We saw at the beginning of this chapter that a surprising number of different parts of the brain are involved in time perception. Maybe this is because we need to deal with so many different time-frames. We shouldn’t expect the sensation of two clicks on a Savart wheel to be measured in the same way as Michel’s freezing nights in his black ice cave. The German psychologist Ernst Pöppel suggests two different mechanisms – one for short durations and one for long. Others suggest that there might be a whole series of clocks for different durations, clocks which could sometimes overlap. I like to imagine it like a newsroom where each clock is set to a different time zone, but with a clock for every duration. Yet if this were the case then why can the same time-period feel longer if we are listening to a sound than looking at a picture? Would we need a whole new set of clocks for each of our senses?
It is possible that different areas of the brain measure different time-frames using distinct mechanisms. We know from research on the emotions that the brain is not designed like a neat, ceramic phrenological head divided into different sections for different emotions, but that each emotion uses a different combination of brain systems. Could the same be true of timing? Could the brain use different combinations of areas to assess durations of different lengths?
Perhaps the whole notion of a single clock or a series of clocks is too complex. An alternative explanation focuses on concentration. Just as time flies when you’re absorbed in a book, the more complex a task you’re given in a lab, the shorter you estimate that time period to be. So if you are presented with a list of words and asked to spot words beginning with E as well as any words that represent animals, this requires two different skills and more concentration than simply looking for animal words alone. So the more that’s going on, the faster time seems to go by. The Attention Gate Model is an example of this kind of idea.29 The theory is that we have a pacemaker which emits an endless series of pulses in the brain, and a gate that allows our brain to count up every pulse that passes through, just like a shepherd counting sheep as they’re herded through a farm gate. If you’re feeling anxious the pulses speed up, and so more pulses pass through the gate within a given period causing you to believe that more time has passed than really did. In other words time felt as though it slowed down. If you are paying attention to time itself, if you’re in a queue for example or taking part in an experiment where you have been told in advance to estimate a period of time, this also causes more pulses to go through the gate, with the result that time apparently seems to pass more slowly. This theory would also help to explain why time passes slowly during periods of depression. While a person is introspecting (or meditating), their attention is turned inward, every pulse of time is noted and the hours appear to go more slowly.
This makes sense, but why should time go faster when you are busy? Perhaps the brain shares its resources between concentrating on the event at hand and timing it, so when you are distracted time flies by uncounted. This is the basis of what’s known as the resource allocation or time-sharing hypothesis. And with this theory it doesn’t matter what the clock is like – it could consist of a pacemaker or a line of hourglasses or measure the rate at which neurons fire in the brain – the crucial point is that its timing mechanism is disrupted if your attention is displaced. The moment you give people a second task to do the minutes go faster; a watched pot never seems to boil, but go and check your emails and it will be boiling over before you know it. With the Attention Gate Model the more absorbed you are in a task, the less attention you pay to time itself, the pulses slow down, fewer pass through the gate and you believe less time has passed than really has.30 The clever thing about this model is that it is flexible enough to allow for the influence of emotions, and by now it should be becoming clear that time perception and emotion are strongly linked.
HEADING FOR A PRECIPICE BLINDFOLDED
Jonas Langer, a psychologist at Clark University in Illinois, had an idea. He would build a platform on wheels, stand people on it, blindfold them and then ask them to manoeuvre themselves slowly towards the edge of the stairwell where there was a drop of several storeys. He wanted to know whether they would think time had gone faster when they wheeled towards or away from the edge. As it was the 1960s, when the standards of university ethics were lower, no one tried to stop him. If you look at the illustration below, the platform had handrails to hold onto at the side, but no safety barrier at the front. The volunteers could start and stop the platform using a button connected to a motor that propelled it forwards at a steady two miles per hour while Langer and his colleagues steered it from behind. They were required to drive themselves towards the edge of stairwell while blindfolded from two different starting points – the first ‘less dangerous’ point was 20 feet back from the precipice, and the second ‘very dangerous’ point was 15 feet from the precipice. The instructions were to press the button for what they, without counting, estimated to be five seconds. Considering the platform moved at two miles an hour, driving it forwards for the whole five seconds from a starting point of fifteen feet would bring you to within less than eight inches of the precipice. Remarkably eight men and eight women agreed to take part, even after seeing both the drop down into the stairwell and the blindfold. And there was to be no cheating by standing well back on the platform. Each volunteer was made to stand with the toes of their shoes level with the front edge of the platform.
Not surprisingly Langer found that when people were facing danger they pressed the button for a shorter time. He interpreted this as indicating that time slowed down for them due to their fear, so after just 3.6 seconds they thought that five seconds had passed.31 We know from the last chapter and from personal experience that fear does cause time to appear to decelerate, but with this experiment there is of course an alternative explanation. If you know you are moving towards a precipice blindfolded then it is probably sensible to err on the side of caution and to stop the platform a little early. If they had continued moving for six seconds instead of five, unless the experimenter was fast enough to reach the stop button, they would have driven off the edge and fallen down the stairwell.
Having said that, we know from the many laboratory studies conducted since, that emotions do alter time perception. Just as fear makes time go slowly, so do looking at pictures of mutilated bodies or listening to the sound of a woman sobbing.32 It seems that when faced with distressing images, your body and mind ready themselves for fight or flight, so the clock goes faster, more pulses accumulate and it feels as though time went slowly.
As we’ve seen, the passing of time is judged in two ways – prospectively, as it happens, and afterwards, retrospectively. When you judge time prospectively it is easy to see that, as I’ve been discussing, attention and emotion both play a part; but when you look retrospectively and try to guess how long an event took, it is a third factor that shapes your answer – memory. This difference between prospective and retrospective time estimation is very significant, and one that provides solutions to many of time’s mysteries. It gives rise to the phenomenon I’ve dubbed the ‘Holiday Paradox’. This is the common experience of a holiday appearing to go fast at the time, but when you look back afterwards it feels as though you were away for ages. I’ll be returning to this in more detail in Chapter Four.
It is clear that memory is involved in time perception, but there are still disagreements over whether we have a separate working memory just for processing time. Do we have a working memory buffer that allows us to hold temporal information briefly in mind in the same way we can retain a phone number for just long enough to dial it? It is possible that a pacemaker counts the milliseconds while complex memory processes allow us to deal with longer time intervals. Studies of patients with amnesia suggest that the processing of time also shares some neurological pathways in common with the creation or recall of certain types of memory. This link between memory and time perception is backed up by the fact that the tranquiliser Valium impairs both memory and time estimation
skills.
To sum up so far it seems that we have some kind of clock in the brain which counts time and is influenced by these three key factors: attention, emotion and memory. It could consist of a single clock, or a series of special pulses. There’s just one problem: no one can find it.
IS THE BRAIN TIMING ITSELF?
So is it possible that there are no clocks, nor any special pulses for timing? Perhaps the brain is exploiting the activity already going on within it by taking readings of time from the neural networks that are constantly making computations to assess everything from colour to pitch. According to this theory no part of the brain exists in order to actively and deliberately count time. There are no specialised mechanisms for time estimation. Instead time is inferred using the general features of brain circuits that are doing something else – processing spatial information, for example, or recognising a face. A number of neuroscientists are moving in the direction of this sort of idea, and what they want to discover is how the brain might do this. Neurons can produce a steady series of pulses that could be used for timing, but the brain appears to have no mechanism for counting them.
An alternative theory is that we use brain oscillations to time short events. Brain oscillations – the alpha waves of brain activity you can see on an EEG trace – are very short and could lend themselves to the role of the clock. This idea is backed up by the curious experience of going under general anaesthetic. We know that neurons stop oscillating when we’ve been put to sleep, and anyone who has had surgery will tell you that when they woke up it didn’t feel as though any time had passed. This is very different from normal sleep. If the mind does assess the passage of time using these oscillations this could explain why. There’s just one problem with this theory. These oscillations last 30 milliseconds, which would imply that the brain counts in 30 millisecond packets. Yet we are able to count durations which are not divisible by 30 milliseconds.
The French neuroscientist Virginie van Wassenhove believes that any set of neurons in the brain has the potential to help us work out timings; it’s just a question of turning our attention to it. So the activity is going on all the time, but it’s not until we ask the brain to time the difference in duration between two musical notes, for example, that we in effect read off the calculations that we need. It is a bit like estimating the number of people in a room – usually we ignore this information coming into our minds, but if someone asked us to do it, we could. So time is directly, if not always accurately, ‘transparent to consciousness’.33
In his lab in Los Angeles, the neurobiologist Dean Buonomano is using electrophysiological, computational, and psychophysical techniques to attempt to discover how the brain tells the time. On his website you can test your own skills at temporal processing at tiny durations.34 The site plays two pairs of sounds just milliseconds apart and you have to work out which pair had the smaller gap between them, a task not dissimilar to that undertaken by the musicians in Istanbul. Buonomano has an explanation for the finding I mentioned before: that people get better with practise, until they move on to a different duration where they find themselves back at square one. Their skills can generalise to other senses, but not to other time-frames. He argues that in order to succeed at this task, the brain treats the sounds like ripples created by a stone thrown into a pond. The ripples remain for a few moments after the stone has sunk, like a memory trace of what was happening just before. When a second stone is thrown in, the pattern of its ripples is influenced by the waves caused by the first, allowing the water to momentarily hold a record of both events. So in the brain, the first tone activates some neurons, leaving them in a new state, then along comes the second tone and, because the neurons are in a new state, the response is different. It’s as though the ripples left by the first tone provide a new context for the next tone. In the sound task, the brain is able to compare the patterns of activity caused by the first pair of sounds with the pattern from the second, before assessing which was shortest. So we don’t need a ticking clock to do the measurement because it is the patterns of activity in the brain themselves that do the timing. He calls this a state-dependent network. The tasks sound easy, but on my first attempt at his test I scored 23/30, not especially good considering that a score of 15 could be achieved through random guesswork alone. Luckily life doesn’t require us to perform this exact task, although millisecond timing is so crucial to the production and understanding of language that skills at judging this time-frame could contribute to linguistic ability. Next researchers are hoping to discover whether deficiencies in timing might underlie conditions such as dyslexia. This could explain the unusual relationship with time experienced by people like Eleanor, who is persistently late because she has no accurate sense of time passing. Is it possible that it is the precise timing of the movements of a pen on the page or of reading a series of letters that allows us to write and read accurately?35
Experiments using three sound tones lend weight to the idea that we don’t need a specialised central mechanism to estimate time, that instead we read it from the activity of the neurons that are there doing other jobs. Volunteers are asked to judge the interval between two tones while ignoring a third irrelevant tone that is played first. If the brain has its own stopwatch this shouldn’t be a problem; you just reset the stopwatch after tone one and time the interval between the second and third tones. But this isn’t what happens. The third tone does confuse people, suggesting that neural activity not specifically designed to measure time does the timing and that this timing is thrown off course by the introduction of an extra tone. This makes the system imperfect, but the advantage is its flexibility. In theory the system can time anything that’s happening, whichever sense it comes from. Crucially a third note is not distracting if the pitch is different. This makes me wonder whether different sets of neurons are used to do the timing for every different note played.
The same David Eagleman who throws people backwards off buildings has another idea which, like Buonomano’s theory, relies on the idea that our brain cells possess inherent properties of timing. When you look at a picture it takes a certain amount of energy for the neurons in your brain to identify what you are seeing. Think back to the giraffe/mango task, where people who are shown a series of pictures of giraffes with a surprise mango in the middle insist that the mango was present for longer than each giraffe. Registering the giraffe the first time uses a certain amount of energy. On seeing an identical giraffe picture the brain doesn’t need to waste as much energy considering it. Eagleman’s theory is that our sense of duration is based on the amount of neural energy used. So the first picture of the giraffe would take more energy and therefore seem longer, while subsequent giraffe pictures use less energy and seem shorter. Then up pops the mango. It’s new and requires more energy for the brain to register what it is, so it appears to be present for longer. In terms of the evidence for this idea, it is true that rates of neuronal firing do increase when there’s a new picture and drop off when the pictures repeat. Whether this is exactly how we calculate timings remains to be seen, but it is certainly a plausible idea. We know that novelty does play a part in timing, even at longer durations. If you arrive in a new city and walk from your hotel to a restaurant, the processing of all the new sights and sounds will use up a lot of neuronal energy, which will give the impression that the journey has taken a long time. The walk back, along a route now more familiar, will seem shorter.
The idea that neuronal activity itself might be used to measure time could also explain the difficulties that people with schizophrenia can have with time perception. Unlike voice-hearing or delusional thinking, this symptom tends to be less well-known, but some people find they are no longer able to view the present while simultaneously remembering the recent past and anticipating the future. The philosopher Edmund Husserl believed that holding these three time-frames in mind was essential for consciousness and for giving us a firm sense of reality. In schizophrenia this can be disrupted, making time
feel unreal. People with schizophrenia find it hard to spot the oddball in an experiment or even to detect flickering lights. Their neural responses suggest that everything they see seems fresh and new. Usually when you show someone the same giraffe again and again their neural response diminishes, but not with people with schizophrenia.36
We can predict the timing of everything from the swinging of a pendulum to a car door closing without trapping our fingers. We don’t even notice these small timing judgements that we make hundreds of times a day. But imagine how unsettling it would be if they started to appear to be out of kilter. Add to this the disturbance of your thoughts. If you have lost all cues to temporal reality and have no way of ordering your thoughts chronologically to identify which are memories, which are daydreams and which is the reality here and now, it is no surprise that a psychotic episode can feel terrifyingly disorientating. Philosopher and neuroscientist Dan Lloyd goes as far as to suggest that a disorder of temporal perception might even account for some of the symptoms clustered under the diagnosis of schizophrenia. This makes sense. I’ve already mentioned the influence of dopamine in time perception and one of the theories for the cause of schizophrenia, ‘the dopamine hypothesis’, implicates this neurotransmitter. It is possible that dopamine in effect sets the mind’s clock, dictating the rate of the pulses, and that some symptoms of schizophrenia could result from a disruption to the clock.
Eagleman’s theory can also explain the stopped clock illusion. The initial tick seems longer because it’s the first time your brain has registered the movement of the second hand, then the neuronal firing and the energy used drops off and so does the perception of time passing as the second hand continues on around the clock face. Likewise a bright light that’s turned on momentarily appears to last longer than one that is more faint and an interval filled with a complex piece of music seems longer than one with a simpler piece. Is this because we are timing them via the energy used to process them?
Time Warped Page 7
