Skip to main content

Knowing Me, Knowing You: How Social Intuition Goes Awry in Autism

How social intuition goes awry in individuals who have autism

AT THE END of Casablanca, when Humphrey Bogart finally tells Ingrid Bergman to get on the plane back to her husband, the young mother watching the afternoon TV movie sheds a tear. Instinctively, her two-year-old tries to comfort her by offering his teddy bear to her. Both the mother and child are displaying intuitive awareness of others’ mental states and emotions.

Social intuition comes naturally to most of us, but not all. Autism is a developmental disorder that affects around one in 500 individuals (although this figure appears to be on the rise and depends largely on how you define it). In general, autism can be thought of as a disorder with three major disabilities: a profound lack of social skills, poor communication and repetitive behaviors. It is regarded as a spectrum disorder because it covers a broad range and individuals vary in the extent to which they are affected. All those with the disorder share problems with social intuition, however.

Individuals with autism have a problem with socializing because they lack a repertoire of developmental social skills that enable humans to become expert mind readers. Not mind reading in the way Spock from Star Trek could do, but rather the capacity to infer what others are thinking in different circumstances. Over the course of early childhood typical youngsters increasingly become more sophisticated at understanding that other people have mental states that motivate their behavior. For example, if you leave your bag in the office, then I know that you believe it to be there even though the cleaner has handed it in to lost and found. I can understand you hold a false belief. This ability is called having a “theory of mind,” and it is a natural ability in typical children. By the time the average child is around four years old, he or she interprets other people as being goal-directed and purposeful and as having preferences, desires, beliefs and even misconceptions. Without this repertoire of social skills, a human is effectively mind blind—unable to understand what others are thinking and why they do the things they do.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Not only do typical children become intuitive mind readers, but they also become agony aunts as well. They begin to understand others’ sadness, joy, disappointment and jealousy as emotional correlates of the behaviors that make humans do the things they do. Again, by four years of age, children have become expert at working the social arena. They will copy, imitate, mimic and generally empathize with others, thereby signaling that they, too, are part of the social circles that we all must join to become members of the tribe. They share the same socially contagious behaviors of crying, yawning, smiling, laughing and pulling disgusted faces that signal they share the same emotional experiences of those around them.

Baffled by Behavior
No wonder individuals with autism find direct social interaction frightening. If you cannot figure out other people, then such interaction must be intensely baffling and stressful. They often do not like direct eye contact, do not prefer to look at faces compared with other things, do not copy, do not mimic, do not yawn when others yawn or retch when others retch, or laugh or join in with the rich tapestry of social signals we share as a species. This inability may be why individuals with autism generally withdraw into activities that do not involve other people.

The incidence of autism is higher in identical twins, who share nearly 100 percent of their genes, compared with fraternal twins, who share only 50 percent, which indicates that there is a genetic component to the disorder. Also, the greater incidence in males compared with females strongly implicates a biological basis. To date, tantalizing evidence exists based on brain-imaging studies that regions in the prefrontal cortex—most notably the frontoinsular and the anterior cingulate cortex, which are activated by social interaction in normal individuals—are relatively inactive in individuals with autism. Autopsy data also indicate that the frontoinsular and the anterior cingulate cortex structures are abnormal in autism disorder.

At the end of Casablanca, when Humphrey Bogart finally tells Ingrid Bergman to get on the plane back to her husband, the young mother watching the afternoon TV movie sheds a tear. Instinctively, her two-year-old tries to comfort her by offering his teddy bear to her. Both the mother and child are displaying intuitive awareness of others’ mental states and emotions.

Social intuition comes naturally to most of us, but not all. Autism is a developmental disorder that affects around one in 500 individuals (although this figure appears to be on the rise and depends largely on how you define it). In general, autism can be thought of as a disorder with three major disabilities: a profound lack of social skills, poor communication and repetitive behaviors. It is regarded as a spectrum disorder because it covers a broad range and individuals vary in the extent to which they are affected. All those with the disorder share problems with social intuition, however.

Individuals with autism have a problem with socializing because they lack a repertoire of developmental social skills that enable humans to become expert mind readers. Not mind reading in the way Spock from Star Trek could do, but rather the capacity to infer what others are thinking in different circumstances. Over the course of early childhood typical youngsters increasingly become more sophisticated at understanding that other people have mental states that motivate their behavior. For example, if you leave your bag in the office, then I know that you believe it to be there even though the cleaner has handed it in to lost and found. I can understand you hold a false belief. This ability is called having a “theory of mind,” and it is a natural ability in typical children. By the time the average child is around four years old, he or she interprets other people as being goal-directed and purposeful and as having preferences, desires, beliefs and even misconceptions. Without this repertoire of social skills, a human is effectively mind blind—unable to understand what others are thinking and why they do the things they do.

Not only do typical children become intuitive mind readers, but they also become agony aunts as well. They begin to understand others’ sadness, joy, disappointment and jealousy as emotional correlates of the behaviors that make humans do the things they do. Again, by four years of age, children have become expert at working the social arena. They will copy, imitate, mimic and generally empathize with others, thereby signaling that they, too, are part of the social circles that we all must join to become members of the tribe. They share the same socially contagious behaviors of crying, yawning, smiling, laughing and pulling disgusted faces that signal they share the same emotional experiences of those around them.

Baffled by Behavior
No wonder individuals with autism find direct social interaction frightening. If you cannot figure out other people, then such interaction must be intensely baffling and stressful. They often do not like direct eye contact, do not prefer to look at faces compared with other things, do not copy, do not mimic, do not yawn when others yawn or retch when others retch, or laugh or join in with the rich tapestry of social signals we share as a species. This inability may be why individuals with autism generally withdraw into activities that do not involve other people.

The incidence of autism is higher in identical twins, who share nearly 100 percent of their genes, compared with fraternal twins, who share only 50 percent, which indicates that there is a genetic component to the disorder. Also, the greater incidence in males compared with females strongly implicates a biological basis. To date, tantalizing evidence exists based on brain-imaging studies that regions in the prefrontal cortex—most notably the frontoinsular and the anterior cingulate cortex, which are activated by social interaction in normal individuals—are relatively inactive in individuals with autism. Autopsy data also indicate that the frontoinsular and the anterior cingulate cortex structures are abnormal in autism disorder.

John Allman of the California Institute of Technology thinks that much of this social deficit may come down to a lack of a special class of spindle neurons, sometimes called Von Economo neurons after their discoverer, who made the observation in 1925. Spindle neurons consist of a very large bipolar neuron that is found only in the frontoinsular and anterior cingulate cortex and thought to provide the interconnection between brain regions that are activated by social learning. This location may explain why spindle neurons have been found solely in species that are particularly social, including all the great apes, elephants, and whales and ­dolphins.

Humans have the biggest population of spindle neurons located in the frontoinsular and anterior cingulate cortex areas—the same regions that may be disrupted in autism spectrum disorder. Spindle neurons are thought to work by keeping track of social experiences, leading to a rapid appreciation of similar situations in the future. They provide the basis of intuitive social learning when we watch and copy others. It may be no coincidence that the density of spindle neurons in these social regions increases from infancy to reach adult levels somewhere around the fourth birthday in typical children, the watershed when most child development experts agree that there is noticeable change in social intuition skills. This may also explain why individuals with autism, who have disrupted frontoinsular and anterior cingulate cortical areas, have difficulty working out what the rest of us just know without having to think very much.

John Allman of the California Institute of Technology thinks that much of this social deficit may come down to a lack of a special class of spindle neurons, sometimes called Von Economo neurons after their discoverer, who made the observation in 1925. Spindle neurons consist of a very large bipolar neuron that is found only in the frontoinsular and anterior cingulate cortex and thought to provide the interconnection between brain regions that are activated by social learning. This location may explain why spindle neurons have been found solely in species that are particularly social, including all the great apes, elephants, and whales and ­dolphins.

Humans have the biggest population of spindle neurons located in the frontoinsular and anterior cingulate cortex areas—the same regions that may be disrupted in autism spectrum disorder. Spindle neurons are thought to work by keeping track of social experiences, leading to a rapid appreciation of similar situations in the future. They provide the basis of intuitive social learning when we watch and copy others. It may be no coincidence that the density of spindle neurons in these social regions increases from infancy to reach adult levels somewhere around the fourth birthday in typical children, the watershed when most child development experts agree that there is noticeable change in social intuition skills. This may also explain why individuals with autism, who have disrupted frontoinsular and anterior cingulate cortical areas, have difficulty working out what the rest of us just know without having to think very much.

(Further Reading)

  • Autism: Explaining the Enigma. Second edition. Uta Frith. Wiley-Blackwell, 2003.

  • The Curious Incident of the Dog in the Night-Time. Mark Haddon. Illustrated edition. Vintage, 2004.

SA Mind Vol 22 Issue 1This article was originally published with the title “Perspectives: Knowing Me, Knowing You” in SA Mind Vol. 22 No. 1 (), p. 16
doi:10.1038/scientificamericanmind0311-16