Hopefully the fact this post has “behaviour intervention” in the title should act as a content warning to anyone who might find a discussion about ABA triggering, but just to be sure, here’s one now.

–

–

Not long ago, I was complaining about the way advocates and researchers/professionals often talk past one another in the field of ABA.  I grumbled that terms like “ABA” were constantly being understood in different ways, so that even if these groups could get together and have a conversation – no easy thing in itself, given the siloed nature of these communities – it wouldn’t be clear that they could actually make progress. As it is, it’s all to easy for ABA professionals to dismiss autistic people’s arguments because the way autistic people understand “ABA” and related terms differs from how professionals do, and vice versa.

This post is an attempt to help with that.  I’m trying to identify and describe some major concepts that might be important to someone who (for example) wants to write a critical discussion of the ethical merits of ABA and not be dismissed by ABA people on the grounds that the discussion misrepresents what ABA is.

To be clear, I’m not going to talk here about many of the terms used by BCBAs as they go about their daily work.  Like any field, ABA has developed a lot of strange jargon, but you won’t find any talk of “manding” in this post.  I am not trying to write a guide about how to talk like a BCBA – not least because I’m not personally any sort of expert on that jargon. I’m only going to cover major concepts that I think are relevant to a broad-scale, wide-ranging critical discussion of the ethical merits of behaviour interventions.

In addition to of course apologizing to anyone who had traumatic experiences of intervention – which I am of course not trying to invalidate – I’m also going to apologize in advance to anyone who is horrified to see their life’s work crudely summarized in a few oversimplified paragraphs.  This is not meant to fully represent the complexity of this field – which I won’t claim to fully appreciate myself as an outsider – but just to point people in the right direction.

With all that having been said, let’s proceed!

Behaviourism

Behaviourism – including radical behaviourism and other variants – are scholarly and psychological theories, and not intervention practices.  These theories were in the vogue in North America from around the 1910s to the 1970s, but nobody really respects them nowadays. 

Behaviourists basically tried to reduce the very complex world of the human mind to something manageable by focusing on observable behaviour, and how behaviour is affected by events in the environment, rather than the unobservable mind itself.  It was an approach that helped simplify the world of psychology research and it led to some interesting experiments, but it quickly became clear that it was impossible to get very far trying to understand the human mind without actually studying the human mind.  As a result, behaviourist theories nowadays only get a quick overview in introductory psychology textbooks.  Psychology, as a field, has moved on from behaviourism.

I sometimes see articles attacking ABA by pointing out that behaviourism has been discarded as a theory.  I think critical scholars and advocates should be wary of focusing too much on this line of attack.  Modern ABA researchers certainly don’t accept behaviourist theories, and yet they continue to develop ABA interventions that don’t in any way depend on behaviourist theoretical assumptions – indeed, many of these latest interventions overtly depend on an understanding of human development that rejects behaviourist theory.

Applied Behaviour Analysis (ABA)

So, what is ABA anyway?

Back when behaviourism was still a respected scholarly school, researchers did a number of experiments exploring how behaviour could be changed through manipulating environmental inputs.  For example, they used operant conditioning to change behaviour by associating certain behaviours with rewards and punishments.  Even if behaviourism is dead as a theory, the results of these experiments are still sound – of course people’s behaviour can be shaped by rewards and other stimuli from their environments, for better or worse. No anti-ABA advocate disputes that.  Therefore, Applied Behaviour Analysis (ABA) principles can still be used by people seeking to modify behaviour, even if we reject behaviourism as a theory.

In this sense, ABA isn’t a term for any particular style of intervention – it’s a broad set of techniques that can be used by pretty much any intervention seeking to modify behaviour through changing people’s environments.

However, today, the term “ABA” itself has become widely associated with a set of more specific autism intervention practices – practices that have been frequently denounced by autistic people as being (among other things) psychologically damaging, traumatizing, focused on suppression of autistic behaviour, and aimed at increasing compliance and suppressing self-determination.  (More on these practices later.)  Consequently, “ABA” as a term is toxic in most autistic communities, and there are some autistic people who find even the term triggering (hence the content warning).

However, calls by these advocates to “#BanAllABA” don’t usually have much impact in research and professional circles, for the simple reason that ABA isn’t a specific intervention practice.  How could we ban ABA, professionals and researchers ask?  After all, humans naturally use ABA principles all the time – without necessarily realizing it – when we informally try to modify one another’s behaviours.  What would it even mean to ban ABA?  To many, the very idea is incoherent.

To be fair, interventions that explicitly incorporate the language of ABA tend to make false claims to objectivity that aren’t present in everyday human use of behavioural principles, but even so the idea of banning all ABA is unlikely to make much of an impression on those who view ABA as a broad set of principles rather than a concrete intervention or set of interventions.

Ultimately, I personally think that the term “ABA” is not a helpful one.  On the one hand, it is widely associated with specific approaches to autism intervention.  On the other hand, it still technically has this broader meaning in professional and academic circles.  It’s completely toxic in much of the autistic community, but in other circles autism intervention programs literally need to be designated as “ABA” in order to receive funding.  Its meaning and its connotation are both bitterly contested.  As such, the term “ABA” doesn’t really do the job of a term, which is to convey information.

Discrete Trial Training (DTT)

Discrete trial training was the first explicitly ABA-based intervention to be widely adopted in the autism field.  There were several experiments with behaviour intervention and autism beginning in the 1960s, but the field only really took off in 1987, when Ivar Lovaas published his (in)famous paper claiming that 47% of young children in an intensive intervention group achieved “normal intellectual and educational functioning” and indeed became “indistinguishable from their normal friends,” compared to just 2% of children in the control group.  The intervention in question was discrete trial training (DTT) delivered at an extremely high intensity of 40 hours/week.

This study was a source of hope for families of newly-diagnosed autistic toddlers who were essentially being told by clinicians that their children were doomed to lives of failure, misery, and isolation as a result of their autism, and many of these families banded together and campaigned for widespread availability of high-intensity behaviour intervention.

DTT in its original form is a very directive, compliance-oriented approach.  The interventionist sits down with the child – typically at a table – and proceeds to use arbitrary rewards (e.g., preferred foods, toys, etc. – things not related to the topic of the “lesson”) to encourage the autistic child to behave in certain ways.  In general, this has included suppression of autistic behaviours (such as stimming) as well as encouragement of more typical behaviours (such as eye contact or spoken language).

If you look at old videotapes of classic DTT from the 1990s, they are horrifying – nearly unbearable to watch.  Screaming, protesting autistic children are seated in front of impassive interventionists until at last the autistics, exhausted and defeated, bow to the demands of the interventionists.  Moreover, back then, aversives/punishments were still widely used, although this has since changed (even the infamous Judge Rotenberg Center was recently banned from subjecting neurodivergent people to electric shocks).

Modern DTT has mellowed somewhat, but it is still widely criticized – not just by autistic advocates, but also by many researchers – for being directive, authoritarian, prompt-dependent, uninformed by the developmental science of how actually children learn, and insufficiently focused on generalization of skills across multiple settings.

In some research and professional circles, people still think that DTT might be useful for autistic individuals with lower scores on standardized developmental assessments. The argument is that these people could need more structured and non-naturalistic teaching opportunities than autistics with more subtle autistic characteristics, but there’s little evidence to support this assertion.  DTT seems to be no more effective than competing naturalistic developmental behaviour interventions (NDBIs) in promoting the outcomes researchers usually focus on (e.g., communication skills), and everything said by autistic advocates suggests it has the danger to come with serious costs in domains such as mental health and self-determination.

Naturalistic Developmental Behaviour Interventions (NDBIs)

So, what are these NDBIs?  The term “NDBI” itself was only coined relatively recently, but it refers to a variety of approximately similar behaviour intervention curriculums like PRT, ESDM, JASPER, and others.

There are several key differences between NDBIs and DTT.  These are important for advocates to bear in mind, as I’ve read many attacks on “ABA” that criticize “ABA” in general for aspects of DTT interventions that aren’t actually present in NDBIs. Such attacks on “ABA” are of course unlikely to make an impression on NDBI researchers.

First, NDBIs are informed by developmental science.  Whereas DTT has strongly behaviourist roots, NDBIs explicitly reject behaviourism and instead bring in the insights of modern psychology.  At their core, NDBIs are based on the idea that all children need opportunities to learn through social interaction in their naturalistic environments, and the intervention is about setting up those opportunities.

NDBIs are also – at least theoretically – child-led.  Granted, there’s an inherent asymmetry of power between a 30-year-old interventionist and a 3-year-old child that is difficult to eliminate, but where DTT imposes structured learning opportunities in a manner dictated by the interventionist, professional-child interactions in NDBIs are built around the child’s existing interests.  As a result, NDBI interventions will usually look more like developmental interventions (such as FloorTime) than DTT.

However, I don’t want to suggest that NDBIs are perfect.  For example, the way NDBIs incorporate developmental science is very much focused on the idea that the autistic child is on an atypical autistic developmental trajectory that needs to be made more typical.  NDBIs frequently incorporate subjective, value-laden pathology language of “deficit,” “disorder,” and so on.  This needs to change.  We need to recognize that our goal should be to put the individual on their best autistic developmental trajectory, not a typical one.  We need to move away from ideas of pathology and normalization and towards autism acceptance.

One can also question NDBI strategies that force autistic children to engage in social interactions in order to obtain toys they want to play with.  For example, NDBI interventionists sometimes move things that autistic kids want out of their reach and only relinquish the objects if the children engage in interactions with them – but of course, autistic people report social interactions can be stressful, so is forcing interaction in this way OK?  This problem also raises the related question of intervention intensity.  At this point, we just don’t have a lot of good evidence about how much intervention is enough, but a recent study by Sally Rogers et al. shows no real advantage of 25 hours/week of intervention over 15 hours/week.

Overall, my own personal opinion – which is just a personal opinion – is that these sorts of problems can be reformed out of NDBI curriculums.  We can eliminate pathology language, we can change our goals to promote the best autistic developmental trajectory, and we can work on how to do this while making sure we aren’t stressing out autistic children by imposing excessive interaction.  This would then leave us with an intervention capable of teaching young autistic kids useful adaptive skills in a way that doesn’t cause stress or damage their identities.  These programs can be integrated – as, indeed, they already are – with augmentative and alternative communication (AAC), so that we aren’t trying to force verbal speech on anyone.

However, we would run into a key problem when we tried to take these reformed NDBIs into the community.  The different manualized NDBI curricula tend to be studied by researchers, and they tend not to be used by professionals in the “real world.”  In communities, professionals might use eclectic approaches that mix up elements of NDBIs with aspects of DTT.  Thus, even if families do their research and figure out what all the different practices and acronyms are, this knowledge won’t necessarily help them find a good program in their community.

Registered Behaviour Technician (RBT)

Moreover, it’s important to know that, at the end of the day, if it’s not an unpaid parent/caregiver delivering intervention, early interventions for young kids in communities is usually delivered by a Registered Behaviour Technician (RBT).

RBTs exist to minimize the expenses associated with early intervention.  They tend to be college undergrads or recent graduates looking for work (for example, many of the research assistants in autism research labs also become RBTs), and they are given only a little bit of training. I understand that this training is usually focused on basic behavioural principles within an uncritical, simplistic, medical model-based framework.

In theory, RBTs are supposed to be supervised by Board-Certified Behaviour Analysts (BCBAs), but in practice there are so many RBTs per BCBA that meaningful supervision is limited.  Indeed, one could also interrogate the training that BCBAs themselves receive and whether this training incorporates sufficient exposure to critical approaches!

As a result, even if sophisticated and progressive techniques can be developed in research, it’s unclear whether communities have the capacity to implement them.

Privatization

Moreover, in many jurisdictions – across the vast majority of North America, certainly – early intervention for young autistic children is delivered by for-profit companies.

I personally fear that these companies’ profit motives can end up conflicting with the goal of delivering high-quality, stress-free programming to promote autistic children’s learning and adaptive skills without attempting to normalize them or suppress autistic behaviour.  At the very least, it’s unclear that a for-profit company would want to accept the expenses associated with hiring adequately trained, adequately paid staff.  For-profit companies may also have an incentive to fearmonger and promote deficit-based approaches to autism in order to take advantage of parents’ natural anxieties about the futures of their young, newly-diagnosed children.

There are some alternative models.  For example, the Canadian province of Nova Scotia offers a PRT program (remember, PRT is a type of NDBI) delivered by adequately trained and paid staff, but the Nova Scotia program is chronically under-capacity.

I have often wondered if non-profit organizations might be a better approach – one not vulnerable to the same limitations as government-run or fully privatized programs.

Another alternative might be to include autistic children in an inclusive, government-funded preschool system organized along lines comparable to the current school system (except, of course, with a clear right to access integrated settings).  However, this would of course require widespread social reforms going well beyond autism.

Positive Behaviour Support (PBS)

But so far we’ve focused on different models and methods of early interventions aiming to promote learning of key developmental skills.  Another common ABA-based intervention in the community is Positive Behaviour Support (PBS), which is often used in settings like schools, and I can’t conclude this post without addressing it as well.

PBS is fundamentally about managing so-called “challenging behaviours”.  However, instead of using DTT principles to reward people for desired behaviours, PBS is concerned with examining the environment through a “functional behaviour assessment” and essentially looking for anything that might be causing the behaviour.  This environmental cause can then be removed, which should stop the behaviour.

However, I do find this sort of behaviour management talk problematic.  Of course I understand that parents, teachers, and others need to not be encountering “challenging behaviours” all the time, but we also need to take into account the views and perspective of the autistic individual.

Fortunately, if we did so, we might find that the goal of eliminating challenging behaviour would often also be conducive to the well-being of the autistic individual.  First of all, some challenging behaviours – especially self-injurious ones – are inherently damaging.  Furthermore, many of these challenging behaviours might reflect anxiety or stress in the autistic individual. (Anxiety is autistic people with intellectual disabilities is incredibly under-recognized, but very common.)  Alternatively, they might reflect barriers to communication.

Personally, I do think the basic ideas at the core of PBS have a lot of potential.  The techniques are powerful, and they do focus squarely on modifying environments around autistic people rather than on modifying autistic people.  This environmental focus has the potential to be very consistent with the demands of autistic individuals who want to see more done to change neurotypical society to be more accepting and inclusive for autistic people.  Therefore, I think that we could probably reform PBS by changing its goals, such that instead of just talking about suppressing “challenging behaviour,” we could also talk about promoting good mental health, promoting communication, and so on.

In short: instead of focusing merely on the analysis of behaviour, we could reform PBS to include empathy for the perspective of the autistic person.