Misleading face-based judgment of cognitive level in intellectual disability: The case of trisomy 21 (Down syndrome)
Introduction
People readily infer traits (e.g., personality traits, intelligence/abilities) from physical – and specifically facial – characteristics (e.g., Ambady and Skowronski, 2008, Fiske and Neuberg, 1990, Hassin and Trope, 2000), which may have important consequences at various levels, from interpersonal interaction to election outcomes (Antonakis & Dalgas, 2009).
Most research on the relationship between social judgments (i.e., social inferences of personality traits and abilities) and variations in facial appearance focused on racial stereotypes. Here, we examine this relationship in the context of trisomy 21 (Down syndrome – hereafter labeled as t21), the most frequent genetic disorder associated with intellectual disability (e.g., Carlier and Ayoun, 2007, Morris and Alberman, 2009, Olsen et al., 2003).
Individuals with t21 form a group characterized by distinct physical and facial features (e.g., round face, oblique palpebral fissures, flat nasal bridge, epicanthal fold, and short broad hands – Lee & Jackson, 1972). As the specific alleles on the extra copy of human chromosome 21 (HSA21) differ for each individual with t21, individual differences in facial dysmorphology arise, whose salience varies from one individual to another. This may drive differences in social inferences about people with this chromosomal disorder.
Research on the “what-is-beautiful-is-good” stereotype (Dion, Berscheid, & Walster, 1972), for example, has shown that physically attractive people are associated with desirable personality traits and are also treated more favorably by others (e.g., Eagly et al., 1991, Feingold, 1992). Babyfaceness is another well-researched example (Zebrowitz, Olson, & Hoffman, 1993). Adults with relatively babyish facial characteristics (e.g., round faces, large and round eyes, small chin, high eyebrows, and thick, pudgy lips) are believed to have more baby-like traits (e.g., weak and naïve) than targets with more mature facial features, and also tend to elicit spontaneously warm and protective responses (for a recent review, see Hugenberg & Wilson, 2013).
Since the first decades of the 20th century, the research on person perception (not restricted to the physical attractiveness or babyfaceness stereotypes) shows that people tend to infer similar traits from given faces, while there is generally no significant correlation between facial features or physiognomic inferences and the traits individuals actually possess (Ambady and Skowronski, 2008, Feingold, 1992, Huguet et al., 2004, Jackson et al., 1995, Kleisner et al., 2014, Richetin et al., 2004, Zebrowitz and Montepare, 2008). Although social inferences from faces generally lead to invalid conclusions (at least in the lack of any information on targets’ actual behavior), the belief in the predictive power of physiognomic information yet is still popular, has a consistent effect on decisions, and creates overconfidence in judgments (Hassin & Trope, 2000).
Because physiognomic information can also be powerful cues to social category membership (e.g., facial features associated to racial categories), category-based stereotyping may ensue (Fiske & Neuberg, 1990). Of particular interest for the present paper, facial feature-based stereotyping may actually operate both between and within categories (Blair et al., 2004a, Blair et al., 2004b, Blair et al., 2002, Maddox, 2004), resulting in greater stereotyping of people with more facial phenotypic features of the target category. Blair et al. (2002), for example, found that people with more Afrocentric facial features were presumed more likely to have personality traits that are stereotypic of African Americans than people with less Afrocentric features. Blair, Judd, and Chapleau (2004) also found that prisoners with more Afrocentric features received more severe sentences than people with less Afrocentric features. In subsequent research (Blair, Judd, & Fallman, 2004), participants seemed to be able to control some aspects of race-based stereotyping, but appeared unaware of and unable to control (within-race) stereotyping based on Afrocentric features. Thus, it seems that individuals who are categorized as members of the same group may be stereotyped to different degrees, depending on the extent to which they manifest their group's facial-phenotypic appearance.
During the first half of the twentieth century, individual differences in physical appearance of “mongoloïd “people have been extensively studied. Because these people are also characterized by large variability in psychological traits such as intelligence (Carr, 1995, Gibson, 1978, Rondal et al., 1999) many attempts were made to find relationships between physical stigmata of “mongoloïd people” and their intelligence and social functioning. Correlations between physical stigmata and IQ were sometimes negative, sometimes positive, and most of them were actually not significant (for reviews see Baumeister and Williams, 1967, Gibson, 1978). Methodological issues and especially the reliability of diagnosis of the individuals included in the studies may explain this stalemate. It must be remembered that the discovery of the extra copy of the HSA21 as the genetic cause of the Down syndrome was published in 1959 (Lejeune, Gauthier, & Turpin, 1959). It took several years before the systematic use of chromosomal analysis to make the diagnosis. It is known now that the sole possible classification of different subtypes of t21 results from the HSA21 triplicated fragments, or from the triplicated allelic forms (see Lyle et al., 2009, Roubertoux and Carlier, 2010). Thus, these earlier research were doomed to failure.
Following a different approach, Strauss, Mintzker, Feuerstein, Wexler, and Rand (1988) assessed the relationships between facial attractiveness and ratings of psychological traits such as intelligence and friendliness. They took photographs of faces of children and young adults with t21 before and after plastic surgery – a procedure intended to reduce facial stigmata. Typical peers judged the two sets of face photographs. The non-operated faces were judged to be less attractive, less friendly, less good-hearted and less intelligent than the faces of typical children. An overall improvement was recorded for all four scores assessing post-operative faces, and showed a linear relationship between the change in the attractiveness rating and the change in the intelligence rating.
Cunningham, Turner, Sloper, and Knussen (1991) explored the possible associations between, on one hand, physical appearance ratings (general and facial appearance) and attractiveness ratings made by the child's classroom teacher and, on the other hand, measures of children's intellectual level, academic attainments, and behavioral problems (among other things). Relying on a large sample of children with t21 (123 families), these authors found no evidence that “appearance had any association with, or influence on, the children's mental or communicative development, behavior problems, independent functioning or social life, nor on parental stress, quality of family life, perceived difficulty of child behavior problems or general satisfaction with life” (p. 291). On this basis, the authors disputed the assumption that the physical characteristics of children with t21 have a negative effect on how these children are perceived and treated by other people.
However, the nine facial items of the facial appearance index used by Cunningham et al. (1991) did not include the two most frequent facial signs of t21: the flat nasal bridge and oblique palpebral fissures (Lee & Jackson, 1972). This index, therefore, falls short of measuring the facial typical characteristics of t21. Furthermore, Cunningham et al. (1991) left open whether the physical and facial appearance of children with t21 influence people's social inferences about them (e.g., their personality and/or intelligence). Here we focused specifically – still in children with t21 – on the relationship between facial-phenotypic appearance of this genetic disorder, social judgments of intelligence, and actual cognitive level. Combining research on face perception and social stereotypes, we expected that facial appearance should drive lay people's inferences about intelligence, so that children perceived as more typical of t21 based on their facial features should be judged less intelligent. Assuming that social inferences from faces generally lead to invalid conclusions we also expected that neither these inferences nor the perception of phenotypic appearance should be related to children's actual cognitive test scores.
Section snippets
Participants
Participants were 80 students from Aix-Marseille University, 50 women and 30 men (M = 20.46, SD = 2.01; age range 18–26 years), who agreed to participate as raters in our research presented as a study on face perception. They were mostly undergraduates from the biological and the “hard sciences” (52.5%), humanities (25%), economics and law (11.3%), and social and human sciences (11.2% – no students from psychology). The current study was approved by the Ethics Committee of University of Provence,
Preliminary analyses
Each photographed children was associated with 13 scores: The raters’ mean score of face typicality, the 10 mean scores of the psychological traits (eight personality traits and two cognitive traits) – see Table 1, and two performance scores (receptive vocabulary and non-verbal reasoning). As expected given the sample of children retained for the present study, all faces obtained a typicality score higher than one (“no t21 features”). More importantly, the inter-rater agreement (inter-rater
Discussion
Dosage imbalance of genes of HSA21 produces an increase in facial developmental instability compared to euploid individuals, giving rise to the characteristic facial appearance associated with the syndrome (Starbuck, Cole, Reeves, & Richtsmeier, 2013). Differences in facial dysmorphology are frequent however and would be perceived. Considering that people readily infer psychological traits from physical and specifically facial characteristics, we expected that children perceived as more typical
Acknowledgments
This research was supported by Aix-Marseille Université, CNRS (LPC UMR 7290), and a grant from Fondation Jérôme Lejeune to M. Carlier.
References (65)
- et al.
An item analysis of Raven's Colored Progressive Matrices among participants with Down syndrome
Research in Developmental Disabilities
(2010) - et al.
A continuum of impression formation, from category-based to individuating processes: Influences of information and motivation on attention and interpretation
Advances in Experimental Social Psychology
(1990) - et al.
Counter-stereotypic beliefs in math do not protect school girls from stereotype threat
Journal of Experimental Social Psychology
(2009) - et al.
Influence of general and specific autobiographical recall on subsequent behavior: The case of cognitive performance
Journal of Experimental Social Psychology
(2011) - et al.
Enumeration skills in Down syndrome
Research in Developmental Disabilites
(2013) - et al.
Stereotype susceptibility in children: Effects of identity activation on quantitative performance
Psychological Science
(2001) - et al.
- et al.
Predicting elections: Child's play!
Science (New York, N.Y.)
(2009) - et al.
Relationship of physical stigmata to intellectual functioning in mongolism
American Journal of Mental Deficiency
(1967) - et al.
The influence of Afrocentric facial features in criminal sentencing
Psychological Science
(2004)
The automaticity of race and Afrocentric facial features in social judgments
Journal of Personality and Social Psychology
The role of Afrocentric features in person perception: Judging by features and categories
Journal of Personality and Social Psychology
A dual process model of impression formation
Power failure: Why small sample size undermines the reliability of neuroscience
Nature Reviews Neuroscience
Déficiences intellectuelles et intégration sociale
Genetic and environmental influences on intellectual disability in childhood
Down's syndrome children growing up
Extending the concept of stereotype threat to social class: The intellectual underperformance of students from low socioeconimic backgrounds
Personality and Social Psychology Bulletin
Stereotype threat undermines intellectual performance by triggering a disruptive mental load
Personality and Social Psychology Bulletin
Is the appearance of children with Down syndrome associated with their development and social functioning?
Developmental Medicine & Child Neurology
What is beautiful is good
Journal of Personality and Social Psychology
Self-theories: Their role in motivation, personality, and development
What is beautiful is good, but…: A meta-analytic review of research on the physical attractiveness stereotype
Psychological Bulletin
Tracking subtle stereotypes of children with trisomy 21: From facial-feature-based to implicit stereotyping
PLoS ONE
Stéréotype social de la trisomie 21 et variables individuelles: sexe, familiarité
Good-looking people are not what we think
Psychological Bulletin
Craniofacial maturity and perceived personality in children with Down syndrome
American Journal on Mental Retardation
Down's syndrome: The psychology of mongolism
Facing faces: Studies on the cognitive aspects of physiognomy
Journal of Personality and Social Psychology
A solution to limitations of cognitive testing in children with intellectual disabilities: The case of fragile X syndrome
Journal of Neurodevelopmental Disorders
Direct and indirect behavioral effects of different genetic disorders of mental retardation
American Journal of Mental Retardation
Faces are central to social cognition
Cited by (5)
A reassessment of Jackson’s checklist and identification of two Down syndrome sub-phenotypes
2022, Scientific ReportsDrug-Targeted Genomes: Mutability of Ion Channels and GPCRs
2022, BiomedicinesEthical dangers of facial phenotyping through photography in psychiatric genomics studies
2019, Journal of Medical Ethics