Discover, August 2008Link
Whenever I lose my watch, I take my sweet time to get a new one. I savor the freedom from my compulsion to carve my days into minute-size fragments. But my liberty has its limits. Even if I get rid of the clock strapped to my wrist, I cannot escape the one in my head. The human brain keeps time, from the flicker of milliseconds to the languorous unfurling of hours and days and years. It’s the product of hundreds of millions of years of evolution.
Keeping track of time is essential for perceiving what’s happening around us and responding to it. In order to tell where a voice is coming from, we time how long it takes for the sound to reach both ears. And when we respond to the voice by speaking ourselves, we need precise timing to make ourselves understood. Our muscles in the mouth, tongue, and throat must all twitch in carefully timed choreography. It’s just a brief pause that makes the difference between “Excuse me while I kiss the sky” and “Excuse me while I kiss this guy.”
Scientists are finding that telling time is also important to animals. At the University of Edinburgh, researchers built fake flowers with sugar inside to reveal how hummingbirds tell time. After hummingbirds drink nectar from real flowers, it takes time for the flowers to replenish their supply. The Scottish researchers refilled some of their fake flowers every 10 minutes and others every 20. Hummingbirds quickly learned just how long they had to wait before coming back to each kind. Scientists at the University of Georgia have discovered that rats do an excellent job of telling time too. They can be conditioned to wait two days after a meal to poke their noses into a trough and be rewarded with food.
For 40 years, psychologists thought that humans and animals kept time with a biological version of a stopwatch. Somewhere in the brain, a regular series of pulses was being generated. When the brain needed to time some event, a gate opened and the pulses moved into some kind of counting device.
One reason this clock model was so compelling: Psychologists could use it to explain how our perception of time changes. Think about how your feeling of time slows down as you see a car crash on the road ahead, how it speeds up when you’re wheeling around a dance floor in love. Psychologists argued that these experiences tweaked the pulse generator, speeding up the flow of pulses or slowing it down.
But the fact is that the biology of the brain just doesn’t work like the clocks we’re familiar with. Neurons can do a good job of producing a steady series of pulses. They don’t have what it takes to count pulses accurately for seconds or minutes or more. The mistakes we make in telling time also raise doubts about the clock models. If our brains really did work that way, we ought to do a better job of estimating long periods of time than short ones. Any individual pulse from the hypothetical clock would be a little bit slow or fast. Over a short time, the brain would accumulate just a few pulses, and so the error could be significant. The many pulses that pile up over long stretches of time should cancel their errors out. Unfortunately, that’s not the case. As we estimate longer stretches of time, the range of errors gets bigger as well.
These days, new kinds of experiments using everything from computer simulations to brain scans to genetically engineered mice are helping unlock the nature of mental time. And their results show that the brain does not use a single stopwatch. Instead, it has several ways to tell time, and none of them seems to work like a conventional clock.
Dean Buonomano, a neuroscientist at UCLA, argues that in order to perceive time in fractions of a second, our brains tell time as if they were observing ripples on a pond. Let’s say you are listening to a chirping bird. Two of its chirps are separated by a tenth of a second. The first chirp triggers a spike of voltage in some auditory neurons, which in turn causes some other neurons to fire as well. The signals reverberate among the neurons for about half a second, just as it takes time for the ripples from a rock thrown into a pond to disappear. When the second chirp comes, the neurons have not yet settled down. As a result, the second chirp creates a different pattern of signals. Buonomano argues that our brains can compare the second pattern to the first to tell how much time has passed. The brain needs no clock because time is encoded in the way neurons behave.
If Buonomano turns out to be right, he will have explained only our fastest time telling, because after half a second, the brain’s ripples dissipate. On the scale of seconds to hours, the brain must use some other strategy. Warren Meck of Duke University argues that the brain measures long stretches of time by producing pulses. But the brain does not then count the pulses in the way a clock does. Instead, Meck suspects, it does something more elegant. It listens to the pulses as if they were music.
Meck first began to develop his musical model when he discovered how to rob rats of their perception of time. He had only to destroy certain clumps of neurons deep inside the brain. Some of these neurons, known as medium spiny neurons, are unlike any other neurons in the brain. Each one is linked to as many as 30,000 other neurons. And those linked neurons can be found throughout the cortex, the outer rind of the brain that handles much of the brain’s most sophisticated information processing. Certain neurons come from regions that handle vision, others from areas that apply rules to what we perceive, and so on. By receiving so many signals from all over the brain, Meck believes, the medium spiny neurons give us a sense of time.
Imagine you are listening to a 10-second tone. At the beginning of the tone, neurons around your cortex reset themselves, so that they all begin to fire in sync. But some fire faster than others, and so at any moment some are active and some are quiet. From one moment to the next, a medium spiny neuron receives a unique pattern of signals from the neurons that link to it. The pattern changes like chords on a piano. When the 10 seconds are over, the medium spiny neuron can simply “listen” to the chord to tell how much time has passed.
Meck has found support for his model by recording the electrical activities of neurons and in other researchers’ studies on people with a skewed sense of time. Certain neurotransmitters, such as dopamine, control pulsing neurons. Drugs such as cocaine and methamphetamine alter the brain by flooding it with dopamine, and studies have shown that they also change the second-to-second perception of time. In one experiment at UCLA, reported in 2007, scientists rang a bell after 53 seconds of silence. Healthy people estimated on average that 67 seconds had passed. Stimulant addicts guessed 91 seconds. Other drugs have the opposite effect on dopamine and compress the subjective experience of time.
IN REAL TIME
Even in a healthy brain, time is elastic. Staring at an angry face for five seconds feels longer than staring at a neutral one. It may be no coincidence that the pulse-generating neurons are directly wired into regions of the brain that handle emotionally charged sights and sounds. And recent experiments by Amelia Hunt at Harvard University hint that we may actually backdate our mental time line every time we move our eyes.
Recently, Hunt had people stare straight ahead with a ticking clock off to one side. She asked people to move their eyes over to the clock and make a note of the time when they had done so. On average, they reported seeing the clock about four hundredths of a second before their eyes actually arrived there.
Moving time backward may actually serve us well, by letting us cope with an imperfect nervous system. Each of our retinas has a small patch of densely packed, light-sensitive cells called the fovea. In order to get a detailed picture of our surroundings, we have to jerk our eyes around several times a second so that the fovea can scan them. On its own, this stream of signals from our eyes would produce a jarring series of jump cuts. Our brains manufacture the illusion of a seamless flow of reality. In the course of that editing, we may need to fudge the time line--both in anticipation of an event and after the fact.
But the most radical reworking of time may come as we inscribe it in our memories. We recall not just what happened but when. We can recall how much time has passed since an event occurred by tapping into our memories. Injuries and surgeries that destroy a particular part of the brain can give some hints about how the brain records time in memory. French scientists in 2007 reported their study of a group of patients who had suffered damage to a region known as the left temporal lobe. The patients watched a documentary, and a familiar object appeared on the screen, then reappeared a few minutes later. The patients had to guess how much time had passed. On average, the patients thought an 8-minute period was roughly 13. (Normal subjects were off by only about a minute.)
These experiments are helping scientists zero in on the regions of the brain that store memories of time. Exactly how those regions record time is still mysterious. It’s one thing to listen in on the brain’s music, recognizing chords that mark the passage of five minutes. But how do the brain’s memory-related neurons then archive those five minutes so that they can be recalled later?
At Humboldt University of Berlin in Germany, scientists have been building a model of how memory may store time. When neurons produce a regular cycle of signals, some signals come a little sooner and some come a little later. The researchers propose that as neurons pass these signals along, they can add tiny advances, some bigger than others. With these tiny wobbles, the brain can compress memories of time from several seconds down to hundredths of a second--a small enough package to store for later retrieval.
As it stores time in memories, the brain may alter it in another way that is even more radical. It may record time so that our brains recall events in backward order. Scientists at MIT discovered reverse memories in an experiment on rats. They had rats run down a track and then stop to eat food at the end. When rats (and humans) become more familiar with a place, individual neurons start becoming active when the rats reach particular spots. The scientists identified “place cells” that fired when the rats moved to different spots along the track. When the rats stopped to eat, the scientists eavesdropped on their brains again. They heard the place neurons fire again--probably as the memories of the track were becoming stronger in the rat brain. But the place neurons at the end of the track fired first, and the ones at the beginning of the track fired last. It’s possible that we reverse time in our memories in order to focus our brains on goals (for the MIT rats, the goal was the food at the end of the track).
We are not free from time, in other words, but we are not its slaves. We stretch and twist it to serve our own needs. Time, in other words, is just a tool.
Copyright 2008 Carl Zimmer. All rights reserved