The autistic-adult community's verdict on ABA is consistent and damning. The clinical establishment's verdict, until recently, was that it was the gold standard. Both verdicts contain real information, and refusing to engage with either honestly is how the conversation stays stuck. Here is the version that doesn't flatter anyone.
What ABA got right
Ivar Lovaas's early UCLA work in the 1960s identified something the clinical establishment had largely missed: autistic children responded to structured, repeated, consequence-shaped instruction in ways the prevailing psychoanalytic theories of the time could not explain. The Lovaas research, for all its later-documented harms, was the first sustained attempt to treat autistic children as learners rather than as broken-and-unreachable. That reframing was real, and it mattered.
The methodological discipline ABA brought — measurable goals, documented progress, observable behavior change — was a meaningful improvement over the previous standard of "the parents did something wrong, send the child to a residential institution." For families who had nowhere else to turn, ABA was the first thing that looked like a treatment. That history is not nothing.
What ABA got catastrophically wrong
The harms accumulated in the decades after the early work. They are extensively documented in autistic-adult testimony and in increasingly the peer-reviewed literature:
- The use of aversives. Lovaas's original program included punishment — slapping, shouting, and in the case of the Judge Rotenberg Center, electric shock. The shock devices were still in use in 2020 before the FDA finally banned them (a ban subsequently overturned in court). The framing that autistic behavior was a thing to be punished out of children produced trauma that survivors are still naming.
- The goal of "indistinguishability." Lovaas's stated goal was making autistic children "indistinguishable from their typically-developing peers." The unstated cost of indistinguishability was a lifetime of masking — which the literature now documents produces autistic burnout, identity rupture, and PTSD-spectrum symptoms in adulthood.
- The suppression of stims. Stimming is one of the most reliable regulation behaviors autistic people have. ABA's standard practice of extinguishing stims removed the regulation tool, which produced the dysregulation it then tried to behaviorally address. The autistic-adult community has been clear: stims are not problems.
- The hyper-compliance teaching. Modern ABA programs continue to use "quiet hands" and "compliance" as goals — teaching autistic children to obey adult instructions without question. The downstream effects on safety (autistic adults who can't say no, can't refuse coercion, can't advocate for themselves) are now well-documented.
- The 40-hour-week intensity. Lovaas-style "intensive early intervention" was 40+ hours per week, every week, on a young child whose nervous system was already operating at maximum regulatory capacity. The schedule itself was harm.
The PTSD-symptom research (Kupferstein 2018, follow-up studies through 2024) found that autistic adults exposed to ABA as children show clinically significant PTSD symptoms at significantly higher rates than autistic adults who did not receive ABA. The harm is not anecdotal at this point.
What modern ABA defenders say
Modern ABA programs argue that contemporary practice is different from Lovaas. Aversives are out. Naturalistic settings are in. Goals are more flexible. Parent buy-in is required. The autistic-adult community responds: the underlying compliance-based framework is the same. The behavior is still the unit of change. The autistic person is still the one expected to do the adapting. Calling it "naturalistic" doesn't change what it is.
This is the impasse. ABA defenders point to specific gentler practitioners. The autistic-adult community points to the systemic incentives that still reward indistinguishability and still extinguish stims. Both can be true in individual cases. The pattern is the pattern.
Why Autism Acceptance World does not refer to ABA
Because the autistic-adult community's verdict is consistent, accumulating, and load-bearing. Because the alternatives exist and work: occupational therapy oriented to sensory regulation, speech-language pathology oriented to communication-in-any-modality, DIR/Floortime, RDI, neurodiversity-affirming counseling. Because there is no reason to send autistic children into a 40-hour-week of compliance training when the same outcomes are reachable through approaches that don't require extinguishing the autistic features that aren't actually problems.
Autism Acceptance World does not promote ABA, will not promote ABA, and will not refer families to ABA providers. Families who choose ABA after considering this position have made their choice — we don't relitigate it with them. But we are not neutral on the question.
For the alternatives we do support, see the Autism Acceptance World Parenting from Acceptance, Not Recovery resource, the First 90 Days After Diagnosis playbook, and the position paper Autism Acceptance World's Position on ABA.
— Cash