Ms. Bluhm:

Thank you for your comment.

As for your personal validation, one item is not be enough. It could be a lucky guess. If words of common usage in your household were employed in your test, there would have to be enough of them that lucky hits could be distinguished from genuine answers. We're facing a "restricted range" problem. We see FC validations in which the facilitators are given the list of possible answers in advance. This artificially pushes up the apparent success rate by causing the facilitators to never guess off-list words. Since in FC World lucky hits are deemed to be "correctly passed information," even a statistical failure can be said to contain some genuine communication. By that standard, Clever Hans was communicating genuine human like thoughts on the 8% of the trials that the right answer was tapped out.

You're going to have to start with about 10 items, and have an unbiased method of shuffling them--easier said than done as research on card shuffling shows. If there is some ability to independently type or point successfully, you may need more items in order to distinguish those answers actually produced by the communicator and those produced by the facilitator. (That would raise the question of why FC was being used at all, and not any of the many methods available to people who have problems communicating, but retain some ability to do so.) And because what is technically called "stimulus leakage" is a strong possibility among people have worked together a long time, items should be presented "blind" such that the questioner doesn't know the answer. Other subtle cues might provide clues about the word being asked. One serious problem with such tests is the response of the questioner, who can give feedback on the success of each item by unconscious facial expressions--satisfaction or disappointment, for instance. If each word is used only once, then elimination will allow the facilitator to narrow the guesses as the test progresses. Questioners might also take slightly longer with previously missed words or difficult ones than ones that were correctly guessed. I was at an FC event where the proposed test did exactly that and more. The facilitator went through a stock list of words, eliminating each one from the list as guesses were made, all while indicating whether each guess was right or wrong. Thus, anyone with even an average memory could always guess the last one. He also suggested throwing out the first few items as "warm ups." Of course, that threw out the answers most likely to be wrong just because the possibilities had not been narrowed. Throwing out the first few meant that even pure random guessing from remaining items from the list would produce a statistically significant positive result!

The friend? Well, I too have seen such a person, maybe the same person, who could supposedly type independently. The problem was that her facilitator was signaling where to move on the keyboard with hand movements: up slightly meant move one way across the keyboard; down mean move the other. Other movements signaled when the communicator was over the right letter, and then when to press. This was very subtle, and it is not surprising others didn't notice it because most people focus on the typing rather than the movements of facilitator. Misdirection, like in a magic trick. That apparent independent typing is real typing is strange thing to believe anyway. First of all, "independent" means to FC advocates something different than it does to the rest of the world. In FC, a communicator can be "independent" even if someone is always present to hold or otherwise manipulate the letter board or typing device. So we find statements such as this, "The 2nd participant [Sue Rubin] was considered an independent typist. Although she did not require any physical touch, she did require a supporter to hold the typing device while she typed" (Robledo & Donnellan, 2009, p. 300). Well, what I see when others hold the boards or typing devices is the needed letter is moved slightly toward the putative communicator's finger, with the communicator responding by a larger motion in the opposite direction toward the letter the facilitator intends to have typed. (This is especially common in Rapid Prompting.) And, people just forget themselves. I've seen attorneys, who should know better, acknowledge that there could be a "Clever Hans" situation going on in FC, but then claim that typing without physical contact is proof of authorship. This is an easy one. Was anyone holding Clever Han's hoof when he tapped out the answers? No. Very obviously physical contact is not required for cueing. The same goes for Lady Wonder, who was literally an "independent typist" in FC parlance because they had built for her an horse-sized typewriter.

http://tinyurl.com/4jkmaop

It would be wonderful if something as simple as FC worked. It can't pass the most basic tests, involving the simplest communicative responses, or the most linguistically natural ones, under the most artificial or most natural conditions. It is simply not an effective method. Even so, you believe it is. Thus, I suppose one big question is why, if you have and know of genuine instances of FC, you're not applying for James Randi's Million dollar prize, or, barring an interest in money, settling the matter by participating in formal versions of the test you have already done.

Thank you again for your comments.

Jim