"Influence" and "Control" In Facilitated Communication
by Alan Kurtz
WHOSE WORDS ARE THESE?
Skeptical readers might legitimately ask if the person whose name appears above actually wrote this article. After all, most of you did not see it being typed. Conceivably someone other than Alan Kurtz may have written these words and attached his name to it. A number of factors probably lead you to regard this possibility as unlikely including: 1) the style is consistent with things Kurtz has written previously, 2) Kurtz has demonstrated some competence in writing articles, and 3) as editor of the newsletter Kurtz would probably reject articles written by someone else with his name on it.
Those of you accepting the fact that Alan Kurtz wrote this article might go one step further. You might ask: "Are these are really Kurtz' words or has he been influenced by others?" You might believe that these words are merely echoes of things that others have said or written. Conspiracy theorists might think that others are telling Kurtz what to say.
OK, I admit it! Those of you doubting the originality of my writing are right. I have been influenced. Michael McSheehan and Rae Sonnemeier influenced me when they spoke at a conference two years ago. Some of the concepts that I use, such as "co- construction" and the "collaborative model of communication" are their words and not mine. Articles that I have read by many authors also influence what I say. I may even use particular words or expressions that I have heard others use, unaware of their roots. Research classes that I have taken, conversations that I have had, and debates in which I have participated all influence the words I choose to use. The fact that our secretary, Kelly, is a published author also influences my writing.
You are influencing me. Although I cannot see your reactions to my words I can imagine the various smiles, smirks, scowls, retorts, and body language that might be produced as you read this. My anticipation of your response influences each word that I type.
What then makes me the author? With all this influence coming from so many directions affecting what I write, how I write, and even the words I choose to use, how can I take any responsibility (or blame) for this article?
I am the author of this article (although it might be difficult to prove) because I am the one who chooses which words to include. Even if I shamelessly plagiarize someone else's work I am choosing which words and ideas to copy. When others make suggestions about my writing, I still choose to accept or reject those suggestions.
FC users are influenced by their communication partners. In addition to the influence communication partners typically exert upon one another, there are some forms of influence unique to facilitation. Facilitators provide a wide variety of subtle physical, emotional, and verbal supports. These include physical support of a person's arm, pulling back between responses, and providing additional structure when necessary. Without this influence communication can be difficult or impossible.
Although facilitators provide some unique forms of influence, FC users should make the final choice about what to say. When trying to determine authorship, the same standard applied to my writing above can be used for evaluating the communication of an FC user. If an FC user chooses which words to type and is the final arbiter about what is communicated, s/he is the author.
Unlike most of us, FC users may face questions about authorship each time they communicate. Typically, FC users do not have a history of demonstrating competence with language - making it more difficult for others to accept their communication at face value. This issue is further complicated by the fact that facilitators can consciously or unconsciously make choices for the user. Facilitators can physically guide a person to a target or provide other cues that can result in the facilitator being the actual source of the communication. Guiding or cueing is facilitator control and must be distinguished from appropriate influence. Control can also occur when facilitators fail to clarify or incorrectly interpret a response.
CONTROL CAN OCCUR IN A NUMBER OF DIFFERENT WAYS. THESE INCLUDE:
There has been a tendency to blame the facilitator when cueing occurs, although cueing may not always be initiated by the facilitator. Sometimes FC users actively seek cues from their facilitator (Crossley, 1993). An individual I used to facilitate sometimes moved his hand slowly in small tight circles, hovering over the board waiting for some indication from me about which direction he should move. When I said to him "You need to find the letter yourself," he initiated movement toward the keyboard on his own. The influence I provided in this case helped ensure that he authored the message.
Recommendations for avoiding control or cueing include:
Facilitators may control some but not all communication. In the original study on the validity of FC in Australia, the Intellectual Disabilities Review Panel (IDRP) made the following observation: " Influence occurred with a client who demonstrated valid uninfluenced responses to other items. It appeared that a given assistant could influence some client responses and leave others uninfluenced" (IDRP, 1989). A second individual, who could not demonstrate that he was the author of any of his communication in that test went on to type independently (Biklen, 1992).
FC speakers may say different things to different people. This should not be surprising. We all adjust our message to our audience. My writing in this article is influenced by my perceptions of the reading audience. A young woman with autism, typing independently, told me that she favored a particular candidate in the 1992 presidential race. She told her parents she favored a different candidate. No one controlled her typing but her communication partners did influence what she said.
INFLUENCE, CONTROL, AND THE VALIDATION CONTROVERSY
The distinction between influence and control is very important in any discussion about the validity of facilitated communication. Facilitators can influence communication without controlling it. (My writing might be influenced by things written by Doug Biklen, Judy Duchan, Gina Green or Howard Shane. These authors are not controlling what I write.) On the other hand researchers must be careful, as they apply experimental controls, to not eliminate appropriate forms of influence that may be an integral part of the way partners interact.
In many of the FC validity studies changes were made in the technique to control for "facilitator influence." In some cases this meant eliminating key components of FC. For instance, Smith and Belcher (1993) instructed facilitators not to provide correction at any stage of training. In doing so, they changed the technique so dramatically that they were no longer studying facilitated communication. In spite of this fact the researchers concluded: "These results suggest that the communications previously reported from individuals with autism may have been influenced by facilitator coactivity" (p.181). It was unclear why their results should have any bearing on previous reports about FC.
In a similar experiment, Smith, Haas, and Belcher (1 994) tested the validity of FC using a procedure in which they prevented facilitators from using fundamental components of FC in each testing condition. Again, the researchers were clearly not testing facilitated communication, invalidating any of their generalizations. They had eliminated appropriate and essential forms of influence. This did not prevent the authors from claiming that they had contributed to the literature demonstrating "facilitator control" in facilitated communication.
Unfortunately most of the validity studies published to date fail to make the important distinction between influence and control . Many studies claim to demonstrate "facilitator influence" without providing any definition (Cabay, 1994; Eberlin, McConnachie, lbel, and Volpe, 1993; Hudson, Melita, &Arnold, 1993; Moore, Donovan, Hudson, Dykstra, & Lawrence, 1993; Smith & Belcher, 1993; Vasquez, 1995, Wheeler, Jacobson, Paglieri, & Schwartz, 1993). This has led to a great deal of confusion about what is actually demonstrated in these studies. In the discussion sections researchers frequently treat their demonstrations of "facilitator influence" as if they are demonstrations of "facilitator control" when they are not.
For example, Eberlin, McConnachie, lbel, Volpe (1 993) describe a study in which they tested 21 subjects after only 20 hours of experience with FC. They report that, in a single test involving vocabulary words and a short personal interview, 8 subjects could answer some questions when the facilitator knew the answer. These subjects could not answer those questions when the facilitator did not know the answer. According to the authors, "There was evidence that some facilitators influence the subject's typed responses while facilitating when they knew the content of questions and/or answers" (emphasis added). Unfortunately, the authors never define "influence." If they mean that the facilitator controlled the user's responses, they clearly failed to demonstrate control. If they mean that facilitator knowledge of the correct answers influenced the ability of a small number of persons to communicate, this is correct. This information tells us nothing about the nature of the influence or whether it was, infact, facilitator control. Neither does it explain why 3 individuals actually did better in blind trials or why 10 others appeared to be uninfluenced!
Several published studies did demonstrate facilitator control of responses in some trials within highly contrived settings (e.g. Hudson, Melita, & Arnold, 1993; Montee, Mittenberger, & Wittrock, 1995; Shane & Kearns, 1994, Wheeler, Jacobson, Paglieri, & Schwartz, 1993). In these studies procedures were used that Green and Shane (1994) refer to as "double-blind." FC users were asked to identify stimuli that may have been the same or different from stimuli seen or heard by their facilitators. When FC users typed the name of things to which their facilitators were exposed and they were not, it was considered an indication of facilitator control.
Undoubtedly some of the facilitators in these studies control at least some of the communication of the people they facilitate. There are, however, a number of reasons for questioning the generalizability of these results and the validity and reliability of this "double- blind" approach in general. These include:
-All of these studies had only a small number of trials in only one or two sessions. Cardinal & Hanson (in press) demonstrated that FC users need to practice the testing protocol to be successful. Gene Marcus' experience supports this view (Shevin & Marcus, In press). After initially being unable to identify any items in "double- blind" trials, Marcus identified 100% of the items after practicing the protocol for 15 months.
-All of these studies ignored caveats from the authors of the original "double-blind" test warning that judgments can not be made about validity based on a single test (IDRP,1989)
- Vasquez (1995) reports that a great deal of variation exists among these studies in the extent to which facilitators seem to control answers. This suggests that these tests are unreliable. It also suggests that the quality of facilitation may vary greatly - possibly the result of variations in facilitator training, facilitator and user practice, and the degree of facilitator supervision.
- The internal validity of "double-blind" tests can be questioned because they have failed to detect valid communication for persons who later prove that they can author their messages with FC.
-The number of people tested was small. Except Smith, Haas, and Belcher (1994) (a study, as noted above, in which the facilitators did not test FC) the studies looked at individuals from populations very different from those described in the qualitative literature (Kurtz & Depoy, 1995). Any generalizations to the population of FC users at large are unjustified.
We do not know how participation in a "double-blind" study affects the way facilitated communication partners interact. We do not know how it changes the way in which facilitators and users influence each other as they create a message. We do not know if cueing is more likely in this kind of testing situation than it is in natural conversations. We do know that people who could not demonstrate their competence the first time they were tested in this manner are now able to clearly demonstrate their ability to communicate messages uncontrolled by their facilitators (Shevin & Marcus, In press; Biklen, 1992).
APPROPRIATE INFLUENCE
Many of the quantitative studies of facilitated communication eliminate appropriate forms of facilitator influence as they attempt to control for facilitator control of responses. Some of these studies alter the technique in very subtle ways. Others use protocols that seem to either ignore or misinterpret early descriptions of FC in the qualitative literature (e.g. Biklen, 1990, Biklen, Morton, Saha, Duncan, Gold, Hardardottir, Karna, O'Connor, & Shridevi, 1991; Biklen & Schubert, 1991; Crossley, 1992).
So far, only two controlled studies preserved the complex interaction and influence typical of FC. In both Sheehan & Mattouzi (In press) and Olney (1995) facilitators were allowed to vary the amount of support and the type of support they provided from trial to trial. In both studies FC users were also given some choice in determining the type of activities in which they would participate. In addition to validating the communication of some of the participants (Olney - 5 of 9 and Sheehan & Mattouzi - 3 of 3) both studies provided additional information on the kinds of supports people need to communicate and to pass messages.
The early qualitative literature has also been supplemented by new studies examining the nature of support and influence in FC. Sabin and Donnellan (1993) studied the complex interaction between the facilitator and the user. They examined the physical and verbal supports facilitators provide, the decisions facilitators make, and how the facilitator helps the user negotiate meaning. They found that the facilitator may provide assistance in 11 message formulation and transmission" and "in repairing breakdowns, introducing topics, expanding on topics, and assisting in guessing and predicting." (p. 207).
Several researchers have attempted to look at facilitated communication from the perspective of a collaborative model of communication. (Duchan, 1993; McSheehan & Sonnemeier, 1994). The collaborative model was distinguished from a conduit view of communication in which communication is "formulated by one partner and then sent to another over an invisible conduit" (Duchan, 1993, p.1111). Evidence indicates that non-disabled communication partners collaborate in many ways when communicating as they "co-construct" conversation (Duchan, 1993; Sonnemeier, 1993). Body language, facial responsiveness, and the synchronization of movements, are among the many ways partners communicate with each other. This same co-construction of messages takes place during facilitated communication although some of the particular mechanics of that collaboration may be unique to FC.
Relevant FC research must begin to treat facilitated communication as a complex and interactive phenomenon. Communication between any two partners is never simply the one directional transmission of a message from one person to another. It is a mutual, collaborative process in which conversants continually influence each other. Facilitators influence their partners' communication just as all communication partners co-construct conversations. Those requiring facilitation may need certain unique forms of physical and emotional influence. This fact alone should not lead researchers to dismiss it as invalid.
SELECTED REFERENCES
Due to space constraints, only some of the references referred to in the above article have been listed. For a complete reference list, write to: Center for Community Inclusion, University of Maine, 5717 Corbett Hall, Orono, ME 04469-5717, or call (207)581-1084.
Biklen, D. (1992). Autism orthodoxy versus free speech: A reply to Cummins and Prior. Harvard Educational Review, 62(2), 242-255.
Cabay, M. (1994). Brief Report: A controlled evaluation of facilitated communication using open-ended and fill-in questions. Joumal of Autism and Developmental Disorders, 24(4), 517-527.
Green, G., & Shane, H. C. (1 994). Science, reason, and facilitated communication. The Joumal of The Association for Persons with Severe Handicaps, 19(3), 151-172.
Hudson, A., Melita, B., & Arnold, N. (1993). Brief report: A case study assessing the validity of facilitated communication. Joumal of Autism and Developmental Disorders, 23, 165-173.
Kurtz, A. & Depoy, E. (1995, November). A review and critical analysis of research on the validity of facilitated communication. Presentation at TASH Conference, San Francisco, CA.
McSheehan, M. & Sonnemeier, R. (1 994, May). Quality training and FC: Relating how we learn and how we support. Mini-seminar at Annual Conference on FC. Syracuse, NY.
Montee, B.B., Miltenberger, R.G., & Wittrock, D. (1995). An experimental analysis of facilitated communication. Joumal of Applied Behavior Analysis, 28,189-200.
Olney, M. (1995). Controlled evaluation of facilitated communication: Results and discussion of a validation study. Presentation at the Fourth Annual Facilitated Communication Conference, Syracuse, New York.
Sheehan, C., & Mattouzzi, R. (In press). An investigation of the validity of facilitated communication through the disclosure of unknown information. Mental Retardation.
Smith M.D. & Belcher R.G. (1994) Facilitated Communication: The Effects of facilitator knowledge and level of assistance on output. Joumal of Autism and Developmental Disorders, 24, (3) 357-367.
Vasquez, C. (1995). Failure to confirm the word-retrieval problem
hypothesis in facilitated communication. Joumal of Autism and
Developmental Disorders, 25, 597-61 1.
The purpose of Facilitated Communication In Maine is to promote the appropriate use of facilitated communication through education, technical assistance, and support to people with disabilities, parents, educators, speech and language pathologists support providers, and other interested individuals. The project provides up-to-date information on current best practices, introductory and advanced workshops on the technique, resources regarding theoretical and practical components of facilitated communication and ongoing support to a network of resource persons who provide local education and support to other facilitators.
In complying with the letter and spirit of applicable laws and in pursuing its own goals of pluralism, the University of Maine shall not discriminate on the grounds of race, color, religion, sex, sexual orientation, national origin or citizen status, age, disability, or veterans status in employment education, and all other areas of the University. The University provides reasonable accommodations to qualified individuals with disabilities upon request. This publication is available on audiotape and upon request this material will be made available in other alternative formats to accommodate the needs of individuals with disabilities.