Beginning in the late 1960s, psychologist Paul Ekman began arguing that certain facial expressions were universal.  He and his colleagues developed cross-cultural experiments that showed how people in very different societies both used their own faces and “read” other people’s faces in the same way.  No matter what their background, people expressed six basic emotions (happiness, sadness, disgust, fear, surprise, and anger) using the same facial muscles; when they saw those muscles come into play on someone else’s face, they were generally able to interpret correctly the emotion the other person was feeling.[1]   Ekman and his colleagues developed and refined FACS, a system for systematically coding the movements of facial muscles, and then EMFACS, a system for interpreting spontaneous displays of emotion, using the movement of facial muscles.[2]  Both systems have been widely used in psychological research, as well as in other contexts. Other coding systems have been developed by other researchers, based on their own theories of emotion and its physical expression.[3]  One of these other systems was used at Cambridge University in the development of Mindreading™, a computer program intended to help children with autism learn to read neurotypical faces in the same way neurotypicals do.  Mindreading™ is the gold standard for this effort, but many other computer programs and phone apps have similar goals.[4]

There is a market for such products because one of the hallmarks of autism is difficulty in interpreting neurotypical people’s emotions.  While Ekman claimed that the ability to read the six basic emotions was universal, most autistic people cannot instinctively do this.  However, many autistic adults have learned– through study and practice–to interpret facial expressions quite well.  In fact, a fascinating series of mystery novels by British author Estelle Ryan is based on this learned ability.  Her autistic protagonist, Genevieve Lenard, has studied psychology (and presumably something like the EMFACS system) and has become so adept at reading the fine details of facial movement and body language that she is employed by an art insurance company to detect people lying in videotaped conversations.  Lenard is the fictional exception, however.  While many autistics can learn to detect the basic emotions in people’s faces, more complex emotions, such as embarrassment, generally elude them.[5]

Autistic struggles with facial recognition have attracted a great deal of attention from psychologists, neurologists and other researchers.  A huge body of scientific literature exists on this phenomenon and its neurological causes.  However, a closely related problem has received no scientific attention at all, despite its inherent interest.  And because this problem has not been studied, we do not know whether the failure of neurotypicals to read autistic faces is because autistic people express emotion using different facial muscles (in which case, Ekman’s claims for “universality” fall apart), or because neurotypicals have a deficit of some sort in the ability to read faces different from their own.

“Flat affect” (or the less severe “blunted” and “restricted affect”) is common among autistic people.  Their faces simply move less than those of neurotypicals; they may talk with stiff lips, without using other parts of the face—in other words, without facial expressions recognizable to neurotypicals.  The latter, however, have a genetic predisposition to expect particular facial expressions during certain kinds of social interactions.   This is true across cultures, at least for the basic emotions. [6]  Someone who does not produce the “right” expression at the right time, or whose face simply remains immobile, is experienced by neurotypicals as somehow “off”—as “weird” or downright “creepy.”  Consider, for example, this interaction between autistic blogger Cynthia Kim and a little boy visiting her house.  The two played together enjoyably in the morning, and the boy wanted to sit next to her at lunch.  But after lunch he told Kim: “you scare me.”  Kim pondered the child’s reaction for several months, and finally concluded that it was the lack of expression on her face that he had found so frightening:  “The technical term for this is flat affect, which means that a person displays reduced emotional expressiveness. It takes a five-year-old to put it in plain English though: you scare me.”[7]

Neurotypicals often see autistic people’s relative facial immobility as reflective of either a sinister masking of emotion, or more commonly as an “unnatural” lack of emotion.  The assumption that autistics don’t feel emotions (or even pain) remains, sadly, commonplace in today’s American society.  Interestingly, this assumption seems particularly common and unquestioned among younger neurotypicals.  A student Prezi presentation on Asperger’s syndrome from April 2016 states unequivocally that “Someone with Asperger’s feels no emotion and does not really care about a lot of anything.”[8]  “Imagine a world where you feel no emotion,” writes another student reviewing Mark Haddon’s The Curious Incident of the Dog in the Nighttime, a novel written from the perspective of an autistic teen (whose emotions are actually described in the novel!).[9]  But the stereotype of autistics as emotionless is also promoted by many adults—including some who really should know better.  Guillermo Sapiro is an engineering professor at Duke University, who recently worked with a professor of Child Psychology to develop a phone app designed to detect warning signs of autism in young children.  In talking to reporters about the app, Sapiro stated that, “Lack of emotion and social sharing are possible characteristics of childhood autism.”[10]  Presumably Sapiro had learned something about autism from his colleague in psychology, and he may have intended to say “lack of emotional expression.”  Nevertheless, his claim that autistic children “lack” emotion was widely repeated in the press, and it feeds the broader cultural stereotype of the emotionless autistic.

Not all autistics display “flat affect,” however.  Some of them “make faces”—that is, they produce facial expressions commonly considered inappropriate (except when produced by young children, playing with other children).  Autistic adults, as well as children, may purse their lips, scrunch up their noses, frown deeply, stick out their tongues, etc., often while making non-verbal noises of one kind or another.  Sometimes, they are doing this on purpose, to express anger or disgust, or (in the case of children) just to be annoying.  In other cases, however, such facial movements may be unplanned and even unnoticed by the person making them.  One little boy got into trouble for “making faces” at his “respite person.”  When his mother asked him why he was doing that, he replied, “Mommy, sometimes I just make faces for no reason. I didn’t know I was making faces at Miss X. I was just making them. Sometimes it just happens and I don’t know why it’s happening. I don’t make them for any reason.”[11]  “Inappropriate” faces are not necessarily intentional or expressive of any particular emotion.  Nevertheless, they often offend others, such as the respite person in this story.

Concerned about this possibility, neurotypical parents often try to make their autistic children at least aware of what they’re doing.  The mother just mentioned tried to explain gently to her son why people might “think he is trying to tell them something with just his face,” even though he didn’t mean to.  She would stop him whenever he moved his face in ways she considered potentially problematic, and ask him whether he was trying to tell her something with his face.  Self-awareness of “making faces” seems like a useful skill for a parent to teach his or her child.  Many other neurotypical parents, however, are more interested in policing their autistic children’s faces, striving to make them appear “normal.”  An online support group for parents of autistic children discussed how to train the kids (with cookies) not to produce the “ugly” faces, but only the ones the parents considered “cute,” or “appropriate.”[12]  Some scientists also want to “normalize” autistic faces:  a professor at Virginia Tech, for example, has developed a computer to teach children with autism not only how to recognize other people’s facial expressions, but also to “reciprocate” them.[13]  I will have more to say about the “normalization agenda” in a future post.

For the moment, however, I would simply note that neurotypicals appear to have just as much trouble reading autistic faces, as autistics do reading neurotypical faces.[14]  They may interpret the expressions on autistic faces incorrectly, as evidence of anger, lack of interest, disgust, mischievous intent, or even insanity.  And they may interpret “restricted,” “blunted” or “flat affect” as evidence that the autistic person feels no emotion at all.  Very little research has been done on this phenomenon, even though it is not only inherently interesting, but also has powerful practical implications for the well-being of autistics living in society.



[1] Paul Ekman, “Universals and Cultural Differences in Facial Expressions of Emotion,” Nebraska Symposium on Motivation, 19 (1971), pp. 207-82; Paul Ekman, Wallace Friesen, et al., “Universals and Cultural Differences in the Judgments of Facial Expressions of Emotion,” Journal of Personality and Social Psychology 53 (1987), 712-17.

[2] Paul Ekman and Erika Rosenberg, What the Face Reveals:  Basic and Applied Studies of Spontaneous Expression Using the Facial Action Coding System (FACS), 2nd ed. (Oxford, 2005), describes the development and use of these systems.

[3] Karsten Wolf, “Measuring Facial Expression of Emotion,” Dialogues in Clinical Neuroscience 17 (2015), 457-62.

[4] The program is distributed by Jessica Kingsley, a publisher who specializes in works on autism:  http://www.jkp.com/mindreading.

[5] Ofer Golan, Jacqueline Hill and Simon Baron-Cohen, “The Cambridge Mindreading (CAM) Face-Voice Battery: Testing Complex Emotion Recognition in Adults with and without Asperger Syndrome,” Journal of Autism and Developmental Disorders 36 (2006), 169-83; see also Ofer Golan,Yana Sinai-Gavrilov, and Simon Baron Cohen, “The Cambridge Mindreading Face-Voice Battery for Children (CAM-C): complex emotion recognition in children with and without autism spectrum conditions,” Molecular Autism 6 (2015), 22.

[6] Karen Schmitt and Jeffrey Cohn, “Human Facial Expressions as Adaptations:  Evolutionary Questions in Facial Expression Research,” American Journal of Physical Anthropology 44 (2001), Supplement:  Yearbook of Physical Anthropology, 3-24—see especially pg. 15 on the social difficulties of those who display flat affect.

[7] “You Scare Me,” on the Musings of an Aspie blog:


[8] https://prezi.com/7nmzsborbhrl/aspergers-syndrome/

[9] http://www.notrequiredreading.com/books/curiousincident.php

[10] https://ssri.duke.edu/news/now-app-could-help-diagnose-autism-children

[11] http://www.modernmom.com/ee91278c-3b3d-11e3-be8a-bc764e04a41e.html

[12] http://www.mdjunction.com/forums/autism-discussions/general-support/1596632-does-any-ones-child-make-faces

[13] https://www.ece.vt.edu/news/ar16/making-faces.php

[14] Rebecca Brewer, Federica Biotti, et al., (“Can Neurotypical Individuals Read Autistic Facial Expressions?  Atypical Production of Emotional Facial Expression in Autism Spectrum Disorders,” accepted for publication in Autism Research ) have shown that autistic individuals produce such idiosyncratic facial expressions that neither neurotypicals nor other people with autism can read them reliably.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s