Why this page exists
Applied Behaviour Analysis is the most widely funded autism intervention in the world. Its origins matter because they shape the assumptions embedded in the framework, the institutional structures that sustain it, and the claims made on its behalf. This page examines those origins through the work of Ole Ivar Lovaas (1927â2010), the psychologist most responsible for establishing ABA as an autism intervention, and follows the thread from his early experiments to the present day.
The page on ABA and sensory processing addresses the specific intersection with sensory behaviour. This page provides the historical and institutional context.
Early life: a revised history
For decades, the accepted account of Lovaasâs early life held that he and his family were forced labourers during the German occupation of Norway (1940â1945). Lovaas himself told this story, and it appeared in profiles and obituaries. A 2025 paper by the Norwegian historian Ă smund Borgen Gjerde overturned it. Gjerdeâs archival research showed that the Løvaas family were members of Nasjonal Samling, Norwayâs collaborationist fascist party. Lovaasâs father, Ernst Albert Løvaas, held a prominent position in the propaganda apparatus of the German-installed government. Lovaas himself was an active participant in, and a local leader of, the fascist youth movement. He was between thirteen and eighteen years old during the occupation.
This does not mean that ABA is fascist. It does mean that the origin story told by the fieldâs founder was false, and that the field built on that story without verifying it. Gjerdeâs paper was published in History of the Human Sciences in 2025; the fieldâs response is still developing. The parallel with Hans Aspergerâs wartime record, revealed by Edith Sheffer and Herwig Czech in 2018, is difficult to ignore: two foundational figures in the history of autism, both with concealed connections to fascist regimes. In neither case does the biographical revelation invalidate all subsequent work, but in both cases it raises questions about what assumptions were carried forward unexamined.
The shock experiments (1960s)
Lovaas established the Young Autism Project at UCLA in 1962. His early work applied B.F. Skinnerâs operant conditioning principles to autistic children, using electric shock, slapping, and shouting as aversive stimuli to suppress unwanted behaviour and establish adult figures as safety cues.
In 1965, Life magazine published a photographic essay titled âScreams, Slaps, and Loveâ documenting Lovaasâs methods. The photographs showed a nine-year-old girl receiving electric shocks for non-compliance and a seven-year-old boy being slapped in the face for inattention during a speech lesson. The article presented these as a âsurprising, shocking treatmentâ that âhelps far-gone mental cripples.â The language is instructive: it tells us what autistic children were understood to be in the institutional culture that produced ABA.
Lovaas, Schaeffer, and Simmons published the underlying research the same year: âBuilding social behavior in autistic children by use of electric shockâ (1965). The paper reported that shocking autistic children produced a âdramatic increase in social contact with adult caretakers.â The logic was direct: make the childâs world painful, then make the adult the only route to safety.
Lovaas later abandoned aversives, telling CBS in 1994 that autistic people âare so used to pain that they can adapt to almost any kind of aversive you give them.â The phrasing is revealing. It does not express regret that the pain was inflicted; it expresses frustration that it stopped working.
Bernard Rimland, a parent-advocate who supported Lovaasâs work, noted in 1978 that the programme was â98% positive reinforcementâ and that Life had selectively photographed only the aversive moments. This is likely true. It is also true that the aversive moments were real, were published in peer-reviewed journals, and were the foundation on which the broader programme was built.
The Feminine Boy Project (1970s)
In 1974, Lovaas and George Rekers published the first in a series of papers describing the use of operant conditioning to modify gender-nonconforming behaviour in young boys. The research was conducted under the banner of UCLAâs Feminine Boy Project, a collaboration between Lovaas, Rekers, and the sexologist Richard Green.
The most documented case is Kirk Andrew Murphy (referred to as âKraigâ in the published papers), who was five years old when treatment began. The behavioural protocol used blue chips (redeemable for treats) to reward masculine behaviour and red chips to punish feminine behaviour. At home, with Rekersâs approval, each red chip was converted into a physical beating. The published results claimed success: the boyâs feminine behaviours were suppressed.
Kirk Murphy died by suicide in 2003, at the age of 38. His family attributed his death to the trauma inflicted during the study. His brother described the long-term effects of the intervention in interviews with the journalist Jim Burroway and in a 2011 CNN investigation.
The methodological and ethical connections between the Feminine Boy Project and the Young Autism Project are not incidental. Both used the same operant conditioning framework. Both targeted behaviours considered deviant by the culture of the time. Both defined success as making the subject appear normal. Both were conducted by the same researcher at the same institution. The Journal of Applied Behavior Analysis published an expression of concern about the Rekers and Lovaas (1974) paper in 2020, acknowledging that âsocietal changesâ had shifted how the work was viewed (Critchfield, 2020).
Autistic and LGBTQ+ advocacy groups have pointed to this shared lineage as evidence that ABA and conversion therapy are not merely historically connected but structurally analogous: both apply behaviourist techniques to suppress expressions of identity that deviate from a culturally defined norm.
The 1987 study: âindistinguishable from their peersâ
Lovaasâs most influential publication, âBehavioral treatment and normal educational and intellectual functioning in young autistic childrenâ (1987), reported on 19 children who received 40 hours per week of one-to-one behavioural treatment for several years, beginning in their early childhood. The study claimed that 47% of the treatment group achieved ânormal intellectual and educational functioningâ and were âindistinguishable from their normal friends.â A 1993 follow-up by McEachin, Smith, and Lovaas reported that these gains were maintained.
The claim of âindistinguishable from their peersâ became the foundational marketing proposition of the ABA industry. It was cited in insurance mandates, parent advocacy campaigns, and professional training materials as evidence that intensive early intervention could effectively cure autism.
The study has been criticised on multiple grounds:
Assignment was not random. Children were assigned to treatment and control groups based on therapist availability, not randomisation. This introduces selection bias at the most basic level.
Pre- and post-treatment measures differed. Different assessment instruments were used before and after treatment, making comparison unreliable (Jordan, Jones & Murray, 1998; Howlin, 1997).
The âindistinguishableâ claim rested on weak measurement. The Clinical Rating Scale used to assess outcomes was unpublished, administered by a graduate student in a 20-minute interview, with no blinding and no evidence of reliability or validity (Shea, 2004). No measures of teacher or peer opinion were reported, despite the claim that subjects were indistinguishable from peers.
Prorated mental age scores were used rather than standard IQ measures, a psychometrically weak approach.
The study was conducted over 14 years (1970â1984) but published as if it were a single cohort study. The long time span raises questions about consistency of treatment and measurement.
Replication has been partial at best. Subsequent studies have found positive effects from intensive behavioural intervention but have not replicated the dramatic ârecoveryâ claims of the 1987 paper. The What Works Clearinghouse (IES, 2010) rated the Lovaas model as having limited evidence, noting that the 1987 study did not meet its evidence standards.
Eric Schopler, developer of the TEACCH programme, called the study âstrictly off the wallâ and questioned whether the subjects or their families were representative. Schopler pointed out that parents who commit to 40 hours per week of home-based treatment are, by definition, not a random sample.
The 1987 study has never been retracted. It continues to be cited in support of ABA mandates.
The institutional legacy
The insurance-industrial complex
The âindistinguishableâ claim transformed ABA from one approach among many into the dominant autism intervention in the United States. Insurance mandates for ABA therapy now exist in all 50 US states. The Behavior Analyst Certification Board (BACB) oversees a workforce of over 60,000 Board Certified Behaviour Analysts. The economic structure, billions of dollars in annual insurance reimbursement, creates institutional resistance to fundamental critique.
This matters for understanding why the debate about ABA is so intractable. The question is not only âdoes this work?â but âwhat happens to an industry built on a foundational study with serious methodological problems, delivering a treatment whose origins include aversive shock and conversion therapy?â
The Judge Rotenberg Center
The use of electric shock on autistic people did not end with Lovaasâs 1960s experiments. The Judge Rotenberg Center (JRC) in Canton, Massachusetts continues to use graduated electronic decelerators (GEDs), devices that deliver electric shocks to the skin, as a behavioural intervention for people with intellectual and developmental disabilities.
A 2012 video showing a JRC resident being shocked 31 times while strapped to a restraint board provoked international outrage. The FDA banned the devices in 2020, but a federal appeals court overturned the ban, ruling that the FDA could not interfere with the practice of medicine. In 2024, the FDA proposed a new ban under revised legal authority. As of early 2026, the rule has been delayed; a final rule is expected in mid-2026. Congressional riders have been attached to funding bills to prevent the FDA from acting.
The JRC is not an ABA provider in any mainstream sense, and the BACB has distanced itself from the centre. Its continued operation nonetheless represents a living legacy of the principle that aversive conditioning is an acceptable response to autistic behaviour when other approaches have been deemed insufficient.
The conversion therapy connection
Multiple jurisdictions have banned or are considering bans on conversion therapy for LGBTQ+ people. The question of whether ABA constitutes a form of conversion therapy for autistic people has been raised by advocacy organisations including the Autistic Self Advocacy Network and Autistic Collaboration. The argument is structural rather than analogical: ABA and conversion therapy share a common intellectual origin (Lovaasâs UCLA programme), a common methodology (operant conditioning to suppress identity-linked behaviours), and a common definition of success (indistinguishability from the norm).
The counterargument from within the ABA field is that modern ABA has evolved beyond its origins, that contemporary practice focuses on quality of life rather than normalisation, and that the association with conversion therapy is a guilt-by-association fallacy (Leaf et al., 2022). This argument has some force. It also requires the field to account for why its foundational study has never been retracted, why its foundational figureâs false autobiography was never questioned, and why its institutional structure continues to define success in terms derived from the normalisation agenda of the 1960s and 1970s.
What this means for the wiki
This wiki does not take the position that all behavioural intervention is harmful or that every ABA practitioner intends harm. It does take the position that the history matters. A fieldâs origins do not determine its future, but they shape its assumptions, its institutional structures, and the questions it fails to ask. The history of ABA reveals a recurring pattern: behaviours that deviate from cultural norms are targeted for suppression, the suppression is presented as treatment, and the evidence base for the treatment is constructed by people with a theoretical and economic stake in its continuation.
The question for any intervention that targets autistic behaviour is the one posed by the Foucauldian critique: whose norms are being enforced, by what mechanisms, and in whose interest?
Key sources
- Lovaas, O.I., Schaeffer, B. & Simmons, J.Q. (1965). Building social behavior in autistic children by use of electric shock. Journal of Experimental Research in Personality, 1(2), 99â109.
- Lovaas, O.I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55(1), 3â9. doi: 10.1037/0022-006X.55.1.3
- McEachin, J.J., Smith, T. & Lovaas, O.I. (1993). Long-term outcome for children with autism who received early intensive behavioral treatment. American Journal on Mental Retardation, 97(4), 359â372.
- Rekers, G.A. & Lovaas, O.I. (1974). Behavioral treatment of deviant sex-role behaviors in a male child. Journal of Applied Behavior Analysis, 7(2), 173â190. doi: 10.1901/jaba.1974.7-173
- Shea, V. (2004). A perspective on the research literature related to early intensive behavioral intervention (Lovaas) for young children with autism. Autism, 8(4), 349â367. doi: 10.1177/1362361304047223
- Howlin, P. (1997). Prognosis in autism: do specialist treatments affect long-term outcome? European Child & Adolescent Psychiatry, 6(2), 55â72.
- Jordan, R., Jones, G. & Murray, D. (1998). Educational Interventions for Children with Autism: A Literature Review of Recent and Current Research. London: DfEE.
- Critchfield, T.S. (2020). Editorâs note: Societal changes and expression of concern about Rekers and Lovaas (1974). Journal of Applied Behavior Analysis, 53(4), 2668â2670. doi: 10.1002/jaba.768
- Gjerde, Ă .B. (2025). Hidden Nazi past: Ole Ivar Lovaas during the German occupation of Norway. History of the Human Sciences. doi: 10.1177/09526951251324102
- Leaf, J.B. et al. (2022). Concerns about ABA-based intervention: An evaluation and recommendations. Journal of Autism and Developmental Disorders, 52, 2838â2853. doi: 10.1007/s10803-021-05137-y
- Institute of Education Sciences (2010). WWC Intervention Report: Lovaas Model of Applied Behavior Analysis. Washington, DC: What Works Clearinghouse.