In which Vikram Jaswal fails to establish the authenticity of S2C for non-speakers
In a recent episode of Barry Prizant’s Uniquely Human podcast entitled “Establishing the Authenticity of Supported Communication for Non-Speakers,” Vikram Jaswal, a psychology professor at the University of Virginia, attempts to make a case for the variant of facilitated communication (FC) known as S2C.
This is Jaswal’s first appearance on Prizant’s podcast, and the episode centers on an eye-tracking study that he co-authored four years ago (Jaswal, 2020) that purportedly shows evidence for S2C. The most interesting parts of the podcast, however, are the parts where Jaswal discusses IQ levels in non-speaking autism, facilitator influence, communicative independence, and what Jaswal (and Prizant) have to say about his critics, AKA “the naysayers.” The podcast also features a co-host, David Finch (“Dave”), who was diagnosed with Asperger’s syndrome by his wife.
Jaswal begins by recapping his experience with autism, characterizing himself as “new to autism” even as recently as 2015, when he was invited to talk about autism research at a summer institute—despite the fact that “I didn’t… know much about autism at the time.” The person who extended this rather surprising invitation was Elizabeth Vosseller, the founder of the summer institute and the “inventor” of S2C (or, more precisely, the person who, after being trained in another variant of FC known as the Rapid Prompting Method or RPM, rebranded that method as Spelling to Communicate or S2C). The summer institute was for “nine or ten non-speaking autistic young adults all of whom at the time were communicating by pointing to letters on a letterboard held vertically by a trained communication and regulation partner”—the hallmark of S2C. For Jaswal, witnessing this variant of FC was “a transformative experience,” and “To say that I was blown away is an understatement”. He adds:
It’s changed my life, it’s changed my research program, it’s changed how I think about the world, and about psychology and communication.
What Jaswal leaves out here (and pretty much everywhere else, including in the disclosure statement in his eye-tracking study) is that he has a non-speaking autistic daughter who uses S2C. According to a piece in the Washington Post written by Jaswal and his wife, his daughter was in second grade in 2016. This means that Jaswal was, in 2015, neither “new to autism” nor an unbiased observer
Indeed, by the time he attended Vosseller’s summer institute, as he makes clear on the podcast, Jaswal was already convinced that much of the longstanding research on autism was wrong. In particular, he was (and is) convinced that there is no Theory of Mind impairment in autism and that there is no cognitive impairment in non-speaking autism. Further strengthening these convictions were the messages he saw being extracted from the non-speaking attendees of Vosseller’s summer institute via S2C, and, perhaps, some additional interactions with other autistic people over the years:
Over and over again as I’ve gotten to know autistic people, of course they understand other people’s beliefs goals desires and they recognize that behaviors are motivated by underlying intentions.
To the extent that Jaswal’s (informal and subjective) impressions are based on messages extracted from non-speaking individuals via S2C, all the available evidence indicates that the people demonstrating these Theory of Mind skills are actually the non-autistic facilitators, or what S2C proponents call the “communication and regulations partners” (CRPs). Without a message-passing test that would establish who is authoring these messages, we can’t know.
As for the Theory of Mind research, Jaswal states that the experiments “were designed in a particular way that was not suitable to allow autistic folks to demonstrate their underlying” skills: claims that he has elaborated in a 2018 article which has been critiqued here and in an earlier post on this blog.
As far as the research showing low IQ in non-speaking or minimally-speaking autism goes (see Anderson et al., 2007; Wittke et al., 2017; Wodka et al, 2013), Jaswal has not published any critiques. Instead, he simply asks “How can somebody who doesn’t have a way to express themselves using language-based communication participate in a lot of standard IQ tests?” He has apparently forgotten that, in the standard IQ tests used with children, only the verbal IQ subsections of those tests require verbal communication. Indeed, that’s why studies of IQ in non-speaking children tend to focus only on nonverbal IQ, aka Performance IQ. The studies listed above are no exception, and they find that nonverbal IQ is significantly impaired in non-speaking autism.
Jaswal, however, cites a Canadian study that, he claims, challenges these results: a study by a “really progressive group of Canadian autism scientists” that used a “strengths-based approach” to investigate IQ in “a group of elementary school-aged non-speaking kids.” Jaswal notes that their assessments required neither verbal responses, nor motor responses “that are different from those that they would otherwise use in their daily lives.” The study Jaswal is most likely referring to—the only one that fits this description—is Courchesne et al, 2015. The study he’s referring to, Jaswal claims, finds that most of these kids “actually scored within the normal range on the strengths-based IQ tests.”
The IQ test that Courchesne et al. used was a version of the Raven’s Progressive Matrices test: a test that places minimal verbal demands on those being tested, even for comprehending the test directions (which, importantly, can be an issue even with test questions that don’t require verbal responses). For that very reason, the Raven’s Progressive Matrices test is, in fact, a common way to assess IQ in autistic individuals with language impairments.
Seemingly consistent with Jaswal’s statement that the study’s IQ test didn’t require motor responses other than those that the children could perform in their daily lives, the version of the test that the researchers used, the Raven’s Colored Progressive Matrices (RCPM), doesn’t require pointing: something that many of the minimally-speaking autistic participants were unable to do. But, while there is evidence for a lack of (or reduced) pointing in autism, there isn’t evidence, as I’ve discussed in other posts, that lack of pointing in autism is due to motor issues. Proponents of S2C, of course, insist otherwise: for them, “unreliable pointing” results from fine motor difficulties that, in turn, justify the need for held-up letterboards. Pointing, however, isn’t just a motor activity; it’s also a communicative gesture, and reduced communicative gesturing is a hallmark of autism (Manwaring et al., 2018). Furthermore, in lieu of pointing, the RCPM test requires a different fine motor skill: placing one of “six movable pieces” into the empty space that represents the missing part of a multi-dimensional pattern.
What about the study’s results? Contrary to Jaswal’s claim that most of the non-speakers scored within the normal range, the researchers found that:
Of the 26 (out of 30) minimally-speaking autistic participants who completed the test, there was an average raw score of 18.61 out of 36.
Their non-autistic counterparts had an average raw score of 28.5 out of 36.
While eight of the autistic participants performed at or above the 50th percentile, with three at the 90th percentile, “autistic children's RCPM performance differed according to their reported spoken language level” with “autistic children using two-word phrases performed better… than those using no words at all.”
So much for Jaswal’s claim that a Canadian study using a strength-based assessment on non-speaking autistic elementary school students shows that most non-speakers scored within the normal range.
Moving on from Theory of Mind and IQ, Jaswal and Prizant rehash some of the other pro-S2C, evidence-free talking points:
That individuals with autism have trouble controlling their bodies (see the sensory-motor section of our misleading articles page).
That the big challenge in autism isn’t helping individuals improve actual communication skills, but giving them access to effective ways to communicate (as if all there’s no evidence of widespread communication challenges in autism, and as if there’s any evidence that responsible professionals are systematically depriving autistic people of effective ways to communicate).
That it’s the people outside the world of S2C, not S2C proponents, who suppress and ignore the communicative gestures made by non-speaking individuals (see Beals, 2023).
The discussion then turns to the facilitator, or “communication and regulations partner” (CRP). Citing the messages facilitated out of S2C-ed individuals by their CRPs as testimonials, Jaswal claims that the purpose of the CRP is for “attentional, regulatory, emotional support.” He suggests that the S2Ced individuals’ emotional needs may stem, not just from co-occurring conditions like anxiety, but also from the “trauma” of having been long denied an effective means of communication. As for the possibility that CRPs might be the ones depriving autistic individuals of effective communication by controlling the messages, Jaswal minimizes this in several ways.
First, he claims that influence is “just part of the way communication works”. He cites Prizant’s affirmative head nods and states that these “giv[e] me the confidence to continue”—as if nodding while someone speaks is comparable to the influence that facilitators exert on message content. While acknowledging the concerns of FC critics about CRP influence on messages, he resolves this concern by (1) saying that this is something the non-speakers are “absolutely aware of,” and (2) changing the topic to how hurtful these concerns must be to non-speakers:
Imagine now that you’ve worked really hard on the skills to be able to communicate by pointing to letters with the assistance of a CRP and, and now you learn that there are people who don’t believe what you’re saying, that there are people who want to deny you access that method and related methods of communication. That’s only going to contribute further to somebody’s trauma.
At this point the conversation turns to Jaswal’s 2020 eye-tracking study (critiques of which are here and here), and Jaswal proceeds to repeat his mischaracterization of the ways FC critics say that facilitators can cue (and control) letter selection. That is, Jaswal depicts this cueing not as the automatic, non-conscious phenomenon that studies of FC and related phenomena have unearthed (see, e.g., Spitz, 1997) but as something that involves conscious interpretation and deliberate decision-making and would take more time than what he characterizes as the “fast clip” (he also calls it “very fast” typing) at which many of the participants selected letters. (Their actual speed was, on average, and as Jaswal confirms here, one letter per second, which most people outside of the Spellerverse would consider quite slow, even for single-finger typing[1]). With Prizant and Dave laughing along and perhaps also head nodding in agreement, Jaswal states:
There is some reason to think that there are limits on the human information processing ability to make use of cues from the environment in order to drive our behavior, so one example is like, you know, I don’t know about you, but like if you’re at a restaurant with your family and, uh, you know, one member of your family wants you to notice person’s shoes at the table next to you they might make a subtle cue—nod their head in a particular direction or roll their eyes over there and you have to notice the cue, interpret its meaning and then like decide how am I going to act based on that cue. So these things—this detection of the cue, this… noticing that it occurred, this… figuring out what it means, and then this action program after having understood it—like, those things take time. So each of those steps takes time. And so as I have spent time with non-speaking autistic people who spell, many of whom spell so quickly and so accurately that I think it’s just… beyond belief that they’re simply responding to cues that many of the rest of us wouldn’t be able to notice sitting next to them that the CRP or communication and regulations partner is offering.
It doesn’t seem to occur to Jaswal that some of the cues that critics are talking about require no interpretation or deliberation. These include:
The CRP shifting the board so that the person’s outstretched index finger approaches a particular letter.
The CRP moving their free hand in a movement that mirrors the movement that would be involved in a particular letter selection, doing so milliseconds before the S2Ced person makes that movement and that letter selection (see here for examples and discussion).
These cues are less like the deliberate, consciously interpreted restaurant cues Jaswal imagines, and more like the tactile pressure cues that immediately tell horses exactly where to turn and how fast to go, or the vestibular and proprioceptive cues that immediately tell cyclists exactly how to maintain their balance. These phenomena occur in well under a second, are largely non-conscious, and yet completely control a horse’s or a person’s movements.
Nor does it seem to occur to Jaswal that these sorts of cues are qualitatively different from the facial and social cues that many studies show autistic people to be under-sensitive to. On one hand, Jaswal dismisses these findings as purportedly as problematic as the Theory of Mind findings. On the other hand, he brings them up in connection with CRP cueing, claiming that, to the extent that FC critics believe these findings (we do), they would contradict our claims about CRP cueing (they don’t).
Jaswal also repeats another of his study’s key misconceptions: that since intentional acts involve looking, acts that involve looking are intentional. This is what allows him to conclude that, since the participants were, on average, looking ahead at the next letter a half-second before they typed it, they must have been intentionally pointing to that letter, and that it therefore couldn’t be the case that the CRP was moving the board. But looking ahead neither logically entails intentionality nor logically rules out board movements, nor does any of this rule out the possibility that the participants’ eyes were responding to the same cues that their fingers were—and/or were simply looking at whichever letter the CRP (however unintentionally) moved into their line of sight.
Prizant, Barry, and Dave then move on to the “naysayers.” Jaswal claims that they/we haven’t come up with any alternative explanations for his data, apparently unaware of the published critiques that do just that (Beals, 2021; Vyse, 2021). Dave, who calls himself an engineer but whom Prizant calls a “scientist,” proceeds to emphasize the importance of data. Prizant, confusing skepticism with non-evidence-based beliefs and lived experience with evidence [2], claims that it’s the naysayers who are the ones who are making assumptions and Jaswal who is being conservative with respect to the alleged evidence.
The conversation then turns to the CRP—with Jaswal citing research on how, in stressful situations, the physical presence of people we’re close to helps to keep us regulated. Connecting this to non-speakers, he notes how anxiety-provoking it must be when a school suddenly swaps out a child’s aide with a new aide. Jaswal doesn’t extend this to CRPs, but Dave does, and presumably Jaswal would agree that it’s also potentially anxiety-provoking when parent suddenly swaps out one CRP for another. Dave, for his part, compares the CRP to the “buddy” that company executives apparently always bring with them to meetings:
I can’t think of a single executive who rolls into a very high-level meeting without his little buddy there, whether it’s the attorney or the regional salesperson who’s there as a safety person to make sure that person’s message gets across. And we don’t challenge that.
Finally, the conversation turns to the possibility of AI taking over some of the CRP’s roles, and Jaswal expresses certainty that, while some of the CRP’s roles might be automated and offloaded to AI, he thinks that there will always be a CRP present. And, of course, as long as a CRP is present, the likelihood persists that the S2C-ed messages are not the authentic communications of the S2Ced individual—even if the actual author is now some combination of the CRP and the AI.
Nonetheless, Jaswal is confident, not only that there is no significant influence by CRPs on messages, but also that some of the S2Ced individuals now type independently. But he also makes it clear that those “independent typers” do not actually type independently of their CRPs:
Among those who type independently they still benefit from the presence of another person who can serve as that… regulation partner and who is able to redirect them when their ADHD leads them astray, for example, or when their OCD leads them to focus on something that person who can help them get back on track.
Apparently, even when S2Ced individuals recover from their alleged fine motor control issues, dyspraxias, and mind-body disconnects, there’s still a lingering (diagnosed or undiagnosed?) ADHD or OCD…or?—anything to justify the continued presence, and influence, of the CRP.
Prizant and Jaswal conclude the podcast with a celebration of interdependence:
Prizant: “Self-determination comes with support… You need that trusting person around you, that safety net, to take risks and get out there.”
Jaswal: “We are interdependent… We are all relying on each other. Why should it be any different for a non-speaking autistic person?”
Dave adds that we depend on each other to help us see our blind spots, whereupon Prizant concludes the podcast with an allusion to “the elephant in the room”: presuming competence.
Presuming competence—that ill-advised but persistent mantra of FC/RPM/S2C proponents used to deter people from asking who is directing the messages and to justify their refusal to participate in rigorous tests that could answer this question. Fittingly juxtaposed, here, with blind spots.
NOTES
1. Perhaps in an attempt to explain away the slowness of the typing, Prizant compares it to typing in the confirmation code or frequent flyer code into an airport scanner: “You’re scanning the screen and visually searching out the next number in the code.” To this, Jaswal replies: “That’s a really good analogy.” But of course, typing out your actual thoughts using the familiar letter sequences of conventional spellings is quite different from typing in the arbitrary sequences in a confirmation code.
2. Prizant cites, as evidence, “Talking to dozens of families and parents and people who are spellers and typers which has been totally dismissed as any aspect of evidence here by the naysayers.”
REFERENCES
Anderson, D. K., Lord, C., Risi, S., DiLavore, P. S., Shulman, C., Thurm, A., et al. (2007). Patterns of growth in verbal abilities among children with autism spectrum disorder. J. Consult. Clin. Psychol. 75, 594–604. doi: 10.1037/0022- doi:10.1080/17489539.2 021.1918890 006X.75.4.594
Beals, K. (2021). A recent eye-tracking study fails to reveal agency in assisted autistic communication. Evidence-Based Communication Assessment and Intervention, 15(1), 46–51.
Beals, K. P. (2023). Sit down, shut up. Index on Censorship, 52(2), 60-61. https://doi.org/10.1177/03064220231183821
Courchesne, V., Meilleur, A. A., Poulin-Lord, M. P., Dawson, M., & Soulières, I. (2015). Autistic children at risk of being underestimated: school-based pilot study of a strength-informed assessment. Molecular autism, 6, 12. https://doi.org/10.1186/s13229-015-0006-3
Jaswal, V. K., Wayne, A., & Golino, H. (2020). Eye-tracking reveals agency in assisted autistic communication. Scientific Reports, 10(1), 7882. doi:10.103841598-020-64553-9 PMID:32398782
Manwaring, S. S., Stevens, A. L., Mowdood, A., & Lackey, M. (2018). A scoping review of deictic gesture use in toddlers with or at-risk for autism spectrum disorder. Autism & Developmental Language Impairments, 3. https://doi.org/10.1177/2396941517751891
Spitz, H. (1997). Nonconscious Movements: From Mystical Messages To Facilitated Communication. Routledge.
Vyse, Stuart. (2020, May 20). Of Eye Movements and Autism: The Latest Chapter in A Continuing Controversy. Skeptical Inquirer.
Wittke, K., Mastergeorge, A. M., Ozonoff, S., Rogers, S. J., & Naigles, L. R. (2017). Grammatical language impairment in autism spectrum disorder: Exploring language phenotypes beyond Standardized Testing. Frontiers in Psychology, 8, 532. doi:10.3389/fpsyg.2017.00532 PMID:28458643
Wodka, E. L., Mathy, P., and Kalb, L. (2013). Predictors of phrase and fluent speech in children with autism and severe language delay. Pediatrics 131, 1–7. doi: 10.1542/peds.2012-2221
