Children with Williams Syndrome don’t form racial stereotypes

WILLIAMS Syndrome (WS) is a rare neurodevelopmental disorder caused by the deletion of about 28 genes from the long arm of chromosome 7. It is characterized by mild to moderate mental retardation and “elfin” facial features. Most strikingly, individuals with WS exhibit highly gregarious social behaviour: they approach strangers readily and indiscriminately, behaving as if everybody were their friend. And, according to a study published today in the journal Current Biology, they are the only known group of individuals who do not form racial stereotypes.

Most of us stereotype others implicitly and automatically – we assume that the characteristics of an individual can be generalized to the group to which they belong, whether it be race, religion, social class or some other group. One factor thought to be important in the formation of stereotypes is fear – we give preferential treatment to those perceived to be like us, while being wary of those perceived to be different. Children with WS are socially fearless. Although they perceive some people to be more approachable than others, they apparently have difficulty inhibiting their compulsion towards social interaction.They therefore provide a unique opportunity to examine the contribution of social fear to the formation of stereotypes.

Andreia Santos and her colleagues investigated racial and gender stereotyping in 20 children with WS, aged 7 to 16 years, and 20 healthy controls of about the same age. They used a revised version of Preschool Racial Attitude Measure II, a standardized method which has been shown to reliably assess childrens’ racial and gender attitudes. The test consists of a series of pictures, each of which contains two human figures and is accompanied by a brief story containing either a positive or a negative description. Each child was shown a total of 18 such pictures, and asked each time to point to the person in the picture that the accompanying story was about.

Williams syndrome racial bias.JPG

Typically, children exhibit an “in-group” bias when presented with this test – they tend to attribute positive characteristics to the figures perceived to be of the same ethnic group as themselves, and negative characteristics to the figures perceived to be different. And the children in the control group confirmed this. They showed a strong bias towards their own ethnic group, attributing positive characteristics to the figures with pinkish-tan skin and blonde hair (the “Caucasians”) more often than to those with brown skin and black hair (the “non-Caucasians”). By contrast, no racial bias was evident in the children with WS – they attributed the positive and negative characteristics equally to the Caucasian and non-Caucasian characters. Sex-role bias, however, was identical in the two two groups.

The researchers suggest that it is the absence of social fear in children with WS tha makes them less likely to form racial stereotypes. Their tendency to be hypersociable and friendly towards everyone could lead to their lack of racial bias. Conversely, their failure to stereotype others according to race could contribute to their highly sociable behaviour. On the other hand, preservation of sex-role stereotyping in WS suggests that social fear does not play a big role in this type of stereotype; it may instead occur as a result of other cognitive processes, such as social learning.

Earlier work by the same group of researchers has shown that hypersociability and lack of social fear in individuals with WS is associated with reduced activity in the amygdala in response to social threats, and to reduced interactions between the amygdala and fusiforn face area (FFA). The amygdala is well known to be involved in fear, and the FFA, as its name suggests, responds selectively to faces. These two structures, together with the prefrontal cortex, are normally thought to encode race information, and it has been shown increased FFA activation is associated with viewing same-race faces.

All of this suggests that the apparent lack of racial bias in children with WS occurs because of reduced activity in the amygdala and FFA and impaired interactions between the two, which causes the threat signal normally elicited by someone from a different social group to be diminished. This hypothesis could be investigated further, using “novel” ethnicities to which participants have not previosuly been exposed. More generally, the findings show that the neural mechanisms underlying racial and sex-role stereotypes are distinct from one another, and also that the underlying genetics are different. They also suggest that ways of minimizing social fear could be an effective way of reducing racism and other forms of prejudice.


Santos, A., et al. (2010). Absence of racial, but not gender, stereotyping in Williams syndrome children. Curr. Biol. 20. DOI: 10.1016/j.cub.2010.02.009.

Meyer-Lindenberg, A., et al. (2006). Neural mechanisms in Williams syndrome: a unique window to genetic influences on cognition and behaviour. Nat. Rev. Neurosci. 7: 380-393 [PDF].

Mobbs, D., et al (2006). Frontostriatal Dysfunction During Response Inhibition in Williams Syndrome. Biol. Psychiatry 62: 256-261 [PDF].

About these ads

20 thoughts on “Children with Williams Syndrome don’t form racial stereotypes

  1. So here’s a thought. If a single human child was raised in a place where many different alien creatures lived, with no other humans present, would he/she develop racial stereotypes? Would there need to be other humans for this to develop?
    Just curious…

  2. “…failure to stereotype others according to race could contribute to their highly sociable behaviour. On the other hand, preservation of sex-role stereotyping in WS suggests that social fear does not play a big role in this type of stereotype; it may instead occur as a result of other cognitive processes, such as social learning.”
    This doesn’t make sense. If these kids are so socially “fearless” wouldn’t they be just as likely to socialize with children of the opposite sex as they are to socialize with children of a different race? Similarly, while I agree that sex-roles stereotypes are acquired largely by social learning, this seems to discount that racial biases are also acquired by social learning.

  3. Jenna – then this study implies that there is a difference in the way sex biases and race biases are acquired, eh? That’s the most interesting part.

  4. Were only white children included in the study?
    The non-Williams children apparently had “in-group bias”, but all preferred the “Caucasians” over the “non-Caucasians”. So either (a) there were no non-white children in the study or (b) the non-white children also prefered the “Caucasian” figures and something other than in-group bias is the issue (internalising negative stereotypes against their own race?)

  5. Jenna: You’ve left out the preceding sentence, so your snippet doesn’t make sense because it’s taken out of context.
    Sorcha: Good question. There’s no mention of the childrens’ ethnicity in the paper, and I assumed that they were all white.

  6. Non-whites have been shown to have a preference in general for caucasians in tests like the Harvard IAT. It seems to be a societal bias separate from the in-group bias which would explain why non-white children, if there were any in the control group, still had a preference for the caucasian pictures.

  7. Research using two groups of people who are different in some important way is problematic for making inferences about the effect of some independent variable (fear) on a dependent variable (prejudice). Nonetheless, the research suggests an experiment might be useful.
    One could, for example, show two variations of a movie involving novel ethnic groups (as suggested in the article). One movie would be a horror movie and the other a comedy. Subjects would then rate one of the following: a novel ethnic group involved in the movie prominently involved in the horror or comedy, a novel ethnic group in the movie but not so involved, a novel ethnic group not in the movie, and a real ethnic group not involved in the movie.
    Fear is very difficult to operationalize in an experiment because of limits to what one can be done to subjects.

  8. @Jenna & Janice: Modern psychology has taken it to understand, through studies using infants, that even children possess an instinctive bias to seek out people who have similar physical features to there own. When placed in a room with images of adults of various ethnicities, even an infant will crawl towards the image that looks the most like them. I’m assuming that this instinct must obviously expand as a child becomes able to understand language, and as they point out in the survey:
    “Typically, children exhibit an “in-group” bias when presented with this test – they tend to attribute positive characteristics to the figures perceived to be of the same ethnic group as themselves, and negative characteristics to the figures perceived to be different.”
    So yes, from out earliest days as humans, when discrimination between appearances was a survival technique, we are programmed to look more favorably at our own race. In actually answering your questions: Janice, the child would still probably look favorably upon someone looking similar to them if presented with one, along with another human of different race. And Jenna, I hope this explanation makes a little more sense of this for you.

  9. I’d like to find out more about this supposed “in-group” bias. My kids are of more than one “race”:
    In a room full of people who are each of one of the “races” my kids are “made of,” whom would they go towards?
    How do “multi-racial” kids score on the other tests mentioned in this post?
    Where would I find this out?
    Thanks

  10. Jenna brings up the most important point with this piece in highlighting that social learning is used to explain sex biases but not for racial biases. This is an unfortunate oversight by Santos, Meyer-Lindenberg, and Christine.
    However, the other side of this was also overlooked. That is, neurogenetic underpinnings were used to explain racial biases, but there was no mention of neurobiology’s potential role in the existence of gender biases as well. In all likelihood, both race and gender have a biological or physiological component as well as a social or environmental component that interact in various positive and negative feedback loops giving rise to similar stereotypes with differential levels of intensity and phenomenology.
    Therefore, what needs to be further explored here is not only the potentially unique neural mechanisms underlying gender biases and other evolutionary advantageous social perception tools, but also, the potentially novel biopsychosocial synergies that are responsible for the dominance of certain types of social heuristics over others.
    Also, an interesting follow-up study would be to give these same participants, or a similar pool of WS children, the Implicit Association Test (IAT) to see if the results come out differently, i. e., even more robustly. The reason one might expect this outcome is because the Preschool Racial Attitudes Measure (PRAM II) measures racial biases somewhat explicitly and it is very possible that children may have unknowingly gave more socially desirable answers in this study (which were not very “desirable” at all). Implicit findings may be even more dazzling.

  11. Intriguing that FFA is activated more for people of one’s own race. I wonder if this is the source of the “All the _____ look alike to me” line?

  12. I recently blogged about the physiology behind how xenophobia develops.
    http://daedalus2u.blogspot.com/2010/03/physiology-behind-xenophobia.html
    I see feelings of xenophobia as a “normal” reaction to being unable to understand someone because they don’t share “the same” body language expression of emotion. You are unable to “read” them, and so you default to xenophobia via the uncanny valley effect.
    The thought experiment that Janice suggests has been done with animals, to produce mules, a male donkey (a jack) needs to mate with a female horse. To do that, a mule jack is needed. A mule jack is produced by taking a male donkey foal and have it grow up only associating with horses.
    I suspect that in humans, the source of the sexual stereotypes is different than the source of ethnic stereotypes. There is a period of development where children will avoid opposite gender children of the same age, especially boys. I suspect that this particular trait evolved for male children to avoid being killed by older males due to sexual jealousy.
    I am pretty sure that the development of racial prejudice is all learned. An infant can learn essentially any human language as a first language with no accent.

  13. I think Daedalus2u is exactly right about development of racial prejudice being all learned. We know nothing about the children being tested and what sort of societal influence they have been exposed to between birth and age 7-16. I’m sure that youth with William Syndrome have had a very different societal influence through their experiences between the ages of birth and 7-16. I’m not a scientist so please correct me on where I’m wrong in my critique of this study, but from what I can tell this study seems useless. In fact it appears to give an excuse for people forming racial stereotypes suggesting that it’s in our genes rather than a problem that has been built into the infrastructure of our society and culture.

  14. Chris Phoenix observation about “All the _____ look alike to me” is a good one. In my early 20′s, I had a best friend at work who was black (and I am white). We were both scientists, and eventually developed enough trust to frankly discuss our reactions to getting to know and love someone of a different race. We both agreed that our initial reaction to seeing each other put us “on guard,” but that after several years went by, this effect diminished to be negligible. I have often wondered if one of the best effects of “integration” is to train a part of our brains to recognize and accept another race and have a more “fellow-feeling” about a person of another race.

  15. First of all, I think all of you people who criticises this are correct in doing so. I was even disgusted by the way a Scientific American article phrased the findings of this article as “a genetic link to racism”. Like as one other commentor said “giving an excuse” to people who want to form racial stereotypes.
    Second why were there no non-white kids used? As far as I am concerned all this study has proven is that white kids without Williams syndrome are more racist than white kids with Williams syndrome nothing more! The only thing nice I can say is that at least this blog ends on a positive note in that if we get to the bottom of social fear in our society we can get to the bottom of social issues like racism.

Comments are closed.