4 Oct 2013

Seeing is believing? An optical illusion that challenges decades-old assumptions

Our brains are wired to make things up. To make sense of the physical world around us, the brain takes bits of information received from the senses and, like an artist painting a landscape, creates a unique mental picture shaped by its experiences. Without this ability to process sensory information (called perception) we wouldn’t be able to see in three dimensions, understand someone speaking in a noisy room, or even watch a film at the cinema. But there is a caveat: the brain can sometimes make mistakes, and optical illusions are one example.

Optical illusions are not only entertaining but they can also help scientists learn more about how our brains work. Researchers from the University Paris Descartes, in France, have now discovered an optical illusion that challenges decades-old assumptions about how the brain perceives movement.

You see it, but it isn’t there
Long before Walt Disney brought cartoons to the wider public in the late 1920s, motion pictures and animations were being made by quickly running a series of images in some sort of projector device. This rapid display of still images creates an illusion of movement, or as it’s technically called, ‘apparent motion’.
 
"A couple walzing" by Eadweard Muybridge (1830-1904) projected with a phenakistoscope.

Several decades of research on how people perceive apparent motion have established a few solid principles on the way our brains process movement. However, an optical illusion discovered serendipitously as a bug in a computer program recently challenged a couple of these principles.

“The minimal-motion principle has extensive support not only in psychology but also in neurophysiology, and underlies nearly every computer motion-detection and motion-perception algorithm. What we have found is a blatant violation of this principle,” says Mark Wexler, an experimental psychologist at the University Paris Descartes who first reported the so-called ‘high-phi’ illusion a couple of years ago.

In this strange illusion, when a moving scene on a screen is interrupted by a random image, the observer sees an illusory fast backwards ‘jump’, as though the image has a hiccup (you can see the illusion here). Somehow, although our eyes detect the still random image, our brain turns that visual cue into fast motion (high-phi jump). But how does the high-phi illusion challenge the minimal-motion principle?

This principle states that when we look at many different motions simultaneously (which happens pretty much all the time), our conflicted brains will ‘choose’ to see the slowest one. We can spot this effect in the barber pole illusion- the pole’s movement is horizontal and quite fast, but what we see is a vertical slow movement. In the high-phi illusion, the brain always perceives the fast jump, even though it has the choice of perceiving slow motion, or even no motion at all.

The best way to understand what’s going on is to watch the illusion in slow motion (by pressing the button on the right), which reveals what we should really be seeing.
 
High-phi illusion. The fast jumps are an illusion, in slow motion you see
the real image (Credit: Mark Wexler et al. PNAS 2013)

Breaking the limit
To better understand the high-phi illusion, Wexler and colleagues asked volunteers to watch blob patterns rotating on a computer screen and then measure the size of the high-phi jumps they saw by using a visual probe. The results of these experiments were published in March in the journal Proceedings of the National Academy of Sciences.

The team found that high-phi jumps could be triggered using different tricks, for example, by reverting the contrast of the blob pattern, or by rotating it in large steps, rather than continuously. And this is where the high-phi illusion breaks yet another paradigm.

The theory predicted that the observers wouldn’t be able to detect any movement when the patterns were shifted in large steps, above a certain cut off distance- the upper displacement limit or dmax. However, this is not what Wexler’s team found. “Below dmax, the steps should be seen as what they are, more or less. Above dmax, on the other hand, you're supposed to not perceive motion, just noise. This is not what happens, though: you perceive the high-phi jump,” Wexler explains. Strangely enough, the high-phi jumps have a maximum size that is “very closely correlated with the dmax limit: people who have higher dmax limits, also see a larger high-phi jump.”

John Perrone, an experimental psychologist from The University of Waikato, in New Zealand, who was not involved in the study says “The study describes a really interesting motion phenomenon that helps constrain theories and models of motion processing in humans. […] The authors have spent a lot of time also carefully testing the various parameters that influence the [high-phi] effect. These results will be useful to theoreticians and modellers and will help provide clues as to what mechanisms underlie the effect.”

So what are the neural mechanisms behind this optical illusion?

“Good question: if we only knew! We can only describe what’s happening: the brain seems to have a default of very fast motion. I wish that a physiologist would find a neural correlate to this effect,” Wexler says.

Reference:
Wexler M., Glennerster A., Cavanagh P., Ito H. & Seno T. (2013). Default perception of high-speed motion, Proceedings of the National Academy of Sciences, 110 (17) 7080-7085. DOI:

This article was published in Lab Times on 4-10-2013. You can read it here.