You must submit to my will in order to get a big smile
February 14, 2020 7:19 AM   Subscribe

If it turns out that, contrary to widespread assumptions, behavior modification techniques aren’t supported by solid data even when used with autistic kids, why would we persist in manipulating anyone with positive reinforcement? A rigorous new meta-analysis utterly debunks the claim that applied behavior analysis (ABA) therapy is the only intervention for children with autism that’s “evidence-based.” In fact, it raises serious questions about whether ABA merits that description at all. Previously on ABA from an autistic point of view: [1], [2].
posted by sciatrix (15 comments total) 31 users marked this as a favorite
 
The actual paper is buried, so here is the abstract, but it's worth mentioning that it looks like only the abstract is available and the paper isn't available until November. I have access to all APA papers through EBSCO, but I cannot access the full paper either. If anyone has it I would love to read it.

I'm agnostic about the findings of the paper until I can actually read it, but I think the findings are a bit more nuanced than what is claimed here. When non randomized control trials were examined, ABA, naturalistic developmental behavioral intervention, and developmental approaches were all effective, as measured by the outcomes considered in this study. When only RCTs were considered, ABA was not found to be effective. When only RCTs with no detection bias were considered, no therapies at all were found to be effective.

I agree that it can be said, based again just on the abstract, that it is not true that ABA is the only evidence-based therapy, as is often claimed. It can also be said based on these results that there are no evidence-based therapies at all. But again - meta-analyses have their own statistical methods that are complicated and worthy of scrutiny, and it's hard to say because the methods aren't available.

I've worked on a team with psychologists who use ABA as a sort of last ditch effort to try and reduce severe violent and aggressive behaviors in children who are generally not diagnosed with ASD but with other aggression disorders. Parents who are at their wits' end come from all over the country to this specialty clinic for two week long treatments. Anecdotally I have seen it work wonders, but it's also important to understand that the goal in these treatments at this particular clinic is to improve safety for the child and for family members by reducing violent and injurious behaviors, rather than try and force "normal" behaviors or stop harmless repetitive behaviors or force eye contact or anything like that. The psychologists I know who use ABA would not recommend it to try and make a child with ASD be more like a neurotypical person, but I understand it is sometimes used this way which is absolutely problematic.

It is important to take meta-analyses very seriously as they are the highest level of evidence we typically have in clinical settings, and I'm looking forward to reading this paper.
posted by Lutoslawski at 8:01 AM on February 14 [14 favorites]


I don't know anything about ABA, but I was an aspie kid. Every time someone in charge of me- usually a teacher- tried to reward good behavior with tokens, I knew exactly what they were doing. I never reacted well to someone trying to buy good behavior to try and control me.

If I'd been forced into something like ABA, it would have been a total catastrophe.
posted by BungaDunga at 8:21 AM on February 14 [6 favorites]


I actually do have institutional access to a preprint; you can see it here.
posted by sciatrix at 8:24 AM on February 14 [9 favorites]


The psychologists I know who use ABA would not recommend it to try and make a child with ASD be more like a neurotypical person, but I understand it is sometimes used this way which is absolutely problematic.

"Sometimes" is rather understating it. I don't doubt that there are useful applications for it, but within the world of autism it is mostly if not all but exclusively used to try and make people more palatable to neurotypicals by suppressing their autistic traits.
posted by Dysk at 8:25 AM on February 14 [14 favorites]


I don't know anything about ABA, but I was an aspie kid. Every time someone in charge of me- usually a teacher- tried to reward good behavior with tokens, I knew exactly what they were doing. I never reacted well to someone trying to buy good behavior to try and control me.

If I'd been forced into something like ABA, it would have been a total catastrophe.


Two points: first, ABA is generally used for more low functioning individuals, typically nonverbal where you can't figure out what they want (what is "reinforcing") by simply asking them. Two, if a reinforcer such as a token economy isn't working, then it isn't a reinforcer. We all have reinforcers, but not all can be feasibly used in a behavioral plan. Feeling in control of one's self could be a reinforcer but it might be hard for a teacher to provide it systematically. At some point however natural consequences take over, kids get bigger, go away, and may or may not find reinforcers that work for them in the "real world".
posted by fraxil at 8:46 AM on February 14 [1 favorite]


Thank you for posting the paper. Indeed, this is a rigorous and very good meta-analyses.

I will point out two important things.

With respect to ABA specifically: It should be reiterated that we exclusively synthesized findings from randomized and nonrandomized group design studies of interventions for children with ASD. By excluding studies with single group pretest-posttest designs and SSDs, we have omitted a substantial body of research that has been used to draw conclusions about evidence-based practice, particularly in regards to the effectiveness of behavior analytic approaches. In fact, as of 2015, the majority of the available studies of intervention techniques for children with autism employed SSD (Wong etal., 2015), though our review and other reviews published since attest to the recent precipitous increase in group design literature published in this field (French & Kennedy, 2017).

Excluding single-subject designs and single group pre/post designs is reasonable for their purposes. But these do make up a lot of the apparent evidence-base for ABA, and this particular analysis ignores those, and that should be mentioned in takeaways from this study.

Second, more than concluding that therapies do not work, this analysis indicates that we lack rigorous, well-designed studies of therapy effectiveness. Which is true of basically all clinical practices from pharmaceuticals to speech therapy and which is a huge problem in general and one unlikely to go away given the slow chipping away at the NIH and such.

This is not in any way a comment on the effectiveness or ineffectiveness of ABA, but just to remind folks that it's important when talking about findings from papers to actually look at what the paper found and how they found it.
posted by Lutoslawski at 8:54 AM on February 14 [4 favorites]


Parents will continue to turn to ABA until we as a society realize that families of high-support-needs toddlers and children (no matter the child's diagnosis) need assistance, respite care, and support. They need childcare, and in home help, and a way to access therapy for themselves.

Until we, as a society, make those things available, make support available and easy to access, ABA and similar therapies will continue to thrive, evidence be damned.
posted by anastasiav at 9:01 AM on February 14 [19 favorites]


Please don't use functioning labels. They're inaccurate and harmful.
posted by stoneweaver at 9:43 AM on February 14 [20 favorites]


I'll also note that I have preemptively removed this thread from my recent activity because people debating whether or not ABA is ok is harmful to my mental health and it's pretty clear that's where this thread is already heading. Please do read the previously's linked on this post before you go down that path.
posted by stoneweaver at 9:46 AM on February 14 [11 favorites]


Two points: first, ABA is generally used for more low functioning individuals, typically nonverbal where you can't figure out what they want (what is "reinforcing") by simply asking them.

Your blithe confidence in a clear distinction between these categories of people is certainly bracing. I often have trouble distilling things down into words when my resources are low and I'm overtaxed, though.

The use of "them" here is also very, very telling. So is "individuals."

Two, if a reinforcer such as a token economy isn't working, then it isn't a reinforcer. We all have reinforcers, but not all can be feasibly used in a behavioral plan.

Please read the linked piece putting reinforcers into context. Seriously. Engage with the linked blog post, because it discusses reinforcers at length--and the ways that over-reinforcing can badly backfire.
posted by sciatrix at 10:09 AM on February 14 [10 favorites]


[few comments removed - this is very much not going to become a thread for re-litigation of self-diagnosis for ASD. Please respect the neurodiversity of this community when commenting in this thread.]
posted by jessamyn (staff) at 11:15 AM on February 14 [14 favorites]


Contrary to popular belief, autistic people are fully formed individuals, and not animals. We do not require training.
posted by a power-tie-wearing she-capitalist at 11:18 AM on February 14 [13 favorites]


love 2 have u back jessamyn 🙂
posted by sixswitch at 11:53 AM on February 14 [9 favorites]


Anecdotally, I'm grateful for the ABA therapy I participated in as a child. My ability to walk and talk among neurotypicals is predicated on the hard work that a lot of good people put into me. I am a top-notch masker, and this skill has generally been to my profit in interfacing with society.
posted by Construction Concern at 2:19 PM on February 14 [2 favorites]


Behavior therapy is very effective for many problems. Behavioral activation is one of the most effective treatments for depression; exposure therapy works wonders for anxiety and PTSD; dialectical behavior therapy reduces suicidality and self-harm. I use a lot of behavioral techniques with my clients, because they work. Not always, not for everyone--but for enough people that I really can't discount it, even though I hate Skinnerian behaviorism.

The thing is, those techniques only work because they're fully based on the client's goals with their consent and direction. Behavior therapy only works if it's self-directed. Behavioral activation doesn't work if you force depressed people to exercise. Exposure therapy doesn't work if you hold people down and shove their fear in their face. And someone is going to say: but, we have to do that with kids, because they don't know what's good for them! Bullshit. Even exposure therapy for kids explicitly requires the kid be on board and agree to the exposures. It won't work otherwise! I mean, it might change the behavior, yeah. But they're still going to be terrified. And that's not treatment. That's not a cure. That's just abuse.

Please understand, this is not me saying that "well ABA is fine if you do it right." ABA is rotten to the core, because by its very nature, there is no self-direction whatsoever. It's someone else analyzing your behavior and deciding what to do to change it. In exposure therapy, you come up with the fear hierarchy. You decide what your exposure homework is going to be. You do it, for you, not for the therapist/parent/whoever. With kids, the parents and therapist help, but it's still under the kid's control. ABA is all about taking away control. That's what makes ABA abusive.

I haven't reconciled how I feel about ABA for self-injurious behavior (the same way I haven't figured out how I feel about institutionalizing people who try to commit suicide), but that is not the vast majority of ABA that's happening, and it's disingenuous to pretend, "We have to do ABA so the poor kids don't hurt themselves." ABA "targets" a hell of a lot more than that. The majority of people with autism do not engage in self-injurious behavior, and yet it's the "gold standard" for autism, and up until a little while ago, it was the only autism therapy insurance would cover.

Two points: first, ABA is generally used for more low functioning individuals, typically nonverbal where you can't figure out what they want (what is "reinforcing") by simply asking them.

There are many alternative communication methods that parents and professionals don't even try, often with the misguided fear that it will limit their verbal speech production (because of course, communicating any other way is viewed as inferior). Also, if your client can't tell you whether or not what you're doing is hurting them, you more obligated to be sure you're doing what they want/need, not less.
posted by brook horse at 8:08 PM on February 14 [12 favorites]


« Older Living as a whole person instead of an idea of a...   |   Even graduate students think poop is funny ... Newer »


You are not currently logged in. Log in or create a new account to post comments.