Young specialist consulting with her experienced colleague while pointing at one of evidence of crime on board

Double Standards: ABA vs. Facilitated Communication10 min read

There were a couple of things that spurred my dive into research regarding facil­i­tated com­mu­ni­ca­tion (FC) and other methods of alter­na­tive and aug­men­ta­tive com­mu­ni­ca­tion (AAC). One that I have men­tioned before is that I watched a few doc­u­men­taries fea­turing autistic people who spell or type to com­mu­ni­cate.

Another was that I very quickly learned about applied behav­iour analysis (ABA) when I became involved with the autistic com­mu­nity, par­tic­u­larly about its poten­tial for damage to young neu­ro­di­ver­gent minds. I imme­di­ately thought back to those doc­u­men­taries I’d seen and I thought to myself, “I bet a lot of these kids with ‘chal­lenging behav­iour’ are just frus­trated that they can’t com­mu­ni­cate with anyone. Why aren’t we pushing com­mu­ni­ca­tion first?” That’s when I dis­cov­ered that FC wasn’t con­sid­ered “evidence-based” and started doing my research.

Critics con­tinue to claim that I believe that FC can be effi­ca­cious because I allegedly have no under­standing of the sci­ence and allegedly refuse to ques­tion it. They are wrong. I cer­tainly had a moment of crisis where I con­sid­ered whether I had been duped into believing some­thing that couldn’t be true. It was through reading both sides that I came to my con­clu­sion. Meanwhile, critics of FC (and of other AAC methods) have simply come up with more com­pli­cated “rea­soning” as to why it can’t pos­sibly work ever. (See: shifting the goal­posts.)

However, there was some­thing pecu­liar that I noticed throughout my researching. Whenever I argued for the effi­cacy of cer­tain AAC methods, I would often get crit­i­cisms of the evi­dence that sup­ports it… using argu­ments that, if applied to ABA, would not hold up at all. This was espe­cially pecu­liar con­sid­ering the people who crit­i­cize FC usu­ally pro­mote ABA.

Far be it for me to accuse anyone of being an ABA shill, but I do see a pat­tern of refusing to rec­og­nize the poten­tial for effi­cacy in spe­cific methods of learning while com­pletely ignoring evi­dence of inef­fi­cacy in the one that you do. Let’s go through those together using facil­i­tated com­mu­ni­ca­tion as an example.

They say FC can only be val­i­dated via double-blinded exper­i­ments in con­trolled set­tings…

The uni­versal way of val­i­dating someone’s iden­tity is to ask them ques­tions that someone who isn’t them would not know. We do this all the time when we deal with sen­si­tive infor­ma­tion. People who com­mu­ni­cate with AAC also pass infor­ma­tion that their sup­port people never knew. Ask anyone who per­son­ally knows someone who uses AAC, and they can verify this for you.

Critics take the posi­tion that the only way to ensure author­ship of the words pro­duced through FC is for the method to be val­i­dated in double-blinded exper­i­ments in con­trolled set­tings. It doesn’t matter if you’ve passed infor­ma­tion out­side of the con­trolled set­tings (which has occurred in many qual­i­ta­tive sci­en­tific studies); it has to be this method, nothing else.

(Skeptics con­stantly shift the goal­posts for what con­sti­tutes proper evi­dence, though, and I’m sure once there is at least one study avail­able, we’ll be hearing critics saying that one study doesn’t trump the other ones…)

…while ignoring the fact that evi­dence using the sci­en­tific “gold stan­dard” for val­i­dating ABA out­comes (ran­dom­ized con­trolled trials, or RCTs) is scant.

Meanwhile, when con­sid­ering the evi­dence of ABA prin­ci­ples being used to train skills, the “gold stan­dard” of evi­dence would be via long-term ran­dom­ized con­trolled trials. Essentially, autistic chil­dren would have to be ran­domly divided into two groups, one receiving an ABA type of therapy and one receiving a dif­ferent type (or no therapy at all). The people observing would also have to be unaware of which therapy (or lack thereof) the sub­jects were receiving, in order to pre­vent bias in the results.

You might think that, for a field that insists it is evidence-based when­ever it receives crit­i­cism, there would be a lot of rig­orous studies of this kind to back up those claims. However, ABA studies using ran­dom­ized con­trolled trials are actu­ally few and far between.

Lovaas’s 1987 study is con­sid­ered the first of this kind (though the sub­jects weren’t actu­ally ran­domly assigned to groups due to parental objec­tions and eth­ical con­sid­er­a­tions). Regardless, 47% of the group receiving inten­sive behav­ioural therapy “achieved normal intel­lec­tual and edu­ca­tional func­tioning.” This number is often used as the defin­i­tive proof that ABA works. I’ll come back to this study later.

There have only been a few studies done since then that qualify. One of the rea­sons is cost. RCT-type studies are very expen­sive (in the mil­lions). Another is that ABA is already assumed to be the stan­dard of care for treating autism spec­trum dis­order and with­holding ABA from autistic chil­dren is “uneth­ical.” You see the conun­drum we’ve found our­selves in here.

A 2009 meta-analysis of applied behavior inter­ven­tion found inad­e­quate evi­dence for out­comes. A 2018 Cochrane review found evi­dence in favour of early inten­sive behav­ioural inter­ven­tion to be “weak.” Behavioural ana­lysts refuted these find­ings, of course, claiming that the very strict stan­dards left out the vast majority of research sup­porting pos­i­tive out­comes for ABA through other exper­i­mental methods.

When I make that same argu­ment for FC — that the pre­pon­der­ance of peer-reviewed arti­cles sup­ports valid author­ship and that the sys­tem­atic reviews claiming it as invalid ignore them entirely — I’m called anti-science.


They say FC should not be pro­moted because the poten­tial for abuse is too high… 

Whenever FC is brought up, I get bom­barded by exam­ples of legal cases in which FC users made abuse alle­ga­tions or in which their com­mu­ni­ca­tion method was used to jus­tify murder, in one case.

In the latter case, the woman was con­victed of manslaughter for killing her son. I, of course, agree with this ruling. However, the mother in this case also had psy­chi­atric issues and no formal training in FC. In fact, in sev­eral of the cases where alle­ga­tions were made, the facil­i­tator had no formal training.

These are awful sit­u­a­tions. I have sym­pathy for every person who is accused of some­thing they didn’t do. However, false alle­ga­tions are not unique to FC. Authorship tests were not com­pleted in every one of these cases, but even if author­ship had been val­i­dated, every person is still capable of lying.

Furthermore, there have been abuse alle­ga­tions made through FC which were later ver­i­fied. Downplaying these alle­ga­tions instead of inves­ti­gating them would have left the chil­dren here much more vul­ner­able than removing their com­mu­ni­ca­tion method alto­gether would have.

As well, the com­plete denial of any effi­cacy of FC also results in abuse. Sharisa Kochmeister, a tire­less autistic self-advocate and former pres­i­dent of the Autism National Committee, was removed from her family’s care against her will and placed into a group home, despite using her method of com­mu­ni­cating (typing with one finger) to ask for a lawyer.

But of course, critics aren’t inter­ested in cases like hers. As I have said before, when people talk about pro­tecting non­speakers from “abuse,” they are more con­cerned with what non­speakers might say about others than they are with the right to com­mu­ni­cate.

…while ignoring the fact that ABA-type methods have been linked to both phys­ical and psy­cho­log­ical abuse.

Flipping back to ABA and the double stan­dards applied, there is plenty of evi­dence of abuse in ABA-type methods. Physical abuse occurs through the use of aver­sives, and psy­cho­log­ical abuse occurs through long-term use of ABA-style inten­sive “therapy,” which is what most autistic chil­dren are “pre­scribed” as the stan­dard of care.

Among advo­cates, the most well-known use of aver­sives as phys­ical abuse is at the Judge Rotenberg Center. Purportedly, the JRC is one of the only insti­tu­tions to still use aver­sives as part of behav­ioural therapy, and one of the only ones still using elec­tric shocks as a form of behav­iour mod­i­fi­ca­tion. By human rights stan­dards, it qual­i­fies as tor­ture.

Despite the asser­tion from behav­ioural ther­a­pists that aver­sives are no longer in use with modern ABA, the JRC has con­tinued to have a pres­ence at the Association for Behavior Analysis International’s annual con­ven­tion. This year’s mate­rials included sev­eral dis­cus­sions on the alleged merits of skin shock as pun­ish­ment.

As for psy­cho­log­ical abuse in ABA, I per­son­ally wrote an article about it, so I won’t rehash it here. This article in Cogent Psychology was my main source of inspi­ra­tion.

Interestingly, non­speakers who later learned to type or spell to com­mu­ni­cate have voiced their thoughts on ABA and how it was pretty much use­less for them. I can’t help but wonder if the pat­tern of non­speakers against ABA is one of the rea­sons ABA pro­po­nents are unwilling to accept alter­na­tive com­mu­ni­ca­tion methods as valid.

They down­play tes­ti­mo­nials from the thou­sands of par­ents who rec­og­nize their child’s authentic com­mu­ni­ca­tion… 

United for Communication Choice has esti­mated via informal polling that there are about 5,000 chil­dren in the United States who com­mu­ni­cate by typing or pointing to let­ter­boards. That’s thou­sands of par­ents seeing their non­speaking (or par­tially speaking) chil­dren improving lit­eracy skills and motor func­tioning, and plenty of anec­dotal evi­dence for val­i­dated author­ship.

According to these critics, these par­ents are vic­tims of emo­tional manip­u­la­tion and false adver­tising on the part of those pro­moting FC, RPM, or another method. It’s that whole con­de­scending “Of course, par­ents want to believe their kids are talking to them, but I, a person who has never met those kids, know better.”

…while claiming that tes­ti­mo­nials from the thou­sands of par­ents who are con­vinced that ABA helped their kids develop are valid evi­dence of its effi­cacy.

It doesn’t matter how many times autistic adults write about how ABA has been dam­aging to their psyche. Proponents are still going to point at the par­ents who sup­port it for their chil­dren as proof that it works (again, without any studies com­paring devel­op­ment with ABA treat­ment to devel­op­ment with no treat­ment, or recog­ni­tion that some studies have found that the number of hours has no rela­tion to diag­nostic out­come.)

Have I men­tioned that there isn’t any evi­dence regarding the psy­cho­log­ical well-being of autistic chil­dren exposed to long-term ABA? Most, if not all, of the evi­dence is based on parental reporting and obser­va­tions of behav­iour. They report on whether behav­iours changed or not. They don’t report on whether or not the child’s psy­cho­log­ical well-being has improved. I wonder why that is.

Remember when I said I would come back to Lovaas? The 47% number was achieved with the use of aver­sives as pun­ish­ment, and Lovaas cred­ited aver­sives with the achieve­ment. Any use of that number as proof for the effi­cacy of “modern” ABA (which I am told does not use aver­sives) is actual false adver­tising.

I keep hearing from local people that ABA in my area can cost up to $80,000 a year. $80,000 a year for a “therapy” which could poten­tially be phys­i­cally abu­sive and is almost cer­tainly psy­cho­log­i­cally abu­sive (and par­tic­u­larly so for non­speaking chil­dren). It would take me four years to make that much money! I can’t imagine the kind of debt some par­ents must have gone into because these so-called experts claimed that their chil­dren would never reach cer­tain devel­op­mental mile­stones if they aren’t imme­di­ately put into inten­sive one-on-one behav­ioural therapy.

And where’s the evi­dence sup­porting that? Oh, right, it would be uneth­ical to deprive autistic chil­dren of ABA to test for that evi­dence. How con­ve­nient.

Is there some kind of agenda at play? It’s hard to say for sure.

Ultimately, though, when you com­pare between these two methods of teaching (and this is, in fact, what they both are), I have a few things to note…

ABA methods…

  • mea­sure intel­li­gence by obser­va­tions of behav­iour
  • no pre­sump­tion of com­pe­tence is made
  • con­sider move­ments and ability to make speech as reflec­tions of intel­li­gence
  • have min­imal under­standing of autism (no spe­cial­iza­tion required to become an RBT, for example) or base it on the out­dated clas­si­fi­ca­tion as a behav­ioural dis­order
  • often dis­re­gard research regarding motor func­tioning and sen­sory dif­fer­ences
  • foster depen­dence (cre­ating an over-reliance on prompting and external rewards)
  • social relat­ed­ness is down­played; one must per­form socially accept­able behav­iours to be accepted, often without under­standing of those behav­iours

Spelling and typing methods…

  • mea­sure intel­li­gence by helping them com­mu­ni­cate their thoughts and desires in methods suited to their needs
  • pre­sump­tion of com­pe­tence is made
  • con­sider move­ments and ability to make speech in the con­text of someone who may not have reli­able con­trol over motor func­tioning
  • recog­ni­tion of motor func­tioning and sen­sory dif­fer­ences expe­ri­enced by autis­tics and con­firmed by recent research
  • foster inde­pen­dence (fading of prompts is an essen­tial part of RPM and other spelling methods; unsup­ported typing is the end goal for FC users, when pos­sible)
  • social relat­ed­ness is encour­aged; one can use their indi­vidual method of com­mu­ni­ca­tion to socialize with and under­stand others

Between ABA-type “ther­a­pies” and FC training, both methods do poorly in terms of evi­dence when they’re held to “gold stan­dard” testing, and both methods rely strongly on anec­dotal and indi­vid­u­al­ized evi­dence for effi­cacy. But only one of the two is widely denounced as “pseu­do­science.”

Among autistic self-advocacy groups and indi­vidual activists, ABA is almost uni­ver­sally con­demned, while com­mu­ni­ca­tion methods are encour­aged. Among those who rec­og­nize the value of neu­ro­di­ver­gent brains, ABA is uneth­ical and dam­ages intrinsic moti­va­tion, but com­mu­ni­ca­tion methods, while still not fully reli­able some­times, rep­re­sent the chance for autonomy and accep­tance of dif­fer­ences. Among non­speakers, ABA is use­less, and real com­mu­ni­ca­tion is freedom.

One of the asser­tions made by ABA pro­po­nents is that those of us who are opposed to behav­iourist prin­ci­ples in teaching autistic chil­dren are “high-functioning” and that ABA is most helpful for “low-functioning” autis­tics. Yet, when someone whose pri­mary com­mu­ni­ca­tion method is through typing or spelling says that ABA did nothing for them, they are silenced, ignored, or oth­er­wise con­de­scended to, by being told their words actu­ally came from someone else.

Again… how con­ve­nient.

Editor’s Note:
In cre­ating this article, I ref­er­enced quite a bit of infor­ma­tion com­piled by autistic researcher Michelle Dawson, and I would be remiss if I didn’t give her credit for all of the excel­lent work she has done, par­tic­u­larly in eth­ical vio­la­tions sur­rounding behav­ioural therapy.

This article is also pub­lished at the author’s blog, NeuroInsurgent.

Ren Everett

Latest posts by Ren Everett (see all)


  1. If both methods do poorly in testing sit­u­a­tions then they should both be rejected.

    1. RPM has not been tested. Again, this is flawed rea­soning to con­flate two ways to com­mu­ni­cate.

      1. Then it shouldn’t be used until it’s been prop­erly tested.

    2. Author

      With respect, I dis­agree. Your com­ment reflects an ide­ology of “sci­en­tism” — which you may or may not have heard of, but essen­tially it means that you are arguing that if some­thing hasn’t been tested via the sci­en­tific method, it can never be trusted. This is not true. There are many methods by which you can val­i­date some­thing that don’t involve con­trolled set­tings.

      Like I said in this very article, “People who com­mu­ni­cate with AAC also pass infor­ma­tion that their sup­port people never knew. Ask anyone who per­son­ally knows someone who uses AAC, and they can verify this for you.” You don’t need sci­ence to verify AAC for per­sonal use; the people who know the person will be able to tell if it is useful or not.

      Meanwhile, ABA-principled ther­a­pies are based on obser­va­tional accounts of people out­side of the “patient,” do not take into account sub­jec­tive expe­ri­ences of autism, and are widely dis­cour­aged as abu­sive by the people they are sup­posed to help. If you take that wide­spread denounce­ment and com­bine it with lack of rig­orous evi­dence, cor­rect, it should be rejected.

      FC, on the other hand, should instead just be approached with cau­tion. It doesn’t work for everyone; anyone involved with FC will freely acknowl­edge this and have been doing so for decades. But it does work for some people. Those people should *not* be denied the chance to com­mu­ni­cate based on lack of sci­en­tific testing. How long do you think these people should stay silent — par­tic­u­larly when their method of com­mu­ni­ca­tion has already been dragged through the mud by skep­tics who don’t care to under­stand apraxia of speech or the process of how they learned to type?

      Another example is that sen­sory inte­gra­tion or sen­sory diets are not con­sid­ered “evidence-based” but there is plenty of anec­dotal evi­dence for them being helpful. If it is very clearly helping someone and clearly not harming them, why should it be rejected?

      It’s a bal­ancing act, not a black-and-white issue.

      1. I know what sci­en­tism means but I fail to see how it’s rel­e­vant here. This is some­thing that falls within the sphere of sci­ence. I want med­i­cines and ther­a­pies to pass sci­en­tific scrutiny.

          1. Oh now it’s a method of teaching? Because the way defenders talk about it, you’d think it was a therapy. Feels like you’re shifting the goal­posts here.

          2. Author

            What do you mean “now”? FC has been called “facil­i­tated com­mu­ni­ca­tion training” for decades in Australia.

  2. I’m glad I saw this. I’ve just fin­ished reading a novel that has an autistic child as the child of the two main char­ac­ters (The Perfect Wife, by JP Delaney) and the author has an autistic adult son. He (the char­acter as well as the author’s son) is sub­jected to ABA and aver­sives. When I wrote my review I shall def­i­nitely be ref­er­encing this post.

  3. I ‘enjoyed’ this read Ren, thanks for pulling it together. It alarms me that any parent would want to put the amount of effort into changing their child in the way that ABA pur­ports to do. How could we ever start without an assump­tion of com­pe­tence, intel­li­gence and empathy? As a parent of an both autistic and NT chil­dren, I believe it’s my job to help them find the ways for them to interact and engage with the rest of the world and society that allow them to be the best ver­sions of them­selves. I hope to bring the same approach to any neu­ro­di­ver­gent indi­vidual I work with as well.

  4. I’ve read both sides of the argu­ment thus far, and per­son­ally I wonder if there isn’t a much sim­pler solu­tion right under our noses that nobody has thought of yet.

    Eye-tracking tech­nology has the poten­tial to do every­thing facil­i­tated com­mu­ni­ca­tion and its rel­a­tives are capable of doing, and as it does not require facil­i­ta­tion at all (being AI-based), so it rules out the pos­si­bility of the ideo­motor effect or any kind of attempt at deceit from a facil­i­tator altering what­ever is being com­mu­ni­cated. Lying would still be pos­sible of course, but it would at min­imum allow the assur­ance that it is the autistic person alone who is com­mu­ni­cating.

    Additionally, it avoids the alle­ga­tions of not being “evidence-based” because it has already been found to be effec­tive in other con­di­tions that impair communication- it is com­monly used in cases of locked-in syn­drome, an obscure neu­ro­log­ical con­di­tion in which the person affected by it is con­scious but inca­pable of com­mu­ni­cating due to total paral­ysis of nearly all vol­un­tary mus­cles. The eye mus­cles are typ­i­cally unaf­fected by this con­di­tion, allowing them to com­mu­ni­cate using eye move­ments alone. Incidentally, this also means that eye-tracking could be effec­tive even for autis­tics whose motor con­trol is so unre­li­able that typing may not be an option.

  5. Somehow I missed until now this second follow-up article by Ren Everett dated July 30, 2019. I have yet to read either follow-up article about Facilitated Communication.

  6. And of course I want to be noti­fied about new com­ments via email.

  7. Then I checked this web­site fur­ther and found that in September there are 4 entries about arti­cles on non­speaking autis­tics deleted from Wikipedia — Tito, Lucy Blackman, Sue Rubin and Amy Sequenzia plus article of Anne McDonald was merged into Rosemary Crossley article. I don’t know when I will have time to read all these entries but I would like to thank Ren Everett for his efforts and the others who helped him. Ben and I are now trying to focus on the issue of research fraud by some of the oppo­nents of Facilitated Communication, an issue which I believe I first pub­licly raised in 2011 as a com­ment on a joint blog of Kim Wombles and James Todd, where my com­ment was later deleted but I saved to my email. I largely got side­tracked, as I often do, but I will try to pursue this issue now.  

  8. I have now read all three July 2019 entries by Ren Everett about Facilitated Communication.

Talk to us... what are you thinking?