The connection between schizophrenia and autism is not purely based on the historical overlap of these diagnoses, but includes clinical connections that are observed in treating individuals with these two disorders. As Kanner noted, individuals with schizophrenia manifest a variety of negative symptoms including social withdrawal which make them appear clinically similar to individuals with autism. The symptoms related to social dysfunction in both disorders can also be severely disabling, and include limited motivation to engage in social interaction, flat affect, and the lack of social and emotional reciprocity (Woodbury-Smith et al. 2010). On tasks of social cognition, patients with both autism and schizophrenia show similar functional abnormalities in multiple brain regions, including the cerebellum, insular cortex, and fusiform gyrus (Meyer et al. 2011). In autism-related irritability, psychiatrists often prescribe antipsychotic medications, originally developed to treat schizophrenia (Woodbury-Smith et al. 2010; Melville et al. 2008). In both disorders, symptoms can be very resistant to both pharmacologic and psychotherapeutic intervention.

The long-term outcomes in both autism and schizophrenia were not known to Kanner when he made his initial distinction between the two disorders. Over time, further observations have been made by following these individuals longitudinally. Interestingly, it has been found that individuals with autism also commonly suffer from psychosis and are actually ten times as likely as the general population to develop new onset psychotic symptoms in a given year (Melville et al. 2008). Several genes have been discovered that may be associated with both autism and schizophrenia indicating, perhaps, a common neurodevelopmental pathway or disease process (Burbach and Van der Zwaag 2009). For example, some authors suggest that a common etiology of maternal infection or inflammation may be involved in both disorders (Meyer et al. 2011). Others have observed that individuals with autism have a significant decrease in life expectancy which is also characteristically seen in individuals with schizophrenia (Shavelle and Strauss 1998), although the mechanism underlying this outcome is not clearly understood.

Kanner made the crucial observation that individuals with autism differed from those with schizophrenia in that they lack a period of normal development prior to the onset of symptoms (Gallo 2010). This observation is the reason that autism is categorized as a pervasive developmental disorder. Symptoms of autism are present throughout early development and diagnosis is typically made in early childhood. Further differentiation includes the concrete and reality-based thought process seen in autism. In addition, hallucinations and delusions, which are the hallmark symptoms of schizophrenia, are often not appreciable in individuals with autism.

When brain imaging technology developed in the 1970s research methods turned to finding differences among and between individuals with mental illness. On a structural level, neuroimaging studies have shown that individuals with autism have increased white matter, gray matter and cerebellar white matter volumes (Stanley 2002). This is in contrast to gray matter reduction in the frontal and temporal lobes found in individuals with schizophrenia (Stanley 2002). Though there are some functional similarities in both patient populations, as mentioned previously, neuroimaging studies have noted significant structural differences.

In addition to differences in developmental history and neuroanatomy, there are also many distinctions between autism and schizophrenia in other domains. In 1972, Sir Michael Rutter published one of the most important articles to define differences in these disorders and enabled clinicians for the first time to see the distinction between them. He noted that, although genetic factors were prominent in both disorders, parents of children with autism were often highly functional, intelligent, and of higher social class (Rutter 1972). Schizophrenia, in contrast, is one of the most inheritable disorders in all of psychiatry. Genetic studies in autism show that schizophrenia is exceedingly rare in families with autism and vice versa (Rutter 1972).

Regarding cognitive functioning, individuals with autism are much more likely to suffer from intellectual disability than individuals with schizophrenia and individuals with schizophrenia are more likely to suffer from cognitive decline over the course of their lifetime (Rutter 1972). With respect to comorbid conditions, individuals with autism are more likely to have epilepsy (Rutter 1972). Regarding distribution within the population, autism differentially affects male children at a rate of 4:1 while the gender distribution in schizophrenia is evenly split between the two sexes (Rutter 1972). Observing disease course, marked remission of symptoms can occur in schizophrenia and remission is exceedingly uncommon in autism (Rutter 1972). Much of the evidence which supported Kanner’s initial hypothesis did not become available until 30 years after his publication, when Rutter used additional research findings to fully differentiate the two disorders.

Based on a series of 11 children (eight boys and three girls) described in his initial publication, Kanner provided the following observations which became his initial criteria for a diagnosis of autism (Jordan 1999; Kanner 1943).

Later, Kanner revised and simplified his diagnostic criteria in order to develop a more general categorization method (Jordan 1999). His goal was to define the most basic features underlying the disorder. Kanner felt that some features of autism (such as repetitive language) could be explained as deriving from higher order symptoms (Jordan 1999). In 1956, Kanner proposed three criteria for the diagnosis of autism. These included extreme isolation, insistence on sameness, and an onset prior to 2 years of age. Simplifying these ideas even further, Kanner stated that the one central feature of autism was the patient’s inability to relate to others from early on in life (Volkmar and Lord 1998). This concept, which he initially called “profound autistic withdrawal” and later referred to as “extreme isolation,” was the highest order symptom that explained all of the others (Jordan 1999). Thus, in Kanner’s eyes, the term “autism” was intended to fundamentally describe a disorder of social relatedness.

It is notable that from his initial observations, Leo Kanner made several assumptions which later proved to be false. This is not meant to detract from the revolutionary accomplishment of his initial observations. To the contrary, many of these false assumptions have carried over to current practice and are still commonly held beliefs by mental health providers. We seek here to clarify some of these assumptions which lack empiric evidence.

Kanner assumed that children with autism had normal or above normal intelligence (Jordan 1999). More recent observations, however, show that around 75 % of individuals with autistic disorder also meet diagnostic criteria for intellectual disability (Volkmar and Lord 1998). Some research has sought to find strengths of individuals with autism based on Kanner’s observations of superior memory. However, the scores of individuals with autism on cognitive tests of local verses global processing has been mixed without the clear advantage in so-called “rote memory” observed by Kanner (Spek et al. 2011). More recent estimates have shown that the so-called “idiot-savants,” or individuals with autism or intellectual disability who display extraordinary talents in highly specialized areas, are exceedingly rare and constitute only around 1 in 2,000 individuals with autism (Heaton and Wallace 2004).

Another Kanner observation shown to be inaccurate was that autism is not associated with medical illnesses. Sixty years of observation and research have shown that this is also false. As was previously stated, as many as 25 % of children who are diagnosed with autism will go on to develop epilepsy (Volkmar and Lord 1998). Additionally, numerous correlations have been found associating autism with deficits in hearing, vision, prenatal exposure to valproic acid, and genetic conditions such as fragile X syndrome.

Early explanations for autism blamed parents and minimized biological factors. This may have had origins in Kanner’s observations of parents of the children in his original case series. However, assumptions about the etiology of autism further developed as a response to his initial publication. “Refrigerator Mothers” who withheld affection and failed to bond with their children were believed to be coupled with overachieving though emotionally distant fathers in order to yield autistic offspring (Volkmar and Lord 1998). Later research repudiated the theory that autism was caused by a dysfunctional child-parent relationship, showing no increased incidence of autism in orphanages in which children were brought up deprived of human contact (Volkmar and Lord 1998). Many children who were raised in social isolation had delays in language development but lacked ritualistic behavior (Rutter and Bartak 1971). Although some of what Kanner originally observed has not weathered the test of time, it is difficult to overestimate Kanner’s impact on the diagnosis of autism. We now understand autism to be a complex neuropsychiatric condition associated with diffuse central nervous system dysfunction and multiple medical conditions .

Rutter, like Kanner before him, noted many abnormalities in the speech of individuals with autism (Rutter and Bartak 1971). He identified pronoun reversal and echolalia as the most prominent features in the speech of individuals with autism. Rutter also noted a general paucity of speech reflecting little motivation to engage others in conversation. In addition, Rutter found that individuals with autism frequently had abnormalities in tone and pronunciation.

Rutter identified social dysfunction as the “key feature” of autism in agreement with Kanner’s previous work (Rutter and Bartak 1971). In his paper, he provides several examples of how this feature is manifested, including poor eye contact, little variation of facial expression, and a lack of interest in people. He saw that individuals with autism lacked normal development of sympathy and empathy and did not effectively express their emotions. This was seen very early in babies preferring not to cuddle with parents and developing very little attachment to them. Later on, these same children did not spontaneously seek out others and did not engage in normal play.

Rutter delineated four ways in which individuals with autism act in ritualistic or compulsive ways (Rutter and Bartak 1971). Some had an unvarying attachment to unusual objects. Others had restricted interests and still others quasi-obsessive ritualistic behaviors. Rutter also identified the resistance to change, which Kanner had initially identified as a separate diagnostic category, falling within the realm of ritualistic or compulsive phenomenon.

Rutter noted that while the vast majority of patients developed symptoms of autism within the first year of their life, some patients developed symptoms later on (Rutter and Bartak 1971). He chose to restrict the diagnosis of autism to those individuals who presented symptoms prior to 30 months. Rutter somewhat arbitrarily chose this number since it captured the population of patients he identified to have autism and excluded other patients similarly impaired by other illnesses, such as early onset schizophrenia.

The Diagnostic and Statistical Manual of Mental Disorders (DSM) did not accept the diagnosis of autism until the third edition which was published in 1980 (Volkmar and Lord 1998; American Psychiatric Association 1980). This reflected the body of work Rutter had accumulated over the previous decade since his initial publication in 1971 (Volkmar and Lord 1998). Though initially called “Infantile Autism” in the Diagnostic and Statistical Manual—Third Edition Revised (DSM-III-R; American Psychiatric Association 1987) changed the name to autistic disorder and removed the onset requirement which was initially present in the DSM-III (Volkmar and Lord 1998). Aside from requiring the onset of symptoms prior to 36 months, reinstituted after DSM-III-R, the criteria in the Diagnostic and Statistical Manual—Fourth Edition (DSM-IV; American Psychiatric Association 1994) reflected the core ideas proposed by Rutter. The three symptom clusters of social dysfunction, speech delay, and ritualistic or stereotypic behavior were all preserved in this version, as well as in the Diagnostic and Statistical Manual—Fourth Edition Text Revision (DSM-IV-TR; American Psychiatric Association 2000). As Kanner understood, autism fundamentally is a disorder of social relatedness. This emphasis is reflected by requiring two criteria from this symptom cluster while requiring just one each from the other clusters (American Psychiatric Association 2000) .

Diagnostic criteria for 299.00 autistic disorder (American Psychiatric Association, 2000).

(A) A total of six (or more) items from (1), (2), and (3), with at least two from (1), and one each from (2) and (3):