Marina Sarris

Date Published: December 10, 2025

In 1919, a small boy named Archie Castro entered an institution in West Virginia. Archie spoke only five words, wandered into dangerous situations, and didn’t respond to his mother’s voice. He was labeled an “idiot” and “insane,” the descriptions of that era, according to “In a Different Key: The Story of Autism.” Children like him often ended up in dismal places like this, separated from their families and communities.¹

A few years later, a psychiatrist in Soviet Russia, Grunya Sukhareva, reported on children like Archie, with a condition we would today recognize as autism. But her findings didn’t reach far outside her homeland, and certainly not into Archie’s world.²

When Castro was in his early 30s, two doctors published separate accounts of children with symptoms like his. American Leo Kanner called the condition “autistic disturbance.” Austrian Hans Asperger, who focused on milder symptoms, labeled it “autistic psychopathy.” But no one went looking for children with autism, let alone for adults like Castro. Autism was thought to be rare.

Today we know that autism does not mean disturbed or psychopathic. And doctors have broadened autism’s definition since then. People with a 21^st Century diagnosis of “autism spectrum disorder” include those who need significant support, like Castro did, as well as people with graduate degrees who need relatively little help.

Autism now affects 1 in 31 children in the United States. Understanding its history can help untangle questions about what autism is and why many people are now diagnosed with it.

The Beginnings of Autism

Did autism exist before doctors named and described it, and published their ideas in journals? Yes, say researchers and doctors. Many health and developmental conditions existed before they had their current names and diagnosis codes.

“The whole field of psychology is a recent invention,” explains Connie Anderson, Ph.D., health sciences department chair and director of the autism studies graduate certificate at Towson University in Maryland.

Institutions for the “insane” and “feeble-minded” – as they were then called – grew in the 19^th Century in the United States.¹ No one knows how many of the children and adults sent there had autism or other misunderstood conditions. What brought them together is that they had a stigmatizing disability or difference.

How Autism Became Linked to Schizophrenia

For much of the 20^th Century, autism was linked to a very stigmatized condition, one that includes hallucinations and delusions. In 1911, a Swiss psychiatrist, Eugen Bleuler, used autism to describe a symptom of schizophrenia.³ For Bleuler, autism described being so focused on one’s thoughts that it was hard to communicate and interact typically with people and situations.⁴

Today we know that schizophrenia and autism are not related, but Bleuler’s description lingered. From the 1930s to the 1970s, doctors diagnosed schizophrenia in children whose symptoms were similar to what we today call autism.^1,5

In 1943, Baltimore psychiatrist Leo Kanner mentioned this when he described 11 children with “autistic disturbances.” These children lacked social awareness, had speech and language delays, feared noise, and had repetitive behaviors. “It is quite possible that some such children have been viewed as feebleminded or schizophrenic,” he explained. But Kanner insisted that he was describing something different, a “unique ‘syndrome.'”⁶

“Kanner tried really hard to distinguish autism from schizophrenia, but he got a lot of pushback from specialists in childhood schizophrenia,” Anderson explains. As a result, he ended up including autism in the schizophrenia chapter of his textbook, as described in the book “NeuroTribes,”⁷ she says.

The Rise of the Refrigerator Mother

Kanner also observed and described the parents of his autistic patients, and not favorably. “In the whole group, there are very few really warmhearted fathers and mothers,” his paper said.⁶ This description would haunt parents for decades.

No one considered whether these parents were stressed or autistic themselves. In the decades that followed, philosopher Bruno Bettelheim used Kanner’s observations to popularize the myth of the “refrigerator mother.” This was a woman so cold and uncaring that her rejection of her baby caused the child’s autism.

In 1971, Bettelheim went on a popular TV talk show and said autism was “the most severe psychotic disturbance of childhood known to man.” The mothers of autistic children even wished that their children did not exist, Bettelheim claimed.¹

Soon, new research helped to banish the refrigerator mother theory. In the 1970s, researchers published a study of twins that suggested that genes were strongly connected to autism.^8,9

But decades of blaming mothers for autism left a mark not easily erased. “It still affects people,” Anderson says. “There’s also a mother-blaming penchant in the whole society – it’s part of the legacy of Freud. Maybe that’s what is left over.”

More than a century ago, Sigmund Freud, M.D., blamed overprotective mothers for triggering certain psychological conditions in their children.

Re-Thinking Autism’s Definition

Few people had an autism diagnosis in the 1970s. Doctors believed only people with severe developmental delays, who resembled those Kanner had described, could have autism. “In the 1970s, the diagnosis of autism had to be proved beyond a reasonable doubt,” according to “Autism Frontiers: Clinical Issues and Innovations.”¹⁰

Autistic people gradually became more visible. The U.S. disability rights movement challenged the poor treatment and low expectations for people with disabilities. State institutions began the slow process of closing and transferring people into homes in their communities. In 1975, parents won the federal right for children with disabilities to attend public schools and receive special education.

American psychiatrists, who publish a manual for diagnosing various conditions, began rethinking their ideas about autism. In 1980, they removed autism from the schizophrenia category and created a new one for it. To be diagnosed, children had to have severe problems in communication, a “lack of responsiveness to other people (autism),” and “bizarre responses” to their environment.¹¹

Expanding the Borders of Autism

American psychiatrists expanded the borders of autism in the 1980s and 1990s, pushing gradually toward the concept of a spectrum condition that we know today.

In 1987, they created Pervasive Developmental Disorder-Not Otherwise Specified, a diagnosis for people who did not meet all the requirements for autism. In 1994, they added another diagnosis under the autism umbrella, this one for children who did not have significant speech and learning delays. They named it Asperger syndrome after the doctor who described this part of the spectrum. These labels would be combined into one, autism spectrum disorder, in 2013.

As the definition of autism expanded, so did the number of people diagnosed with it.

In the early 1990s, school systems added autism to the list of diagnoses that qualified for special education. Before then, autistic children could receive these services under different labels, such as intellectual disability, learning disability, or the catch-all, “other health impaired.” Their special education label could now be switched to autism. The number of schoolchildren with the new autism label rose dramatically in the 1990s.^1,12

Looking for, and Counting, Children With Autism

Doctors began actively looking for autism. In 2007, the American Academy of Pediatrics recommended that pediatricians routinely screen all children for autism at 18 months and 24 months of age.

In 2000, the U.S. government also began counting autistic children in a sample of regions across the country. Autism rates have risen steadily since then, although they vary by location. Some of this variation may be due to differences in autism awareness, screening, and services in each region.

For example, the highest autism rate (5 percent) is found in San Diego, California, which has a Get SET Early program. Created by SPARK study researcher Karen Pierce, Ph.D., the program works with parents and pediatricians to promote early diagnosis and intervention. Doctors screen babies beginning at 12 months and may refer them for evaluations and services. By contrast, the lowest autism rate (1 percent) is found in Laredo, Texas, which does not have a Get SET Early program actively looking for toddlers with autism.¹³

Research into autism also expanded. Most scientists agree that both our genes and the environment contribute to autism. They have found variations in more than 400 genes and parts of chromosomes that cause autism.

They have also found that some conditions and environmental exposures during pregnancy and birth may increase the chances of autism. These factors include prenatal exposure to pesticides, air pollution, diabetes, fever, and infections; premature birth; and parents’ older ages at conception. Many studies have found no link to vaccines.

What About Autistic Adults?

There is evidence that some adults who missed out on an autism diagnosis during childhood are getting diagnosed now. In the SPARK study, the most frequently reported age of autism diagnosis for independent adults is 36. Many of these adults were born before doctors changed the definition of autism to include them, and before pediatricians began routine autism screenings.

A different study found large increases in the autism rate of adults from 2011 to 2022. The researchers examined the health records of 12 million children and adults in the U.S. They found the largest relative increase in autism rates, 450 percent, in adults ages 26 to 34 during that time period. The autism rate of those aged 35 to 44 rose almost 340 percent.¹⁴

“To at least some extent, we believe that the increase among those age groups is due to first-time diagnoses during the time period we studied,” says the paper’s lead author, Luke P. Grosvenor, Ph.D., of Kaiser Permanente Northern California’s Division of Research. But, he points out, his research team did not have data on how old autistic people were when they were first diagnosed.

As for Archie Castro, the boy sent to an institution in 1919, he finally got an autism diagnosis in the late 1980s, when he was in his 70s. Castro spent the last decade of his life in a group home, where he learned to dress himself and tidy his room.¹ During that time, the 1990s, doctors were diagnosing a record number of autistic children. Those children would grow up in a world that viewed them much differently than Castro’s did.

Interested in joining SPARK? Here’s what you should know.

Photo credit: iStock

References

1. Donvan J. and C. Zucker In a different key: The story of autism. New York: Crown Publishers (2016)
2. Zeldovich L. The Transmitter (2018). Accessed Dec. 4, 2025
3. Evans B. Hist. Human Sci. 26, 3-31 (2013) PubMed
4. Maatz A. et al. Dialogues Clin. Neurosci. 17, 43-49 (2015) PubMed
5. Fellowes S. Hist. Philos. Life Sci. 46, 29 (2024) PubMed
6. Kanner L. Nerv. Child 2, 217-250 (1943)
7. Silberman S. NeuroTribes: The legacy of autism and the future of neurodiversity. New York: Avery (2015)
8. Folstein S. and M. Rutter Nature 265, 726-728 (1977) PubMed
9. Folstein S. and M. Rutter J. Child Psychol. Psychiatry 18, 297-321 (1977) PubMed
10. Shapiro B. and P. Accardo (Eds.) Autism frontiers: Clinical issues and innovations. (pp. xi-xiii). Baltimore: Brookes Publishing (2008)
11. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd edition. Washington, DC: APA (1980)
12. Shattuck P.T. Pediatrics 117, 1028-1037 (2006) PubMed
13. Fombonne E. Eur. J. Epidemiol. 40, 981-994 (2025) PubMed
14. Grosvenor L.P. et al. JAMA Netw. Open 7, e2442218 (2024) PubMed