20 March 2000

Yes, We Do Live In The Past

by Kate Melville

If you think you're living in the past, you're right -- and science can tell you just how far behind the times you are. According to a new Salk study, it's at least 80 milliseconds, just slightly longer than the blink of an eye.

"What you think you're seeing at any given moment is actually influenced by the future," said David Eagleman, lead author of a study in the current issue of Science. "This doesn't mean the brain is clairvoyant, however."

He compared the timing of conscious perception to the broadcasting of a live television show, "which is actually not live. The show is delayed by about three seconds, so it can be edited if something happens. The brain does the same thing."

Using a visual illusion known as the flash-lag phenomenon, Eagleman and Salk Professor Terrence Sejnowski showed that the human brain appears to construct conscious awareness in an after-the-fact fashion, which they term postdiction. Their findings counter a leading hypothesis that visual awareness is predictive, extrapolating ahead of perceived events.

"In fact," said Sejnowski, "it looks like the conscious mind is just catching up on past information."

The flash-lag phenomenon was initially noted in 1958 and, more recently, recognized as a potential tool to probe puzzles of visual perception. Imagine you're watching a moving ring or hoop, and a light flashes in the center of the ring.

"Although the flash is physically in the center of the ring," said Eagleman, "it is perceived to lag behind the ring. You can sometimes see this if you look at an airplane at night -- the blinking lights may appear to be lagging behind the plane."

One popular hypothesis held that this was because our brains assume the ring will continue in its path of motion and extrapolate its position forward.

That was a reasonable theory, according to Eagleman. "Let's suppose there's a moving object in the world, and by the time the light from that object hits my retina and gets processed in my brain, the object has already moved on. So if you want to see things where they really are, maybe the visual system needs to extrapolate and guess where things will be in the future. It was intuitively very appealing. But I had reasons to doubt it was true."

To test this theory, Eagleman and Sejnowski devised a set of simple experiments. Instead of continuing the ring's movement through space, at the instant of the flash they stopped or reversed its motion.

"If the predictive hypothesis is correct," said Eagleman, "one would expect the same result in each case -- that is, the flash should appear to trail behind because your brain is assuming the ring will continue in its path."

Instead, they found that perception of the flash was dependent on where the ring moved after the flash. If the ring stopped, subjects reported the flash to be dead center. If it reversed, the flash lagged in the other direction.

"That's a wacky result," said Eagleman. "It means that your brain collects information into the future of an event before it commits to what it thinks it saw at the time of the event."

The researchers followed these experiments with a set in which the ring was at a complete standstill at the time of the flash; immediately afterward it moved in one direction or another. They obtained the same results -- the illusory displacement depended on where the ring moved after the flash occurred.

"The flash always appears to trail the movement, when in reality it occupies the center of the ring," said Eagleman.

And how long does the brain have to polish the past?

"If I were to show a subject a flash and move the ring two weeks later, there would be no effect on perception," said Eagleman. "So I asked: how long could I delay movement after the flash and still get the flash-lag effect?"

That window turned out to be 80 milliseconds -- a trice by our conscious clocks -- but long enough to be clearly measured in the laboratory. Eagleman pointed out that this is an averaged number: "I don't know, perhaps fighter pilots live less in the past than the rest of us."

Sejnowski added: "Now that we know our brain is stealing time from our visual awareness, we can begin to ask why. More surprises may be in store as we look for this missing time gap in the brain itself."