Consciousness and Cognition 63 (2018) 99–109

Contents lists available at ScienceDirect

Consciousness and Cognition

journal homepage: www.elsevier.com/locate/concog

Review article

Cognitions about time affect perception, behavior, and physiology

T
– A review on effects of external clock-speed manipulations
⁎
Sven Thönes , Stefan Arnau, Edmund Wascher
Leibniz Research Centre for Working Environment and Human Factors, Dortmund, Germany

A R T IC LE I N F O ABS TRA CT

Keywords: Our understanding of (altered) time perception may benefit from investigating its potential ef-
Time perception fects upon other psychological and physiological parameters. To date, however, only a few rather
Time passage isolated studies have experimentally manipulated the subjective passage of time or the amount of
Duration apparently elapsed duration in order to investigate such effects. Based on our review of these
Cognition
studies, first, altered time perception can be induced effectively by means of rigged (accelerated
External clock speed
or decelerated) external clocks, second, these clock-speed manipulations remain unnoticed by
Appraisal theories
Motivation most participants, and third, several psychological, cognitive, behavioral, and physiological
Mental fatigue variables can be affected, e.g., fatigue, hunger, pain perception. However, the existing studies on
time-perception manipulation do not systematically relate to each other and the underlying
mechanisms of the observed effects are poorly understood. Based on cognitive appraisal theories
and the given empirical results, we propose a cognitive framework for interpreting and ex-
plaining the effects of manipulations of time perception.

0. Introduction

Time is regarded as a fundamental property of our world and an accurate perception of time is relevant for successful guidance of
behavior and well-being. Though, our perception of time is highly subjective and context-dependent. For decades, time perception
has been studied extensively as a dependent variable in psychological experiments. By measuring, for example, the perceived
duration of a visual or auditory stimulus, or by asking participants how fast time seems to pass in different experimental conditions,
various external and internal factors have been isolated that affect our sense of duration and the subjective speed of time passage (for
a recent review, see Matthews & Meck, 2016). However, time perception in terms of perceived elapsed duration or the apparent speed
of time passage has rarely been treated as an independent or mediating variable in experimental settings. And yet, the investigation of
time perception as a fundamental psychological process in terms of its impact on other cognitive and physiological variables bears a
large potential in experimental research. On a theoretical level, models of time perception, such as the prominent pacemaker-
accumulator models (Gibbon, Church, & Meck, 1984; Treisman, 1963), may be refined as being more complex with multidirectional
causal relationships between clock components on the one hand and attentional and memory processes on the other hand. Basically,
these pacemaker-accumulator models assume an internal clock that consists of a pacemaker emitting pulses and an accumulator
collecting or counting these pulses. The amount of collected pulses is positively correlated with the perceived duration of an event,
i.e., the more pulses are accumulated, the longer the perceived duration. The clock device is integrated into an information processing
system comprising attentional and memory components. Accordingly, it is assumed and has been shown frequently that cognitive
processes can affect perceived time (e.g., Roy, Grondin, & Roy, 2012; Thönes & Hecht, 2017; Tobin & Grondin, 2012). However, it

⁎
Corresponding author at: Leibniz Research Centre for Working Environment and Human Factors, Ardeystr. 67, 44139 Dortmund, Germany.
E-mail addresses: thoenes@ifado.de (S. Thönes), arnau@ifado.de (S. Arnau), wascher@ifado.de (E. Wascher).

https://doi.org/10.1016/j.concog.2018.06.014
Received 20 February 2018; Received in revised form 11 June 2018; Accepted 20 June 2018
Available online 30 June 2018
1053-8100/ © 2018 Elsevier Inc. All rights reserved.
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

remains uninvestigated whether the perception of time, e.g., the state of the internal clock, may also affect other perceptual and
cognitive processes. Therefore, on a more applied level, systematic manipulations of time perception or a directed induction of
specific time-related cognitions may be used to improve cognitive performance and facilitate motivation (e.g., ‘time is passing quickly
– the task is interesting’).
Surprisingly, the idea to investigate perceived time as an independent variable was brought forward in the 1960 s already (Rotter,
1969) but it has not been pursued systematically. Only a few rather isolated studies have investigated time perception as an in-
dependent variable and reported potential effects of covert manipulations of external (physical) clocks on a variety of dependent
measures, such as mental fatigue, boredom, hunger, and pain perception (Craik & Sarbin, 1963; Lewis, Lobban, & Shaw, 1956;
London & Monello, 1974; McGrath & O'Hanlon, 1967; Nelson, Nilsson, & Johnson, 1984; Park, Pagnini, Reece, Phillips, & Langer,
2016; Pomares, Creac’h, Faillenot, Convers & Peyron, 2011; Rotter, 1969; Schachter & Gross, 1968; Snyder, Schulz, & Jones, 1974).
In its first part, the present paper reviews the methods and results reported by the past studies and aims at answering three
important questions: First, do participants become aware of covert external clock manipulations and what can be done in order to
prevent participants from questioning a rigged clock’s accuracy? Second, is it possible to systematically induce an accelerated or
decelerated passage of time by means of external clock-speed manipulations? And third, which specific perceptual, cognitive, and
physiological parameters may be affected by external clock-speed manipulations?
In the second part, based on the results from our review, we propose a basic cognitive framework that aims at explaining the
effects of clock-speed manipulations in terms of attributional processes. Moreover, we will discuss potential future perspectives for
this particular line of research with a focus on its significance in relation to current models of mental fatigue (Boksem & Tops, 2008).
By using the term (external) clock-speed manipulation in the present paper, we refer to accelerated and decelerated clocks in the
participants’ environment. The term should not be conflated with the concept of an internal clock as commonly used in the time
perception literature (e.g., Meck, 1983; Wearden, 2008).

1. Background and review of the past literature

In modern societies, we are surrounded by a continuously growing number of time keepers. Over the last decades, classic clocks
and wristwatches have been complemented by a variety of technical devices, such as computers, mobile phones, and domestic
appliances, that accurately display the current time. Importantly, the validity of the temporal cues provided by these social clocks is
generally accepted and serves as an organizing principle within our societies: People can fix appointments, plan dates, and structure
their everyday life because they have learnt to rely on clocks without questioning their accuracy and regularity. In many western
regions, this strong adaptation to clocks is ‘challenged’ twice a year by systematic clock manipulations according to the concept of
daylight saving time. In spring, clocks are adjusted forward one hour in order to save daylight in the evening. In fall, clocks are
adjusted backward to standard time. Indeed, people seem to react sensibly to these adjustments (e.g., Kantermann, Juda, Merrow, &
Roenneberg, 2007). Therefore, the practice of daylight saving time is discussed controversially and negative effects on physiological,
cognitive, and behavioral parameters, such as changes in health behavior and sleep quality (Harrison, 2013) as well as increases in
traffic accidents (Coren, 1996), have been reported.
Apart from potential effects of clock adjustments according to the procedure of daylight saving time, which are presumably
mediated by disturbed circadian rhythms but cannot easily be investigated in a controlled manner, systematic clock manipulations
have rarely been studied in an experimental setting. Interestingly, however, such investigations may contribute to our understanding
of time perception and its impact on other cognitive and perceptual processes. The strong relevance of clocks in our society, our trust
in their accuracy, and our sensitivity to social clock adjustments (e.g., daylight saving time, travel through time zones) suggest that
manipulations of external clocks, i.e. perceived time, can affect a variety of perceptual, cognitive, and even physiological parameters.
In an experimental setting, clock manipulations need to be applied cautiously because the participants should not become aware
of the manipulation. Therefore, discrete and obvious shifts in apparent time (e.g., ± 1 h inside vs. outside a laboratory) do not appear
applicable. Previous studies on effects of clock manipulations have focused on adjustments of clock speed. Accelerated and de-
celerated clocks have been used to manipulate perceived time by inducing the impression of more or less elapsed duration or altered
speed of time passage.
We reviewed the relevant past literature and identified 10 studies that manipulated participants’ time perception by means of
clock-speed manipulations in order to investigate potential effects on psychological and/or physiological parameters (see Table 1). In
all these studies, time passage or seemingly elapsed duration were manipulated by means of accelerated and/or decelerated clocks
that were placed in the laboratory and to which the participants had to attend to (implicitly or explicitly). Without telling the
participants, the hands of the presented clocks moved faster or slower depending on the specific experimental condition. Crucially, on
the one hand, manipulations of clock speed need to be strong enough to affect the psychological (and physiological) measures of
interest, on the other hand, the speed manipulations may not become obvious to the participants. Therefore, and with regard to our
first question, clock-speed needs to be adjusted cautiously.

1.1. (When) do participants become aware of clock-speed manipulations?

The possible effects of experimental clock-speed manipulations most likely depend on successfully keeping the manipulation
concealed. Participants need to ‘believe’ in the clock in order to get the impression that time is running or dragging. Presumably,
participants will ignore a clock if they suspect it to be inaccurate.
To keep a clock manipulation concealed, experimenters need to instruct the participants to remove their watches (and other clock

100
 S. Thönes et al.

Table 1
Overview of past studies manipulating external clock speed. Information on sample size, degree of clock-speed manipulation, dependent variables, general results, effect-size estimates, and the number of
participants that suspected clock manipulation.
Study Sample Clock-speed manipulation Dependent Variables/Tasks Results Effect-size estimate Manipulation noticed?

*
Lewis et al. N=8 109%; inducing a 22-hours circadian Rate/Cycle of urine production No effect of clock speed on urine n.a. No
(1956) rhythm (22 h appeared to be 24) for six (during c-sm.) production; normal rhythm persisted in
weeks; 7 subjects
wristwatches structuring the days (limitation: no control group)
Craik and Sarbin N = 29 Initially 100%; Tapping tempo (during c-sm.); Faster/slower tapping with faster/ Tapping: Two out of 29 participants
(1963) second half: 50% vs. 200% between Dotting tempo (during c-sm.); slower clock; d = 1.00 Dotting:
subj. Verbal time estimate (post c-sm.) Faster/slower dotting with faster/ d = 1.13 Time
(obj. duration manipulated: about slower clock; estimate:
30 min.) clock position: wall; explicit Overestimation/underestimation d = 0.67 (based on
function in the exp. task with faster/slower clock reported Fs and ns)
*
McGrath and N = 48 No clock vs. 200% vs. 100% vs. 50% Vigilance Mixed results, participants may be less n.a. No
O'Hanlon between subj.; (in signal-detection task during c- vigilant with fast clock (limitations:
(1967) (obj. duration manipulated: 60 min/120 in sm.) differences between groups in pre-test;
50% cond.) clock position: wall (passive) confounding of clock-speed and
objective experiment duration)
Schachter and N = 38; (50% Initially 50% vs. 200% between subj.; Eating behavior (amount eaten Control participants ate more with n.a. n.a.
Gross (1968) obese) second half: 100% (obj. duration during c-sm.) slow vs. fast clock; obese participants

101
manipulated: about 60 min); clock ate more with fast vs. slow clock
position: wall (passive)
Rotter (1969) N = 88 100% vs. 50% vs. 25%vs. 12.5% Rating: how interesting was No effect of clock speed on interest n.a. No
between subj. (obj. duration reading material (post c-sm.); rating; Time passage:
manipulated: 32 min) clock position: Rating of time passage (post c- Strong (compatible) effects on time- d > 1.00 Hand
wall; explicit function in the exp. task sm.); Rating of speed of clock-hand passage and hand-speed ratings. speed:
movement (post c-sm.) d > 2.00
(100% vs. 12.5%
comparisons based
on Fs and ns)
London and N = 32 150% vs. 50% between subj. (obj. Boredom (questionnaire) Participants were sign. less bored in n.a. One out of 32 participants
Monello duration manipulated: 20 min) clock (post c-sm.) the 150% clock-speed condition;
(1974) position: wall; explicit function in the Rating of time passage Time passed sign. faster in the 150%
exp. task (while writing stories about clock-speed condition
presented pictures)
(post c-sm.)
Snyder et al. N = 36 150% vs. 50% between subj. (obj. Fatigue indicated by reaction times More fatigue (slower responses) after Reaction Time: Three out of 36 participants
(1974) duration manipulated: 8 min); before and after a pursuit rotor apparently long task relative to d = 0.87 (excluded); some more reported
clock position: wall; explicit function in task (‘Placebo’ effects of apparent apparently short task. No effect on (based on reported suspicion about the clock’s
the exp. task duration of effort?) self-reported fatigue Ms and SDs) accuracy after having been
(post c-sm.); debriefed
Fatigue ratings (post c-sm.)
(continued on next page)
Consciousness and Cognition 63 (2018) 99–109
 Table 1 (continued)

Study Sample Clock-speed manipulation Dependent Variables/Tasks Results Effect-size estimate Manipulation noticed?
S. Thönes et al.

Nelson et al. N = 140 150% vs. 100% vs. 50% between subj. Fatigue (Feeling Tone Check List); Specific interactions with n.a. Three out of 140 participants
(1984) (under different conditions of (during c-sm.) temperature: indicated suspicion about the
temperature [13 °C vs. normal room Mood Less fatigue in cool air with slow vs. clock’s accuracy (not excluded)
temperature] and luminance); (obj. (Mood Adjective Check List); fast clock,
duration manipulated: 50 min); clock (during c-sm.) More fatigue in warm air with slow
position: desk; Productivity vs. fast clock;
explicit function in the exp. task (Number of words written, number Better mood in cool air with fast vs.
of stories written); slow clock, better mood in warm air
(during c-sm.) with slow vs. fast clock;
Mixed results for productivity
Pomares et al. N = 36 100% vs. 75% within subj. (obj. Pain intensity (thermal In parts 1 and 2, pain intensity was 1. 30 s: No
(2011) durations manipulated [three parts]: 30, stimulation) higher when the duration (of 30 and dz = 0.462. 25 s:
25, 15 s); (during c-sm.) 25 s) was apparently longer; dz = 0.713. 15 s:
clock position: no effect in part 3 (15 s) dz = 0.12
screen; explicit function in the exp. task (based on reported F
and N)
Park et al. (2016) N = 46 (100% 200% vs. 100% vs. 50% between subj.; Blood glucose level (BGL); Sign. stronger BGL-decrease in fast BGL-decrease: No
diagnosed with (obj. duration manipulated: 90 min (post c-sm.) Verbal time estimate; vs. normal vs. slow cond.; d = 1.76 Verbal time
type two diabetes) video gaming); (post c-sm.) Stress (rating); Sign. larger estimates in fast vs. estimate:
clock position: desk; explicit function in (post c-sm.) normal vs. slow cond.; d = 3.86 Stress:
the exp. task Hunger (scale); (post c-sm.) Pos. correlation between duration d = 0.21 Hunger:
estimate and BGL-decrease; d = 1.31
No differences between cond. in (fast vs. slow

102
reported stress; comparisons based
Sign. more hunger reported in fast vs. on Ms and SEs, SDs)
normal vs. slow cond.

Note. Clock-speed manipulation in % relative to objective clock speed; effect-size estimates based on reported M- and SD-values or F-, t-, and n-values (between-subjects designs: Cohen’s d; within-subjects
designs: Cohen’s dz) (see Lakens, 2013; Rosnow & Rosenthal, 1996); ‘n.a.’: not available; ‘c-sm.’: external clock-speed manipulation; * and italic indicates studies with severe methodological limitations: no
control group in Lewis et al. (1956), confounding of clock-speed and objective experiment duration as well as significant between-group differences in pretest in McGrath and O'Hanlon (1967).
Consciousness and Cognition 63 (2018) 99–109
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

devices, such as mobile phones) and provide a convincing cover story why participants should do so. Usually, this has and can be
done by simply referring to sensitive electrical devices in the laboratory, which may interfere with watches, mobile phones, etc.
Interestingly, half of the previous studies reported that the clock-speed manipulations remained fully unnoticed by their participants
(Lewis et al., 1956; McGrath & O'Hanlon, 1967; Park et al., 2016; Pomares et al., 2011; Rotter, 1969). Snyder et al. (1974) reported
the highest rate of participants having noticed the manipulation: Three out of 36 participants questioned the clock’s accuracy, which
is still less than 10%. Accordingly, keeping the clock manipulations concealed does not seem to be a substantial methodological
challenge. This is particularly surprising as clock speed has been manipulated quite drastically in the past studies. Most studies
manipulated clock speed between two or more groups of participants. Usually, in one group, clock speed was reduced to 50% of
normal speed, while in a second group it was increased to 150% (London & Monello, 1974; Snyder et al., 1974) or 200% (Craik &
Sarbin, 1963). In order to relate possible effects to a baseline condition, some studies also included a control group working at normal
(100%) clock speed (McGrath & O'Hanlon, 1967; Nelson et al., 1984; Park et al., 2016), while others manipulated clock speed
additionally within groups: Craik and Sarbin (1963), for example, presented a normal clock in the initial phase of their experiment
before accelerating/decelerating its speed in group one and two, respectively. Snyder et al. (1974) first presented the manipulated
clocks (50% vs. 150%) before accelerating/decelerating their speed to 100% in both groups. A pure within-subject design was applied
by Pomares et al. (2011), who reduced clock speed by 25% after an initial phase of normal clock speed. Taken together, irrespective
of the specific experimental design (between vs. within-subject manipulations vs. mixed designs), even strong manipulations of clock
speed (up to a decrease of 87.5%; Rotter, 1969) remain unnoticed by the vast majority of participants. This result is even more
compelling given the wide range of (veridical) durations of the different experiments (e.g., from several minutes (Snyder et al., 1974)
up to weeks (Lewis et al., 1956)) and the heterogeneity of dependent variables involved (see Section 1.3). In general, participants
seem to attribute an impression of strange time passage rather to internal processes, such as flow, boredom, or an inability to estimate
time (Craik & Sarbin, 1963), whereas they do not question a clock’s ‘objectivity’. As outlined above, this may be a result of the high
significance of clocks within our culture. Through decades, people have learnt that clocks are working reliably and with high ac-
curacy. Clocks are the important pacemakers in our societies, relying on their ticking is necessary in everyday life. Thus, we do not
easily suspect a clock to be inaccurate, but rather attribute a strange impression of time passage to internal processes or the situa-
tional context.
Based on the existing literature, we cannot rule out that manipulated clocks can affect even those participants that are aware of
the manipulation. In fact, people sometimes use manipulated (inaccurate) clocks that are, for example, 5 min ahead of the veridical
time in order to be early for meetings. In this context at least, manipulated clocks affect a subjects’ behavior even though the subject is
explicitly aware of the manipulation. Especially in situations where no other temporal cues are available, information provided by
clocks that are obviously inaccurate may be used to guide behavior and evaluate internal processes. These considerations could be
tested in experiments with overt compared to concealed clock manipulations.

1.2. Do clock-speed manipulations affect time passage and duration judgments?

In order to be an effective independent variable, covert clock-speed manipulations need to affect measures of time perception,
such as judgments of time passage and duration. For example, if a clock runs faster than normal and the participant does not doubt its
accuracy, the subjective speed of time passage should be accelerated and retrospective time estimates should indicate an over-
estimation of elapsed duration. If the clock runs slower, time passage should be decelerated and elapsed duration underestimated.
Both these measures of time perception can be assessed easily. Ratings on the perceived speed of time passage can be obtained simply
by asking the participant how quickly time passed during the experiment. Usually, 7 or 9 point Likert scales or visual analogue scales
ranging between ‘very slow’ and ‘very fast’ are used to assess ratings of time passage (Oberfeld, Thönes, Palayoor, & Hecht, 2014;
Wearden, 2015). Retrospective time estimates can be obtained after the experiment by asking the participant to indicate the ex-
periment’s duration in temporal units, such as minutes and seconds (e.g., Thönes & Wittmann, 2016). Four of the past studies
obtained these measures of time perception and consistently reported that a faster/slower clock induced an accelerated/decelerated
passage of time (London & Monello, 1974; Rotter, 1969) and corresponding over-/underestimations of duration (Craik & Sarbin,
1963; Park et al., 2016). These effects were statistically significant and large in size. The determined effect-size estimates (Cohen's d;
not available for London & Monello, 1974) ranged between 0.67 and 3.86, confirming that covert clock-speed manipulations can
systematically induce an altered perception of time in terms of passage and elapsed duration.
The between-study differences in effect size may be explained by the particular task(s) the participants had to work on in the
different experiments. Depending on the specific character of the task (e.g., boring vs. demanding), the participants may have paid
more or less attention to the rigged clock. Accordingly, the clock had more or less effect on the participants’ time perception.
However, based on the method reports of the previous studies that had obtained measures of time perception, there are no obvious
work-load differences between studies that could explain the effect-size differences. Park et al. (2016), who reported the largest
effect, applied ‘simple video games’, while Rotter (1969) instructed the participants to read text material and Craik and Sarbin (1963)
included different dotting and tapping tasks in their experiment. Moreover, in all these studies the clocks had explicit functions within
the experimental protocol and the participants were instructed to check the clock’s time regularly (e.g., ‘write down the time after
each part of the experiment’). These instructions may have weakened potentially moderating effects of the specific experimental tasks
on the effects of clock-speed manipulation on time passage and duration judgments. The different strengths of clock-speed effects on
measures of time perception more likely depend on differences in the veridical durations of the experiments. More precisely, the
effect of a rigged clock may accumulate during the experiment, therefore being pronounced in longer rather than shorter experi-
ments. This may explain the (extremely) large clock effect on verbal time estimates in the 90-minutes experiment by Park et al.

103
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

(2016) in comparison to the considerably smaller (but still substantial) effects obtained in the 30- and 32-minutes experiments by
Rotter (1969), and Craik and Sarbin (1963), respectively. Irrespective of the between-study effect-size differences, all four studies that
obtained time passage and duration judgments consistently reported that time perception can be manipulated successfully by means
of covertly rigged clocks. In terms of a manipulation check, this may be a precondition for the interpretation of possible effects of
clock-speed manipulations on perceptual, cognitive, or physiological parameters beyond the mere perception of time.

1.3. Which parameters can be affected by clock-speed manipulations?

In terms of the dependent measures of interest, the previous studies do not systematically relate to each other. While a couple of
studies focused on clock-speed effects on different (rather) physiological parameters, such as blood glucose level (Park et al., 2016),
pain (Pomares et al., 2011), and urine flow (Lewis et al., 1956), others investigated perceptual, cognitive, and behavioral outcomes.
Craik and Sarbin (1963), for example, investigated the participants’ ‘personal tempo’ of dotting (producing pencil dots in rows of
circles as fast as possible) and tapping (producing rows of dashes at a preferred pace), which were affected by the altered clock speed
in a compatible way. In both tasks, the participants were faster when the clock speed increased and slower when the clock speed
decreased. In the cognitive domain, particular emphasis was put on research questions related to the field of fatigue (Nelson et al.,
1984; Snyder et al., 1974), boredom (London & Monello, 1974; Rotter, 1969), and vigilance (McGrath & O'Hanlon, 1967). Does a
decelerated clock, which induces a slowed passage of time, lead to higher levels of fatigue and boredom, and do subjects become
more vigilant in the presence of a faster ticking clock? Indeed, these relationships seem plausible given, first, the impression that (vice
versa) time drags when we are bored (Watt, 1991) or tired (Wearden et al., 2014), and second, the recent discussion on the close
interweaving of vigilance, fatigue, and boredom (Arnau, Möckel, Rinkenauer, & Wascher, 2017; Wascher et al., 2016). According to
Pattyn, Neyt, Henderickx, and Soetens (2008), for example, a decrement in vigilance may be attributed to fatigue on the one hand or
boredom on the other hand, depending on the situational context. Based on the previous studies, however, the question of whether
clock-speed manipulations systematically affect levels of boredom, fatigue, and vigilance cannot be answered sufficiently yet.
Whereas London and Monello (1974) reported significantly lower boredom ratings in a story-writing task in a fast-clock condition as
compared to a slow-clock condition, there were no comparable differences in ‘interest ratings’ on reading material in the study by
Rotter (1969). Fatigue levels as measured in terms of a decrement in reaction times after an 8-minutes pursuit rotor task were
significantly increased after an apparently long task (fast clock) as compared to an apparently short task (slow clock) in Snyder et al.’s
study (1974) – an observation that is not really compatible with the results from the boredom studies. Moreover, Nelson et al. (1984)
reported rather mixed effects of clock speed on fatigue, which were moderated by additional variables, such as room temperature and
luminance, whereas McGrath and O'Hanlon (1967) observed that participants were less vigilant in the presence of a fast relative to a
slow clock. Taken together, these results are rather inconsistent, which may be due to several reasons. First, the dependent variables
selected were quite diverse (e.g., questionnaires, reaction times, interest ratings) and may not all be reliable and valid indicators of
boredom, fatigue, and vigilance. Choosing more recent and established measures may increase the replicability of the effects. Second,
clock-speed effects on cognitive and motivational parameters may strongly depend on the specific experimental context. How long
does the experiment take? Is there any feedback or reward involved? Is the task demanding or rather simple? We will refer to this
discussion in more detail in part 2. And third, some of the (early) studies exhibit severe methodological shortcomings that could have
contributed to the inconsistent and partly unexpected results. For instance, the vigilance effect reported by McGrath and O'Hanlon
(1967) needs to be interpreted with caution due to significant between-group differences in the pretest. Moreover, in veridical
duration, the slow-clock condition lasted twice as long as the normal-clock, fast-clock, and no-clock conditions, meaning a potential
confound of clock speed and duration of the experiment.
Apart from these considerations, clock-speed effects seem to be highly ‘cognitive’. The reported (mostly large) effects of apparent
duration on blood glucose level, pain perception, hunger, boredom, and fatigue can be interpreted in terms of attributional processes
that can even affect physiological parameters (e.g., ‘much time has passed – I should be tired and hungry’ – blood glucose level drops;
‘time goes by extremely slowly – the task must be boring’ – task performance drops, etc.). This suggests that simple clock-speed
manipulations may affect a wide range of additional parameters that are conceivably shaped by cognitive processes. Apart from
explicit clocks, even simple temporal cues can exhibit strong effects on several affective parameters. Sackett, Meyvis, Nelson,
Converse, and Sackett (2010) asked participants to evaluate different cognitive tasks, music, and noises. The participants’ evaluations
were more positive when an apparently task-relevant progress bar or seemingly valid information on the duration of the task/music/
noise induced the impression of fast passage of time. The authors concluded that (invalid) pieces of temporal information can provide
a metacognitive cue that people attribute to their enjoyment of a specific experience.
Based on the results from the previous studies, we can now clearly answer the three initial questions: First, even strong (within-
subjects) manipulations of clock speed remain unnoticed by a vast majority of participants. Second, systematic clock manipulations
effectively induce an accelerated or decelerated passage of time. And third, depending on the experimental context, the induced
speed of time passage may affect a variety of cognitive, affective, motivational, and even physiological parameters. With regard to the
third aspect, well-conducted future studies are needed to systematically investigate the effects of clock-speed manipulations as well as
its moderating (context) factors and underlying mechanisms.

2. A cognitive framework for clock-speed effects and future perspectives

How does clock speed affect perception, cognition, and even physiological parameters? What are the underlying mechanisms of
these effects? These questions cannot be answered adequately yet and additional research is needed. What we propose is an

104
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

explanatory framework for the observed effects, which needs to be tested and developed further by future studies.
With regard to the terminology used, it can be argued that not time perception but clock speed represents the independent
variable in the experiments. Strictly speaking, experimenters manipulated the speed of physical clocks in order to induce a sub-
jectively altered speed of time or the impression of more or less elapsed duration, which is not a direct manipulation of time per-
ception. Time perception would be viewed rather as a variable that mediates the effects of clock-speed manipulations. Accordingly,
clock-speed effects on psychological, behavioral, and physiological processes are based on cognitions about time (‘temporal cogni-
tions’) and may be interpreted in terms of cognitive appraisal theories (Schachter & Singer, 1962; Seth, 2013). Cognitive appraisal
theories assume that physiological states and changes are evaluated based on cognitions that are currently available to the subject.
For instance, Schachter and Singer (1962) demonstrated that injections of adrenaline, a manipulation of physiological arousal, lead to
emotional reactions of either anger or joy depending on the experimental context (an irritated or elated confederate). Environmental
(context) factors provide a frame for the cognitions and believes that are available to the subject.
In the context of our study, the crucial cognitive component is represented by the subjects’ believes about time passage and/or
elapsed duration. Accordingly, the subjects’ temporal cognitions shape their perceptions and evaluations of specific tasks and internal
states, such as hunger or vigilance, and can thereby influence task performance and behavior in general. Some of the past studies
already interpreted their results with reference to cognitive appraisal theories.
Schachter and Gross (1968), for example, proposed the passage of time as one of many possible external cues that can regulate
(eating) behavior. On a more general level, cyclic behavioral patterns, such as food intake, can be triggered by valid or invalid
external cues. London and Monello (1974) extended this view by more explicitly referring to cognitive appraisal of given temporal
cues in the context of boredom. They concluded that subjects evaluate a task based on cognitions that are available to them. If clock-
speed manipulations induce a substantial discrepancy between the amount of time that has passed (apparently) and the amount of
time felt to have passed, this discrepancy provides the basis for cognitions, such as ‘time is passing slowly – the task is boring’. As
many other external cues, (manipulated) pieces of temporal information are considered to be capable of influencing behavior by
prompting attributions (Sackett et al., 2010). Especially violations of (temporal) expectations may initiate processes of sense making
or metacognition in order to explain discrepancies between internal and external cues (e.g., the clock tells me that only 12 min have
passed but it feels like half an hour. How is that possible? The task must be extremely boring’).
These interpretations of clock-speed effects in terms of metacognitive processes and theories of cognitive appraisal are plausible.
However, with regard to time perception and the crucial differentiation between perception of time passage and perception of
duration (Droit-Volet & Wearden, 2016; Thönes & Oberfeld, 2015), a more refined analysis is needed. As we will point out, this
differentiation is highly relevant because a focus on time passage will lead to different temporal cognitions and subsequent effects
than a focus on elapsed duration.
Most importantly, it needs to be considered whether tasks or internal states are evaluated during or after the phase of clock-speed
manipulation and whether the clock has explicit task-relevance. During the experimental phase when the clock is accelerated or
decelerated and not explicitly task-relevant, temporal cognitions should be related to the passage of time. When the clock is ac-
celerated, the subject gets the impression that time is passing quickly. The temporal cognition ‘time is running’ should induce positive
emotions, leading to a positive evaluation of the ongoing task (‘the task is interesting and entertaining’). It may even induce the
experience of flow with positive effects on task performance (Csikszentmihalyi & Csikszentmihalyi, 1992). When the clock is de-
celerated, the subject gets the impression that time is passing slowly. The temporal cognition ‘time drags’, should induce feelings of
unpleasantness leading to a negative evaluation of the task (‘the task is boring and dull’; e.g., London & Monello, 1974) and a
potential decline in task performance. After the phase of clock-speed manipulations and also when the clock is task-relevant during
the experimental phase (e.g., the participant is instructed to note how much duration has passed after a specific part of the ex-
periment), temporal cognitions should be related less to the perceived passage of time but more and more to the amount of (ap-
parently) elapsed duration. The subject believes that a short (slow clock condition) or a long (fast clock condition) period of time has
elapsed. These cognitions should differently affect the perception and evaluation especially of internal states. For example, depending
on the apparently elapsed duration, the subject may feel more or less hungry (Park et al., 2016), tired, and exhausted (Snyder et al.,
1974). In either case, it has to be noted that (post-experimental) questionnaires on time perception can bias temporal cognitions
towards elapsed duration (e.g., ‘How long did the task take?'; Craik & Sarbin, 1963; Park et al., 2016) or time passage (e.g., ‘How
quickly did time pass during the task?’; London & Monello, 1974; Rotter, 1969).
Additionally, it needs to be considered that the specific role of the rigged clock within the task represents a crucial factor. If the
clock is task-relevant and the participants are instructed, for example, to stop working on the task after a specific duration has elapsed
(Snyder et al., 1974), attention will be (partly) shifted to elapsed duration even during the experimental phase. In such a case,
temporal cognitions during the task may be difficult to control, i.e., the experimenter cannot (easily) determine whether cognitions
are related rather to time passage or to duration. Accordingly, more ambiguous effects of clock-speed manipulations on task eva-
luation and performance are to be expected in experiments with task-relevant clocks. Moreover, with task-relevant clocks during the
experimental phase, the availability and therefore the effects of temporal cognitions should be highly task-dependent. For example, if
the task (seemingly) extends over the course of the experiment and the participants get the impression that the task should be solved
within a given time (time-pressured circumstances), apparently shorter elapsed duration should lead to positive cognitions, such as,
‘the task is easy – I am good at it’. In this case, task performance would increase, whereas opposite effects should emerge with
accelerated task-relevant clocks in (apparently) time-critical tasks. If the task does not seem to measure performance, these effects
should not occur.
Additionally, as discussed in the first part, the size of clock-speed effects most likely depends on the attentional demands of the
task applied and the veridical duration of the experiment. The more demanding and the shorter a specific task, the less attention will

105
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

be paid to a (rigged) clock. Therefore, the effects of clock speed may not be observed in highly demanding and rather short tasks,
while they should be pronounced in longer experiments including tasks of low to medium difficulty. Especially in tasks with low
attentional demands, where participants can pay much attention to a clock, cognitions may be difficult to control even with task-
irrelevant clocks. As participants have enough cognitive resources left, they may tend to ‘think’ about the clock and what it is
indicating. As in time-critical tasks with relevant clocks, this can induce cognitions about time passage and duration at the same time.
Taken together, effects of clock-speed manipulations depend on several (moderating) factors, which need to be considered
carefully when planning experiments or interpreting results from past studies (see Fig. 1). The most challenging aspect may be to
predict and successfully control which specific temporal cognitions will arise in the experimental setting. First, it needs to be con-
sidered whether temporal cognitions are related to the perceived passage of time or to apparently elapsed duration. As outlined
above, during the phase of clock-speed manipulation, temporal cognitions are related to the perceived passage of time, directly
affecting measures of task evaluation and performance. However, if the manipulated clock becomes task-relevant and time is explicit,
the temporal cognitions will (also) refer to elapsed duration, with more complex effects on task evaluation and performance. After the
phase of clock-speed manipulation, temporal cognitions should be related to duration rather than to time passage and the perception
of internal states will be affected rather than the evaluation of the task, which has already been completed. Second, the specific
properties of the task (during the phase of clock-speed manipulation), the degree of clock-speed manipulation, the assessment of time-
related questionnaires, and the objective duration of the experiment should be regarded as important moderating factors. As outlined
above, temporal cognitions and therefore clock-speed effects on task evaluation and performance should also depend on attentional
demands and on whether the task explicitly measures time-critical performance or not. Under such time-pressured circumstances,
which are common in everyday life, clocks are task-relevant and even during the task temporal cognitions should be biased to elapsed
duration. Accordingly, in time-critical tasks, decelerated clocks indicating shorter elapsed duration should induce positive cognitions
about task and performance.
By considering cognitions about time passage and elapsed duration (temporal cognitions) as the relevant mediating variables, the
outlined framework helps to explain the effects of rigged clocks on perceptual, cognitive, and behavioral measures. Even influences
on physiological parameters may be explained via this route similar to placebo and nocebo effects (e.g., Turner, Deyo, Loeser, Von
Korff, & Fordyce, 1994), which may also depend on expectations and processes of reappraisal and attention (Tracey, 2010). In fact,
Snyder et al. (1974) interpreted their results on effects of clock speed on fatigue, which were moderated be expectations about
duration (‘internal standards’), in a similar way. Considering such moderating factors can explain some of the seemingly inconsistent
results from the past studies. However, due to the large (methodological) heterogeneity of the previous studies in terms of quality,
tasks applied, objective durations of manipulation, and dependent measures, it is not yet possible to empirically evaluate the fra-
mework as outlined and illustrated above.
Based on its testable predictions (e.g., ‘easy tasks are less boring when an accelerated (task-irrelevant) clock is presented’), the
framework can and needs to be tested systematically by future studies. In order to manage the complex interweaving of the different
moderating factors, standardized and well-structured experiments with a limited number of independent variables are preferable.
Moreover, results can be interpreted most straightforward when clocks are task irrelevant, dependent variables are measured after
clock-speed manipulations, and time-related questions clearly bias temporal cognitions either to time passage or to duration.

Future directions and practical implications

Especially with regard to the phenomenon of mental fatigue, research on clock-speed manipulations and temporal cognitions may
contribute to the progression and modification of recent cognitive and neural models. According to Boksem and Tops (2008), for
example, mental fatigue represents an adaptive signal indexing a negative outcome of a (subconscious) comparison of the expected
benefits and the energy already spent at a given time during task performance. As soon as the energetical costs outweigh the potential
rewards, the feeling of fatigue emerges in order to ‘prevent’ the organism from further spending (see also, Boksem, Meijman, & Lorist,
2006). This is the case, for example, when tasks have to be performed for prolonged periods of time without gaining proportionate
benefits. Now, in clock-speed experiments, the apparent duration of a task, that is, the apparent costs, can be manipulated while
keeping the actual task (the costs) perfectly constant. This approach is promising for investigating the framework by Boksem and
Tops (2008) since manipulations of the objective costs of task performance most likely affect also motivational aspects of the task (the
reward side of the equation). When only the subjective costs are manipulated, the predictions of the model are straightforward: An
accelerated clock should lead to an overestimation of elapsed duration. Accordingly, the energetical costs for completing the task
should be overestimated as well, resulting in higher levels of reported fatigue after the experiment. During an experiment, higher
levels of mental fatigue are to be expected in an experimental condition with a decelerated clock. This is because the perceived work-
load, that is the costs, should be higher when the same objective amount of work is done within subjectively less time. Again, this
example shows that the differentiation between cognitions related to time passage and duration is crucial.
Apart from these theoretical considerations, research on effects of time perception on cognitive and physiological variables bears
important practical implications and a potential for application, for example, in working environments. Among the most popular
phenomena in the context of time perception are those related to attentional processes (Brown & Merchant, 2007; Brown, 1997) and
affective states (Droit-Volet & Meck, 2007). As discussed in the previous sections, it is well known that time seems to fly when we are
engaged in an interesting task and in joyful situations, whereas time drags when we (have time to) focus on time or when we are in a
negative affective state including boredom (Watt, 1991) and depression (Thönes & Oberfeld, 2015). Now, the promising practical
implications of the interweaving of affective states, attentional processes, and time perception may be revealed when viewing the
relationships vice versa as in the present article. As it is possible to ‘manipulate’ the subjective passage of time, thereby inducing

106
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

Fig. 1. Flow chart illustrating relevant (moderating) factors in experiments with clock-speed manipulations. Note. Circles indicate moderating
factors. Clock-speed effects should be pronounced in longer experiments with stronger clock-speed manipulations and less attentionally demanding
tasks. Post-experimental time-related questions can bias temporal cognitions towards time passage (e.g., ‘How quickly did time pass during the
task?’) or elapsed duration (e.g., ‘How long did the task take?’). Empirical studies reporting clear results are integrated at the bottom line fitting the
model’s predictions.

specific cognitive and affective states, we may improve cognitive performance and facilitate motivation or even the experience of
flow by means of simple (rigged) clocks.
In this regard, clock-speed experiments may also contribute to our theoretical understanding of the causal relations between

107
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

arousal, attention, and time perception as being more complex (multidirectional: arousal ⇔ time perception; attention ⇔ time
perception) than suggested by the prominent pacemaker-accumulator models of time perception (unidirectional: arousal ⇒ time
perception; attention ⇒ time perception) (Gibbon et al., 1984; Zakay & Block, 1996).
At the same time, there are practical limitations and ethical questions that need to be discussed. Do clock-speed effects still occur
when participants are aware of the manipulation? Do effects decrease when clock speed is altered repeatedly? In which specific real-
world settings are clock-speed manipulations applicable, effective, and justifiable? Apart from clock-speed manipulations, the simple
visibility and salience of clocks and time cues in general (e.g., in working environments) may influence behavior, mood, and mo-
tivation. For example, as the experience of flow entails a feeling of timelessness (Mainemelis, 2001), salient clocks may prevent
people from getting into flow-like mental states and should therefore be avoided. As with regard to the theoretical considerations, the
possible practical perspectives and limitations need to be investigated systematically by future studies.

3. Declaration of interest

None.

References

Arnau, S., Möckel, T., Rinkenauer, G., & Wascher, E. (2017). The interconnection of mental fatigue and aging: An EEG study. International Journal of Psychophysiology,
117, 17–25.
Boksem, M. A., Meijman, T. F., & Lorist, M. M. (2006). Mental fatigue, motivation and action monitoring. Biological Psychology, 72(2), 123–132.
Boksem, M. A., & Tops, M. (2008). Mental fatigue: Costs and benefits. Brain Research Reviews, 59(1), 125–139.
Brown, S. W. (1997). Attentional resources in timing: Interference effects in concurrent temporal and nontemporal working memory tasks. Perception and
Psychophysics, 59(7), 1118–1140.
Brown, S. W., & Merchant, S. M. (2007). Processing resources in timing and sequencing tasks. Perception and Psychophysics, 69(3), 439–449.
Coren, S. (1996). Daylight savings time and traffic accidents. New England Journal of Medicine, 334(14), 924–925.
Craik, K. H., & Sarbin, T. R. (1963). Effect of covert alterations of clock rate upon time estimations and personal tempo. Perceptual and Motor Skills, 16(2), 597–610.
Csikszentmihalyi, M., & Csikszentmihalyi, I. S. (1992). Optimal experience: Psychological studies of flow in consciousness. Cambridge University Press.
Droit-Volet, S., & Meck, W. H. (2007). How emotions colour our perception of time. Trends in Cognitive Sciences, 11(12), 504–513.
Droit-Volet, S., & Wearden, J. (2016). Passage of time judgments are not duration judgments: Evidence from a study using experience sampling methodology. Frontiers
in Psychology, 7(176).
Gibbon, J., Church, R. M., & Meck, W. H. (1984). Scalar timing in memory. Annals of the New York Academy of Sciences, 423(May), 52–77.
Harrison, Y. (2013). The impact of daylight saving time on sleep and related behaviours. Sleep Medicine Reviews, 17(4), 285–292.
Kantermann, T., Juda, M., Merrow, M., & Roenneberg, T. (2007). The human circadian clock's seasonal adjustment is disrupted by daylight saving time. Current
Biology, 17(22), 1996–2000.
Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4.
Lewis, P. R., Lobban, M. C., & Shaw, T. I. (1956). Patterns of urine flow in human subjects during a prolonged period of life on a 22-hour day. The Journal of Physiology,
133(3), 659–669.
London, H., & Monello, L. (1974). Cognitive manipulation of boredom. In H. London, & R. E. Nisbett (Eds.). Thought and feeling: Cognitive alteration of feeling states (pp.
74–82). Oxford: Aldine.
Mainemelis, C. (2001). When the muse takes it all: A model for the experience of timelessness in organizations. Academy of Management Review, 26(4), 548–565.
Matthews, W., & Meck, W. (2016). Temporal cognition: Connecting subjective time to perception, attention, and memory. Psychological Bulletin, 142(8), 865–907.
McGrath, J. J., & O'Hanlon, J. (1967). Temporal orientation and vigilance performance. Acta Psychologica, 27, 410–419.
Meck, W. H. (1983). Selective adjustment of the speed of internal clock and memory processes. Journal of Experimental Psychology-Animal Behavior Processes, 9(2),
171–201.
Nelson, T. M., Nilsson, T. H., & Johnson, M. (1984). Interaction of temperature, illuminance and apparent time on sedentary work fatigue. Ergonomics, 27(1), 89–101.
Oberfeld, D., Thönes, S., Palayoor, B. J., & Hecht, H. (2014). Depression does not affect time perception and time-to-contact estimation. Frontiers in Psychology, 5.
Park, C., Pagnini, F., Reece, A., Phillips, D., & Langer, E. (2016). Blood sugar level follows perceived time rather than actual time in people with type 2 diabetes.
Proceedings of the National Academy of Sciences, 113(29), 8168–8170.
Pattyn, N., Neyt, X., Henderickx, D., & Soetens, E. (2008). Psychophysiological investigation of vigilance decrement: Boredom or cognitive fatigue? Physiology &
Behavior, 93(1), 369–378.
Pomares, F. B., Creac’h, C., Faillenot, I., Convers, P., & Peyron, R. (2011). How a clock can change your pain? The illusion of duration and pain perception. PAIN,
152(1), 230–234.
Rosnow, R. L., & Rosenthal, R. (1996). Computing contrasts, effect sizes, and counternulls on other people's published data: General procedures for research con-
sumers. Psychological Methods, 1(4), 331–340.
Rotter, G. S. (1969). Clock-speed as an independent variable in psychological research. The Journal of General Psychology, 81(1), 45–52.
Roy, M., Grondin, S., & Roy, M. A. (2012). Time perception disorders are related to working memory impairment in schizophrenia. Psychiatry Research, 200(2–3),
159–166.
Sackett, A. M., Meyvis, T., Nelson, L. D., Converse, B. A., & Sackett, A. L. (2010). You’re having fun when time flies: The hedonic consequences of subjective time
progression. Psychological Science, 21(1), 111–117.
Schachter, S., & Gross, L. P. (1968). Manipulated time and eating behavior. Journal of Personality and Social Psychology, 10(2), 98–106.
Schachter, S., & Singer, J. (1962). Cognitive, social, and physiological determinants of emotional state. Psychological Review, 69(5), 379–399.
Seth, A. K. (2013). Interoceptive inference, emotion, and the embodied self. Trends in Cognitive Sciences, 17(11), 565–573.
Snyder, M., Schulz, R., & Jones, E. E. (1974). Expectancy and apparent duration as determinants of fatigue. Journal of Personality and Social Psychology, 29(3), 426–434.
Thönes, S., & Hecht, H. (2017). Counting does not improve the accuracy of long time productions. Attention, Perception, & Psychophysics, 79(8), 2576–2589.
Thönes, S., & Oberfeld, D. (2015). Time perception in depression: A meta-analysis. Journal of Affective Disorders, 175, 359–372.
Thönes, S., & Wittmann, M. (2016). Time perception in yogic mindfulness meditation—Effects on retrospective duration judgments and time passage. Psychology of
Consciousness: Theory, Research, and Practice, 3(4), 316–325.
Tobin, S., & Grondin, S. (2012). Time perception is enhanced by task duration knowledge: Evidence from experienced swimmers. Memory and Cognition, 40(8),
1339–1351.
Tracey, I. (2010). Getting the pain you expect: Mechanisms of placebo, nocebo and reappraisal effects in humans. Nature Medicine, 16(11), 1277–1283.
Treisman, M. (1963). Temporal discrimination and the indifference interval: Implications for a model of the “internal clock”. Psychological Monographs, 77(13), 1–31.
Turner, J. A., Deyo, R. A., Loeser, J. D., Von Korff, M., & Fordyce, W. E. (1994). The importance of placebo effects in pain treatment and research. Jama, 271(20),
1609–1614.
Wascher, E., Heppner, H., Kobald, S. O., Arnau, S., Getzmann, S., & Möckel, T. (2016). Age-sensitive effects of enduring work with alternating cognitive and physical

108
S. Thönes et al. Consciousness and Cognition 63 (2018) 99–109

load. A study applying mobile EEG in a real life working scenario. Frontiers in Human Neuroscience, 9(711).
Watt, J. D. (1991). Effect of boredom proneness on time perception. Psychological Reports, 69(1), 323–327.
Wearden, J. H. (2008). Slowing down an internal clock: Implications for accounts of performance on four timing tasks. Quarterly Journal of Experimental Psychology,
61(2), 263–274.
Wearden, J. H. (2015). Passage of time judgements. Consciousness and Cognition, 38, 165–171.
Wearden, J. H., O’Donoghue, A., Ogden, R., & Montgomery, C. (2014). Subjective duration in the laboratory and the world outside. In V. Arstila, & D. Lloyd (Eds.).
Subjective time: The philosophy, psychology, and neuroscience of temporality (pp. 287–306). .
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. 129–
141). Amsterdam and New York: Elsevier.

109