What magic tricks should teach us about tomorrow's technology
Do you believe in magic? Whether you do or not in the abstract is actually less important than you might think, because your senses have already decided that they’re believers.
Dr Gustav Kuhn, a reader in psychology at Goldsmiths and author of Experiencing the Impossible: The Science of Magic, has been practising magic since he was a child. Now he spends his days examining how the tricks work, and the alarming thing is that often the magicians’ best hunches are only half correct or even completely off. “Although magicians are very good at performing the trick and knowing what tricks work, their explanation for why they work are not necessarily correct,” he tells TechRadar, following his performance/lecture to an audience of skeptics at New Scientist Live.
“With magic, there are huge perceptual memory or reasoning distortions and they can give us really interesting insights into how the human mind works.” The long and short of it is that your perceptions are inherently faulty, with your brain providing helpful shortcuts that generally help you to function… unless someone is deliberately trying to deceive you, that is.
At one point in his talk, Kuhn bounces a ball twice before pretending to bounce it a third time. It’s not exactly David Copperfield, but it does the job as a demonstration. “About two thirds of people experience a ball that’s moving up and then disappearing,” he says. “You’re seeing something that clearly hasn’t happened.”
In short, it takes about a 10th of a second for eye stimulus to reach the brain, and that’s too long if you’re in danger. As a result, the visual system replicates what you’re expecting to see – hence some people really do see the ball flying into the air and disappearing.
Another example: just by adding a flicker to two images, Kuhn is able to make the majority of the audience fail to spot a fairly obvious difference between two pictures. Remove the flicker, and it’s apparent to everyone – which is probably why you should never play those pub quiz machine Spot the Difference games.
“People are oblivious to most of the things going on in their environment,” Kuhn explains. “And so once you've actually got control over their attention, you've got pretty much full control over what they see and what they miss.”
Case in point, this attention study from 1999. It’s only a short video: follow the instructions and see how you get on.
Tests like this and use of eye tracking software tell us something important. “What you’re seeing is not necessarily related to where you’re looking, but to where you’re attending with your mind,” Kuhn says. “Even within the magic literature magicians pretty much assume that attention and eye movements are very very closely linked.” Recent research suggests that simply isn’t the case.
The lessons for tech
These things are generally not considered when it comes to technology. We now know, for example, that talking on a phone – even with a hands-free kit – is just as dangerous as drink driving, yet one is legal while the other isn’t (this isn’t replicated when talking to passengers, by the way, because passengers can see where the driver is situated, and generally have the good sense to shut up during periods where maximum concentration is required.)
By the same token there’s another piece of technology which should never have left the design table: Google Glass. “Is it a good idea to develop human interfaces that allow you to present information onto glasses while you’re interacting with the world?” he asks during the talk.
“No, it really isn’t. It might make intuitive sense because people can keep an eye on the task, but doing so will distract them and they simply won’t be able to see it. Most importantly, they’re not aware that they won’t be able to see it.”
It’s for this reason that Kuhn jumped at the chance to do his talk – very similar to the one we heard at New Scientist Live – to a bunch of Google employees. You can watch the whole thing below if you like: we still can’t work out what’s going on at 03:52, despite watching it a whole bunch of times.
“I was interested because Google Glass has been a bit of a bugbear of mine, because it's just such a bad idea,” Kuhn recalls. “It's so terrible, it's really terrifying and I was quite keen to go and give a talk to Google to just highlight some of these limitations.” As it turns out, they were “as oblivious to these limitations as the general public.”
That’s not a criticism of Google, by the way: it’s everyone. Kuhn tells us the story of how eye trackers were used in the 1990s to measure saccadic eye movements, and how people would simply miss obvious things despite looking directly at them. “You could make these really huge changes to a scene and people just wouldn’t notice them,” he says. But because it was the 90s and eye trackers were expensive, people just wouldn’t believe the findings when presented at vision conferences.
“The fact that even visual scientists didn't believe that you'd miss these kind of changes illustrates just how counterintuitive a lot of these findings actually are,” Kuhn says. “I wouldn't expect someone working at Google or in a tech company to be aware of this unless you've actually experienced it because it is very, very counterintuitive.”
Google Glass may be dead and buried, but the Heads Up Display technology lives on, especially in car dashboards, and it’s just as bad an idea as it was on the eyes. “If you're developing new technologies, people are very much focused on the actual code and the technology without really thinking about the user experience,” Kuhn says. “And to measure that you really need psychologists.”
At this point, you may be getting flashbacks to a piece we wrote last year about our minds not being evolved enough to deal with social media. We put this hypothesis to Kuhn, and his response is pretty unequivocal: “No, they're not.”
Magic for good
It’s in this spirit that Kuhn set up the MAGIC Lab at Goldsmiths to empirically probe deeper into how magical techniques can improve our lives. It may sound flippant (MAGIC is actually an acronym for Minds Attention and General Illusion Cognition) but there are actually very real reasons to take an interest.
Right now, Kuhn and the MAGIC Lab are working on everything from applying magical suggestion to gaming narratives to conjuring up artificial intelligence that can trick humans.
“We're actually training up computer bots so they can deceive humans to look at what impact that this form of deception will have on people,” Kuhn explains. “We’re testing some of these magic principles, but also looking at what how do you how do people react when computers suddenly start cheating because we've got a lot of trust in computers.”
The gaming narrative stuff is even more interesting and works on a concept up the magician’s sleeve called 'forcing', where you believe you’ve made a choice via free will, but it’s actually been heavily influenced.
Imagine a situation where there are four cards on the table and you’re asked to touch one of them – Kuhn tells us that 60% of people will pick card number three. “If you ask people how likely others are to choose the same card as them, they go ‘well, probably about 30% or so’, so they vastly underestimate the extent to which they are influenced by these biases.”
And what about dual screens? Back when Microsoft unveiled the two-screened Surface Duo, Panos Panay made a big deal about the science behind double displays, saying: “We absolutely know scientifically that you will be more productive on two screens – much more than one screen ever could do… that seam down the middle lights up the mind in ways that’s almost impossible to explain because you have to feel it.”
We put that quote to Kuhn who is, it’s fair to say, unconvinced. “I find that quite hard to believe. The two-screen thing it's actually really highly unproductive,” he says. “When you disengage your attention and re-engage it, that requires lots of cognitive resources. So the best thing to do is to really shut down all other devices, concentrate on getting one thing finished and then move on to the next. Any new tool that encourages this attentional switching… I don't think it's a good idea, personally.”
But again, intuitively you might believe this to be wrong. That’s why magic is so helpful as an aid for testing: not only does is make cognitive psychology more accessible, but it challenges our perceptions in every sense. “The reason why magic works is because these limitations are so surprising and often counterintuitive,” Kuhn says. “Even for me as an individual who works as a magician and a scientist, I find it quite hard to fully appreciate just how wrong my intuitions about perception are.”
And how does the Magic Circle feel about the candid way Kuhn discusses why magic tricks work? “Yeah, the Magic Circle are generally not very happy with some of the ways in which we discuss this,” he concedes. His contention, backed up by a survey at a recent exhibition at the Wellcome Collection, is that some understanding of the science behind the tricks enhances people’s enjoyment of them.
That probably won’t wash with the powers that be in the Magic Circle, but for Kuhn there are bigger issues at stake. “A lot of my projects started off with wrong hypotheses,” he says. And if that’s the case with a life-long magician, then what hope for people who are accidentally meddling with forces beyond their comprehension?