Is facilitated communication evidence based?
The short version: partly, and the parts that are supported by evidence are the parts the public conversation tends to ignore.
You probably arrived here searching for facilitated communication, which is a term used under the broader umbrella of Assisted Communication (AC) — which also includes S2C, RPM, the Spellers Method, FC and Supported Typing. Each of these methods has distinct approaches, while they each share the foundational principles of presuming competence, authentic communication, and developing independence.
"Evidence based" sounds like a single yes-or-no question. It isn't. The phrase covers a spectrum from "strong randomized trials" to "consistent observational data" to "plausible mechanism with growing support." Most clinical fields live in the middle of that spectrum. Facilitated communication — and the broader assisted communication field it sits inside — is no different.
What the evidence supports
- The neuromotor framework. Whole-body apraxia in nonspeaking autism is well-characterized in functional imaging and motor-learning research. The mechanism the field invokes to explain why supported typing works is consistent with the underlying neuroscience.
- Eye-tracking. Jaswal et al. (2020) recorded gaze-before-point patterns in nonspeaking autistic typists during open-ended communication. Those patterns are not consistent with facilitator authorship.
- Independent typists. A growing population started with physical support and now types without it. Their continued output — books, college courses, public testimony, careers — is the cleanest authorship evidence available.
- Outcomes on Spelling to Communicate and Spellers Method. Long-form clinical observation and program-level outcome data continue to accumulate. See the evidence base summary for clinicians.
What the evidence doesn't yet settle
- Large randomized trials. They don't exist at scale, and the standard designs are structurally unfair to apraxic learners. Designing fair tests is itself an open research question.
- Long-term cross-method comparisons. We don't have head-to-head data on which approaches work best for which motor profiles.
- The exact role of partner familiarity, regulation, and warm-up. Clinically obvious. Empirically under-quantified.
The honest summary
The field has a growing, mixed evidence base. Some claims are well-supported. Some are still open. None of that is unusual for a clinical area serving a small, heterogeneous population. What's unusual is the confidence with which "no evidence" gets repeated about a literature its critics have largely stopped reading after 1995.
Frequently asked questions
- Is facilitated communication evidence based?
- Partly. 'Evidence based' is a spectrum, not a binary. Some claims about assisted communication are well-supported (eye-tracking, motor-planning neuroscience, independent-typist outcomes). Others are open empirical questions. The honest answer is that the field has a growing, mixed evidence base — not a settled negative verdict and not a finished positive one.
- Has facilitated communication ever worked?
- Yes. The clearest evidence is the population of typists who began with physical support and now type independently. They write books, complete college coursework, testify publicly, and hold jobs. That population exists because the underlying motor learning happened. If the method had never worked for anyone, those typists would not be there.
- What research supports it?
- Jaswal et al. (2020) eye-tracking work on nonspeaking autistic typists; functional imaging and motor-learning research on whole-body apraxia; long-form outcome studies on Spelling to Communicate and the Spellers Method; and the existence of a growing cohort of independent typists whose authorship is not in question. The literature is smaller than for mainstream AAC and larger than the 'no evidence' framing suggests.
- What does the research not yet support?
- Large-scale randomized trials don't exist in this field — partly because the population is small and heterogeneous, partly because the standard study designs (novel partner, time-pressured, single-trial) actively suppress performance in apraxic bodies for reasons unrelated to authorship. Designing tests that are fair to the population is itself an open research question.