How emotions cloud our sense of time

Humans are pretty good at estimating durations of time. For instance, on hearing a tone with a certain duration, say half a second, people will give an estimate close to this. If I were to ask you how long it was since you logged on, you would have a fairly good idea. What enables us to be so good at estimating duration? By exploring situations where our ability to estimate the duration of time is distorted, we may begin to understand which mechanisms underlie our sense of time.

Consider high-adrenaline situations, such as a car crash. People often report that they could see the other car coming towards them 'as if in slow motion'. Noyes and Kletti (1977) investigated depersonalisation in response to life-threatening danger using personal interviews. One of their participants recalled an accident which happened a few years previously. The accident happened when he had been driving at 60mph in an old car when the steering became ineffective. He states:
'…My mind speeded up. Time seemed drawn out. It seemed like five minutes before the car came to a stop when, in reality, it was only a matter of a few seconds'. (p.376)

When people are enjoying themselves, the reverse experience of time is found, with time going really fast. However, Stetson et al. (2007) argue that the perception of time speeding or slowing is due to the 'function of recollection', not perception (see also Zakay, this issue). My focus is the interaction between increased arousal and accompanying psychological mood, which may have a combined effect on judgments of durations.

Research has supported the idea that emotional states systematically induce error in time estimation. For example, Angrilli et al. (1997) showed participants scenes that were high-intensity-pleasant (e.g. kissing couple), high-intensity-unpleasant (e.g. baby with eye tumour), low-intensity-pleasant (e.g. smiling baby) and low intensity-unpleasant (e.g. large spider) and asked them to estimate how long they lasted. Participants shown the low-intensity scenes tended to underestimate the duration of unpleasant scenes more than the pleasant ones. Interestingly, for the high-intensity scenes, there was an overestimation of the duration of unpleasant scenes compared to pleasant scenes. Similarly, Droit-Volet et al. (2004) found that the duration of pictures of emotional faces (expressing anger, happiness and sadness) was overestimated compared to neutral-baseline facial expressions.

This effect of arousal on timing judgements is found even in children. Gil et al. (2007) found that in children of three, five and eight years the perception of the duration of angry (compared to neutral) faces was overestimated – replicating findings in adults (Droit-Volet et al., 2004; Effron et al., 2006).

It is all very well showing that emotional visual stimuli can distort our perception of time, but what is the mechanism underlying such an effect? The leading contemporary theory of time perception is the information-processing model of scalar expectancy theory (SET) (Gibbon, 1977), which suggests that we have an internal pacemaker which produces the raw material for time representation. The pacemaker emits pulses and a switch controls how many pulses enter into an accumulator which stores the number of pulses during the event to be timed. With increasing stimulus duration, more pulses are accumulated and this increases the judgement of the duration. The information-processing model of SET also states that temporal judgements rely on memory and decision stages.

So how can this model explain the effects of emotional visual stimuli on time perception? There are two ways in which pacemaker-accumulator internal clock theories might explain how arousal impacts on the internal clock (Gibbon, 1977). The first is that arousal causes an increase in the speed of the pacemaker. The alternative explanation is that the effects could be due to changes in the functioning of the switch of the internal clock, the link between the pacemaker and the accumulator (Church, 1984). So, emotional stimuli may cause the switch to close faster or open later than usual. However, the switch is independent of the durations to be judged, so the effect on the switch as a result of emotional stimuli would not increase in proportion to duration of the stimulus. This is in conflict with findings from timing studies which have demonstrated that the effect of a clock speed change is in direct proportion to the stimulus duration.

Individually and combined, both these theories can plausibly explain how the number of pulses accumulated during the event to be judged can be increased as a result of arousal. However, the speed of the pacemaker is multiplicative with real time. This means that the effect of a stimuli would have more of an impact at longer durations compared to shorter when the speed of the pacemaker is increased. In other words, there is an actual manipulation of clock speed rather than a simple bias effect. If it was simply a bias effect caused by for instance, arousal, the increased speed of the pacemaker would be at a constant rate across all durations simply because of the presence of the arousing stimuli. Indeed this was supported by Gil et al. (2007), who showed that the overestimation of angry faces compared with neutral faces was greater for the long stimulus durations than for short ones. Since the switch latency theory would predict a time overestimation that is independent of the durations to be judged, it appears that this is not a feasible explanation for the effect of emotion-provoking stimuli.

Individual differences

Armed with the knowledge that emotional stimuli can distort our perception of time, one might naturally go on to suspect that there is a link between individual differences in negative emotionality and overestimation due to negative facial expressions. For example, people whose life experience has hardened them to situations of anger or conflict may not experience the same degree of stimulation from the negative facial expressions and therefore may not demonstrate the same level of overestimation of such stimulus.

Tipples (2008) investigated whether overestimation of angry and fearful expressions was greater in individuals who consistently reported high levels of negative emotional arousal. A previous study (Calder et al., 2001) had found that ratings of pleasantness and arousal are equal for fearful and angry facial expressions, so if an arousal-sensitive mechanism is responsible for the effects of emotional facial expressions on temporal estimations, similar degrees of overestimation for both angry and fearful expressions would be expected. To examine these issues, Tipples trained participants to discriminate between a 'short' (400ms) and 'long' (1600ms) durations and then gave them the emotional facial expression stimuli for durations selected from five intermediate values (600, 800, 1000, 1200 and 1400ms). They had to judge whether these intermediate durations were more similar to either the short or long standard durations. Tipples found that there was a relationship between individual differences in negative emotionality and increased levels of overestimation due to both angry and fearful expressions, but not happy expressions. So, the greater the level of self-reported negative emotionality the greater the overestimation.

However, Fox et al. (2001) found that when participants are anxious, they usually spend longer looking at angry and fearful expressions relative to neutral and happy expressions – which they argue is due to the 'delayed disengagement of attention' in individuals with anxiety. So, in a situation where an individual is distracted away from the passing of time (such as in a situation where they are enjoying themselves), they will 'lose' pulses, leading to a shorter judgement of duration. More research is needed to explore the effect of individual differences in negative emotionality on timing abilities to see whether the presentation of angry or fearful facial expressions leads to an overestimation of time.

It is important to point out here that the type of discrete emotion used is crucial in timing judgements. If the effects of facial expressions on time perception vary only as a function of the aroused state or facial expression of the sender, then Tipples (2008) should have found that fearful and angry expressions produced the same degree of overestimation. Yet he found that angry facial expressions produced a greater overestimation of time, relative to neutral facial expressions, compared to both fearful and happy expressions. And others have found that disgust, although rated as arousing, induced no timing distortions (Droit-Volet & Gil, 2009; Droit-Volet & Meck, 2007; Gil & Droit-Volet, 2009). Perhaps this is because the basic function of disgust is to avoid the ingestion of foods that may make us ill, so it is 'inner directed' as opposed to, say, a social signal of potential aggression such as anger.

Attention has recently turned to shame, with Gil and Droit-Volet (in press) finding that participants from the age of eight years, who recognise the facial expressions of shame, underestimate their presentation time compared to that of neutral faces. However, no time distortion was found in the children who did not recognise the ashamed faces or in those younger children who did. This suggests that shame is a self-conscious emotion which develops as we get older to involve an attentional mechanism. Pulses are then lost when our attention is directed away from the passage of time, in this case towards ourselves and what we are feeling, leading to an underestimation of time.

Food-elicited emotions

We have seen how emotional face stimuli can distort our timing abilities, but what about other kinds of emotion-provoking stimuli? Could it be that even food can produce an emotion which alters our perception of time?

Experiences of this nature prompted Gil et al. (2009) to investigate this very question. Based on pretest findings, three liked food pictures (cream cake, dark chocolate and French bread) and three disliked food pictures (blood sausage, beef sausage with vegetables and dried beef sausage) were chosen. The neutral stimulus comprised of a picture of a white oval with similar perceptual characteristics to the plate used in the food stimuli pictures. The presentation duration of food pictures was underestimated compared with the presentation duration of the neutral picture, and this underestimation was more marked for the disliked than for the liked food pictures. Gil et al. (2009) suggest that time underestimation in this instance, is due to an attentional-bias mechanism (i.e. Zakay & Block, 1996). If we consider these findings from an evolutionary perspective, it is not surprising that the disliked food pictures captured more attentional resources (as evidenced by greater underestimation) than the liked food pictures. Potentially, disliked food represents an increased danger to health and therefore is given more attention.

The role of embodiment 

Food also crops up in other fascinating studies on how even inanimate stimuli can get under our skin and influence rhythms and time. For example, Zhong and DeVoe (2010) revealed that even an unconscious exposure to fast-food symbols can increase participants' reading speed, despite being under no time pressure. Is it possible that we 'embody' arbitrary stimuli like fast food and fast-food symbols, which represent a culture of getting food fast and eating it quickly?

The idea that some form of embodiment results in the overestimation of the duration of emotional stimuli was supported by Effron et al. (2006). They asked people to estimate the duration of angry, happy and neutral faces by comparing them to two durations learned in a previous training phase. To inhibit the imitation of the faces and therefore possible effects of embodiment, individuals in the experimental group (but not the control group) were told to hold a pen in their mouth, thus limiting facial movement. When people were allowed to imitate faces, they tended to overestimate the duration of angry faces (and to a smaller degree, happy faces) compared to neutral faces. Also when imitation was inhibited (by inserting the pen in the mouth), estimates of the angry, happy and neutral faces were equal. Therefore, it seems that the embodiment of emotional expressions may have a role in our estimation of duration.

So the next question is how exactly does the perception of emotional expressions, such as fear or anger, result in the speeding up of the internal pacemaker in the observer? We also need to consider the possibility that the speeding up of the internal clock may not be the perception of anger per se, but the fear produced in the observer. However, numerous studies do suggest that when someone observes an emotion, their neural system activates in the same way as if they were producing the emotion themselves.

Conclusion

Why are we relatively inaccurate in making timing judgements during emotional experiences, despite having a complex internal clock mechanism? Numerous studies support the idea that emotional faces have an effect on humans' estimates of duration. For instance, highly arousing stimuli such as angry faces result in an overestimation of duration, and it is suggested that this is because of an increase in the speed of the internal clock. Various explanations suggested to explain the effect such stimuli may have on the internal clock processes include an arousal-sensitive mechanism, an attentional mechanism and the idea of the effect of the embodiment of emotions.

- Clare Allely is a Research Assistant at the University of Glasgow.

References

Angrilli, A., Cherubini, P., Pavese, A. & Manfredini, S. (1997). The influence of affective factors on time perception. Perception and Psychophysics, 59, 972–982.
Calder, A.J., Burton, A.M., Miller, P. et al. (2001). A principal component analysis of facial expressions. Vision Research, 41, 1179–1208.
Church, R.M. (1984). Properties of the internal clock. Annals of the New York Academy of Sciences, 423, Issue, Timing and time perception, 566–582.
Droit-Volet, S., Brunot, S. & Niedenthal, P.M. (2004). Perception of the duration of emotional events. Cognition and Emotion, 18, 849–858.
Droit-Volet, S. & Gil, S. (2009). The time–emotion paradox. Philosophical Transactions of The Royal Society B, 364, 1943–1953.
Droit-Volet, S. & Meck, W.H. (2007). How emotions colour our perception of time. Trends in Cognitive Sciences, 11, 504–513.
Effron, D.A., Niedenthal, P.M., Gil, S. & Droit-Volet, S. (2006). Embodied temporal perception of emotion. Emotion, 6, 1–9.
Fox, E., Russo, R., Bowles, R. & Dutton, K. (2001). Do threatening stimuli draw or hold visual attention in subclinical anxiety? Journal of Experimental Psychology: General, 130, 681–700.
Gibbon, J. (1977). Scalar expectancy theory and Weber's Law in animal timing. Psychological Review, 84, 279–335.
Gil, S. & Droit-Volet, S. (2009). How do emotional facial expression influence our perception of time? Attention, Representation, and Human Performance: Integration of Cognition, Emotion and Motivation. PsyPress.
Gil, S. & Droit-Volet, S. (in press). Time perception in response to ashamed faces in children and adults. Scandinavian Journal of Psychology.
Gil, S., Niedenthal, P.M. & Droit-Volet, S. (2007). Anger and time perception in children. Emotion, 7, 219–225.
Gil, S., Rousset, S., & Droit-Volet, S. (2009). How liked and disliked foods affect time perception. Emotion, 9, 457–463.
Noyes, R. Jr. & Kletti, R. (1977). Depersonalization in response to life-threatening danger. Comprehensive Psychiatry, 18, 375–384.
Stetson C., Fiesta M.P. & Eagleman, D.M. (2007). Does time really slow down during a frightening event? PLoS One, 2, 1–3.
Tipples, J. (2008). Negative emotionality influences the effects of emotion on time perception. Emotion, 8, 127–131.
Zakay, D., & Block, R.A. (1996). The role of attention in time estimation processes. In M.A. Pastor & J. Artieda (Eds.) Time, internal clocks and movement (pp.143–164). Amsterdam: Elsevier Science.
Zhong, C-B. & DeVoe, S.E. (2010). You are how you eat: Fast food and impatience. Psychological Science, 21, 619–622.