Assent in ABA: what the BACB Ethics Code requires and what critics are asking for

The BACB Ethics Code for Behavior Analysts has applied to all BCBA and BCaBA certificants since January 1, 2022, and the current PDF is the August 2024 update¹. In its first edit to the Code, announced in the March 2023 newsletter, the BACB added a definition of assent to the glossary²: vocal or nonvocal verbal behavior that can be taken to indicate willingness to participate in research or behavioral services by individuals who cannot provide informed consent, for example because of age or intellectual impairments². The same definition appears in the current Code¹.

Informed consent is the permission given by an individual with the legal right to consent, and a legally authorized representative consents for someone who cannot¹. Standard 2.11 makes you responsible for explaining, obtaining, reobtaining and documenting required informed consent, and for obtaining assent from clients when applicable¹. Three consequences follow from the text:

• Assent does not replace consent. The guardian’s informed consent is still required, for example before starting assessments or interventions and when making substantial changes¹.
• Assent is for clients who cannot consent. A client who can give informed consent does so, with the right to decline or withdraw at any time¹.
• The Code sets no method. When a research review committee or a service organization requires assent, those entities provide the parameters for assessing it².

Other standards shape how assent works in daily practice¹:

• 2.14: interventions prioritize positive reinforcement procedures and weigh risks, side effects and client and stakeholder preference.
• 2.15: restrictive or punishment-based procedures come only after less intrusive means have not worked, or when the team determines that the risk of harm to the client outweighs the risk of the intervention, with any required review, such as a human rights committee.
• 2.09: clients and stakeholders take part in selecting goals.
• Core principles: treat others with compassion, dignity and respect, which includes promoting clients’ self-determination to the best of their abilities.

RBTs follow the RBT Ethics Code (2.0). It shares the same four core principles, requires work under close supervision and allows restrictive or punishment-based procedures only when they are in a documented plan and the supervisor has verified the RBT’s competence³. Florida Medicaid adds its own limit: as part of this benefit, its coverage policy does not cover procedures that use seclusion or manual, mechanical or chemical restraint to control behaviors⁴. Other states and plans set their own rules.

The Autistic Self Advocacy Network (ASAN), an advocacy organization, states its position on its website. In its words, some therapies try to make autistic people seem “normal” or “less autistic”, the most common being ABA, which “uses rewards and punishments to train autistic people to act non-autistic” and can hurt autistic people⁵. ASAN does not back any single therapy, and it says any therapy should help autistic people get what they want and need, “not what other people think we need”⁵.

A 2026 article in Behavior Analysis in Practice reports a dialogue in which two ABA scholars put questions to two doctoral scholars in Critical Autism Studies⁶. The ABA authors summarize criticisms from the neurodiversity movement that their team reviewed in 2024: erasing autistic identity, overfocusing on compliance, reducing people to overt behavior, leaving autistic voices out of research and insisting on ABA as the only option⁶. As they describe it, a key position of Critical Autism Studies is that professionals who serve autistic people must learn from autistic lived experience and center autistic voices in services⁶. The Critical Autism Studies scholars asked for 4 changes: continuing dialogue with the autistic community, adoption of the neurodiversity paradigm by regulators, providers and researchers, functional analyses centered on autism, and reporting consent, assent and conflicts of interest in all ABA autism research⁶. Their bottom line is that support for autistic people, whatever it is called, needs “to serve, rather than traumatize”⁶.

Survey data add a measure of these views. A 2024 online survey asked autistic adults, caregivers and early childhood practitioners about practices used with autistic children; 226 of the 660 respondents were autistic⁷. Respondents rated self-determination goals and antecedent interventions as highly acceptable, rated goals that promote masking and some forms of extinction low, and named the child as the most important voice in decisions⁷.

The answers are not uniform:

• Leaf and colleagues, a group that included BCBAs, psychologists, parents and autistic coauthors, reviewed the concerns and urged behavior analysts to stay “compassionately skeptical” of broad claims that ABA is abusive, while showing compassion and learning from lived experiences, without negating what they call the substantial evidence supporting ABA-based interventions⁸. A correction later noted that several of the authors provide ABA research, training and services⁹.
• Breaux and Smith described assent-based practice, in which a learner’s withdrawal of assent is honored, whether it is a vocal “no” or a nonvocal behavior¹⁰.
• Newcomb and Wine objected to the implication that assent must be part of every form of treatment, and asked who is authorized to decide how assent is assessed¹¹. Breaux and Smith replied that they strongly disagree with limiting learner assent to research¹².

• Harm claims. Leaf and colleagues identify two peer-reviewed publications that appear to support claims of abuse or PTSD⁸. One, Kupferstein (2018), surveyed 460 respondents with a self-designed questionnaire to evaluate a correlation, and reported that 46% met the threshold for PTSD after ABA⁸. A correlation alone cannot show that ABA caused the symptoms. Leaf and colleagues say reliable data that ABA caused PTSD, anxiety or depression are lacking⁸. The Critical Autism Studies scholars answer that the criticism of that study came from within the ABA industry without a commitment to fund further research, and call for transparent studies co-produced with autistic scholars⁶.
• Assent research. Few published studies mention or use assent in ABA, as Breaux and Smith note¹⁰¹².
• Whose voices. The 2026 dialogue did not include nonspeaking or minimally speaking autistic people, a limit its authors acknowledge⁶.

The Code in force since 2022, with the definition of assent added in its first edit, is the current text, and the BACB leaves the assessment method to organizations and review committees¹².

The CMS toolkit for state Medicaid and CHIP programs, released in August 2026, asks states whether they have acknowledged self-advocate perspectives and whether coverage language emphasizes safety, communication, self-advocacy and mental health outcomes rather than behavior “normalization”¹³. It also asks whether parents can decline ABA in favor of other services¹³. CMS says the toolkit does not establish new federal requirements¹⁴. This article is not legal or clinical advice: for your caseload, check your organization’s policy and your payer contract, and ask the BACB when the Code’s wording leaves you in doubt.