Time and information processing

One of the most persuasive pieces of evidence that humans possess an internal clock comes from the fact that the rate of this clock can be altered. The speed of the internal clock can be altered by changing body temperature (Wearden & Penton-Voak, 1995), or by pharmacological manipulation (Meck, 1983). The nature of the stimulus 'to-be-timed' can also alter the rate of this clock, for example, auditory stimuli versus visual stimuli (Wearden et al., 1998) or filled versus unfilled durations (Wearden, et al., 2007). Of more practical use in experiments has been the discovery that the internal clock can be sped up through the use of repetitive stimulation (Treisman et al. 1990; Penton-Voak et al., 1996), typically in the form of click trains. If a series of clicks are presented before a stimulus to be timed (usually a five-second click-train at 5Hz), then the duration of that stimulus will be overestimated relative to stimuli not preceded by clicks. This overestimation is not a simple bias, but a multiplicative effect, in other words, the degree of overestimation gets larger, as the duration of the stimulus increases.

Interesting as these effects are to the time perception aficionado, what bearing to they have on our everyday experience of time? One way of approaching an understanding of our perception of time is to examine situations in which our perception is radically altered or distorted. In doing so one finds that there are certain situations that people commonly report that time appeared to slow down. These are typically situations of high danger or fear. The classic example is that of a car crash where people often report that the event seemed to happen in slow motion, which would imply a greatly increased internal clock speed. Other examples are from military firefight engagements, where again people report that time seems to slow down, in fact this has entered movies and games as 'bullet-time'. Professional players of high-speed racket sport also report a similar phenomenon, which is termed as being 'in the zone', where they perceive their opponent and the ball moving in slow motion so that they have all the time they need to calculate their return shot. On a more mundane level everyone at some point has experienced time slowing to a crawl in situations of boredom or anxious waiting (the slow tick of the dentist's waiting room clock).

Despite the pervasiveness of these reports what is less clear is what objective reality lies behind phrases such as 'Time seemed to slow down'. There are several possibilities, which are not necessarily mutually exclusive. Firstly it could be a retrospective projection of memory. If one is involved in a dangerous situation such as a car crash, then it is likely to be an event that one plays over and over again in one's mind, and also an event that one that recounts numerous times to other people. It may be that in this repetition the memory of the event becomes embellished and stretched or telescoped in memory, leading to the pervasive feeling that the event occurred in slow motion. Another possibility is that the phrase refers to a purely temporal phenomenon, one's internal clock is sped up and one sees the world slowed down, but one is helplessly trapped in the frozen moment, unable to act in a quicker manner. The third and more intriguing possibility is that both the internal clock and the rate of information processing are sped up, similar to being 'in the zone', people perceive the world in slow motion, but they can act/react more quickly than normal. There would obviously be a huge evolutionary advantage to such speeding up in moments of peril, and this 'slow motion' effect may be a cognitive facet of the flight or fight response.

Investigating these phenomena in an experimental setting is obviously challenging, although there have been some attempts to do so. Most noticeably in recent times has been the well-publicised experiment of David Eagleman (Stetson net al., 2007; there is a very amusing experiment from the 1960s which pre-dates this which is well worth a read: Langer et al., 1961). Eagleman's participants were given a visual display strapped to their wrists that displayed a two-digit number followed by a mask. The display rate was set at a rate at which the number displayed couldn't be perceived. The participant then performed a free-fall into a cargo net, and was asked to view the visual display as they plummeted towards the ground. If the fear of the fall increased the rate of the internal clock and the rate of visual information processing then the prediction was that they may be able to discern the displayed number. Results were inconclusive, with no clear evidence for an increase in visual processing speed. Despite being a less than rigorous test of this idea, the Eagleman experiment was an interesting and intriguing attempt to investigate this phenomenon, and served as inspiration to a series of experiments at our Time Perception Lab at the University of Manchester, together with Professor Wearden from Keele University. We reasoned that instead of using a high-adrenaline situation to try and invoke the time-dilation effect, we had a much better technique at our disposal; the use of click-trains. We had used click trains extensively in dozens of previous experiments and they had proved to be a very reliable method of increasing internal clock speed. So we decided to use the same click-trains to increase information processing speed (Jones, et al., 2011).

In our first experiment we tested click-trains in a simple choice reaction time task. Participants were presented with four boxes on the computer screen, and on each trial a cross would appear in one of the boxes. The participants simply had to react as quickly as possible and push a button when the cross appeared. In order to manipulate the level of processing we had three choice conditions one-choice condition (press any button as soon as the cross appears), a two-choice condition (press one button if cross appears in either of the first two boxes, press another button if it appears in the third or fourth box), and a four-choice condition (press the button corresponding to the box in which the cross appears). We found that reaction times were indeed significantly quicker in the click train conditions compared to silence.  We next wanted to see whether the effect would occur in a task requiring a more sophisticated level of processing, and longer response latencies. We used the same paradigm but this time instead of the participant responding to the appearance of a cross, a maths sum with an answer was displayed. The task was to decide as quickly as possible whether the answer displayed was correct or incorrect. Again we found that reaction times were significantly quicker for the click conditions compared to the silence control.

For our next two experiments we took a slightly different approach, we wanted to see whether the cognitive-enhancing effect would generalise to a memory task, where the response/measurement was not response time. The rationale was that if we briefly presented a visual stimulus to the participant and asked them to recall information from it, they may be able to recall more information if the display was preceded by clicks. To this end we performed a replication of the full report condition of Sperling's 1960 iconic memory task (Sperling, 1960). A 3 x 3 matrix of letters was presented for either 300 or 500ms, and was preceded by a click train or silence. We found that for both durations participants were able to recall significantly more letters in the click condition compared to the silence condition. For our last experiment we performed a replication of another classic experiment, that of Loftus et al. (1985), participants were presented with a total of 80 images, each presented for 200ms, each presentation was again either preceded by five seconds of clicks or five seconds of silence. After presentation of all 80 images there was a recognition task, comprising one hundred and sixty images (50% of which had been previously presented). Participants recognized significantly more of the pictures that had been presented in conjunction with clicks.

Intriguing as these findings are, what conclusions can we draw? To begin with there is the thorny issue of cause and effect. Is the effect of click trains on internal clock speed and on information processing coincidental? Or does internal clock speed determine information processing speed? Perhaps the two are mediated by some third factor such as arousal or attention? This problem is not an easy one to resolve, but we have made some progress. To begin with we repeated all the experiments so far described with white-noise replacing the clicks, this had no cognitive enhancing effect at all (Jones et al., 2011); white noise also has no effect on internal clock speed. This suggests that there is something critical about the repetitive nature of the stimuli, and also indicated that the effect on information processing rate by the clicks is not just due to the presence of any pre-cue stimulation.

In respect of the effect on memory encoding we wanted to know whether the effect was anything to do with time perception, in other words were the participants actually perceiving the display as lasting for subjectively longer when preceded by clicks? The matrix used is not the kind of stimulus that has previously been used to show a speeding up of the internal clock so this was unknown. We therefore repeated the experiment, but this time we had two types of trial, one in which participant recalled as many letters as possible just as before (and we again found an enhancing effect of the clicks), and a second in which they had to estimate how long the matrix was displayed for (we manipulated the display time on each trial). We found that participants did indeed relatively overestimate the duration of the matrix in the click condition compared to the control condition.

All of these findings returns us to the question of how do clicks work? There are three main hypotheses in the timing literature although somewhat surprisingly the issue is rarely addressed. The first of these is 'arousal' although exactly why type of arousal is never fully specified. Clearly click-trains are not arousing in the everyday sense of the word, and without any independent measure this argument becomes tautological, i.e. Why do clicks speed up the internal clock? Because they are arousing. How do we know they are arousing? Because they speed up the internal clock. However, we do know that in some experiments the duration of arousing or emotional stimuli has been shown to be overestimated, although the strength of this effect and whether is multiplicative and not a simple bias is less clear. Recently in our lab we have attempted to see if we can find some independent measure of arousal occasioned by click trains and used a verbal estimation procedure with clicks versus silence. Although we obtained the usual behavioural effect of clicks on the internal clock, we found no evidence of physiological arousal on GSR, EMG or heart rate.

Another hypothesis in the literature has been that the clicks cause their effect because their frequency is in phase with that of the underlying pacemaker/oscillator component of the internal clock. Thus the clicks in some way 'strengthen' the beats of the internal clock. This hypothesis depends on the perhaps questionable assumption that all participants who show a speeding up of their internal clock by click trains must necessarily have exactly the same frequency (or harmonic) pacemaker. To test this idea we repeated the reaction time task this time using a variety of different click frequencies and found that the enhancing effect was insensitive to frequency, which seems to weaken support for this particular hypothesis. We also repeated verbal estimation with different click frequencies and again found no effect. However it should be noted other studies in the timing literature have found some evidence that the frequency of the clicks can be important in some circumstances (e.g. Burle & Bonnet, 1999).

Another possibility is that the click-trains cause their effect through modulation of brain rhythms or oscillatory activity. Links between these and cognitive performance have been discussed for many years (for a review of these ideas, see Burle et al., 2003). In particular the idea that the alpha rhythm (8–12Hz) plays a key function in information processing has a long history, with work by Surwillo (e.g. 1963) being well publicised and cited in the literature., Some striking results have been obtained, for example, Woodruff (1975) using biofeedback to increase or decrease alpha frequency compared to baseline, found that increases in frequency decreased reaction times, and decreases in frequency increased them. It may therefore be the case that click-trains have some effect on alpha rhythm (although the frequencies used are usually lower than the alpha frequency).

Although links between alpha rhythm and aspects of information processing are sometimes found (Callaway & Yaeger, 1960; Lansin, 1957), other studies (Boddy, 1981; Treisman, 1984) have failed to obtain relations between alpha frequencies and either information processing or timing. More recent work has again suggested evidence of a link between alpha rhythms, information processing and reaction time, although often by way of a complicated interaction of factors. For instance, Klimesch et al. (1996) found that participants with high alpha frequency showed fast reaction times (RT), whereas slow subjects had low alpha frequency for similar results.

One aspect of the experiments using clicks that may be important is that the clicks themselves are never a focus of the experiment from the participant point of view, they are simply passively experienced. There is some work that has examined the effect on a subsequently timed stimuli when the prestimulus does require processing. An example of this comes from a series of experiments by Wearden et al. (2010). Their participants were required to perform a rapid serial addition task and then immediately switch to estimating the duration of a subsequently presented tone; or in another of their experiments they had to produce a duration by terminating a tone with a button press. There were two main predictions, if the usual effect of click trains is caused by some sort of arousal, then one might expect the serial addition task to produce the same effect, i.e. speed up the internal clock resulting in a relative overestimation and underproduction of durations. Alternatively the non-timing task may exert some interference or cost to the timing task, perhaps by affecting attentional processes to the timing task, resulting in the opposite effect to speeding up the internal clock. The results were consistent with the later hypothesis. The detailed theoretical interpretation of their results is beyond the scope of this article, but overall it shows that a manipulation that makes processing worse also seems to make subjective time slower, (the opposite of the click-train effect), so the relationship between subjective time and processing can go in both directions not just one.

Very recent work has found that clicks can also affect judgements of other stimulus dimensions such as line length and number. Interestingly this effect is only present if the stimuli to be estimated are presented in a sequential-additive manner. This implies that the clicks may be affecting some kind of accumulator process (Droit-Volet, 2010).

In conclusion, the cognitive-enhancing effect that we have found with click trains is an interesting but modest effect, what seems to be of more important value is the fact that it suggests a fundamental link between our perception of time and the rate at which we can process information.

- Luke A. Jones is a Lecturer at the University of Manchester.

References

Boddy, J. (1971). The relationship of reaction time to brain wave period: A reevaluation. Electroencephalography and Clinical Neurophysiology, 30, 229-235.
Burle, B. & Bonnet, M. (1999). What's an internal clock for? From temporal information processing to the temporal processing of information. Behavioural Processes, 45, 59–72.
Burle, B., Macar, F., & Bonnet, M. (2003). Behavioral and electrophysiological oscillations in information processing: A tentative synthesis. In H. Helfrich (Ed.), Time and mind II. Information processing perspectives (pp. 209-232). Hildesheim, Germany: Hogrefe and Huber.
Callaway, E., & Yaeger, C. (1960). Relationships between reaction time and electroencephalographic alpha phase. Science, 132, 1765-1766.
Droit-Volet, S. (2010). Speeding up a master clock common to time, number and length? Behavioural Processes, 85, 126–134.
Jones, L.A., Allely, C. & Wearden, J.H. (2011). Click trains and the rate of
information processing; Does 'speeding up' subjective time make other psychological processes run faster? Quarterly Journal of Experimental Psychology, 64, 363–380.
Kilmesch, W., Doppelmayr, M., Schimke, H., & Pachinger, T. (1996). Alpha frequency, reaction time and the speed of processing information.     Journal of Clinical Neurophysiology, 13, 511-518.
Langer, J., Wapner, S. & Werner, H. (1961). The effect of danger upon the experience of time. American Journal of Psychology, 74, 94–97.
Lansin, R.W. (1957). Relation of brain and tremor rhythms to visual reaction time. Electroencephalography and Clinical Neurophysiology, 9, 497-504.
Loftus, G.R., Johnson, C.A. & Shimamura, A.P. (1985). How much is an icon
worth? Journal of Experimental Psychology: Human Performance and Perception, 11, 1–13.
Meck, W.H. (1983). Selective adjustment of the speed of internal clock and
memory processes. Journal of Experimental Psychology: Animal Behavior Processes, 9, 171–201.
Penton-Voak, I.S., Edwards, H., Percival, A. & Wearden, J.H. (1996). Speeding up an internal clock in humans? Effects of click trains on subjective duration. Journal of Experimental Psychology: Animal Behavior Processes, 22, 307-320
Sperling, G. (1960). The information available in brief visual presentations. Psychological Monographs, 74, 1–29.
Stetson, C., Fiesta, M.P. & Eagleman, D.M. (2007) Does time really slow down during a frightening event? PLoS ONE 2(12), e1295. doi: 10.1371/journal.pone.0001295
Surwillo, W. (1963). The relation of simple response time to brain wave frequency and the effects of age. Electroencephalography and Clinical Neurophysiology, 16, 510-514.
Treisman, M. (1984). Temporal rhythms and cerebral rhythms. Annals of the     New York Academy of Sciences, 423, 542-565.
Wearden, J.H., Edwards, H., Fakhri, M. & Percival, A. (1998). Why 'sounds
are judged longer than lights': Application of a model of the internal clock in humans. Quarterly Journal of Experimental Psychology, 51B, 97–120.
Wearden, J.H., Norton, R., Martin, S. & Montford-Bebb, O. (2007). Internal clock processes and the filled duration illusion. Journal of Experimental Psychology: Human Perception and Performance, 33, 716–729.
Wearden, J.H., O'Rourke, S.C., Matchwick, C. et al. (2010). Task switching and subjective duration. Quarterly Journal of Experimental Psychology, 63, 531–543.
Wearden, J.H. & Penton-Voak, I.S. (1995). Feeling the heat: Body temperature and the rate of subjective time, revisited. Quarterly Journal of Experimental Psychology, 48B, 129–141.
Woodruff, D. (1975). Relationships among EEG alpha frequency, reaction time, and age: A biofeedback study. Psychophysiology, 12, 673-681.