In new research published in the journal Frontiers in Psychology, a group of psychologists recruited 420 participants to watch several short, silent videos on their computers and report on what they saw.
In the first set of videos, respondents watched a man named Matthew L. Tompkins — the lead author of the study, a doctoral researcher in psychology at Oxford and, as it turns out, a semiprofessional magician since age 14 — perform some simple magic tricks involving common objects, like this one.
After the simple tricks, the subjects watched a non-magical "trick" — Tompkins eating the crayon, or wearing a poker chip like a monocle. This was to make sure they were paying attention, and that they weren't saying that silly, everyday activities were "magical."
Finally, the subjects watched the one video that was the meat and potatoes of the whole experiment, the piece de resistance — the Prestige — of Tompkins's routine.
Before we go on: What did you just see? Did you see something — an object, maybe a crayon or a coin — disappear? Or did you see a guy waving empty hands, pantomiming a magic trick but not actually performing one?
If you saw something disappear, congratulations: You're one of the 32 percent of people in Tompkins's study who could be induced to see an object that literally doesn't exist.
In that final video, the only trick was that Tompkins wasn't actually doing a trick — he was simply going through the motions. He calls this the "Phantom Vanish Illusion." But the fact that roughly a third of viewers were convinced that they actually saw something disappear says a number of interesting things about the human psyche.
Many magic tricks that we're familiar with operate either by giving the illusion of something disappearing, or of a thing appearing in a place it's not expected to be. One of the reasons the Phantom Vanish Illusion is unique is that it creates, literally out of thin air, the perception of an object that never existed in the first place.
Of course, a 32 percent success rate isn't exactly a grand slam when it comes to magic tricks. No performer wants to do a routine that two-thirds of the audience aren't going to fall for. "You wouldn't design a trick like that for a magic show," Tompkins said in an interview.
Still, the success rate is strikingly high considering that the videos were silent, meaning Tompkins wasn't able to talk, distract viewers, or otherwise plant ideas about the trick in their minds. And the questionnaires viewers filled out didn't contain any leading information that could have prompted people to "remember" phenomena they didn't actually see.
Rather, viewers were simply given the following instruction after each trick: "Please write a description of what was shown in the video." Most of the people who said they saw "something" disappear didn't give any specific indication of what object they saw. But a third of the people who saw an object — or 11 percent of the total sample — said they saw a specific object. In most cases, this was whatever object was being used in the tricks they watched in previous videos: a crayon, a poker chip, a coin, etc.
"Magic effects result from 'hacking' otherwise adaptive perceptual processes to create false fictional experiences that lead to paradoxical experiences," Tompkins and his co-authors write. "At the 'climax' of the Phantom Vanish Trick, the magician clearly shows that both of his hands are empty. Because the spectator does not believe that they could have misperceived an object that was never really there, they are unable to intuit that the true method is even possible."
"When you combine science with magic," Tompkins explained in an interview, "you get access to the nicely reproducible phenomena that magicians have honed for centuries, and you can use the scientific method to actually take them apart and see what's happening."
Tompkins said that approaching magic performance from a scientific standpoint is a relatively new development in psychology. For most of the 20th century, he says, serious researchers shied away from the topic. Academics were supposed to shine light on empirically observable phenomena. After all, what interest would they have in perceptual tricks?
But that attitude has changed in recent years. Tompkins himself is in a unique position to study the effects of magic. He's been a performing magician himself since he was a teen. After becoming interested in the science of how magic tricks fool people, he got into Oxford's experimental psychology doctoral program in part by performing several coin tricks during his admissions interview.
His experiments shine light on some curious facts about human nature. Nearly 1 in 5 American adults say they have seen a ghost, for instance. Tompkins says numbers like these might be a function of the same "eccentricities" in our perceptual systems that magicians have taken advantage of for centuries.
"There are a lot of ways we can perceive things that are seemingly impossible, but which are very normal and predictable eccentricities of our visual awareness," he said. Things like hypnagogic hallucinations, for instance — where dreams intrude on our wakeful consciousness, often either right before we fall asleep or after we wake up.
"If you don't know that those hallucinations are a thing," Tompkins explained, "and you see a weird shadow person walking around the room, the most plausible thing to you is that there's actually a weird shadow person walking around the room."
Tompkins' next trick will be seeing what types of success rates he can get on the Phantom Vanish Illusion by carefully adding other variables — like a spoken monologue during the routine, or leading questions in the post-video interview. His work adds to a growing body of research showing just how limited and ultimately fallible our perceptual abilities really are.
More from Wonkblog: