ABA vs. FC: What ABA knows about autism, instructional needs, and the harmful effects of inadvertent cues

In some ways, ABA and FC are exact opposites.

First, ABA is evidence-based; FC is not.

Second, ABA is based on an understanding of autism that is informed by decades of clinical and neurological research; FC is based on a redefinition of autism that depends on circular reasoning from messages generated via FC. That is, ABA views autism as a neurodevelopmental disorder involving social deficits, communication deficits, and restrictive/repetitive behaviors. FC views autism as a movement disorder in which what look like social and communication deficits are instead the results of some sort of brain-body disconnect whose existence is merely hypothetical.

Third, ABA makes no assumptions about pre-existing skills and knowledge. FC, on the other hand, “presumes competence.”

Fourth, ABA systematically analyzes what needs to be taught and teaches it directly and systematically, subskill by subskill. FC, meanwhile, presuming that language, literacy, and factual knowledge already exist, provides no systematic instruction whatsoever.

But in other ways ABA and FC appear similar. Both are based on prompting/cuing and conditioning. An ABA therapist might present her client with a cookie and a doll and prompt him with “touch cookie”. Over time, the client reliably touches the cookie when asked to. An FC facilitator, meanwhile, provides tactile, auditory, or movement cues that direct his client to type particular letters on a keyboard. Over time, the client reliably touches particular letters when cued to.

In addition, both ABA and FC cues are gradually faded. Initially, the ABA therapist may physically move the child’s hand over to the cookie. Eventually, the child independently selects the cookie from a large array of objects. Initially, the FC therapist holds the child’s typing hand at the wrist. Eventually, she moves the hand from wrist to elbow to shoulder. Or (in the case of the Rapid Prompting Method/RPM variant of FC), initially she holds up one of several letterboards that each contain just a quarter of the alphabet and shifts and whisks them around, with frequent oral prompts (“keep going”, “you can do it”). Eventually, she holds up a board with all 26 letters and her body movements and auditory cues may be subtle enough that outside observers don’t notice them.

But there are three crucial differences between ABA’s cues and those of FC.

First, in ABA, cues are chosen and faded deliberately accordingly to a systematic, predetermined protocol. In FC, cues are subconscious and are faded subconsciously as facilitator and faciltatee each subconsciously adjust to them. That is, while the ABA therapist knows what he’s doing when he directs his client’s hand to the cookie or positions the cookie relative to the doll (he is, after all, following a systematic protocol), the FC facilitator is not aware that he is tensing his hand as his client’s hand approaches the correct letter and then relaxing it suddenly when the letter is reached. And the RPM facilitator is not aware that she is moving the letterboard closer to her client’s outstretched finger as it approaches the target letter and/or moving her own free hand towards the target letter before her client’s hand moves there.

Second, in ABA, the cues are eventually faded away completely—such that eventually anyone anywhere under any circumstances can ask a child to touch a cookie and she will be able to do so. In FC, in contrast, even when cues fade to the point of near invisibility (perhaps the keyboard is stationary and the facilitator isn’t actually touching his client, but merely sitting next to her), the client remains dependent on the facilitator being within visual or auditory range. If all the client’s facilitators leave the room, she is no longer able to type out intelligible responses to novel questions.

Third, one and only one of these frameworks acknowledges the possibility of unintentional cuing and spells out strategies for mitigating it.

Guess which.

In their ABA-based Teaching Language to Children with Autism and Other Developmental Disabilities, Sundberg and Partington write:

Once the child has learned to touch one of two items held in front of him, training should be given on learning to touch one of two items placed on a table. The procedure is similar, but this type of trial is often a little more difficult for a child because it eliminates most inadvertent prompts (e.g., slight position, movement, or eye prompts). (Bold-face mine).

And in the ABA journal Behavioral Analysis in Practice, Grow et al caution that therapists should recognize their susceptibility to giving off inadvertent cues and be careful not to make revealing hand movement or shift their gaze to the correct object before the client identifies it:

Instructors should look directly at the learner's face during the presentation of the trial and avoid any gaze shifts. Instructors should be explicitly trained to criterion to have consistent and accurate eye gaze; however, since people are unlikely to detect subtle patterns evolving in their own behavior, instructors should monitor procedural integrity for common gaze patterns that could inadvertently establish faulty stimulus control [i.e., cue the client]. Instructors should monitor implementation to detect inadvertently looking toward the location for a desired response for receptive instructional programs (e.g., glance at the bookshelf when saying “Get the book,” glance at the wall when saying “Turn on the light,” glance down while saying “Touch your feet”) or the target stimulus in a visual array for conditional discriminations. When presenting arrays for selection responses, the instructor should make eye contact throughout presentation of the stimuli and monitor for hand movements. When the learner's hand moves, instructors should orient their gaze to the learner's hand rather than to the visual array. [Boldface mine]

More generally:

“the task should be presented in a way that eliminates or substantially reduces the risk of the instructor providing inadvertent cues”

Grow, L., LeBlanc, L. Teaching Receptive Language Skills. Behav Analysis Practice 6, 56–75 (2013).

We can only wonder why FC promoters, guidelines, and practitioners don’t show similar concerns about the possibility of subconscious cuing in FC.