A password will be emailed to you.

The Ouija Board remains an object of fascination in the 21st century, despite – or perhaps because of – its ‘vintage’ feel and rudimentary equipment. A board with some letters on it (or in some cases, just ‘yes’ and ‘no’ options), and a movable pointer (planchette) for participants to put their hands on – a simple set-up from which otherworldly communication appears to flow.

Ouija seems like a throwback to the era of 19th century Spiritualism, sitting in a quiet room with friends or family and allowing spirits to take control of the conversation. For some, it’s just a fun game that can have surprising results – for others, it’s considered a dangerous past-time that can invite malevolent forces into your life.

For scientists and skeptics though, it’s a prime example of people fooling themselves, via what is known as ‘the ideomotor effect’. As the Wikipedia article for Ouija states, “the action of the board can be parsimoniously explained by unconscious movements of those controlling the pointer, a psychophysiological phenomenon known as the ideomotor effect.”

But despite the apparent consensus of the ideomotor effect being behind the mystery of the Ouija Board, there has actually been very little scientific research done into determining the mechanisms at work in how the planchette spells out answers to questions asked by Ouija participants.

However, Dr. Eckhard Kruse, an applied computer scientist at Baden-Wuerttemberg Cooperative State University in Germany, has now created a camera- and software-based system that can serve as a tool for future Ouija research.

Kruse’s Ouija tracking system grew out of his fascination with the game and early attempts to record his sessions:

My idea to bring in some technical devices to Ouija sessions was initially triggered by mere convenience. In the course of regular sessions, writing down letters on paper was tedious and often not fast enough. So I switched over to speech recognition…but still prone to errors, when speech recognition picked up questions or voices from other sitters. So, finally I developed a setup that allows complete tracking and spelling support in real time during the Ouija session.

Kruse’s Ouija-tracking set-up utilizes a webcam and Java-based software to detect the position and motion of two colored dots on the planchette in order to track the letters traced out during a session, along with sensors to detect the pressure of fingers on the planchette and its movement.

Eckhard Kruse's Ouija Board motion tracking system
Eckhard Kruse’s Ouija Board motion tracking system

In a paper published in the most recent Journal of Scientific Exploration (“A Camera-Based Tracking System for Ouija Research“), Kruse gives details of his open source tech set-up, along with examples from his 50+ Ouija sessions using it, and how it might inform future research into the topic. For example, he notes the following observations and data obtained through his tracking system:

Throughout the course of the recorded Ouija sessions, a development toward faster and more accurate spelling was observed. Initially, letters were frequently jumbled, focus was on yes–no answers, and total text output was limited to a few hundred characters, often with meaningless phrases. In later sessions, jumbled letters or errors of orthography were rare (typically less than one percent of the spelled characters), and total output increased. For example, a session on 2018/12/30 lasted 2 hours and 10 minutes, and yielded 4,799 characters, on average a character every 1.75 seconds. Due to pauses for questions at the beginning and end of a session, actual spelling was faster, about 1.3 to 1.5 seconds per character. Maximum speed was about 100 cm/s, with almost one character per second.

Kruse also notes that his preliminary data challenges prevailing assumptions about the ideomotor effect, such as the claim that “that meaningful responses from the Ouija board may be an emergent property of interacting and predicting minds that increasingly impose structure on initially random events in Ouija sessions.”

If that was the case, one should expect that once “structure” is established, i.e. a meaningful beginning of a word or sentence, the spelling becomes easier and quicker. Sometimes this was the case in our data, but in general there was no correlation between spelling speed and the number of meaningful letter choices, or, in other terms, the ease with which the next letter could have been guessed. Also, sessions typically started right away with clear and meaningful sentences, which do not at all appear like initial random events that are only gradually molded into German phrases.

The tracking software also allowed researchers to pick up on less obvious movements of the planchette. During one session withe four sitters, the response to a question did not touch on any letters, but instead the planchette simply made a swift motion that none of the sitters initially recognised. After it occurred two more times, one of the sitters suggested it might be a heart – and a check of the motion trace captured by the software confirmed that this was indeed the case.

Motion tracking showing heart shape traced out on Ouija Board
Motion tracking showing heart shape traced out on Ouija Board

Overall, Kruse notes that his preliminary experiments with his prototype Ouija tracking system offer the following useful observations – including the possibility of psi effects assisting in the movement of the planchette:

  • When hitting a letter, details of the motion path (cusps vs. smooth curves) could indicate whether the next letter is already anticipated, hinting at (conscious or unconscious) knowledge of how the spelling will proceed.
  • The time needed to spell the next letter was only weakly related to its “guessability,” i.e. the number of choices to construct meaningful words, in contrast to Andersen et al.’s (2018) statement about imposing “structure on initially random events.”
  • According to the experiments with the touch sensor, it is the sitters’ fingers that are moving first, then the planchette follows. This complies with the ideomotor explanation. If there were psychokinetic effects, it seems likely that it would be the other way around and the fingers would be following the actions of the planchette.
  • Often two sitters were able to move the planchette synchronously, similar to the volitional spelling of a given message. This challenges conventional ideomotor explanations, as these would require some negotiation process regarding the next, unknown target. Even though this might happen unconsciously, it would require some time for information transmission between the sitters and potentially some delay in the action of one sitter, causing similar rotations as when one sitter voluntarily leads the planchette — unless there are psi effects explaining the synchronicity.

For more details on Kruse’s tracking system, the entire article is available through the open access Journal of Scientific Exploration, Volume 33, Issue 2.