conditions. You put the ball under the cup, lift up
the cup, and it’s not there. You put the ball under
the cup again and it’s in your pocket. Now it’s
underneath the cup. Now it has disappeared
again. It’s a great illusion – and the important
thing is you can repeat it again and again.”
Caffaratti connected his test subjects to an
electroencephalography monitor, which measures
electrical activity in the brain, and showed them
the video. He was looking for traces – known as the
event-related potentials (ERPs) – of the perceptual
processes that are at play when we perceive magic.
ERPs are changes in voltage that indicate that the
brain has responded to a specific stimulus (in this
case, the appearance or disappearance of the ball).
What Caffaratti found was that the unexpected
outcome of the magic trick elicited strong ERPs
that were significantly delayed compared with
those triggered by an event that wasn’t magical.
When subjects were shown a series of “normal”
events – the ball placed under the cup, for example,
and still there when the cup was lifted – their
brains showed a particular ERP, known as P300,
approximately 300 milliseconds after the event.
“But when the magic trick happened,” he says,
“a P300 was elicited about 50ms later. We were
seeing that, when it’s magic, it takes more time.”
What the results demonstrated was that an
“impossible” event, such as a magic trick, takes a
greater amount of time to be processed, compared
with something the brain is familiar with.
The fact that what was physically the same
object, in this case, a magician’s ball, elicited
different responses, depending on whether the
object appeared “magically” or not, helped to
confirm a long-held idea in psychology: that
our perception – the way in which we make sense
of the world – relies not only on the raw data
coming in to the brain from our senses, but also
on our pre-existing knowledge of the world,
which is stored in our memory.
Caffaratti had shown what happens when
searching that second stream draws a blank. He
had lifted the lid on the neuroscience of wonder.
am sitting in the back row of a lecture theatre in
Kuhn’s department at Goldsmiths. It’s
the summer holidays, and the faculty
is holding an open day for prospective
students. A hundred 17- and 18-year-
olds trickle in, neatly dressed in school
uniforms. They’re here to hear about
life as a Goldsmiths undergraduate.
But first they are going to be the subjects
of a psychology experiment.
Kuhn stands up and explains what is going
to happen. He says the department is often
approached by people claiming to be psychics.Usually it tells them to go away; this is a place of
science, not superstition. But someone has come to
their attention recently who seems worth a second
look, and we’re going to see him in action. But first
Kuhn would like us to complete a questionnaire. He
passes a four-page handout down the aisles and asks
us to complete page one – rating on a scale of zero
to seven how much we believe in “supernatural”
phenomena such as reincarnation, poltergeists,
communicating with the dead and so on. Once we
have filled it in, he asks us to turn to page two.
Here we are presented with a short paragraph
about what we are about to see. Kuhn asks us
to paraphrase this information in the space
underneath, to ensure that we have understood it
properly. Half of the audience, he tells me later, is
informed that the man we are about to see claims to
have psychic powers. The other half is told that he is
a skilled magician and that everything we’re going
to witness is an illusion. Then the show begins.
The lights dim and a young man takes us through
a wonderfully entertaining psychic routine. He
correctly guesses, to oohs and aahs from the
crowd, how hidden dice thrown by members of the
audience have landed. He tells random people in the
theatre surprisingly intimate details about their
lives. As a finale, he brings someone up on stage and
relays a message from her deceased grandfather.
She reels in astonishment as the audience gasp and
applaud. The show is over, our psychic takes a bow,
and pretty much everyone is wide-eyed.
But not with disbelief. Now we have to fill in the
same questionnaire – reincarnation, poltergeists,
communicating with the dead and so on – as before.
And the results are surprising. Again and again,
when Kuhn stages this experiment, it does not
matter what the audience members have been told
in advance – afterwards, people tend to believe
more strongly in the supernatural than they did
before seeing the show. And remember, these are
prospective psychology students.
“We have shown,” Kuhn says, “that you can
change someone’s belief through a really strong
demonstration of something that people previously
thought impossible or very unlikely. You can move
them down the continuum quite a long way.”
The psychic is introduced as Matt Tompkins,
a semi-professional magician and researcher in
experimental psychology at Oxford; the dice are
revealed to be rigged; the questions are general
enough for Matt to work out something from his
target’s answers (a mentalist trick called “cold
reading”); and the woman who was stunned to hear
from her grandfather is revealed as a stooge.
Kuhn is troubled by results like this. “Magic is all
about giving people false explanations for what you
have done physically,” he says. “Mentalism does
this a lot, pretending to read people’s minds, when
obviously that’s impossible. But our research showsHUGO CAFFARATTI has shown the brain is slower to process the “impossible”