How is this possible? Seriously.

I search his book and found 0 occurrences of “psychosis” and 2 occurrences of “schizophrenia”, one of which might be relevant:

In this regard, let’s conclude this chapter by considering a very common variation from the “typical” human mental state—a variation in which ToM and intentionality seem to be altered, leaving the subjects unable to effectively carry out many of the twenty typical human behaviors listed above (or others like it). We have earlier mentioned autism,52 a mental difficulty affecting children and young adults that some people claim is increasing in frequency and severity in Westernized societies. Whether or not the increase is real (a subject of some controversy), the word autism has been largely replaced by the term autism spectrum disorders (ASD), for the reason that autism is not a single disease. Rather, like “pneumonia” (any kind of disease that fills up the lungs and makes you short of breath), ASDs are a group of diseases that can be lumped together because they have certain common features. The spectrum ranges from individuals who are completely unable to communicate and take care of themselves all the way to so-called high-functioning autistics who are only slightly “abnormal” compared to others (a condition that is sometimes called Asperger’s syndrome). But one common feature that runs through all cases of autism spectrum disorders is the relative or absolute lack of ability to communicate with others at a social level (a social communication deficit). In fact, this is often the first feature noticed by parents, and Leo Kanner’s description of classical autism in 1943 described individuals who had an “inability to relate themselves in the ordinary way to people and situations from the beginning of life.” In other words, autistic subjects partially or completely lack the ability to imagine themselves in the situations of others (sometimes called mind-blindness). Effectively, then, autism spectrum disorders include a difficulty with full “attribution of mental states to others.” In other words, many of these individuals appear to be lacking a full ToM.53

In the latest update of the Diagnostic and Statistical Manual of Mental Disorders, published by the American Psychiatric Association (DSM-5, scheduled for release in 2013), two cardinal features are being defined as typical of ASD:54

1) persistent deficits in social communication and social interaction across contexts, not accounted for by general developmental delays and manifested by all three of the following: deficits in social-emotional reciprocity; deficits in nonverbal communicative behaviors used for social interaction; and deficits in developing and maintaining relationships appropriate to developmental level; and

2) restricted, repetitive patterns of behavior, interests, or activities as manifested by at least two of the following: stereotyped or repetitive speech, movements, or use of objects; excessive adherence to routines; ritualized patterns of verbal or nonverbal behavior or excessive resistance to change; highly restricted, fixated interests that are abnormal in intensity or focus; or abnormal reactivity to sensory input or an unusual interest in sensory aspects of the environment.

To complete the diagnosis, the symptoms must be present in early childhood (but need not be fully manifested until social demands exceed capacities) and must limit and impair everyday functioning.

As you can see, one cardinal criterion is a social communication deficit (effectively, varying degrees of deficits of ToM). In this regard, Simon Baron-Cohen confirmed to me that blushing is anecdotally absent in individuals with autism (but that this needs further study). In a related domain, such individuals also perform more poorly than expected on tests of recognition of faux pas, a recognition that normally entails feeling embarrassed for someone else.55 Baron-Cohen also noted that parents may complain about teenage children with autism undressing in public without embarrassment, something one would never see in a typical child.

The second cardinal criterion is repetitive behaviors that often manifest themselves as unusual talents. A small percentage of individuals who behave this way have savant syndrome, as we discussed earlier—a condition in which remarkable abilities are concentrated in a few narrow domains of human cognition. Of course, there is a tendency to equate the independent existence of unusual talents with being “on the spectrum,” leading to suggestions that many historically famous scientists such as Newton and Einstein were actually autistic. In fact, being “on the spectrum” is even considered to be a positive attribute in places like Silicon Valley, where employers may want to hire bright individuals who prefer to spend most of their time in front of a computer screen rather than socializing with others. This is scant comfort to the parents of the unfortunate children who have severe communication deficits but do not appear to have any obvious special talents. Of course, as Francesca Happé and Uta Frith so eloquently put it in their phrase “the beautiful otherness of autism,”56 these minds are just different from ours, and we should probably be making a greater effort to find and foster such hidden talents.57 One example of this is Jake Barnett, who was featured on the popular TV show 60 Minutes. Jake was diagnosed with autism at age two and became increasingly withdrawn and uninterested in most things. All attempts at conventional therapy failed until his mother happened to take him along to some evening classes involving mathematics, physics, and astronomy. He then began to come out of his shell, asking questions and learning so avidly that he was eventually able to graduate from high school and enter college at a very young age. Jake appears destined for brilliance in mathematics and physics. Interestingly, the process by which he came out of his shell in the setting of mathematics and physics was associated with his also recovering from his most severe autistic symptoms. Perhaps there are many more such hidden talents that have yet to be uncovered among cases of autism. Indeed, one father of an autistic child has created the Specialisterne (which translates from Danish as “the specialists”)—an unusual social enterprise, staffed mostly by people with autism, that provides assessment, training, education, and IT consultancy services. The Specialist People Foundation, which runs Specialisterne, says it has a goal of enabling one million jobs for people with autism and similar challenges through social entrepreneurship, corporate-sector engagement, and a global change in mind-set.58

Another remarkable feature of ASDs are their consistent and widespread occurrence in human populations worldwide, despite the fact that they are poor prescriptions for reproductive fitness. As most severe autistics do not leave progeny, why was the condition not eliminated by natural selection? The current explanation is that most causes of autism represent de novo (new) genetic mutations, which can then be inherited in a family.59 This begs the question of why such mutations should be happening at such high frequencies in otherwise critical genes. One possible explanation is that parents are having children at an older age than they did in previous generations—especially fathers, whose sperm are at greater risk for accumulating random mutations over time.60 Another possibility is that the concentration of high achievers into certain geographic areas and professions (such as the tech sector) may be increasing the possibility of individuals already “on the spectrum” mating with each other.61 In addition to such modern possibilities, our evolutionary transition to a full ToM may be far from completely hardwired (since it requires an environmental input at the right time), and the risk of failure of the ToM mechanism remains very high in the human species. Yet another view is a more positive one, and posits that having group members who were able to ignore social niceties and instead deal with specific problems and difficult tasks (in other words, having members who displayed the features of mild to moderate autism) may have actually been of benefit to early human societies. As Aristotle recognized, humans are the greatest imitators. So once someone in a group is able to invent or discover something, imitation will rapidly spread the new concept through human populations. Thus having a few individuals with less-than-optimal ToM could benefit the group, since their ability to perform restrictive and focused activities would eventually be of value to others. And such individuals might also be the ones who could best memorize and accurately pass down important details of a lengthy oral tradition before the advent of writing. If this scenario is correct, it may help explain the persistence of genes predisposing people to ASDs in all societies and cultures. Severe autism spectrum disorders may well be the price that we pay for the benefits of having milder cases. In keeping with this, Temple Grandin says that if not for the talents of many people with various levels of autism over the years, we humans might still be living in caves!

Of course, such considerations of autism remain limited in discussions of human origins, for obvious reasons. First, one cannot assume that such individuals represent the mental state of humans prior to the final transition to modern cognitive behavior. Second, many of these cases feature other neurological, psychological, or cognitive abnormalities, making it difficult to come to any easy conclusions. Regardless, the social-cognition deficits of autism seem to involve a difficulty with expressing a full ToM. Thus it is reasonable to suggest that a better understanding of autism would also lead to a better understanding of ToM. Recent studies have made it clear that inherited or de novo gene defects, duplications, and deletions play a key role in the causes of autism and that any single genetic cause accounts for less than 1 percent of all cases. It appears that many different defects can trigger the neural processes involved in generating the symptoms of autism. There are many hypotheses regarding these processes, ranging from the “extreme male brain” theory (Baron-Cohen)62 to abnormalities in nerve cell number or pruning (Courchesne and others)63 to defective mirror neurons (Ramachandran) to the social motivation theory (Chevallier and others)64 to autism as the diametric opposite of schizophrenia (Crespi).65 To adequately address these and other theories would be far beyond the scope of this book.

Regardless of which of the theories turn out to be right (and it is very likely that more than one will be), social communication deficits can be equated partly with variable degrees of lack of ToM—a feature we have suggested was a critical step in the evolution of the human mind. Thus, in a sense, some individuals with autism spectrum disorders might be studied to test our primary hypothesis—that the evolution and stabilization of full ToM in human ancestors was originally held back by the recognition of mortality of others, which in turn results in awareness of personal mortality (a distressing and maladaptive situation). If so, one might predict that individuals with severe autism spectrum disorders (i.e., those with very limited ToM) should not be fully aware of the meaning of death. As you can imagine, this is not an easy question to ask or address, and there seem to be no definitive published studies on the matter. But anecdotal evidence suggests that autism spectrum disorders are indeed associated with a relative lack of full understanding of the death of others. One can even find Web-based instructions for parents on how to explain the death of family members to individuals with ASD. There are also no formal studies that ask whether autistic individuals understand their own mortality. Again, it would be difficult to get such a study approved—to directly ask such individuals whether they have a true recognition of their personal mortality. Regardless, it is interesting to ask how many of the twenty random examples of uniquely human features requiring full ToM we listed above are significantly blunted in autism spectrum disorders. It would likely be most of them.

Given how useful and powerful full ToM has been for the success of our species, why did some of the other intelligent and self-aware species not acquire the same ability? Was it just chance? Or a very special one-of-a-kind brain evolution process that was specific to the human species? Or is there simply a nearly insurmountable “wall” that has been holding back all the other species? If so, what is that wall?