July 7, 2016 at 7:38 AM
Many people think of their eyes as operating like a movie camera – capturing everything that happens in front of them and then projecting those images back into the dark room that is the brain.
But scientists say that metaphor isn't accurate. Our brains don't just watch a rolling film of everything in front of our eyes – that much information would quickly overwhelm us. Instead, neurons in the brain pick out specific signals that might be important, while other neurons take those signals and fill in the gaps.
That process mostly functions beautifully, but sometimes it can malfunction in interesting ways — as it does when we view optical illusions. Patterns of color and light that trick our eyes and brains into seeing things that aren't there, optical illusions have been valued as entertaining puzzles since at least the Victorian era. But today, scientists are also studying them for insight into how our brains work.
Gideon P. Caplovitz, a professor of psychology, has been using optical illusions in his research for more than a decade. Along with Matthew T. Harrison of the University of Nevada at Reno, Caplovitz created the optical illusion that won first prize this year in an annual contest sponsored by the Neural Correlate society (above is a gif from their illusion -- a full video is below).
Caplovitz spoke with me about how optical illusions work, and what kinds of fascinating clues they are offering new generation of neuroscientists about the brain. This interview has been edited for length and clarity.
So broadly, what do optical illusions tell us about the brain?
Illusions represent the mistakes the brain makes in interpreting what we're looking at, which arise because of the way our visual systems have evolved. And so by understanding these illusions, we can gain insight into how our visual system works.
We don't experience the world the way a movie camera operates. We experience the output of neurons in our brains that respond to light hitting our retinas. These neurons capture different types of visual information – like color, motion, the position of an object, or elements of its shape or size. And the visual system ultimately has to combine all of this information together in a best guess as to what we're looking at.
There's all sorts of stuff out there in the world that we don't see, like infrared radiation or radio waves. For example, bees have visual systems that are very different from ours, and they experience colors and shapes that we don't see.
We experience the world that we do because of the visual information our brains are able to represent. In general, what we experience is accurate enough to allow us to not fall into holes, or get run over by cars. But visual illusions highlight that sometimes what we see is incorrect.
Your illusion won first prize in the 2016 contest for Best Illusion of the Year. Explain to us exactly what is going on in the video below.
If you watch our illusion, you will see a rotating square or lines rocking back and forth, when in fact there is no rotating square or moving lines. What's happening is that these little gratings — we call them sine wave gratings or Gabor elements — they are rotating on themselves and creating what are called local motion signals that are detected by our visual system. Our neurons then integrate these local motion signals, and that integration can lead to the experience of a rotating square, or these waves, which actually aren't there.
I don't have a rotating square detector in my brain; what I have are neurons that detect local motion, and then other neurons that integrate them. And the output of that is what we experience. So the existence of the illusion is telling us something about the way the brain works.
Now, this all may seem silly, talking about rotating squares or rocking lines. But pretty much from the moment you open your eyes in the morning to the moment you close your eyes at night, unless you're visually impaired, you're visually experiencing the world around you, and it's moving, and you're moving. So this helps answer the question, why do things look the way they do?
Is there a real-world situation in which your brain would be detecting and integrating this kind of information?
The mechanisms that underlie this illusion pretty much underlie our experience of anything that's moving. Where it's particularly tangible would be the experience of seeing a car driving down the road, but from where you're standing, that car is either driving behind bushes or through trees, and you only get bits and pieces of it at any moment in time. But your brain is able to integrate these bits and pieces into a car as a visual whole, and figure out where it's moving and how fast.
There's another illusion that's not exactly the same, but similar, that we've all experienced, which are the red, white and blue stripes of a barbershop pole. Barbers have been exploiting this illusion for its advertising purposes forever. As the cylinder rotates, we have the experience of those stripes moving up, even though they're not.
I was going to ask you about that — are there other famous illusions that rely on similar principles?
There's a long history of them. Karen DeValois, a researcher at Berkeley, first reported this effect in the 1980s. Peter Tse, who was my graduate adviser, won the second-place prize in the illusion contest in 2006 for what he called the infinite regress illusion. Then in 2009, Arthur Shapiro won the first-place prize with his colleagues for an illusion demonstrating virtually the same effect, what he called "The Break of the Curve Ball." He published a paper suggesting that these neural mechanisms underlie the perceived "break of the curve ball" in baseball — the notion that a curve ball to a batter or a spectator appears to be traveling on a straight trajectory, and then at the last moment curves. In fact, it is a continuous curve from the moment it's released from the pitcher's hand.
And motion illusions are just one category of optical illusions, right?
Yes. In 2014, we won the illusion contest with a dynamic size illusion. It's related to a classic size illusion called the Ebbinghaus illusion. Hermann Ebbinghaus was a psychologist who found that if you take a small circle and surround it with big circles, it will look smaller than the same size circle surrounded by small circles. You may have seen this — it's one of those, if your uncle is going to send you some spam email with illusions in it, it often includes an Ebbinghaus illusion.
What my colleague and I discovered is that if you put this into motion, so that the surrounding circles are growing and sinking, the illusion gets really strong. So we call this the dynamic Ebbinghaus illusion, because everything is moving.
And there are some other illusions involving color afterimages?
Yes, there are a whole class of color illusions. There was a spectacular one that won first prize in the illusion contest in 2008, and a variant of it was presented in this year's contest. It was my favorite this year.
You're seeing colors that aren't there, and the way these colors manifest themselves is telling us something about the way the brain processes visual information.
What does it say about the way the brain processes information?
You'd have to ask the creators for the full explanation, but fundamentally, it's saying that the brain isn't processing visual information on a point-by-point basis. Even though the bubble does not have any color, it is dictating what color you see.
I would also like to mention there is a very long history of studying illusions. And probably the most profound illusion that we experience is at the movies. Motion pictures are a continuous sequence of still images that, when played at the correct speed, give rise to continuous motion. The movie experience that we have today is the culmination of almost 200 years of technological development investigating how to create moving images out of still frames.
There's a beautiful entry in the illusion contest this year called the "Silhouette Zoetrope." It's a a modification of one of the earliest picture devices, the zoetrope, which dates to pre-cinematic optical devices in the Victorian era. And I was very, very happy to see this new version of the zoetrope in the contest this year. It's another new trick for an old dog.
In a typical zoetrope, the inside of the cylinder contains a series of images, and as the cylinder revolves you see a short scene play out. But here, the birds are on the outside of the cylinder, and you're just catching a small glimpse of the bird through the cylinder, which creates the illusion of motion?
Right, you get a small slice of it at a time. You had people in the first half of the 19th century making devices like this, some of which were sold as toys to children. It's pretty neat to see them bring that back and add something new to it.
Is that what you what mean by "another new trick for an old dog"? These illusions have been around for a long time, but now they're revealing something very new about how our brains work?
Going up to the Gestalt psychologists in Germany in the 1920s and 1930s, you have scientists describing illusions for their phenomenological aspects, to explain the principles of how we see the world.
But then in the advent of the age of neuroscience, which starts post-1950s, the question arises of what illusions are telling us about the way the brain works. Now we can use illusions as one of the tools in neuroscience's toolbox — in fact, these are sometimes called "the psychologist's electrode," a non-invasive neuroimaging technique. That's in part the source of the resurgence of optical illusions, and one of the reasons why the illusion contest has been as successful as it has been. I think it speaks to the degree to which these illusions are valuable as a tool for neuroscience.
You might also like: