Ever since Einstein, physicists have been telling us that time - this steady tick-tock of the universe - is much weirder than we think. It doesn't flow in a single, linear direction, or beat like a steady metronome. Instead, it depends on all sorts of peculiar cosmi variables. We speed up, time slows down. (Fall into a black hole and time turns into a viscous sludge.) And there's nothing in the mathematical laws of physics that says time can only go forward. In theory, at least, the hands of your clock can tick in both directions.

But if time is so strange, then why does it seem so normal? Why don't we feel all the quantum weirdness? Psychologists and neuroscientists are now beginning to explore the phenomenology of time, beginning not with spacetime but with the fleshy brain. If our' sense of time is largely a cognitive illusion, then where does the illusion come from?

Let's begin by looking at this audacious experiment, led by David Eagleman of Baylor College of Medicine. (Eagleman is also a best-selling novelist - Sum is a brilliant riff on the possibility of an afterlife.) He was interested in why time seems to slow down when we're really scared. (His research was inspired by a childhood fall.) Of course, it's not easy to terrify subjects in a science lab, or to trick people into thinking that they're about to die. (It might also violate a few IRB rules.) And so Eagleman came up with an original experimental paradigm: SCAD jumping, which is often described as bungee jumping without the bungee. A subject is hoisted 150 up in the air and then dropped (hopefully) onto a big net. Jad and Robert of WNYC's Radiolab explain what happened next:

SCAD diving was just what David needed - it was definitely terrifying. But he also needed a way to judge whether his subjects' brains really did go into turbo mode. So, he outfitted everybody with a small electronic device, called a perceptual chronometer, which is basically a clunky wristwatch. It flashes numbers just a little too fast to see. Under normal conditions - standing around on the ground, say - the numbers are just a blur. But David figured, if his subjects' brains were in turbo mode, they would be able to read the numbers.

The falling experience was, just as David had hoped, enough to freak out all of his subjects. "We asked everyone how scary it was, on a scale from 1 to 10," he reports, "and everyone said 10." And all of the subjects reported a slow-motion effect while falling: they consistently over-estimated the time it took to fall. The numbers on the perceptual chronometer? They remained an unreadable blur.

"Turns out, when you're falling you don't actually see in slow motion. It's not equivalent to the way a slow-motion camera would work," David says. "It's something more interesting than that."

According to David, it's all about memory, not turbo perception. "Normally, our memories are like sieves," he says. "We're not writing down most of what's passing through our system." Think about walking down a crowded street: You see a lot of faces, street signs, all kinds of stimuli. Most of this, though, never becomes a part of your memory. But if a car suddenly swerves and heads straight for you, your memory shifts gears. Now it's writing down everything - every cloud, every piece of dirt, every little fleeting thought, anything that might be useful.

This is a deeply Proustian idea. It turns out that our sense of time is deeply entangled with memory, and that when we remember more - when we are sensitive to every madeleine and sip of limeflower tea - we can stretch time out, like a blanket. This suggests that the simplest way to extend our life, squeezing more experience out of this mortal coil, is to be more attentive, more sensitive to the everyday details of the world. The same logic should also apply to our vacations. If we want our time off to last longer, then we should skip the beach naps and instead cram our days full of new things, which we will notice and memorize.

Furthermore, the link between the perception of time and the density of memory can also work in the other direction, so that it's possible to increase our memory by speeding up our internal clock. In 1999, a team of psychologists at the University of Manchester demonstrated that it was possible to tweak our "pacemaker" by exposing people to a sequence of click-trains, or acoustic tones that arrive in rapid progression. It turns out that such click trains accelerate our internal clock - it beats a little bit faster - which means that everything else seems to take just a little longer. (Perhaps this is why, when companies put us on hold, they always play sluggish muzak - the adagio sounds might slow down our clock, thus making the frustrating experience of waiting on the phone pass more quickly.)

A new study, by the same Manchester lab, uses click trains to explore the implications  of this accelerated tick-tock. It turns out that when our internal clock is ticking faster, we don't just perceive the external world as moving slower - we can actually remember more about it. In other words, our sense of time isn't just a perceptual illusion, but instead seems to regulate the pace of information processing in the brain. When it ticks faster, we can process more. It's like getting a faster set of microchips embedded in the cortex. (This suggests that we'd get better at executing very demanding tasks, such as making quarterback decisions in the pocket, or playing Rachmaninoff, if we listened to a series of fast clicks first.) Here's the abstract from the latest paper:

A series of experiments demonstrated that a 5-s train of clicks that have been shown in previous studies to increase the subjective duration of tones they precede (in a manner consistent with "speeding up" timing processes) could also have an effect on information-processing rate. In general, preceding trials by clicks made response times significantly shorter than those for trials without clicks, but white noise had no effects on response times. Experiments 3 and 4 investigated the effects of clicks on performance on memory tasks, using variants of two classic experiments of cognitive psychology: Sperling's (1960) iconic memory task and Loftus, Johnson, and Shimamura's (1985) iconic masking task. In both experiments participants were able to recall or recognize significantly more information from stimuli preceded by clicks than those preceded by silence.

And this returns us to the SCAD jumping paradigm used by David Eagleman. Perhaps the feeling of terror is like a series of clicks, speeding up our clock. We think of time as a constant, but there is nothing constant about it. Even a fleeting feeling can change the pace of everything.