Why does life appear to speed up as people get older?

Abstract

In this study, the influence of contemporaneous and retrospective recall of time pressure on the experience of time was examined. Participants (N = 868) first indicated how fast the previous week, month, year, and 10 years had passed. No effects of age were found, except on the 10-year interval. The participants were subsequently asked how much time pressure they experienced presently and how much time pressure they had experienced 10 years ago. Participants who indicated that they were currently experiencing much time pressure reported that time was passing quickly on the shorter time intervals, whereas participants who indicated that they had been experiencing much time pressure 10 years ago reported that the previous 10 years had passed quickly. Cross-sectional comparisons of past and present time pressure suggested that participants systematically underestimated past time pressure. This memory bias offers an explanation of why life appears to speed up as people get older.

Keywords

Aging autobiographical memory experience of time time estimation time pressure

introduction

The phenomenon that life appears to speed up as people become older has been occupying philosophers and psychologists for at least 130 years (e.g. Guyau, 1890; James, 1890; Janet, 1877, cited in Fraisse, 1963). There are many studies which have shown that people believe or have the feeling that the passage of time is now faster than when they were younger (i.e. Baum et al., 1984; Gallant et al., 1991; Joubert, 1983, 1984, 1990; Lemlich, 1975; Tuckman, 1965; Walker, 1977).

These studies have a within-subjects design. Participants are generally asked how much faster or slower time seems to pass at the present time compared with when they were approximately one-half and one-quarter of their present age. For example, a participant who is currently 40 years old will be asked to indicate how much slower or faster the present time is passing compared to when he or she was 20 and 10 years old. In this example, the participant is thus required to make contemporaneous judgments about experiences that had taken place recently and retrospective judgments about experiences that took place 20 or 30 years ago. These studies show that about 70% of the participants feel that present time passes faster than past time, about 20% feel that present time passes as fast as past time, whereas about 10% feel that present time passes slower than past time. The majority of the participants in these studies were, however, young adults, suggesting that the experience that life speeds up as one gets older does not seem to be unique to middle-aged or older adults.

When cross-sectional (or between-subjects) comparisons about the subjective experience of time are made, age effects have only been found either when very short intervals, such as seconds and minutes, were examined (cf. Block et al., 1998; Craik and Hay, 1999; Lustig, 2003) or when people reflected on how fast the past 10 years had passed (i.e. Friedman and Janssen, 2010; Wittmann and Lehnhoff, 2005). No age effects have been found when different age groups rated the passage of the previous week, month, or year. Since people also expect that time will pass more quickly in the future, when they are twice their current age (Joubert, 1990), it is not clear whether the results of those earlier studies that required the retrospective recall of the subjective passage of time (i.e. Baum et al., 1984; Gallant et al., 1991; Joubert, 1983, 1984, 1990; Lemlich, 1975; Tuckman, 1965; Walker, 1977) reflect a genuine increase or merely a belief that there has been an increase in the subjective passage of time.

In the cross-sectional studies that used very short intervals (cf. Block et al., 1998; Craik and Hay, 1999; Lustig, 2003), young and older adults are usually asked to estimate or produce intervals in the range of seconds or minutes. In estimation tasks, participants are first provided with a stimulus that lasts a certain period and are then subsequently required to estimate the duration of this interval. In production tasks, participants are first provided with the description of an interval and after a stimulus is given that indicates the start of the interval, are then required to indicate when the interval has ended.

Because people’s internal timing mechanism or pacemaker, which, according to these studies, is used to estimate and produce intervals, is assumed to slow down as people become older, it was first expected that older adults would give shorter estimates and longer productions than young adults (Block et al., 1998). It was, however, found that when there are no secondary tasks, older adults actually tend to give longer estimates and shorter productions than young adults, which would suggest that time passes more slowly for older adults. When there are secondary tasks, the outcome is often reversed: older adults tend to give shorter estimates and longer productions than young adults, which would suggest that time passes more quickly for older adults. It was then hypothesized that the second set of experiments is a better reflection of real life, since older adults experience in daily life more cognitive load than young adults (Craik and Hay, 1999; Lustig, 2003).

Although many of these timing studies claim to explain why life speeds up as people become older (i.e. the slowing down of the internal timing mechanism), surprisingly, no study has actually shown that participants who give shorter estimates (or longer productions) also report that time is passing faster than participants who give longer estimates (or shorter productions). The studies simply assume that: (a) older adults experience time as passing more quickly than young adults; and (b) the performance on these estimation and production tasks is related to the subjective experience of time.

Results of cross-sectional studies that used longer intervals question these assumptions. One would expect that individuals or groups who tend to give shorter estimates or longer productions in a laboratory setting (i.e. older adults) would also give higher ratings for the subjective passage of longer time intervals, but Wittmann and Lehnhoff (2005) and Friedman and Janssen (2010) showed that when asked how fast the previous week, month, or year had passed, people from all ages actually report that time is passing fast and that only when people were asked how fast the previous 10 years had passed was there a small effect of age. Moreover, recent studies have shown that responses on timing tasks can easily be relearned with inaccurate feedback (Ryan and Fritz, 2007; Ryan and Robey, 2002), casting even more doubt on the validity of these internal pacemaker explanations.

Very little work is available that tests the many accounts of impressions of the speed of time (for short descriptions of these accounts, see Table 1). Friedman and Janssen (2010) examined three age-related explanations about time acceleration (i.e. a lower number of memorable events in older age, the slowing down of an internal pacemaker, and a decrease in the attentional resources of older adults) and three general explanations about why people of all ages might experience time passing quickly (i.e. forward telescoping, difficulty of recall, and time pressure) but found support for only one of them (i.e. time pressure).

Table 1.

Titles and short descriptions of the accounts for the subjective experience of time

Title explanation	Description explanation
Slowing down of an internal pacemaker	Objective time is compared to an internal biological clock or pacemaker and this clock or pacemaker slows down in older age, which causes, for older adults, objective time to appear to pass faster than subjective time.
Decrease in the attentional resources of older adults	When people know in advance that they are required to judge the duration of an interval, the perceived duration of the interval depends on how attention is allocated. When there is no or an undemanding non-temporal task during the interval, all or much attention is focused on the passage of time and the interval will appear to have passed slowly. When much attention is required by the non-temporal task, the interval will appear to have passed quickly. Since older adults have generally less attentional resources, objective time will appear to pass faster for older adults.
Difficulty of recall	People estimate the duration of an interval by the number of events that can be recalled from this interval. When a person recalls many events from a period, the period will appear to have passed slowly, whereas when a person recalls only a few events from a period, the period will appear to have passed quickly.
Lower number of memorable events in older age	People estimate the duration of an interval by the number of events that can be recalled from this interval, but it is assumed that people experience fewer memorable events as they become older, which causes objective time to appear to pass more quickly in older age.
Forward telescoping	People have the tendency to underestimate the age of events. They tend to date remote events as being more recent, which is called forward telescoping. When people learn that they have underestimated the age of an event, they may feel that objective time is passing faster than their subjective experience of time.
Time pressure	The way that people recall and reconstruct the past and the present affects their judgments about the subjective passage of time in these time periods. People might be able to recall from a recent time period that they were very busy and did not have enough time to complete things, but these events are likely to be forgotten from more remote time periods. The observation that one is currently experiencing more time pressure than one had experienced in the past will give people the experience that objective time is currently passing more quickly than in the past.

In their second experiment, Friedman and Janssen (2010) asked 1766 participants, who were between 16 and 80 years old, to indicate how fast time had passed for them on six different time intervals, ranging from the next hour to the previous 10 years. The overall scores showed a surprisingly small effect of age (explained variance = 1.4%), but the question that asked about how fast the previous 10 years had passed yielded a larger age effect (4.5%). The overall scores were, however, influenced by time pressure (10.0%). Consistent with the time-pressure account, participants who indicated that they were very busy and did not have enough time to complete things reported higher values on the speed-of-time items than participants who gave lower ratings for time pressure. The results also showed that time pressure correlated more strongly with the scores on the shorter time intervals than with the scores on the 10-year interval.

Memory bias

On the basis of these cross-sectional results, a time-pressure explanation for why life appears to speed up as people get older was proposed (Friedman and Janssen, 2010; Joubert, 1983). The phenomenon seems to be caused by a memory bias: the way that people recall and reconstruct the past and the present affects their retrospective judgments. One might be able to remember many instances from the previous year in which one was very busy and did not have enough time to complete things, but many of these mundane events are likely to be forgotten after 10 years. This makes people feel that they are currently experiencing more time pressure than they did in the past. This memory bias, which could be likened to the fading affect bias (Walker et al., 2003), could lead to an underestimation of past time pressure.

Joubert (1984) did not examine time pressure but a related variable: the amount of free time available. He first asked participants to compare the passage of time in the present to the passage of time when they were one-half and one-quarter of their present age. He then asked participants to compare the amount of free time available in the present to the amount of free time available when they were one-half and one-quarter of their present age. Participants reported that present time was passing faster than time in the past and that they currently had less free time than they had in the past. Furthermore, there were correlations between the passage of time and the free time comparisons. Participants who reported that they currently had less free time also reported that time was currently passing faster.

In Friedman and Janssen (2010), it was found that judgments of present time pressure are related to general impressions of the speed of present time. It seems fair to assume that judgments of past time pressure are similarly related to general impressions of the speed of past time. If people indeed underestimate time pressure at younger ages, then this would give the impression that life has sped up. To examine this explanation, the authors developed three hypotheses: (1) recall of present time pressure predicts present time passage; (2) recall of past time pressure predicts past time passage; and (3) people underestimate past time pressure. In the present study, therefore, participants were asked to indicate how fast they expected time to pass or how fast time had passed for them on six time intervals, ranging from the next hour to the previous 10 years. They were also asked to answer four items about being rushed now and four items about being rushed 10 years ago. We expected that participants who indicated that they were currently experiencing much time pressure would report that time is passing quickly on the shorter time intervals, whereas participants who indicated that they had been experiencing much time pressure 10 years ago would report that the previous 10 years had passed quickly.

Judgments about past time pressure are, of course, retrospective, and the retrospective findings of the subjective speed of time (Baum et al., 1984; Gallant et al., 1991; Joubert, 1983, 1984, 1990; Lemlich, 1975; Tuckman, 1965; Walker, 1977) could not be confirmed by Wittmann and Lehnhoff’s (2005) and Friedman and Janssen’s (2010) cross-sectional studies. If the subjective experience of time does indeed speed up as people become older, then the effect of age should have been found on all timescales and not just on the 10-year scale. This study, therefore, will examine whether retrospective ratings of time pressure are consistent with contemporaneous ratings of time pressure. It was expected that participants would underestimate time pressure at younger ages.

Method

Participants

Participants in this study were 868 Japanese volunteers, who were between 16 and 80 years old (M = 44.70, SD = 17.99). They were divided into 13 five-year age groups. Each age group consisted of at least 50 participants. There were slightly more female (N = 438) than male (N = 430) participants, but they were equally divided over the age groups (χ²(12) = 1.29, p = 1.00). Except for the youngest age group (16–20 years), all age groups were well-educated (M = 15.69 years of education, SD = 2.78). This age effect is probably due to the fact that the participants in the youngest age group have not yet completed their education, whereas the participants in the other age groups have.

Participants took either a paper-and-pencil version (N = 104) or one of two online versions of the questionnaire. The first Internet group (N = 576) was recruited by an outside agency, whereas the second Internet group (N = 188) had participated approximately three months earlier in another online study about autobiographical memory (Kawasaki et al., 2011). Participants that had taken the paper-and-pencil version were undergraduates who received course credit for their participation. Participants from the first Internet group were members of a website through which they participated in surveys for small financial rewards. Participants from the second Internet group were volunteers who did not receive any financial compensation.

Materials and procedure

After participants had indicated their date of birth, gender, and level of education, they were first asked to rate how fast time usually passed for them. The participants could choose between “very slow” (–2), “slow” (–1), “neither fast nor slow” (0), “fast” (1), and “very fast” (2). The participants were subsequently asked to rate on identical scales how fast they expected the next hour to pass and how fast the previous week, month, year, and 10 years had passed. These six questions were the same as in Friedman and Janssen’s (2010) study and were among the ones that Wittmann and Lehnhoff (2005) had used.

When the participants had answered the questions about the passage of time, they were given eight statements about being busy, feeling rushed, and being unable to complete things. The participants had to indicate whether they agreed or disagreed with each statement on a seven-point scale that ranged from “completely disagree” (–3) to “completely agree” (3). The first four statements were about being busy now and had been used in Friedman and Janssen (2010: 136, Exp. 2), whereas the four other statements, which were presented on the following page, were about being busy 10 years ago.

The six speed-of-time and eight time-pressure items are given in the Appendix. The questionnaire was translated from English into Japanese and back-translated from Japanese to English by two fluent bilingual speakers (from Japan). The study met all current ethical requirements.

Results

Effects of age

Because the scales of the questionnaire are ordinal, non-parametric tests were used to analyze the results. However, the number of participants is large enough to justify the usage of parametric tests. We conducted the analyses with both non-parametric and parametric tests and found similar results. Because we found only two results that changed when using a different method, we have opted to report the outcomes of the non-parametric tests and the outcomes of the two parametric tests that yielded different results.

The six speed-of-time variables were averaged to form one single measure. As in the previous studies (Friedman and Janssen, 2010; Wittmann and Lehnhoff, 2005), participants from all age groups reported that time was passing quickly (M = 0.89, SD = 0.58). The overall speed-of-time scores fell reliably above the midpoint of the scale (t(867) = 45.77, p < .001). Only 281 of the 5208 (5.4%) responses on the speed-of-time scales were “very slow” or “slow,” whereas 72.7% of the questions were answered with “fast” or “very fast.” The remaining items were answered with “neither fast nor slow.”

An independent-samples Kruskal–Wallis test was used to examine the effect of age on the speed of time. Only a small effect of age group was found (χ²(12) = 23.22, p = .026), with older participants reporting slightly higher subjective speeds of time (r_s = .086, p = .011, explained variance = 0.7%). Using a related-samples Friedman test, it was found that the passage of time was not rated similarly across the six time intervals (χ²(5) = 233.51, p < .001). Six Kruskal–Wallis tests were conducted, but age group only had an effect on the 10-year interval (χ²(12) = 48.96, p < .001). Age correlated with the 10-year interval (r_s = .178, p < .001, explained variance = 3.2%), but not with any of the other individual time intervals (r_ss ≤ .033, ps ≥ .331).

Furthermore, there was a small effect of gender on the overall speed-of-time ratings (χ²(1) = 6.80, p < .01). Women gave higher ratings for the passage of time (M = 0.94, SD = 0.59) than men (M = 0.85, SD = 0.56), but this effect did not influence the previous results, because men and women were equally divided over the age groups. There was no significant effect of educational attainment (p = .062).

Time pressure

To understand better how time pressure affects impressions of the speed of time, the participants indicated how busy they were presently and how busy they had been 10 years ago. Both the four contemporaneous and the four retrospective judgments were averaged to form single measures. The participants reported that they currently felt rushed and did not have sufficient time to complete the things they wanted to do (M = 0.91, SD = 1.31). The mean present time pressure scores fell reliably above the midpoint of the scale (t(867) = 20.55, p < .001). Present time pressure correlated strongly with the overall speed-of-time scores (r_s = .280, p < .001, explained variance = 7.8%) and with all individual speed-of-time items (r_ss ≥ .178, ps < .001), except the 10-year item (r_s = .068, p = .044, explained variance = 0.5%). The correlation between present time pressure and the scores on the 10-year item was no longer significant when a parametric test was used (r = .060, p = .078), but the correlations did not change when we controlled for age group with Kendall’s partial rank correlation coefficients.

Participants reported that they experienced some time pressure 10 years ago (M = 0.23, SD = 1.09). The mean past time pressure scores also fell reliably above the midpoint of the scale (t(867) = 6.34, p < .001). Past time pressure correlated significantly with the overall speed-of-time scores (r_s = .105, p = .002, explained variance = 1.1%). It correlated with the ratings on the 10-year interval (r_s = .185, p < .001, explained variance = 3.3%) and the next-hour interval (r_s = .070. p = .038, explained variance = 0.1%), but not with the other individual speed-of-time items (r_ss ≤ .056, ps ≥ .097).

When parametric tests were used, the correlation between past time pressure and the scores on the next-hour item was no longer significant (r = .053, p = .116). When we controlled these correlations for age group using Kendall’s Tau, they tended to be smaller. The partial correlation between past time pressure and the overall speed-of-time scores, for example, became τ_xy.z = .059, and the partial correlations between past time pressure and the ratings on the 10-year-interval and the next-hour interval became τ_xy.z = .113 and τ_xy.z = .057, respectively. These partial correlations are not accompanied by significance levels.

Furthermore, judgments of past time pressure did not correlate with judgments of present time pressure (r_s = –.011, p = .753), even when it was controlled for age group (τ_xy.z = .039). The correlations between age group, overall and individual speed-of-time items, and present and past time pressure are given in Table 2. These correlations are not corrected for family-wise errors (i.e. no Bonferroni corrections).

Table 2.

Correlation matrix

1.	2.	3.	4.	5.	6.	7.	8.	9.
1. Age group	–
2. Overall speed of time	.086*	–
3. Usually	.017	.744***	–
4. Next hour	.011	.628***	.460***	–
5. Previous week	.008	.733***	.561***	.364***	–
6. Previous month	–.033	.746***	.489***	.333***	.594***	–
7. Previous year	.018	.707***	.462***	.285***	.388***	.485***	–
8. Previous 10 years	.178***	.546***	.299***	.206***	.190***	.293***	.412***	–
9. Present time pressure	–.201***	.280***	.325***	.178***	.234***	.242***	.207***	.068*	–
10. Past time pressure	.420***	.105**	.028	.070*	.056	.035	.017	.185***	–.011

Notes: ^*p < .05; **p < .01; ***p < .001.

Underestimation of past time pressure

Participants reported that, on average, they were busier now than they had been 10 years ago. According to a Wilcoxon’s signed-rank test, the scores for present time pressure were significantly higher than the scores for past time pressure (p < .001). These retrospective results are at odds with the cross-sectional finding that participants report less present time pressure as they became older (χ²(12) = 55.55, p < .001; r_s = –.201, p < .001, explained variance = 4.0%). Because judgments about present time pressure decreased with age and judgments about past time pressure increased with age (χ²(12) = 180.64, p < .001; r_s = .420, p < .001, explained variance = 17.6%), the differences between present and past time pressure decreased with age (χ²(12) = 168.25, p < .001; r_s = –.420, p < .001, explained variance = 17.6%).

It might be that people are not good at recalling past time pressure. In Figure 1, the two mean time pressure scores are given for each age group as a function of present age and age 10 years ago. According to an independent-samples Mann–Whitney U test, present feelings of being busy of participants who were between 16 and 20 years old when they answered the questionnaire (M = 1.23, SD = 1.35) were stronger than past feelings of being busy of participants who were between 26 and 30 years old at the time of testing (M = –0.33, SD = 1.14, p < .001). Comparable differences remain significant until the comparison of the present feelings of participants who were between 41 and 45 years old (M = 1.09, SD = 1.32) and the past feelings of participants who were between 51 and 55 years old (M = 0.67, SD = 0.96, p = .007). Although present feelings of time pressure continued to be stronger than past feelings of time pressure until the comparison between participants who were between 56 and 60 years old (M = 0.84, SD = 1.28) and between 66 and 70 years old (M = 0.77, SD = 0.83), the differences are no longer significant (ps ≥ .093).

Figure 1.

Mean time pressure scores for each age group as a function of the age at time pressure. Solid diamonds represent current time pressure, whereas hollow squares represent past time pressure.

Discussion

Although there are many accounts that predict that time passes more quickly as people get older (cf. Draaisma, 2004; Friedman and Janssen, 2010), there is actually no cross-sectional evidence for substantial age effects in the subjective speed of time when intervals longer than a few minutes or shorter than 10 years are considered. Furthermore, cross-sectional studies that used shorter intervals (cf. Block et al., 1998; Craik and Hay, 1999; Lustig, 2003) have provided no evidence that the performance on those estimation or production tasks is related to general impressions about the speed of time.

Cross-sectional studies that have looked at the effect of age on the subjective speed of longer intervals, such as weeks, months, and years, found that participants of all ages reported that time appears to be passing quickly (Friedman and Janssen, 2010; Wittmann and Lehnhoff, 2005). These two studies only found an age effect when participants were asked how fast the previous 10 years had passed. This age difference was, however, very small and only reached significance because of the large sample sizes of both studies.

The studies of Wittmann and Lehnhoff (2005) and Friedman and Janssen (2010) were conducted in four industrialized Western countries: Austria, Germany, New Zealand, and the Netherlands. In the present study, participants from another industrialized, but non-Western, country (Japan) were asked how fast time usually passes, how fast they expected the next hour to pass, and how fast the previous week, month, year, and 10 years had passed. The effect of age on the overall speed-of-time scores in this Asian sample was, like in those Western samples, very small (explained variance = 0.7%) and caused by the ratings on the 10-year interval. Although the effect on this item is small, it has been shown to be reproducible.

Time pressure

Friedman and Janssen (2010) also examined three age-related explanations about time acceleration (i.e. a lower number of memorable events in older age, the slowing down of an internal pacemaker, and a decrease in the attentional resources of older people) and three general explanations about why people of all ages might experience time passing quickly (i.e. forward telescoping, difficulty of recall, and time pressure) but found support for only one of them (i.e. time pressure). To understand better how time pressure affects impressions of the speed of time, participants in the present study were asked to indicate how busy they were now and how busy they had been 10 years ago. Three hypotheses regarding the time pressure explanation were tested: (1) recall of present time pressure predicts time passage on shorter intervals; (2) recall of past time pressure predicts time passage on longer intervals; and (3) people underestimate past time pressure.

We replicated the earlier findings of Friedman and Janssen (2010) that judgments of present time pressure correlated relatively strongly with shorter time intervals (explained variance = 7.2%). We also found that participants who indicated that they had been experiencing much time pressure 10 years ago reported that the previous 10 years had passed quickly (3.0%). This correlation was slightly higher than the correlation between age and the subjective passage of the previous 10 years (2.9%), suggesting that people not only use impressions of time pressure when reporting how fast the previous week, month, and year have passed, but also when reporting how fast the previous 10 years have passed.

The results of the retrospective (or within-subjects) comparisons of time pressure were contradicted by the results of the cross-sectional (or between-subjects) comparisons. Although many participants indicated that they were currently busier than they had been 10 years ago, the cross-sectional comparisons showed that present time pressure decreased with age. Whereas people are probably able to recall many instances in which they were busy and did not have enough time to complete things from the past 12 months, they might not be able to recall as many of these events from 10 years ago. If we assume that participants are indeed more accurate at estimating present time pressure than past time pressure, then it can be concluded that people underestimate past time pressure.

It is possible that the discrepancy between judgments of present and past time pressure are caused by some sort of recalibration. One’s assessment of time pressure may change with time, especially if one does become busier as one ages. The retrospective judgments may therefore be more accurate than the current ones. This interpretation is unlikely, because if one truly experiences more time pressure as one becomes older, then we would have found a small increase or no change, but no decrease, in the ratings of present time pressure. Furthermore, there is no need for a recalibration, because the mean ratings of present time pressure ranged between 0.36 and 1.30 on a scale that ranges from –3 to 3. If older adults are indeed busier than young adults, they would probably have more often chosen ratings that are closer to “agree” or “completely agree” than their current ratings, which were on average somewhere between “neither agree nor disagree” and “somewhat agree.” Another alternative explanation for the discrepancy between judgments of retrospective and contemporaneous time pressure might be that societal time pressure has in fact increased in the last 10 years. It is, however, unclear whether or why this societal change would only affect young and middle-aged adults and not older adults.

Why does life appear to speed up as people get older?

Most people probably believe or have the feeling that time passes more quickly now than it did when they were younger (Baum et al., 1984; Gallant et al., 1991; Joubert, 1983, 1984, 1990; Lemlich, 1975; Tuckman, 1965; Walker, 1977). Participants in these studies had to indicate how much faster or slower time seemed to pass at the present time compared with when they were approximately one-half and one-quarter of their present age. The results of these studies showed that about 70% of the people feel that time is currently passing faster than it did in the past, but in the present study, we found minimal effects of age when we conducted a cross-sectional comparison of the subjective ratings of the speed of time. Participants of all ages indicated that time appeared to be passing quickly, and an effect of age was only found on the 10-year item. If the subjective experience of time does indeed speed up as people become older, as reported by those studies in which participants had to compare contemporaneous time passage with past time passage, then the effect of age should have been found on all timescales and not just on the 10-year scale.

The results of the present study also revealed a similar inconsistency between age patterns of perceived time pressure: judgments of present time pressure decreased with age, whereas judgments of past time pressure increased with age. If the experiences of present time pressure of older adults are compared with their experiences of time pressure when they were 10 years younger, then the differences are small. These differences would probably be much larger if they would compare their experiences of present time pressure with their experiences of time pressure when they were one-half and one-quarter of their present age. Joubert (1984) asked participants to make this comparison about the amount of free time available and found that participants reported that they had more free time when they were one-half and one-quarter of their present age. The comparisons of present and past time pressure suggest that people tend to underestimate past time pressure, which gives the impression that life appears to speed up as they become older.

The study has some limitations and more research needs to be done. First of all, the results of the present study are correlational, which makes it difficult to draw any conclusions about the causality between time passage and time pressure. A follow-up study could, for example, ask first about time pressure to see whether its effect on the subjective experience of time is enhanced. Furthermore, we did not ask about the passage of time 10 years ago, but about the passage of the previous 10 years. Finally, the effect sizes in the present study are significant but very small, which suggests that there are other factors at play that affect the subjective experience of time.

Conclusions

In this study, we found, as two previous studies had done, only a weak correlation between age and the subjective speed of time, but the impressions of the speed of time were more strongly correlated with time pressure. Participants reported time to be passing more quickly when they had been very busy and did not have sufficient time to complete things. People seem to use present time pressure when reporting how fast the previous week, month, or year has passed and past time pressure when reporting how fast the previous 10 years have passed. Because people tend to underestimate past time pressure, life appears to speed up as they become older.

Footnotes

Acknowledgments

We would like to thank Haruka Koyasu and Ai Uemiya for their help with the translations and Rie Matsunaga for her help with the data collection. We would also like to thank two anonymous reviewers for their helpful comments.

Funding

This study was supported by a grant from the Japanese Society for the Promotion of Science (JSPS) to Steve Janssen.

References

Baum

Boxley

Sokolowski

(1984) Time perception and psychological well-being in the elderly. Psychiatric Quarterly, 56: 54–61.

Block

Zakay

Hancock

(1998) Human aging and duration judgments: A meta-analytic review. Psychology and Aging, 13: 584–596.

Craik

FIM

Hay

(1999) Aging and judgments of duration: Effects of task complexity and method of estimation. Perception & Psychophysics, 61: 549–560.

Draaisma

(2004) Why Life Speeds Up as You Grow Older: How Memory Shapes Our Past, Cambridge: Cambridge University Press.

Fraisse

(1963) The Psychology of Time, New York, NY: Harper and Row.

Friedman

Janssen

SMJ

(2013) Aging and the speed of time. Acta Psychologica, 134: 130–141.

Gallant

Fidler

Dawson

(1991) Subjective time estimation and age. Perceptual and Motor Skills, 72: 1275–1280.

Guyau

J-M

(1890) La genese de l’idee de temps [The Origin of the Idea of Time], Paris: Alcan.

James

(1890) The Principles of Psychology, (Vol. 1)New York: Henry Holt and Co.

10.

Joubert

(1983) Subjective acceleration of time. Perceptual and Motor Skills, 57: 49–50.

11.

Joubert

(1984) Structured time and subjective acceleration of time. Perceptual and Motor Skills, 59: 335–336.

12.

Joubert

(1990) Subjective expectations of the acceleration of time with aging. Perceptual and Motor Skills, 70: 334–334.

13.

Kawasaki

Janssen

SMJ

Inoue

(2011) Temporal distribution of autobiographical memory: Uncovering the reminiscence bump in Japanese young and middle-aged adults. Japanese Psychological Research, 53: 86–96.

14.

Lemlich

(1975) Subjective acceleration of time with aging. Perceptual and Motor Skills, 41: 235–238.

15.

Lustig

(2003) Grandfather’s clock: Attention and interval timing in older adults. In: Meck

(ed.) Functional and Neural Mechanisms of Interval Timing, Boca Raton, FL: CRC Press, pp. 261–293.

16.

Ryan

Fritz

(2007) Erroneous knowledge of results affects decision and memory processes on timing tasks. Journal of Experimental Psychology: Human Perception and Performance, 33: 1468–1482.

17.

Ryan

Robey

(2002) Learning and performance effects of accurate and erroneous knowledge of results on time perception. Acta Psychologica, 111: 83–100.

18.

Tuckman

(1965) Older persons’ judgments of the passage of time over the life span. Geriatrics, 20: 136–140.

19.

Walker

(1977) Time estimation and total subjective time. Perceptual and Motor Skills, 44: 527–532.

20.

Walker

Skowronski

Thompson

(2003) Life is pleasant—and memory helps to keep it that way!. Review of General Psychology, 7: 203–210.

21.

Wittmann

Lehnhoff

(2005) Age effects in perception of time. Psychological Reports, 97: 921–935.