As part of 2016’s Autism Acceptance Month, Apple released an uplifting video called Dillan’s Voice, in which a nonverbal teenager delivers a speech at his graduation, his text turning swiftly to spoken word through his iPad. Before he had the iPad, he says in the video’s voiceover, people thought he didn’t have a mind, that he wasn’t in control. We watch him going for a run as the sun rises, doing pull-ups at the gym, walking alone along a school corridor, his hands flapping, humming occasionally, but it’s a life lived mainly through silence. Then, we move to his middle school graduation ceremony as he steps confidently up to the podium, award medals around his neck. “We are the reality of our thinking lives,” he tells the audience, urging them to open their minds. We hear enthusiastic applause, resounding cheers, the video fades to the Apple logo.
“All my life I wanted so badly to connect with people, but they could not understand because I could not communicate. But now you can hear me,” a voiceover reads. “The iPad helps me see not only my words, but to hold onto my thoughts... No more isolation. I can finally speak with the people that love me.”
Dillan’s Voice features something called “assistive technology”: devices and systems that maintain, increase, or improve the functional capabilities of those with differences and disabilities — including voice recognition software, screen readers, adaptive keyboards, and eye tracking devices. Recent years have seen groundbreaking advances in the assistive technology sphere. Individuals are using iPads and third-party apps and devices to make themselves heard, from Sady Paulson directing and editing features using Switch Control on her Mac, to Charlie, aged six, saying “mommy” for the first time using Proloquo2Go on his iPad.
“Bad science and bad ethics.”
But there was something different about this Apple video: in it, the teen uses Rapid Prompting Method, or RPM, a form of communication that depends on a “facilitator” or “communication partner” standing close by, providing continuous physical and verbal cues. RPM has been labeled pseudoscientific, unethical, and inhumane. Michelle Dawson, an autistic researcher, called it “bad science and bad ethics.” The American Speech-Language-Hearing Association (ASHA) says that “RPM is a technique without any research support.”
Over the next year, Dillan’s Voice generated significant buzz. A single tweet about the video, which does not explicitly discuss RPM, generated over 230 million impressions. It has reached 4.4 million YouTube views. It was featured on the Today Show, highlighted by Mashable, BuzzFeed, The Huffington Post, Forbes, Metro UK, and shared across disability and autism blogs. The there was a follow-up video that also featured RPM, a nomination for a 2017 Webby Award and a series of other Accessibility videos.
This video had reach.
“It is regrettable that this pseudo-scientific method was featured in an Apple promotional video and received worldwide viewing,” says Howard Shane, an associate professor at Harvard Medical School and director of the Autism Language Program and the Center for Communication Enhancement at Boston Children’s Hospital. He adds that it is not supported by research and that its underlying theory is “nonsensical.”
“It is regrettable that this pseudo-scientific method was featured in an Apple promotional video and received worldwide viewing.”
Concern over the RPM system hinges on how it works, and questions remain surrounding whether the method facilitates real first-person communication, or creates dependence on outside cues and interference from a user’s communication partner.
RPM was developed in the early 1990s by Soma Mukhopadhyay, the mother of an autistic child, who, like about one-third of individuals diagnosed with autism, was nonverbal. RPM is based on what’s called a Teach-Ask paradigm, explains the FAQ page of Helping Autism through Learning and Outreach (HALO), the organization that Mukhopadhyay joined in 2005. It works like this: the communication partner, or interlocutor, begins with a few sentences on a subject and then asks a related question:
“A square has 4 sides. What did I say? Does a square have 4 sides, or 3 sides?”
The interlocutor then writes the two possible answers on separate pieces of paper, tapping the choices while reading and spelling them aloud and then “encourages the student to pick up the correct answer.” If they are “auditory learners,” they might not look or read the answers, but instead rely on the tapping to “hear” the position of the correct answer. The guide can also use gestures for “visual learners.” Tearing up paper works, Mukhopadhyay said, as both an auditory and visual prompt. “Soma moves quickly from having the student choose from two choices to three or more, from picking up pieces of paper to having the student point to the answer, and then to pointing to letters to spell the answer,” reads the explainer on HALO’s website.
The process can sound innocuous, if a little chaotic, but the Encyclopedia of Autism Spectrum Disorders says that RPM prompts have, at times, taken on more punitive forms. Sometimes, the encyclopedia says, these include “verbal reprimands, trial termination, physical redirection, slapping or shaking the letter board against the subject’s face or chest, and blocking escape by positioning the subject between the table and walls.”
RPM users develop a startling dependency on facilitators’ prompts
Professor James Todd at Eastern Michigan University has been analyzing pseudoscientific interventions for autism for over a quarter of a century. He recently authored a paper with colleagues from Texas State University, Vanderbilt University, and the University of Kansas, on the shortfalls of RPM as a method of facilitating communication. RPM users develop a startling dependency on facilitators’ prompts, they discovered — so much so that facilitators “generate and subsequently prompt participants to convey messages that may not represent the individual’s genuine desires, thoughts, or emotions,” the study found. RPM participants, Todd and his colleagues conclude, “therefore only seem to communicate.”
Psychologist Dr. Sue Fletcher-Watson at the University of Edinburgh shares these concerns: “If there is a chance that the words being attributed to autistic people actually issue from their non-autistic supporters, this is undeniably an abuse of power.”
“Most facilitators are sincere in their unconscious authorship,” Todd says — though he warns that, “among them are conscious frauds—crooks.” The idea of unconsciously produced behavior, he explains, is easily reproducible. “What should be controversial, indeed unacceptable, is that RPM advocates seem oblivious to these facts, and put their clients in the position of being unwitting, if sincere and loving, forgers.”
Aspects of the system have even failed its founder: in an early experiment, Mukhopadhyay and her son failed a simple message-passing test. The child was unable to answer questions about information his mother did not know.
Furthermore, writing in the journal Developmental Neurorehabilitation, authors from Texas State University-San Marcos and Baylor University concluded that RPM lacks empirical evidence for any claims of educational efficacy or independent communication.
A more supportive study of RPM recently failed to disclose that a key author was on HALO’s scientific advisory board. Meanwhile, Mukhopadhyay was the therapist for all participants and selected the video footage. Even then, the authors state: “We defer, for the moment, the crucial question of whether the communications produced during RPM therapy are genuine.”
This lack of empirical evidence is particularly damaging in the development of autism communication. Dr. Oliver Wendt at Purdue University explains that autism is “a ripe ground for substantive claims since development trajectory very individual – a child could make spontaneous large gains for no known reason.”
RPM lacks empirical evidence for any claims of educational efficacy or independent communication
When reached for comment, an Apple spokesperson declined to comment on why the company decided to depict RPM, but said that Apple built “a wide range of assistive features directly into its products and do not direct users to a specific [communication method].” Indeed, Dillan’s Voice is just one of the many videos on Apple’s Accessibility site; none of the others feature individuals using RPM. Also, users are free to choose whatever third-party communication app they like.
Nevertheless, Dillan’s Voice achieved a profile other videos didn’t. Tim Cook pointed to it as one of his all-time highlights during a CNBC interview in 2016: “I’m most proud of the products that Apple makes that give our customers the ability to do things that they couldn’t have done,” said Cook, referring to “the autistic kid that’s given a voice.”
Featuring RPM detracts from the stunning accessibility technology being developed by Apple and others and their democratization of communication access. Individuals who work regularly with nonverbal, autistic individuals speak highly of the impact of visual computing platforms. “Before the iPad, Leo’s autism made him dependent on others for entertainment, play, learning, and communication,” says Shannon Des Roches Rosa, senior editor of The Thinking Person’s Guide to Autism about her son. “With the iPad, Leo electrifies the air around him with independence and daily new skills......I don’t usually dabble in miracle-speak, but I may erect a tiny altar to Steve Jobs in the corner of our living room.”
Different users have different needs, preferences, and abilities (like the rest of us), and any assistive communication device, low-tech or high-tech, is only as good as the support around its users. Those include families, of course, but also speech-language pathologists, occupational therapists, and other educators. It isn’t a miracle solution in and of itself.
Des Roches Rosa is broadly concerned at the lack of proper communication support and research into ensuring solid pathways for autistic people. “RPM is merely one tool in an understocked toolbox,” she says, “and it gets outsize attention because there are so few other options. It is unconscionable that so little is being done to ensure that minimally speaking autistic people like my son are able to accurately communicate their wants, needs, hopes, and opinions.”
Apple has not featured RPM in subsequent Accessibility videos, and it’s unclear whether the company will be featuring the method during this year’s Autism Acceptance Month in April. Regardless, as Apple continues to operate and expand in the health and wellness sector, it would do well to avoid featuring pseudoscience, even in passing.