The Grand Illusion: How and Why Magic Works

I can still remember the amazement I felt the first time an adult produced a coin from behind my ear. While this is not exactly an advanced magic trick, it is still baffling and entertaining for a five-year-old. Ever since then, no matter how hard I’ve tried to concentrate on watching a magician’s hands, they have always managed to surprise me. I have often wondered how magicians make objects disappear and reappear right in front of our eyes. Magicians have two golden rules: never perform the same trick twice and never reveal how you accomplished a trick. If a British magician discloses any of magic’s deepest secrets, then they are thrown out of the professional guild of magicians, the ‘Magic Circle.’  This is the real-world equivalent of ‘The Ministry of Magic’ in Harry Potter (Rowling, 1997). I’ve always had a burning desire to know what secrets the Magic Circle keeps. Now magicians such as Tell & Peller, who are not members of the Magic Circle, are sharing these secrets with us on television and in the laboratory (Macknik et al., 2008). It seems that my more recent interest in psychology is a natural progression of my childhood fascination with magic. Magicians realized that human perception is highly subjective and manipulatable. It turns out that magic is an entertaining application of cognitive psychology.

Before studying psychology, I never had reason to suspect that what I see is not a direct image of reality, but a world of my creation. Ever since making a pinhole camera in secondary school art class, I’d assumed that our visual system worked via similar principles. The image is focused as light travels through a hole and is projected onto a screen at the back, where it is captured for viewing. For my camera, the image was imprinted onto photographic paper; for my eyes, the conscious mind. 

It is difficult to get away from the feeling that my eyes are windows to the world. It leads to the sense that my consciousness lies somewhere just behind my eyes. However, from biology, we know that the image first falls onto the retina, and from neuroscience that the signals from the retina travel to a region at the back of the brain for visual processing, known as V1 (Hubel & Wilson, 1959 as cited in Wurtz, 2009). This can easily lead to the false belief that the retina is like the charged-coupled-device (CCD) sensor on a digital video camera, with the image projected onto a movie screen on the back of the skull. However, the implausibility of this becomes apparent when you consider the anatomy of the retina itself.

The retina contains two types of photoreceptors: rods and cones. The cone receptors are the ones required for perceiving colour and fine detail, while the rod receptors enable you to see in the dark and are better at detecting motion. The cone receptors are concentrated around a small area of the retina known as the fovea. It is only the portion of the image that hits the foveal area that is seen in colour and sharp focus. At two square millimeters, the foveal area covers one five-hundredth of the retina. The fovea contains a high density of cones, and outside of the fovea, the ratio of cones to rods pretty much drops to zero. To say that the retina contains far more rods than cones is a major understatement.  While there are over one hundred billion rods on the retina, there are only six to seven million cones (Kolb, 2007). Given that cone photoreceptors are required for colour vision, an alien scientist trying to deduce whether humans have colour vision would find strong support for the hypothesis that we only see in black and white, obtaining a p-value of around 0.00005.

Given we experience a detailed world in full colour, what appears in our mind’s eye is very different from what is captured by the retina. There are simply no CCD sensors or movie screen equivalents in our head. The richness of our visual experience, given the poverty of input, seems quite miraculous. Our sense of perfect vision is clearly a trick of the mind, perhaps the grandest illusion of them all (Noe, Pessoa, & Thompson, 2000). Not only does the mind add colour and detail to blurry and black and white images, but it also creatively covers up any unseen parts of the visual field. The process by which it does this is known as ‘filling-in’.

Evidence for filling-in comes from analysis of what we see at the location on the retina where the optic nerves attach. There are no photoreceptors at this location, which is known as the optical blindspot. It is an area thirty percent larger than the fovea and creates blindspots in our peripheral vision at an angle of six degrees towards the centre (Kolb, 2007). When both eyes are open, these spots are covered by the reciprocal eye. However, if what you see is a direct reflection of what is on your retina, then what you should see with one eye closed would contain a patch of ‘blank canvas’. However, we actually see a complete image, and the brain appears to fill in the gap with a continuation of the background. This was first noted by Brewster (1832) in his book, Letters on Natural Magic (Ramachandran & Gregory, 1991).

If you are not familiar with your blindspot, there are many methods for finding it described online (Chulder, n.d.). When I locate my blindspot, it is large enough to cover half of my index finger at arms-length. The underlying process used by the brain when filling in was the subject of an interesting experiment by Ramachandran and Gregory (1991). They asked subjects to fixate on the centre of a display of flickering ‘snow’ on a screen. A small grey square was then displayed at the location of the blindspot. At first participants reported the square was visible, but after about five seconds, it became filled with snow to match the rest of the screen. Once participants could see snow at this location, they then switched the entire screen to the grey. Participants continued to see flickering snow at the location of the blindspot for two to three seconds after the screen change. The experiment revealed how long it takes to update the blind spot if our attention is not drawn to it. The results suggest that the unattended areas of the visual field are updated slowly, if at all. This is why if we are concentrating on a particular part of the visual field, we can miss significant events happening in the background, a phenomenon known as ‘inattentional blindness’ (Mack & Rock, 1998 as cited by Jensen, Yao, Street, & Simons, 2011).

Simon’s & Chabris (1999) investigated inattentional blindness by asking participants to count the number of passes between basketball players dressed in white while ignoring the players wearing black. If you are unfamiliar with this experiment, then grab your phone and search for ‘inattentional blindness’ on YouTube before you continue reading. While busy counting the passes, most people fail to notice the woman dressed in a gorilla suit walking through the middle of the action. Less than half of us notice the gorilla because of the effect of selectively focusing your attention on elements that are white and ignoring those in black. It seems that the automatic alerting system for those who do not spot the gorilla does not judge its appearance sufficiently important to divert attention from keeping count of the passes.  However, were the person in the gorilla suit to break into a sprint, then their automatic alerting network would become highly aroused, and the gorilla would be difficult to miss.

Further investigations of inattentional blindness have shown that the probability of you noticing a change in your visual field is a function of how motivated you are about your current task and how surprising the distraction is (Jenson et al., 2011). This means that what we visually experience depends on our ongoing motivation, the saliency of the distraction and our ability to inhibit the urge to look. Due to individual differences, everyone will see the same scene slightly differently. Therefore, the contents of our visual experience are highly subjective and personalized.

Magicians have used their knowledge of inattentional blindness to entertain us for generations (Magnik et al., 2008). The audience’s shared motivation to work out how the trick works makes manipulating their attention easier. Your desire to beat the magician is often part of the mechanism behind the trick. Magicians manipulate attention using a range of techniques from obvious distractions such as loud noises and flying birds to more subtle methods such as those that exploit our natural biases. For example, magicians discovered that if we see two moving objects, then we always track the larger (Magnik et al., 2008). As well as knowing how to attract attention, magicians also know how to avoid it. This skill is needed to execute a trick undetected once your attention is drawn elsewhere. For example, when doing something that seems innocuous, like adjusting their glasses, they may be performing a sleight-of-hand manoeuvre (Magnik et al., 2008). The final way magicians manipulate attention is through the use of humour. Laughter causes inattentional blindness because it breaks our concentration.

Magicians also take advantage of another limitation of our visual system: our inability to detect substantial changes to things in the background. This allows them to add, remove, or replace items which we only subsequently become aware of when they explicitly draw our attention to them. While inattentional blindness feels somewhat intuitive, our inability to detect substantial changes in our environment can be astonishing. Known is psychology as ‘Change Blindness’; it is well demonstrated in The Colour Changing Card Trick on YouTube (Wiseman, n.d). William James first noted a significant deficit in our ability to notice changes in the late nineteenth century (James, 1890 as cited in Jensen et.al, 2011). However, further investigation into the dimensions of change blindness was not possible until psychologists started using computers to manipulate what was displayed during moments of inattention.

The first researchers to observe change blindness in the laboratory were investigating the cognitive processes underlying reading. In one study, subjects were asked to read text on a computer screen and then, during their saccades (fixational eye movements), parts of the surrounding text were altered (Rayner & Kaiser 1975 as cited in Jensen et al., 2011). While observers saw a screen of rapidly changing words, participants did not notice anything and continued to read undistracted.

Further research found that it was not the saccades per se that caused change blindness but the limits of short-term visual memory (Rensink, O’Regan & Clarke, 1996 as cited in Jensen et al,. 2011). Research showed that any distraction lasting more than one-tenth of a second could prevent the detection of quite significant background changes such as windows and even entire buildings disappearing. While the failure to observe such changes is surprising, it would have served no adaptive benefit from an evolutionary point of view. In the real-world substantial changes do not occur during the blink of an eye. If during our evolutionary history, something was so slow, or small, that we were unable to resolve any movement in 100 milliseconds, then it was not worth worrying about. That is unless it was poisonous, in which case our attention system is fully primed to make sure we don’t miss it.

Change and inattentional blindness only occur for objects in our environment that we subconsciously decide are not important. This is the reason why magicians never repeat tricks. Magic is about surprising the audience, and therein lies the entertainment. If the audience knows where to look, then it makes the trick easier to work out. It seems we have an answer to what the secrets of the magic circle are but what is it that we see in our mind’s eye?

The small size of the foveal region and optical blindspots mean that we see a lot less than we think we do. Our visual experience is the result of the interaction between two streams of visual processing. Information from our rod receptors is processed rapidly and feeds our attention system for the detection of salient objects and motion. Our attention then directs our slower, cone-based processing system that generates a rich and detailed picture for our conscious consumption. What we see is not a precise picture of what is in front of us. It is our mind’s best guess of what is there, based on the information it has available. Despite how little information it seems to have, the fact that photographs always match what we have seen (minus the odd photo-bomber) is evidence that it almost always guesses right.

 Thanks to the mind’s grand illusion, the world we see is not black and white or full of holes but rich, colourful and detailed. Evolution has managed to optimize our visual processing to give us the best possible picture using the least possible cognitive resources. After all, it is no good seeing a lion in magnificent detail if it delays you from sprinting off in the opposite direction. From a wellbeing perspective, it is important we can trust our eyes given that vision dominates conscious perception. However, bear in mind that the images on your mental movie screen are mind-generated (MGI). This means that magicians and cognitive psychology students can play the director.