which early and later racial cognition result from very different psychological mechanisms.
... Studies with infants and children suggest that the ORE is a product of asymmetrical experience with own-and other-race faces during development. The developmental trajectory of the ORE follows a process called perceptual narrowing, through which environmental input tunes our perceptual system towards the predominant visual category (i.e., typically own-race faces) at a cost to less experienced visual categories (i.e., other-race faces; Anzures et al., 2013;Kelly et al., 2007;Sugden, Mohamed-Ali, & Moulson, 2014; also see Chien, Wang, & Huang, 2016 for a perceptual-learning view). In line with this view, Kelly and colleagues found that between 6 and 9 months of age infants' ability to discriminate among own-race faces remains, whereas their ability to discriminate among other-race faces declines (Kelly et al., 2007). ...
Adults show impaired recognition of other-race compared to own-race faces. This other-race effect (ORE) is suggested to be the result of asymmetrical perceptual experience with own- and other-race faces during development. However, it is unclear whether the impact of experience on adults’ ORE differs across development, and whether experience during adulthood can exert similar effects as experience during development. To investigate these questions, we tested face recognition in White adults, East Asian (EA) adults born and raised in Canada, and EA adults who immigrated to Canada at different ages from infancy to adulthood. When recognizing upright faces, White adults and EA immigrants demonstrated a reliable ORE, whereas EA adults born in Canada showed no ORE. These effects were not present when recognizing inverted faces. Notably, age of arrival positively predicts the magnitude of the ORE. Our study highlights the influence of early experience on the ORE and suggests that the ORE appears relatively unmalleable during adolescence and adulthood.
... core knowledge system Spelke, 2004;Spelke & Kinzler, 2007 object agents goal-oriented action number space social partners cooperation reciprocity group cohesion form attend to coalitions Cosmides & Tooby, 2003 in-group out-group race-based preference e.g. Baron & Banaji, 2006 other-race-effect ORE face recognition or discrimination race categorization bias own-race advantage ORA own-race faces other-race faces Meissner & Brigham, 2001 monorace biracial 50% 50% morphing face Peery & Bodenhausen, 2008 40 robust reliable Meissner & Brigham, 2001 in-group vs. out-group 3 ~ 4 Kelly et al., 2005;Kelly et al, 2007aKelly et al, , 2011Chien & Hsu, 2012;Chien, Wang, & Huang, 2016;Hayden, Bhatt, Joseph, & Tanaka, 2007;Sangrigoli & De Schonen, 2004;Tham, Bremner, & Hay, 2015 6 ~ 9 Chien et al., 2016;Chien, 2018;Kelly et al., 2007b;Kelly et al., 2009;Sugden & Marquis, 2017 3 ~ 6 7 ~ 8 race-based social preference 2017 Kinzler & Spelke, 2011Gaither et al., 2014Pauker et al., 2009;Peery & Bodenhausen, 2008 Peery (2) coalitional byproduct hypothesis coalitional psychology Kurzban, 2001 (3) essentialism byproduct hypothesis essentialist inference system Rothbart & Taylor, 1990Gigerenzer, Hoffrage, & Kleinbölting, 1991 The results showed that, as the Asian face component increased, the participants were more likely to categorize the morphed face as Asian. Importantly, for the 50% Asian/White racially ambiguous faces: 7-8 year-olds categorized the faces as Asian whereas 9-10 year-olds, 11-12 year-olds, and adults categorized the face as White. ...
... core knowledge system Spelke, 2004;Spelke & Kinzler, 2007 object agents goal-oriented action number space social partners cooperation reciprocity group cohesion form attend to coalitions Cosmides & Tooby, 2003 in-group out-group race-based preference e.g. Baron & Banaji, 2006 other-race-effect ORE face recognition or discrimination race categorization bias own-race advantage ORA own-race faces other-race faces Meissner & Brigham, 2001 monorace biracial 50% 50% morphing face Peery & Bodenhausen, 2008 40 robust reliable Meissner & Brigham, 2001 in-group vs. out-group 3 ~ 4 Kelly et al., 2005;Kelly et al, 2007aKelly et al, , 2011Chien & Hsu, 2012;Chien, Wang, & Huang, 2016;Hayden, Bhatt, Joseph, & Tanaka, 2007;Sangrigoli & De Schonen, 2004;Tham, Bremner, & Hay, 2015 6 ~ 9 Chien et al., 2016;Chien, 2018;Kelly et al., 2007b;Kelly et al., 2009;Sugden & Marquis, 2017 3 ~ 6 7 ~ 8 race-based social preference 2017 Kinzler & Spelke, 2011Gaither et al., 2014Pauker et al., 2009;Peery & Bodenhausen, 2008 Peery (2) coalitional byproduct hypothesis coalitional psychology Kurzban, 2001 (3) essentialism byproduct hypothesis essentialist inference system Rothbart & Taylor, 1990Gigerenzer, Hoffrage, & Kleinbölting, 1991 The results showed that, as the Asian face component increased, the participants were more likely to categorize the morphed face as Asian. Importantly, for the 50% Asian/White racially ambiguous faces: 7-8 year-olds categorized the faces as Asian whereas 9-10 year-olds, 11-12 year-olds, and adults categorized the face as White. ...
The broadly defined other-race effect (ORE) refers to differential processing for own- and other-race faces, such as own-race face recognition advantage and categorization bias for racially ambiguous faces. The present study adopted bi-racial (East Asian and Caucasian) morph face images as stimuli, aiming to explore the development of race categorization in Taiwanese school-age children and adults. In Experiment 1, we tested 33 adults (17 females) on their race categorization of Asian/White morphed faces. In Experiment 2, we tested 65 school-age children (34 girls), divided into three age groups, 7-8 year-olds (N = 21), 9-10 year-olds (N = 22), and 11-12 year-olds (N = 22), with the same task. The morphed face stimuli contained 11 levels (from A100/C0 (100% Asian) to A0/C100 (100% White) in 10% increment); both children and adults were asked to categorize each of the morphed faces as either Asian or Caucasian. The results showed that, as the Asian face component increased, the participants were more likely to categorize the morphed face as Asian. Importantly, for the 50% Asian/White racially ambiguous faces: 7-8 year-olds categorized the faces as Asian whereas 9-10 year-olds, 11-12 year-olds, and adults categorized the face as White. Moreover, the reaction times for the morphed faces of nearly equal Asian and White components (e.g., A50/C50 or A60/C40) were the longest among all, indicating that those faces were rather difficult to categorize. Lastly, we adopted curve fitting (using a 4-parameter sigmoidal function) to estimate individuals’ threshold of the Asian component for categorizing as “Asian.” The adults’ group mean threshold was 56.01%; the group mean thresholds of 7-8, 9-10, and 11-12 year-olds were 49.52%, 54.27%, and 53.13%, respectively, showing a tendency of increasing threshold from age 7 to 12. In summary, our findings provide a cross-cultural comparison on the development of race categorization in school-age children.
