Ouija, FC, and Eye-Tracking Devices: Review of Andersen et al. 2019
Before I review Episode 9 of the Telepathy Tapes, I want to take a moment to talk about FC and the Ouija board. Both are forms of automatic writing, both are predicated on the belief that movements made by the participants result in the spelling of words by an outside agent. In the case of FC, the “outside agent” is the nonspeaking individual being subjected to the technique and in the case of Ouija, it’s an invisible entity of some sort (deceased relative, spirit guide, etc.).
Image by Colton Sturgeon
Ouija board/planchette use has long been part of a mourning culture: that is, its users are trying to, and believe they can, engage in conversation with a dead loved one (e.g., ancestor, child, parent). I personally think that’s what’s happening with FC use as well. The facilitators, often mothers, create an idealized version of their nonspeaking child and carry on a “conversation” using automatic writing, a letter board, and magical thinking as they project their own thoughts onto their child or client via FC-generated messages. The facilitator either uses the wrist, elbow, shoulder or other body part of their client or loved one to mimic a planchette (as in touch-based FC) or uses the letter board held in the air to serve the same purpose (as in Spelling to Communicate, Rapid Prompting Method or Spellers Method).
Almost from FC’s inception, critics of the technique have pointed out the similarities between FC and the Ouija board. (See References and Recommended Reading below). Douglas Biklen, founder of FC in the United States, knew about this concern when he wrote Communication Unbound, an article he published in 1990 touting the miraculous benefits of FC. In it, he stated:
The behaviors of people labeled autistic are often unusual and appear to reflect the lack of attention and/or awareness of social and communication cues and/or severe intellectual disabilities. Perhaps it is such behaviors, including the on-again, off-again ability or willingness of students to communicate, that cause some people to worry that facilitated communication is no more real than a Ouija board. (p. 299, Biklen, 1990)
Jeff Powell stares at the ceiling while Douglas Biklen and his facilitator stare at the letter board during an FC session. (Image from Prisoners of Silence, 1993)
But, he was wrong. It isn’t the behaviors of the individuals with autism that concerned critics of FC. Rather, it was—and still is—the behaviors of the facilitators that interfere with the individual’s ability to communicate independently that cause critics to question FC use.
A frequent question we get is: how does the facilitator not know they are cueing their clients? Fair question.
I suspect sometimes facilitators do know they’re moving the child’s hand or the letter board in the air, but they are taught in FC workshops to downplay or ignore these concerns.
But, beyond that, a 2019 study by Andersen et al. titled “Predictive minds in Ouija board sessions,” gives researchers clues as to how non-conscious, automatic writing can seem to Ouija board users—or facilitators—like the movements are coming from someone (or something) other than themselves.
Equipment developed by Michael Faraday to explain the ideomotor effect on turning tables. The Illustrated London News, July 16, 1853, page 35. (Wiki Media Commons)
In Andersen et al., the researchers found 40 volunteers at an Ouija board convention who were willing to participate in two relatively short tests and answer a questionnaire about their experiences. The testing was less about authorship than it was about exploring the behaviors of the participants while interacting with the Ouija board and planchette.
The participants were divided into 20 groups of two and asked to participate in two activities or test conditions: “Ouija board” condition and “Voluntary Action” condition. During the two conditions, both participants were equipped with mobile eye tracking technology that allowed researchers to “compare predictive gaze behavior of the individuals” (singly and in combination). In other words, researchers wanted to see if
one or both participants visually anticipated the planchette’s movements by looking at a specific letter on the letter board and/or at the planchette as they spelled out words
the participants’ eye gazes were synchronized while moving the planchette (e.g., did the participants look at the same or different letters?) or was one person (perhaps unwittingly) “leading” the other to a desired letter
the eye gaze/predictive behaviors of the participants varied with a known response (e.g., spelling out the word “Baltimore”) vs. a response that was unknown to at least one of the participants.
The methods used in the testing followed guidelines and regulations in accordance with the National Committee on Health Research Ethics in Denmark. Informed consent was obtained from all the participants.
In the first, “Ouija board” condition, the participants were asked to use the board in their typical fashion, that is by sitting at a table with their fingers resting on a planchette. Then one of them would ask a question (presumably to the “spirit” associated with the Ouija board) and have the “board” respond. The test lasted 10 minutes.
Note: I’m guessing that the person asking a question to the Ouija board would have some idea of the answer he or she was expecting to “appear” even if subconsciously, but the researchers did not directly discuss this issue.
Researchers used data from the eye tracking technology to annotate when the participants 1) looked at the planchette and 2) looked at—or predicted—the upcoming letter. The coders only counted eye gazes that fell exactly on the letter the planchette subsequently reached. This was true for the “voluntary action” condition as well.
Because some of the responses in this (the “Ouija board”) condition generated random strings of letters, the researchers only counted meaningful responses. The examples provided in the report were these:
Participant: “Would you like to spell your first name?” Ouija board: Henry (counted as a meaningful response)
Participant: “Where did you pass away?” Ouija board: STACW (not counted as a meaningful response)
Researchers scored 67 responses out of 143 responses in the Ouija condition as “meaningful.”
From the data collected, researchers were able to determine that at the start of spelling a word in the “Ouija condition,” one participant’s eye gaze would generally track to the target letter of the word being spelled ahead of their partner’s gaze. In other words, one of the two participants would initiate or “lead” the response to the question asked. This would remain true until a point in the spelling of the word where it was evident to both participants what word was being spelled out. Once that point was reached, each participant’s ability to “predict” the next letter(s) of the word (by looking at it) increased and the two participants would visually track letters at more or less the same rate.
Since Anderson et al. gave the example of “Henry” as a response to the question “Would you like to spell your first name?” Let me use that as an example to further explain what I mean.
It’s likely the person who asked the question “Would you like to spell your first name?” had a response in mind (e.g., “Henry”) even if the response wasn’t pre-planned before the session started. All it takes is a split second to come up with an answer.
Note: I think it’s also possible that random strings of letters occurred in this condition when both participants thought they knew the answer to a question but then (subconsciously) tried to lead their partner to spell out different words. For example if Participant 1 thought “Henry” and Participant 2 thought “Michael,” the Ouija-generated spelled response might be something like H-M-E-I-C-N-L and be counted as a meaningless response.
For our example, let’s call the person who asked the question “Participant 1.” And, since Participant 1 now (silently) knows the answer to the question is “Henry,” he or she (perhaps unconsciously) gazes at the letter “H” before his or her partner (Participant 2) does. In this example, Participant 2 doesn’t know the answer, but passively allows the planchette to drift to the letter “H.” Just like with FC, the Ouija board only “works” when participants (facilitators) have eye contact with the letter board.
After learning that the letter “H” was the first letter in the “Ouija board’s” response, Participant 2 would then try to think of names that begin with “H” and try to figure out the next logical letter in the sequence. Again, this thought process would happen very quickly and, perhaps, without much conscious awareness.
Those familiar with the rules of English spelling and grammar would likely know that the letter “H” would be followed by a vowel. But which vowel? H-A, H-E, H-I, H-O, H-U, H-Y could be the start of any number of first names. Again, Participant 1’s “predictive behaviors” (e.g., looking at the next letter in the sequence) would, most likely, lead the planchette (and Participant 2) to the second letter in the sequence, “E.”
”H-E” still leaves room for Participant 2 to wonder if the next letter will be a vowel (as in the name “Heather”) or a consonant (as in the name Heddy). So, it’s likely the choice of the third letter in the sequence would also be controlled by Participant 1. In our example, the sequence selected by the partners using the planchette is “H-E-N.”
But with the addition of the letter “N,” in the sequence, the options for the two participants to spell out a meaningful first name reduce quite a bit (e.g., ruling out names like “Helen” or “Herman”). With the addition of this third letter, it is likely that Participant 2 now has a pretty good idea of what name the “Ouija board” is trying to spell out. Once Participant 2 knows (or thinks he knows) the correct response, his ability to predict the next letter in the sequence increases and aligns with Participant 1’s predictive/eye gaze responses. There’s still room for some error (as we’ll see in a moment), but at this point in spelling out the “Ouija board’s” response to the question “Would you like to spell your first name?,” both participants’ eye gaze and predictive behaviors would be more or less in synch as they finish spelling out the word “Henry.”
In our example, H-E-N-R, then H-E-N-R-Y would be logical guesses for the fourth and fifth letters in the sequence.
Note: If one of the participants had guided the planchette to H-E-N-R-I, perhaps the participants would have stopped, or perhaps they might have continued on to spell H-E-N-R-I-E-T-T-A.
The easier the movements become (due to an increased ability to “predict” the next letter in the sequence), the less resistance the participant feels when moving the planchette, the greater the chances the Ouija board user will feel like the movements are not coming from himself or his partner. In post hoc testing, the researchers discovered that participants’ chances of predicting upcoming letters “increased roughly 5% by each consecutive letter in the meaningful responses.” (See Abdication Patterns in FCed Individuals for an FC comparison).
In the “Voluntary Actions” condition of the Andersen et al. study, the pairs of Ouija board users were asked to spell the word B-A-L-T-I-M-O-R-E. In this condition, both participants were aware of the target response and, unlike in the “Ouija board” condition, their eye gazes were, generally, synchronized throughout the spelling of the word. This makes sense because, from the outset, both individuals knew the target word.
Image included in the Andersen et al. study showing an example of the video used by coders to detect eye gaze. Andersen, et al. , 2018)
Andersen et al. discussed the concept of Sense of Agency (SoA) or an individual’s (subjective) feeling that he or she is “generating and controlling actions in order to influence external events.” To explore this topic, researchers had the participants fill out a questionnaire about their experiences directly after completing both the “Ouija board” and the “Voluntary Actions” conditions. The questionnaire was designed to explore participants’ SoA during the activities.
Overall, the participants most “successful” with the Ouija board experienced a reduced SoA that the researchers described as a “mismatch between the predicted and actual sensory consequences” of moving the planchette. In other words, to the participants, it didn’t feel like they were looking at the letters ahead of time (even when the eye tracking equipment showed that they were) and/or the participants were unaware that they or their (real life) Ouija board partners were moving the planchette toward the desired letter on the board. Instead, the participants attributed the spelled responses to someone (the other participant) or something other than themselves (an entity from the spirit world). Based on the questionnaire, this reduced SoA amounted to a basic lack of self-awareness regarding the participants’ own behaviors during the Ouija board activities. A reduced SoA was most prevalent in individuals who were already predisposed to believe that the Ouija board “can be used to contact entities such as spirits, ghosts, demons, angels, etc.”
Switching to FC for a moment, this idea of a reduced Sense of Agency in facilitators can also be found in a 2003 article called “Clever Hands: Uncontrolled Intelligence in Facilitated Communication.” Wegner et al. argued in that paper that a belief in FC contributes to facilitators’ perceptions that they are not the authors of FC-generated messages. Remember, facilitators are taught to “presume competence” in their clients and not test for authorship, so it is unlikely facilitators would (self-reflectively) detect cueing behaviors in themselves during a facilitation session 1) because, in order to make FC “work” facilitators have to suspend disbelief, and 2) because they’re taught by workshop leaders not to look.
One particularly interesting finding of the Andersen et al. study was that even in Ouija board users who believed responses were “really driven by the subconscious mind of its users” (due to the ideomotor effect) and not some supernatural, invisible entity, the participants reported that their partners were pushing the planchette more than they were themselves. I hear this from facilitators as well. Facilitators may (reluctantly) admit that facilitator cueing is possible, but always in reference to other facilitators and not themselves.
The authors of the Andersen et al. study included a reference to a 2003 article titled “Two Eyes for an Eye: The Neuroscience of Force Escalation” (See Shergill et al. below) that explored the question of why shoving matches between two people can escalate so quickly (e.g., one person gets shoved, maybe even by accident, but responds by pushing the person back with more force. Then the other person responds by pushing back with even more force as the situation gets out of hand).
By using a device called a “force transducer,” researchers applied external pressure to their participants’ fingers and then asked them to replicate how much pressure they’d felt. It turns out that the participants of the study perceived that the pressure exerted on their index finger was greater than it actually was. In other words, when the person was “being pushed” by the force transducer, they perceived it to be a greater amount of force than when they were doing the “pushing.”
I think Andersen et al. mentioned the Shergill study when discussing the use of a planchette, because, if true, that means it is likely that participants underestimate the amount of pressure that they, themselves, are using to move the planchette toward a desired letter and overestimate their partner’s influence over the device. This faulty perception by participants aids in the illusion that the movements are not coming from the planchette user him or herself, but from someone (the other planchette user or users) or something other than themselves (a “spirit” or otherworldly being).
Switching back to FC for a moment, I am reminded of a study by Emiko Kezuka called The Role of Touch in Facilitated Communication that measured the pressure exerted by the facilitator during a touch-based FC session. Kezuka discussed how surprisingly little pressure is necessary for a facilitator to influence letter selection once the momentum of searching for a letter starts (e.g., while the facilitators is holding onto the client’s wrist, elbow or shoulder while the individual is hovering a finger over the letter board). Of course, this situation is different than when two literate adults are using a planchette. In the FC example, it’s likely that only the (literate) facilitator would be directing letter selection, but I think it would be fascinating to replicate the Kezuka study using some of the concepts from the Shergill study. My guess is that facilitators would underestimate the amount of pressure they’re using to “support” their clients’ wrists, elbows, shoulders, etc. during letter selection and overestimate their clients’ participation in the spelling activity.
As I’ve thought about the Andersen et al. study, I wondered if it inspired Vikram Jaswal’s eye tracking study with individuals being subjected to FC. Both sets of researchers used eye-tracking devices to measure the prediction of letters in Ouija board-like activities. Katharine wrote a review of Jaswal’s eye-tracking study, so I won’t go into detail here, but suffice it to say, the study was poorly controlled and raised a lot of questions about the individuals’ ability to predict letters while being subjected to facilitator-dependent techniques.
Personally, I’d love to see Jaswal’s study replicated—only this time put eye tracking devices not just on the participants but on the facilitators as well. This type of testing would be useful for both touch-based and no-touch forms of FC, but I’d be especially interested to know how much the predictive eye gazing behaviors of facilitators influence where they position the letter board in front of their clients in the supposed “no-touch” forms of FC (like Spelling to Communicate or Rapid Prompting Method). I think a facilitator-focused study of FC with the facilitators wearing eye tracking devices would reveal a lot about facilitator influence and control over letter selection and explain why they (the facilitators) have to look at the letter board during letter selection, but their clients and loved ones do not.
References and Recommended Reading
Andersen, M., Nielbo, K.L., Schjoedt, U. et al. (2018). Predictive minds in Ouija board sessions. Phenomenology and the Cognitive Sciences. DOI: 10.1007/s11097-018-9585-8
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. https://doi.org/10.1080/17489539.2021.1918890
Biklen, D. (1990). Communication unbound: Autism and praxis. Harvard Educational Review, 60, 291-315. DOI: 10.1080/21548331.1992.11705407
Kezuka E. (October 1997). The Role of Touch in Facilitated Communication. Journal of Autism and Developmental Disorders, 27(5), 571-593. DOI: 10.1023/A:1025882127478
Shergill, S.S., Bays, P.M., Frith, C.D, and Wolpert, D.M. (2003, July 11). Two Eyes for an Eye: The Neuroscience of Force Escalation. Science. Vol. 301, p. 187.
Spitz, H. (1997). Nonconscious Movements: From Mystical Messages to Facilitated Communication. Routledge. ISBN 978-0805825633
Wegner, DM, Fuller, VA, Sparrow, B. (2003, July). Clever hands: Uncontrolled intelligence in facilitated communication. Journal of Personality and Social Psychology. 85 (1): 5-19. DOI: 10.1037/0022-3514.85.1.5