Is ABA Therapy Harmful? An Honest Answer.

This is the most important question in our field, and the ABA community has not been honest enough about the answer.

The honest answer is: it depends on who delivers it, how they deliver it, and what goals they pursue. ABA therapy has helped millions of individuals build communication, independence, and quality of life. ABA therapy has also harmed people when it was practiced without regard for their dignity, autonomy, and humanity.

Both things are true. Pretending otherwise helps no one.

What Critics Say (and Why It Matters)

The criticisms of ABA come primarily from autistic adults who experienced ABA as children. Their experiences are not hypothetical. They include:

• Being trained to suppress harmless behaviors. Stimming (hand flapping, rocking, spinning) is a natural self-regulatory behavior for many autistic people. Early ABA programs often targeted stimming for elimination because it looked different, not because it was harmful. Many autistic adults report that losing their stims left them without a coping mechanism.
• Compliance as the primary goal. Some ABA programs focused on making children compliant: sit still, make eye contact, follow instructions without question. Critics argue that training children to comply unconditionally makes them vulnerable and teaches them that their own needs and boundaries do not matter.
• Excessive hours. Some early research recommended 40 hours per week of ABA therapy for young children. Critics point out that 40 hours of any therapy is a full-time job for a child who should also be playing, resting, and being a child. The evidence does not support a "more is always better" approach.
• Historical use of punishment. The earliest ABA programs (Lovaas, 1960s-1980s) used aversive procedures including physical punishment to reduce unwanted behaviors. Some of these practices would be considered abuse by current standards. This history is real and cannot be dismissed.
• "Normalizing" instead of supporting. Critics argue that some ABA programs aim to make autistic children look and act non-autistic, rather than helping them build skills they actually need. The goal of making someone "indistinguishable from peers" (a phrase from early ABA research) implies that being visibly autistic is something that needs to be fixed.

What Has Changed

The ABA field has changed significantly since its earliest days, though the change is not universal. Here is what is different:

1. The BACB Ethics Code prohibits harmful practices

The current BACB Ethics Code for Behavior Analysts explicitly requires practitioners to prioritize client welfare, use the least restrictive effective intervention, and avoid procedures that cause harm or distress when alternatives are available. This is a meaningful shift from the era when aversive procedures were standard.

2. Positive reinforcement is the foundation

Modern ABA practice is built on positive reinforcement, not punishment. The emphasis is on teaching new skills (what the person can do) rather than eliminating behaviors (what the person should stop doing). This is a fundamental philosophical shift, not just a technique change.

3. Assent and client autonomy

The concept of "assent" has entered ABA practice. This means the client has the right to participate or withdraw from therapy activities. If a child says no (verbally or through behavior), that communication is respected, not overridden. The child's preferences influence the session. This is a direct response to the compliance criticism.

4. Naturalistic teaching

Modern ABA increasingly uses naturalistic teaching methods (NET, incidental teaching, pivotal response training) that embed learning in play, daily routines, and the child's natural interests. This is a departure from the table-based, drill-heavy approach of early ABA programs.

5. Functional communication over compliance

Rather than teaching a child to comply silently, modern ABA teaches functional communication: giving the child the tools to express their needs, make choices, and advocate for themselves. A child who can say "I need a break" has more agency than a child who has been trained to sit quietly through distress.

6. Preserving harmless behaviors

Many modern BCBAs no longer target stimming or other harmless behaviors for reduction. If a behavior is not dangerous, does not interfere with learning, and serves a self-regulatory function, ethical practitioners leave it alone. The shift is from "making the person look normal" to "helping the person function and thrive."

What Has Not Changed (The Honest Part)

Not every provider has adopted these changes. The ABA field is large, and quality varies enormously:

• Some providers still focus on compliance. The emphasis on "reducing problem behavior" can slide into training compliance if the BCBA is not thoughtful about goals. If every goal on a treatment plan is about what the child should stop doing, that is a red flag.
• Some providers still recommend excessive hours. Recommending 40 hours per week before assessing the child is a billing decision, not a clinical one. The right number of hours depends on the individual child's needs, and more is not always better.
• Private equity ownership has introduced financial incentives that can conflict with clinical judgment. When an ABA agency is owned by investors who need to show growth, there is pressure to increase billable hours. This can lead to recommending more therapy than a child needs. ESBAP tracks provider ownership and ethics.
• Some practitioners are not listening to autistic adults. The autistic community has been clear about what they want from ABA: autonomy, respect, and functional skills that improve quality of life. Some BCBAs and agencies have integrated this feedback. Others have not.
• The RBT workforce is undertrained and underpaid. RBTs are the people delivering most of the therapy. They start at $15-18/hr, receive only 40 hours of initial training, and often lack ongoing professional development. The gap between what good ABA requires and what the workforce is equipped to deliver is real. (The new 12-PDU requirement is a step toward addressing this.)

What Good ABA Looks Like

This is the section that matters most for families evaluating providers. Good ABA practice:

• Uses positive reinforcement as the primary strategy. The child earns access to preferred activities and items for demonstrating skills. Punishment-based procedures are a last resort after positive strategies have been documented and tried.
• Respects the client's right to say no. If a child resists an activity, the response is to understand why (assess the function), not to force compliance. Assent is built into the session structure.
• Includes goals the family and individual choose. Treatment goals should reflect what matters to the person and their family: communication, independence, safety, social connection. Not "quiet hands" or "good sitting."
• Uses naturalistic teaching. Learning happens in play, during routines, and in the community. Not just at a table with flashcards.
• Preserves harmless behaviors. Stimming, preferred routines, and sensory preferences that are not dangerous and do not interfere with functioning are not treatment targets. They are part of who the person is.
• Teaches functional communication. The child learns to express needs, make requests, refuse activities, and advocate for themselves. Communication is the foundational goal.
• Involves parents and caregivers. Parent training is part of the program. You learn strategies to support your child outside of therapy sessions. The goal is generalization, not dependency on the therapist.
• Measures progress with data. The BCBA shows you data at regular intervals. You can see whether the treatment is working. See our provider evaluation guide.
• Is delivered by credentialed, supervised professionals. RBTs with the BACB credential, supervised by BCBAs who observe sessions regularly and provide clinical guidance. Learn about the RBT role.

What Bad ABA Looks Like

• Every goal is about reducing behaviors, not teaching skills
• The child is never allowed to say no or take a break
• Stimming is targeted for elimination without clinical justification
• The provider recommends 40 hours/week before assessing your child
• You are not allowed to observe sessions
• The BCBA is never present during sessions
• There is no parent training component
• Progress is described with vague words ("doing great") instead of data
• The focus is on making the child "indistinguishable" or "normal"
• High staff turnover, frequently changing therapists

What Parents Should Do

• Read the criticisms. Understanding what autistic adults have experienced helps you ask better questions and recognize red flags.
• Evaluate the provider, not just the field. "ABA" is not one thing. It is a set of principles applied by individual practitioners with varying competence, ethics, and philosophy. The provider you choose matters more than the label of the therapy.
• Ask about goals. If the treatment plan focuses on teaching communication, independence, and life skills, that is a good sign. If it focuses primarily on compliance and behavior elimination, ask why.
• Observe sessions. Watch what actually happens. Your child's comfort, engagement, and autonomy during sessions tell you more than any brochure.
• Listen to your child. If your child consistently resists therapy, that is information. It may mean the approach needs to change, the therapist is not a good fit, or the goals are not right. Do not override your child's distress without understanding it.
• Use the ESBAP directory. esbap.org rates providers on ethical practices, not just existence. This helps you find providers who practice the kind of ABA described in the "good ABA" section above.

What Behavior Analysts Should Do

• Listen to autistic adults. Not to defend your practice, but to understand their experience. If someone tells you that ABA harmed them, believe them. Their experience is valid even if your practice looks different today.
• Audit your own goals. Look at your current caseload. How many treatment goals are about teaching skills versus eliminating behaviors? If the ratio is heavily skewed toward reduction, reconsider.
• Build assent into every session. Give clients meaningful choices. Respect "no." Create escape routes that are not contingent on compliance first.
• Stop targeting harmless behaviors. If stimming is not dangerous and does not prevent learning, leave it alone. Your job is to increase quality of life, not to make someone look neurotypical.
• Invest in your RBTs. The people delivering the therapy need more than 40 hours of training and $16/hr. Professional development, quality supervision, and fair compensation produce better outcomes for clients. See the PDU compliance guide for supervisors.

Our Position

Special Learning has been in the ABA field since 2010. We train behavior analysts and behavior technicians. We have a stake in the field's reputation. Here is what we believe:

• ABA is a powerful science that can improve lives when practiced ethically.
• ABA has caused real harm when practiced without regard for the person's autonomy, dignity, and humanity.
• The field has improved. Not enough. Not everywhere. But meaningfully.
• The quality of the provider matters more than the label of the therapy.
• Autistic voices must be part of how ABA evolves. Not as an afterthought, but as a core input.
• Providers who still use aversive procedures, target harmless behaviors, or prioritize compliance over wellbeing should be held accountable.
• ESBAP exists because accountability requires transparency. Rating providers on ethical practices, not just whether they exist, is how the field earns trust.

Related: How to Choose a Provider | What Is an RBT? | Insurance by State | Resources for Parents | RBT FAQ