In just six years, facilitated communication, a technique imported from Australia, has had an unprecedented impact on the treatment of autism. Its premise is simple: Autistics who cannot speak can communicate by typing their thoughts onto a computer keyboard or other device with the help of a facilitator who supports their arms, hands or wrists. The technique, which requires little or no training, has had seemingly astounding results. Reports have circulated of autistic teenagers who did not speak and whose IQ scores had been measured at 20 or 30 -- in the severely mentally retarded range -- who were retested using the technique, known as FC, and scored above 140 -- in the genius range. Some are attending college-level classes with their facilitators. Others have written perfectly punctuated letters to President Clinton, prize-winning poems, essays and lengthy autobiographies -- all products, FC supporters maintain, of "unexpected literacy." School systems across the country, including some in New York, Virginia and California, quickly embraced the technique, as did many parents eager to communicate with their autistic children, often for the first time. The media, led by TV newsmagazine shows including "PrimeTime Live," hailed FC as a "breakthrough." In 1993, the Virginia General Assembly passed an unusual resolution in support of the technique.

Is facilitated communication a long-awaited miracle -- the proverbial key that finally unlocks the door of autism? Or is it, as one influential critic has called it, "the cold fusion of special education" -- the latest in a series of useless fads that have no scientific validity and potentially pernicious results? Or is the truth less spectacular: that FC may work for a minority of people with certain specific motor problems, such as cerebral palsy, but that its general applicability with autistics remains unproved? Whatever one's view, the technique has ignited an extraordinarily bitter debate among mental health workers, advocates for the disabled, educators and the parents of autistic children. "People who believe in it believe in it religiously," said critic Glenn Elliott, chief of child and adolescent psychiatry at the University of California at San Francisco. "And those who don't are furious."

Lack of Evidence

Among the nonbelievers is psychologist Eric Schopler, one of the nation's most respected autism experts and the pioneer of a widely used behavioral method of treating autism. "I think this is the most reckless, irresponsible use of a technique that I've seen in 30 years in this field," said Schopler, a professor of psychiatry at the University of North Carolina. FC, he says, has been propelled by media hype; by naive, idealistic and poorly trained practitioners; and, most understandably, by anguished parents "who've had the experience, or heard of the experience, of having their kid say for the first time, 'I love you, and I've loved you for years.' You don't have to have much imagination to see that that would matter."

Schopler is among a growing chorus of critics who point to the preponderance of scientific evidence that says FC does not work. Since 1991, more than 40 studies published in peer-reviewed journals involving more than 400 people with autism have failed to document FC in all but a handful of cases, according to Bernard Rimland, a research psychologist who founded the Autism Research Institute of San Diego and the Autism Society of America. In most of these studies, the key variable was whether the facilitator knew or could guess the answer. In less than 10 percent of cases, the autistic people in the experiments occasionally spelled a word or named a familiar object correctly when the facilitator didn't know the answer. But in nearly all cases in which the facilitator did not know what the disabled person was being shown, all they produced was gibberish. The widespread conclusion is that any successes reported with FC were due to the facilitator. While facilitator influence was virtually universal, it was, many researchers have concluded, probably unconscious. Facilitators genuinely thought that their clients, not they, were controlling the communication. As a result of these studies, five professional groups, among them the American Psychological Association, the American Academy of Child and Adolescent Psychiatry and the American Speech-Language-Hearing Association, have passed resolutions cautioning against use of the technique and noting that there is no scientific proof that communication via a facilitator is valid. The technique also has been the subject of a year-long investigation by the Federal Trade Commission. Last month two companies that sell FC devices, Crestwood Co. of Glendale, Wisc., which makes the popular Canon Communicator, a hand-held keyboard with a paper-tape output, and the Abovo Co. of Chicopee, Mass., agreed to settle charges that they made "false and unsubstantiated claims" about the devices. In advertisements the companies had claimed that the machines enable autistics and other disabled people to communicate using FC. The consent orders, which are subject to final approval by the FTC, would bar the companies "from misrepresenting that any communication aid enables autistic individuals to communicate through facilitated communication" without "competent and reliable scientific evidence that substantiates the representation."

Passionate Defenders

Proponents of facilitated communication, led by Syracuse University special education professor Douglas Biklen, who introduced the technique in the United States, vehemently disagree with critics. They argue that FC does work, but that it is too fragile to be measured by the traditional scientific methods used in the studies that detractors cite. Those methods, Biklen and others say, frighten disabled people and undermine the process, sometimes by displaying skepticism. New, and as yet unpublished research, Biklen insists, will confirm the existence of FC.

Biklen, a longtime advocate of "full inclusion," the controversial educational philosophy that says that people with disabilities should be mainstreamed into normal classrooms regardless of the severity of their problems, says he is not surprised by the passionate opposition FC has aroused. Medical orthodoxy, he notes, has been proved wrong before.

"Look at Christy Brown in 'My Left Foot,' " he said, citing the case of the Irish writer with cerebral palsy who was wrongly assumed to be mentally retarded. (Until the 1960s all people with cerebral palsy, a brain injury caused by oxygen deprivation before or at birth, were believed to be retarded; in fact, some have normal or superior intelligence.) "Helen Keller was thought to be severely retarded, and then you have a way of communicating and you see what a person is capable of," he added. "The definitions of who's retarded change every 10 years or so." Biklen's view of autism differs radically from that of autism experts. He believes that autism is fundamentally a motor problem, and that people with the disorder suffer from apraxia, an inability to control their bodies, much as people with cerebral palsy do. That, he says, is why autistics engage in bizarre, repetitive actions such as hand-flapping. Facilitation, he says, provides the stability and physical support that enables them to communicate and to demonstrate their true mental abilities: normal, and in many cases superior, intellectual functioning.

Among those who agree with Biklen is Arthur Schawlow, a Nobel Prize-winning physicist at Stanford University whose severely autistic 38-year-old son has been using FC for years. Schawlow dismisses the often cited scientific studies as "phony experiments which were badly planned. There's no way to persuade me that this doesn't work," he said. "I've seen it work." Rimland and other critics of FC dispute Biklen's theory that autism is fundamentally a motor problem. They note that autistic children typically possess excellent fine motor skills. "How is it possible that an autistic kid can pick up the last tiny crumbs of potato chips off a plate but not have sufficient motor coordination to type the letter E?" Rimland asked.

Rimland says that for some people with cerebral palsy and for a small number of nonverbal autistics, FC may have some benefit. "I would guess that maybe 5 percent of nonspeaking autistic individuals who have the capacity for language can't communicate," he said.

Extraordinary Results

In 1990, Biklen published a seminal article about FC in the Harvard Educational Review, a journal published by students at Harvard's Graduate School of Education. He had recently returned from a visit to Australia, where he had witnessed the technique, which was invented for use with cerebral palsy patients. Biklen wrote that when he tried FC with 43 autistic people, some of whom were severely retarded, 41 were able to use the technique immediately. While facilitators held their hands, arms, wrists or elbows, they communicated their thoughts, made witty jokes and held extended conversations using computers and Canon Communicators.

Among those whose communication Biklen facilitated was Jonothan, a nonverbal autistic teenager who was not toilet-trained and who, according to Biklen, "had a history of fits of screaming, regurgitating food, scratching and running away from people."

During one exchange, Jonothan said, "I'm not very quiet right now because I got sat on by Rosie." When asked whether he meant that "literally or metaphorically," Biklen wrote that Jonothan typed "met."

He also cited the case of Polly, 15, who was fond of stuffing wads of dried gum or rocks in her mouth. In her communications with Biklen, Polly had a long discussion about full inclusion and used the words "gauntlet" and "illogical."

Biklen's claims of unexpected -- and precocious -- literacy extend to very young autistic children. In his 1993 book about FC, he describes the case of Bobby, who at the age of 3 3/4, answered questions about "The Three Little Pigs" by typing out the words "house," "sticks," "brick" and "wolf." At that age, it is rare for intellectually gifted children to be able to type a word, let alone spell it correctly.

News of Biklen's article and the remarkable new technique electrified advocates and parents who rushed to buy Canon Communicators and other devices.

School systems across the country, among them Fairfax County's, sent their staffs to Syracuse to be trained at the university's Facilitated Communication Institute, which Biklen directs. Staff members at institutions for the autistic began spending hours every day "facilitating" with their disabled residents.

From the outset, some autism experts were skeptical, both of Biklen's lack of experience in the field of autism and of his outspoken support of the philosophy of full inclusion. Most of all, they found Biklen's claims that large numbers of profoundly retarded people were capable of reading and writing, often at an advanced level, as preposterous as his assertion that they had learned to read through osmosis -- by watching "Sesame Street," reading cereal boxes and road signs. "We thought there might be something to {FC}," recalled psychologist Gina Green, director of research for the New England Center for Autism in Southboro, Mass., who visited Australia to investigate FC. "We didn't think the outlandish claims that it was successful for everybody were credible. ... There's no treatment that has that kind of benefit." Green said she initially suspected, and continues to believe, that FC represents the triumph of anecdote over evidence. To prove that FC works, she says, it is necessary to know what participants were capable of before they began facilitating, such as whether they could type independently. She added that reports by FC advocates that autistic people have communicated information allegedly unknown to facilitators are scientifically unreliable. "In 50 years of research among the developmentally disabled, we know they have a really strong tendency to become dependent on cues, little subtle things like when the teacher looks at the correct picture of an object without even being aware of it," Green said. "Animals pick up these cues."

Many of the staff at the O.D. Heck Developmental Center, an institution for the profoundly autistic in Schenectady, N.Y., were sold on FC. In 1992, they set out to prove to skeptics that FC worked. The staff conducted a simple experiment, which remains the most influential of nearly four dozen studies of FC published so far.

Researchers picked 12 autistic residents deemed to be the most proficient at FC. All had been typing full sentences and engaging in long interactive exchanges for months. There were three tests: Autistic residents were first asked to identify a picture the facilitator could not see, first with facilitation and then independently. In the third test, both participants and facilitators were shown pictures, sometimes of the same and sometimes of different objects. Neither could see the pictures shown to the other.

The results were unequivocal -- and devastating to those who believed in FC. In 180 trials, the only correct answers, through facilitation or independent typing, were given when both facilitator and resident saw the same picture. When they were shown different pictures, the only correct answers were for the pictures shown to the facilitators. The O.D. Heck study, published in 1993, was quickly followed by a spate of studies that reached similar conclusions: Facilitators, not the autistic participants, were doing the communicating and they were not conscious of their influence. These studies have led to a reevaluation of FC by some of its early supporters and by some school systems, which have circumscribed its use or stopped using FC altogether. Nevertheless the research has not dissuaded proponents, including Biklen, who maintain that these studies do not disprove FC. Biklen says that the O.D. Heck study is not a true test of facilitated communication. Traditional scientific experiments, he says, suppress the phenomenon by making autistic people too nervous and rendering them unable, or unwilling, to perform. A more accurate test, he says, is an unpublished study by Donald Cardinal, a special education professor at Chapman University in Whittier, Calif. Cardinal tested 43 disabled students, some of whom were autistic, and all of whom were proficient at FC. Each student, who practiced the experimental protocol for weeks, was shown a series of words selected from a list of 100 while their facilitator was out of the room. The facilitator then entered the room and helped the student type the word he or she had just seen. Cardinal says that slightly more than half of the students could correctly type two out of five words. "There are many people who are now typing independently that once needed FC," Cardinal said. "For some people, FC does seem to work; the question is, when and who and how?"

Green said that until Cardinal's study is published and his results are reviewed independently, it cannot be evaluated.

"Some autistic kids are very good at repeating rote tasks; in fact it's what they like best," she said. "I've said all along I'll believe solid, objective, scientific evidence that FC works. So far, I haven't seen it."

CAPTION: Brian Meredith, left, and Graham Tomlinson are guided in facilitated communication at a Chantilly school by teachers Vicki Kernstine and Sue Jones and autism specialist Rosemarie McGuinness.

CAPTION: Fairfax autism specialist Rosemarie C. McGuinness helps Graham Tomlinson work on a computer.

CAPTION: Another approach to treating autism, the behavioral program called the Lovaas Method, which has shown promising results, will be examined next week in Health. For further information about autism, write to the Autism Society of America, 7910 Woodmont Ave., Suite 650, Bethesda, Md., 20814. Or call the society at 800-328-8476 (800-3-AUTISM).