Benefits of napping in healthy adults: impact of nap length, time of day, age, and experience with napping
Kimberly A. Cote, MSc, PhD, Associate Professor, Psychology Department, Director, Brock Sleep Research Laboratory, Brock University, St. Catharines, Ontario, L2S 3A1, Canada. Tel.: 905-688-5550 (ext. 4806, Office), 905-688-5550 (ext. 3795, Lab); fax: 905-688-6922; e-mail: email@example.com
Napping is a cross-cultural phenomenon which occurs across the lifespan. People vary widely in the frequency with which they nap as well as the improvements in alertness and well-being experienced. The systematic study of daytime napping is important to understand the benefits in alertness and performance that may be accrued from napping. This review paper investigates factors that affect the benefits of napping such as duration and temporal placement of the nap. In addition, the influence of subject characteristics such as age and experience with napping is examined. The focus of the review is on benefits for healthy individuals with regular sleep/wake schedules rather than for people with sleep or medical disorders. The goal of the review is to summarize the type of performance improvements that result from napping, critique the existing studies, and make recommendations for future research.
Although napping is considered to be a normal daily routine for babies and young children, many people continue to take daytime naps across the lifespan. Moreover, napping is a cross-cultural phenomenon. Estimates on the frequency of napping vary considerably. In a compilation of surveys from a number of countries, the frequency of napping (at least once a week) varied from 36% to 80% (Dinges, 1989). The most recent ‘Sleep in America’ poll indicated that 46% of respondents napped at least twice in the last month, with an average nap duration of approximately 1 h (National Sleep Foundation, 2008). The frequency of napping has consistently been reported to increase with advancing age [e.g., Ohayon and Zulley (1999)].
People choose to take naps for a variety of reasons. While some nap in response to sleep loss (i.e., replacement napping) or in preparation for sleep loss (i.e., prophylactic napping), others simply nap for enjoyment (i.e., appetitive napping) (Broughton and Dinges, 1989). Napping is a practical solution to daytime sleepiness for shiftworkers (Gillberg et al., 1996; Oexman et al., 2002; Rosa, 1993; Takahashi and Arito, 2000) and people with sleep disorders [see Takahashi (2003) for review]. But even for individuals who generally get the sleep they need on a nightly basis, napping may lead to considerable benefits in terms of mood, alertness, and cognitive performance.
A number of factors may influence the degree of benefit accrued from a daytime nap. Homeostatic and circadian processes may interact to determine the extent of restoration, e.g., the quality of the prior sleep period, the duration of the nap, the timing of the nap, and the presence of sleep inertia. Other factors such as subject characteristics (e.g., age, gender, individual differences in napping experience, experience with sleep loss, and degree of sleepiness at the time of testing) or test characteristics (e.g., task sensitivity, task difficulty, timing, or instructions) may also influence the benefits of napping. Further, regular nappers may glean more benefits from napping. For example, Evans et al. (1977) categorized nappers and non-nappers based on their frequency of napping and found that only habitual nappers described their naps as restorative in nature.
Why do some people continue to seek out a daily nap across their lifetime, while others seem to evade the need altogether? There are likely individual differences in napping behavior that are related to both sleep need and the degree of benefit gained (e.g., mood enhancement, memory consolidation). This review paper will examine research on the benefits of napping for healthy adults who have typical sleep/wake schedules rather than in those with sleep or circadian rhythm disorders. An overview of the research illustrating the benefits of naps and a description of strategies used to investigate naps is provided. Factors thought to impact the benefits of napping in healthy adults, such as the timing and duration of the nap, sleep inertia, age, and experience with napping are then reviewed in terms of their influence on mood, arousal, and performance.
Benefits of napping
Benefits of napping for waking performance have been confirmed by many researchers (Table 1). For instance, several studies have provided well-documented evidence of the benefits of naps during total sleep deprivation [e.g., Bonnet (1991); O’Connor et al. (2004); Song et al. (2002)] and in nightshift workers (Purnell et al., 2002; Sallinen et al., 1998). Other studies have documented the benefits of naps during partial sleep deprivation [e.g., Gillberg (1984); Takahashi and Arito (2000)]. Appetitive napping has also been shown to have benefits for non-sleep-deprived groups (Betrus, 1986). These studies have employed a variety of measures to show that napping leads to subjective and behavioral improvements. Napping also improves mood and subjective levels of sleepiness and fatigue. It is particularly beneficial to performance on tasks, such as addition, logical reasoning, reaction time, and symbol recognition.
Table 1. Overview of key studies documenting the benefits of napping
|Bonnet (1991)||104 healthy males aged 18–30 years||Between-groups comparison of 2-, 4-, and 8-h naps, with no nap during two nights of sleep deprivation||Naps improved vigilance, addition, logical reasoning, and alertness, only on the first night|
|Song et al. (2002)||8 male medical students aged 20–22 years||30-min naps during 40 h of sleep deprivation||Naps improved reaction time, but no change in accuracy|
|O’Connor et al. (2004)||41 healthy individuals (37 male) aged 21–47 years||Between-groups comparison of two 2-h naps, one 2-h nap, and no nap, per 24 h during 3.7 days of sleep deprivation||Dose–response trend with neurobehavioral performance improving most for the two 2-h nap group, and least for the no-nap group|
|Sallinen et al. (1998)||14 male industry workers aged 31–52 years||Within-subjects comparison of 30- and 50-min naps, at both 01:00 and 04:00 hours, with no nap||Naps reduced number of lapses on reaction time task and reduced physiological sleepiness and subjective fatigue|
|Purnell et al. (2002)||24 male engineers aged 21–59 years||Within-subjects comparison of 20-min and no-nap conditions during shift work||Naps improved performance on reaction time and vigilance tasks but no change in subjective sleepiness|
|Smith et al. (2007)||9 hospital workers (3 male), mean age 45.7 ± 13.2 years||Within-subjects comparison of 30-min and no-nap conditions during shift work||Naps improved psychomotor speed and subjective sleepiness; these improvements persisted to the end of the shift|
|Gillberg (1984)||12 healthy males aged 20–25 years||Within-subjects comparison of 1-h naps at 21:00 and 04:30 hours, with no nap after a night restricted to 4 h of sleep||Naps improved reaction time, reduced sleep latency, and reduced subjective sleepiness|
|Takahashi and Arito (2000)||12 healthy students (7 male), mean age 22.1 ± 1.6 years||Within-subjects comparison of 15-min and no-nap conditions after a night restricted to 4 h of sleep||Naps improved accuracy but reaction time was unaffected|
|Betrus (1986)||17 young adults (8 male), aged 18–32 years||Within-subjects comparison of 10- and 30-min naps, with 30-min bedrest and control-active conditions||Naps increased vigor, decreased fatigue, decreased confusion, decreased reaction time, and increased number of correct additions|
Naps as a counter-measure to sleepiness
A number of studies have investigated benefits of napping in comparison to other counter-measures of sleepiness such as caffeine and stimulant medication. In 1995 Bonnet et al. (1995) compared the effects of a prophylactic nap with those of caffeine, and found that benefits derived from a nap were less variable and lasted longer than those from caffeine. Bonnet and Arand (1994, 2000) and Reyner and Horne (1997) found that the combined treatment of a nap and caffeine was superior to either alone. Hayashi et al. (2003) compared napping, caffeine, bright light, and face-washing as means to combat mid-afternoon sleepiness. They found that the combined treatment of a nap and caffeine was superior in alleviating subjective sleepiness and aiding performance.
Batejat and Lagarde (1999) conducted a similar study, comparing the effects of naps, a stimulant (i.e., modafinil), and the combined treatment on performance during sleep deprivation. The authors found that naps did lead to cognitive performance benefits, and that these benefits were further enhanced when modafinil was combined with a nap. Benefits included improved reaction time and enhanced learning on a tracking task. These results are similar to those found in other studies with the stimulant caffeine.
Caldwell and Caldwell (1998) used a slightly different approach to compare a nap taken using a non-benzodiazepine hypnotic (i.e., zolpidem tartrate) to placebo nap and no-nap conditions. The nap induced by the drug showed greater attenuation of sleep deprivation-related deficits compared with both no-nap and placebo nap conditions. More specifically, both naps decreased sleepiness as measured by the repeated test of sustained wakefulness and Visual Analog Scales; the naps improved mood as measured with Visual Analog Scales and improved vigor scores; and the naps improved reaction time and accuracy on a flight simulation task. Zolpidem-aided naps were equivalent or superior to placebo naps in all cases. The authors hypothesized that this outcome resulted from differences in the amount of total sleep time, where more sleep was obtained with the drug. More specifically, individuals in the zolpidem-nap condition experienced faster sleep onset, less Stage 1, and more Stage 4 compared with the placebo condition.
Predicting benefits from nap EEG
A limited number of studies have investigated sleep architecture as a way to predict the type and degree of benefits gained from a nap. In a study using only habitual nappers, Taub et al. (1976) equated postnap benefits with sleep itself and not with a particular stage of sleep or nap duration. Subsequently, however, Taub (1979) reported that sleep architecture co-varied with performance and fatigue measures. Specifically, he found that Stage 4 sleep was associated with increased sleepiness, and rapid eye movement (REM) sleep with decreased sleepiness. More recently, Hayashi et al. (2005) showed that performance and alertness decreased through the postlunch dip when no nap was taken. Naps containing Stage 2 sleep improved performance and alertness. Naps containing Stage 1 but not Stage 2 sleep were successful in improving subjective alertness and reducing fatigue, but participants experienced performance decrements and showed evidence of decreased objective alertness. Tucker et al. (2006) showed that memory is dependent on the sleep architecture of the nap. Specifically, they showed that a nap helps to improve declarative memory, but that a nap and a period of rest were equivalent in terms of procedural memory improvement. They reported that the degree of declarative memory improvement was related to the amount of slow wave sleep (SWS) in the nap. Despite these recent findings, there is a definite lack of studies investigating both how the architecture of a nap differs from nocturnal sleep architecture, and the effect of sleep architecture on postnap measures. Future research would benefit from continuing to include data about the structure of the nap itself.
More subtle differences in sleep electroencephalogram (EEG) may be investigated using quantitative EEG measures. Few studies have examined quantitative EEG during a nap in relation to outcome or performance. Dijk et al. (1987) used naps to study how sleep EEG varied according to amount of prior wake time, but they did not use any performance measures in their study. They reported that for a sample of six non-habitual nappers, delta and theta power increased with increasing wake time prior to the nap, indicating deeper sleep followed by more time awake. Sleep onset latency, however, followed a circadian pattern, reaching a minimum in mid-afternoon. Hirose and Nagasaka (2003) examined quantitative EEG in eight healthy young adults during a 15-min nap opportunity, and found that increasing theta (4–7 Hz), high alpha (11–13 Hz), and low beta (14–20 Hz) power predicted performance improvements following the nap. The authors suggested that this pattern of EEG during the nap was related to the presence of Stage 2 sleep; that is, the presence of Stage 2 would lead nappers to feel that they had obtained sufficient sleep, in contrast to individuals whose naps consisted of only wakefulness and Stage 1. In another study, Schmidt et al. (2006) used a repeated measures design to investigate learning-dependent changes in the EEG during a 4-h daytime sleep. They compared declarative memory performance on easy and difficult word-pair learning tasks with a control condition. Results indicated that density of sleep spindles and sigma power during the nap was positively correlated with increases in memory performance from pre- to postnap.
These studies suggest that quantitative EEG measures of arousal have real potential to explain the improvements following a nap. It is plausible that the character of the nap may predict how well mood, alertness, and performance will be enhanced after the nap. As with sleep architecture, there is a need to include this finer objective measure of sleep in future studies investigating the benefits of naps.
Measuring nap benefits with quantitative EEG techniques
Some napping studies have also used quantitative electrophysiological measures following a nap as measures of alertness. For example, Macchi et al. (2002) reported that arousal was increased following a nap, as evidenced by lower theta and alpha power. As well, examining waking event-related potentials (ERPs) following a nap allows investigation of information processing and attention changes resulting from the nap. Using quantitative EEG and ERPs, Takahashi and Arito (1998) reported a positive correlation between delta power in non-REM sleep during 15- and 45-min naps and latency of the P300 ERP component 30 min after awakening. The authors suggested that sleep inertia (i.e., grogginess) due to delta activity during the nap prolonged P300 latency shortly after the nap, indicating delayed information processing. Three hours after awakening, P300 latency was shortened but this was not related to quantitative EEG measures. Takahashi et al. (1998) used the same ERP paradigm to determine if nap duration (i.e., 0, 15, or 45 min) had an effect on waking P300 latency or amplitude. They found that P300 latency was shortened after the 15-min nap compared with no nap or the 45-min nap. The authors attributed the lack of improvement after the 45-min nap to sleep inertia. P300 amplitude was not affected by nap duration. Takahashi and Arito (2000) next examined P300 latency and amplitude following a nap taken from 12:30 to 12:45 hours after 4 h of nocturnal sleep. The authors found that napping shortened P300 latency but amplitude did not change compared with the no-nap condition. In both studies, attention was enhanced through napping. Song et al. (2002) showed that taking naps during sleep deprivation prevented P300 latency from becoming delayed as much as during sleep deprivation with no nap, therefore preventing delayed information processing. Thus, EEG and ERPs are useful to describe both the quality and depth of nap sleep as well as levels of attention and information processing following a nap. In sum, various studies have confirmed the beneficial effects of naps on measures, such as vigor, alertness, and cognitive performance. These benefits in healthy adults may be influenced by the following factors which are reviewed below: circadian placement of the nap, duration of the nap, sleep inertia, age, and experience with napping.
Timing of the nap
Investigators have systematically varied the time of day at which a nap is taken. Circadian rhythms ‘reflect 24-h cycles of increases and decreases in a range of biological and physiological functions, including body temperature, heart rate, and hormone secretion’ (Hasher et al., 2002, p. 200). These rhythms are also seen in behavioral indices such as cognitive functioning (Hasher et al., 2002; Higuchi et al., 2000) and mood (Giannotti et al., 2002).
Broughton (1989) summarized that there is well-documented evidence for an increase in sleep tendency in the afternoon, making naps most likely at this time of day. Lavie and Weler (1989) demonstrated that sleep efficiency was better, sleep latency shorter, and amount of SWS greater in a nap taken closer to the afternoon circadian dip in alertness (15:00–17:00 hours) compared with a nap taken during the evening forbidden zone for sleep (19:00–21:00 hours). However, Naitoh (1981) cautioned that naps taken at the wrong time could lead to prolonged sleep inertia.
As naps are influenced by circadian rhythms (Lockley and Skene, 1997), one research strategy involves varying the time of day a nap takes place during sustained wakefulness. For example, in a study by Dinges et al. (1987), 41 healthy, young adults were assigned to one of five nap groups after 6, 18, 30, 42, or 54 h of wakefulness. The groups were all permitted a 2-h nap during 56 h of sleep deprivation. All groups showed improved reaction time performance following the nap, although subjective sleepiness showed the expected increase as sleep loss increased. Naps taken after 6 or 18 h of wakefulness showed more benefits than those taken after 30, 42, or 54 h of wakefulness. As well, naps taken after fewer hours of wakefulness prevented a mean drop in body temperature. Dinges et al. (1987) concluded that the benefits seen after earlier naps resulted from the fact that naps taken earlier prevented the drop in body temperature seen with extended wakefulness. As naps taken later in the day were associated with greater amounts of wakefulness, a 2-h nap may have been insufficient to alleviate the decrements caused by the long period of wakefulness. Hence, later naps were not sufficient to reduce sleep pressure enough to generate noticeable changes.
In a more recent line of research, participants in each of three studies (Hayashi and Hori, 1998; Hayashi et al., 1999a,b) underwent a repeated measures design that consisted of a 20-min nap condition and a no-nap condition at 1-week intervals. As well, in each study, assessments comprised of EEG recordings, Visual Analog Scales for sleepiness and fatigue, performance tests (logical reasoning, alphanumeric detection, addition, and auditory vigilance), and self-rating of performance. Subjective sleepiness, subjective rating of performance, and EEG arousal were improved following a nap in mid-afternoon (14:00 hours; Hayashi et al., 1999b) and preceding the ‘postlunch dip’ (12:20 hours; Hayashi and Hori, 1998; Hayashi et al., 1999a) but objective performance measures were only improved following the later nap. While Dinges et al. (1987) found that naps following less wakefulness were better, Hayashi et al. (1999a) showed that a later nap was better for performance and alertness. Hayashi’s group (1999a) suggested that because Dinges et al. had 12 h between naps and only 2 h separated the timing of the naps taken in their studies, they could not compare the results. The timing of a nap will have a bearing on the type or amount of benefit observed. Moreover, for sleep-satiated individuals, it may be best to choose a later nap time (following more wake time) to experience maximum benefits, while sleep-deprived individuals are more likely to be extremely fatigued, and thus earlier naps are indicated.
Timing of testing
Research has also investigated performance at different clock times following a nap placed at a fixed time to determine how long performance benefits last. For example, in one study (Hayashi et al., 1999a), participants each underwent both a nap and no-nap condition at 12:20 hours, and completed a performance assessment battery at multiple testing times throughout the day (every 20 min from 10:00 to 18:00 hours except for 12:20 to 13:00 hours). As reported above, neither reaction time nor accuracy on logical reasoning, mathematics, or auditory vigilance tasks were improved following the nap. Additionally, although a time of day effect was found for sleepiness, fatigue, and motivation in this study, changes across the day were due to circadian effects alone and were apparent following both the nap and no-nap conditions. It would be interesting to determine the relative contributions of clock time versus duration of postnap wakefulness in these studies.
In summary, research with healthy, young adults, which has focused on the circadian timing of naps has not identified a ‘best’ time to nap, although the circadian dip in alertness (i.e., mid-afternoon) is also when sleep propensity is highest. The best time to nap might depend on factors, such as individual sleep need, stability and timing of sleep/wake schedule, morningness-eveningness tendencies, quality of sleep during the preceding night, quality of sleep during the nap, or amount of prior wakefulness. The benefits of napping, in general, likely depend on the interaction of these factors.
As well as manipulating the timing of a nap, a number of studies have systematically varied the length of a nap to determine the shortest nap that may be taken to achieve maximum benefits [e.g., Helmus et al. (1997); Lumley et al. (1986); Tietzel and Lack (2002a,b)]. At the same time, these studies aimed to determine the longest nap that may be taken without resulting in substantial sleep inertia. Many of these studies aimed at identifying practical recommendations for napping in the workplace.
A series of studies (Brooks and Lack, 2006; Tietzel and Lack, 2001, 2002a,b) has recently focused on identifying the ‘ideal’ nap duration, using healthy, young adults with no history of sleep difficulties as participants. In a first study, Tietzel and Lack (2001) allowed participants to sleep for 10 or 30 min. Benefits were seen following both a 10- and 30-min nap but a delay was needed to see benefits following the longer nap due to sleep inertia. As the longer nap did not prove to be superior, Tietzel and Lack (2001) hypothesized that benefits of napping depended not on the depth or length of sleep, but instead on another process, sleep onset. Here, simply initiating sleep would be sufficient to lead to measurable benefits. This idea was investigated in a subsequent study, where participants limited their sleep on the night prior to the study to 5 h. Tietzel and Lack (2002b) compared a 10-min nap with 30- and 90-s naps to determine if the onset of Stage 1 sleep produced postnap improvements. Thus, in this study, participants were awoken after 30 (i.e., 1 epoch) or 90 (i.e., 3 epochs) seconds of Stage 1 sleep (defined as less than 50% of the epoch characterized by alpha) or after 10 min of sleep from the onset of Stage 1 sleep. The 10-min nap improved subjective and objective alertness, decreased fatigue, increased vigor, and improved performance. No such effects were seen following either length of ultra-short nap, indicating that more than sleep onset is needed to make naps beneficial. These results are consistent with Process S, as described in the two-process model of sleep/wake regulation (Borbely, 1982), which predicts that longer sleep is required for restoration and improvements in waking performance and cognition in particular.
In a third study, Tietzel and Lack (2002a) used a dose–response paradigm to compare 5-, 10-, 20-, and 30-min naps, as well as no nap, to determine if brief naps were as effective as longer naps. They showed that the 10-, 20-, and 30-min naps produced improvements in cognitive performance and alertness, while the 5-min nap and no nap conditions did not. Moreover, the 10-min nap showed immediate benefits, while the 20- and 30-min naps led initially to sleep inertia. Performance was improved 5 and 35 min after the 10-min nap, 95 min after the 20-min nap, and 35 and 155 min after the 30-min nap. Three hours postnap, there were no differences among all five conditions. Thus, benefits following the 10-min nap could be seen immediately after the nap, while a delay following naps of longer durations was needed to allow sleep inertia to dissipate. As well, the benefits following the 10-min nap were superior at all testing times compared with those following either longer nap. Consistent with the two-process model of sleep/wake regulation, it would be expected that longer naps would allow more opportunity for slow wave activity, reducing sleep pressure, thereby producing greater waking benefits. Conversely, increased slow wave activity would also lead to more sleep inertia, thereby delaying benefits. In general, the 10-min nap was more beneficial than longer naps of 20 and 30 min at immediate and delayed testing times, although all three nap lengths were shown to be of some benefit.
Brooks and Lack (2006) published a fourth study using a similar paradigm, with the goal of identifying the underlying EEG changes that accompanied restorative nap benefits. Participants slept for 5, 10, 20, or 30 min in the afternoon or participated in a no-nap condition, all following a night restricted to 5 h of sleep. Results indicated that the 5-min nap was equivalent to the no-nap condition, the 10-min nap produced immediate benefits that were maintained, and the longer naps showed benefits emerging later in the test bout. Benefits were attributed to the presence of delta waves in the nap based on visual scoring of the sleep. Utilizing quantitative EEG in future studies would allow a finer measure of delta activity in the nap.
Lahl et al. (2008) used a dose–response paradigm to investigate the benefit of a daytime nap on memory performance. In an initial study, they reported that naps were superior to a waking condition for memory consolidation, but they were unable to find a relationship between memory performance and any particular aspect of the nap. In a second study, they compared waking with short and long naps and found a dose–response trend such that both naps were superior to waking and that the long nap was superior to the short nap. Again, there was no particular aspect of the nap which predicted improvement in memory.
Other studies have employed participants experiencing daytime sleepiness. Helmus et al. (1997) investigated the effects of nap duration on daytime sleepiness, using the Multiple Sleep Latency Test (MSLT), in narcoleptic, sleep-deprived, and sleep-satiated participants. Narcoleptic and sleep-satiated participants spent 8 h in bed prior to the study, while sleep-deprived participants spent no time in bed. The authors found that all groups had longer sleep latencies on the MSLT following a 120-min nap, showing that sleepiness was reduced following the longer nap compared with a 15-min nap. No evidence was found to determine if the pathologically sleepy group (i.e., narcoleptics), the experimentally sleep-deprived group, and the sleep-satiated group experienced different benefits of the nap.
In a dose–response paradigm, Lumley et al. (1986) employed a sleep latency measure to compare 15-, 30-, 60-, and 120-min recovery naps with a no-nap condition, each following a night of total sleep deprivation, in a repeated measures design with 10 healthy young adults (18 to 32 years old). Lumley et al. discovered that alertness increased with nap length, although alertness levels following the two longest nap durations were equivalent. Lumley et al. remarked that the extra 60 min in the 120-min nap was predominately REM sleep. As this extra sleep did not improve performance over and above the 60-min nap, REM sleep during a nap may not improve alertness measured by sleep latency tests. However, REM sleep may benefit individuals in other ways; for example, performance tasks involving learning and memory might show greater benefits following REM-containing naps [e.g., Mednick et al. (2003)].
In general, based on the research outlined above, healthy young adults should ideally nap for approximately 10 to 20 min [e.g., Hayashi and Hori (1998); Hayashi et al. (1999b); Tietzel and Lack (2002a)]. These short naps are ideal for workplace settings where performance immediately upon awakening is normally required. It seems that naps containing REM sleep may not contribute to alertness per se, but may play a role in specific types of memory consolidation gleaned during a nap.
Longer naps (e.g., sleeping for 30 min or longer) produce sleep inertia, making nap benefits obvious only after a delay. Sleep inertia is characterized by a reduction in the ability to think and perform upon awakening due to sleep. Confusion, grogginess, and deficits to cognitive performance may be associated with this transition (Dinges, 1993). Sleep inertia is associated with longer naps as it is typically thought to result from awakening from SWS (Dinges, 1993). For example, Tietzel and Lack (2001) suggested that although a 10-min nap showed no sleep inertia effects, a 30-min nap did show the effect because 7.5 times more SWS occurred in the longer nap.
In assessing the influence of sleep inertia, it is important also to consider the timing of the nap. Sleep inertia is worse in the circadian trough than the peak, coinciding with the time when sleep is more likely (Caldwell and Caldwell, 1998; Dinges et al., 1985). It is possible that this phenomenon is associated with the stage of sleep from which the individual awakens (i.e., sleep architecture; Tassi et al., 1992) and/or depth of sleep throughout the sleep period (Jewett, 1997). In particular, sleep is deeper at this time (Jewett, 1997), creating more sleep inertia. The amount of wakefulness prior to the nap or the extent of sleep deprivation (Dinges et al., 1987) may also bear on the level of sleep inertia.
Dinges (1993) noted that cognitive impairment due to sleep inertia was worse than that due to sleep deprivation, and that impairment due to sleep inertia is smaller for psychomotor and perceptual tasks than for information processing. In a recent experimental investigation of sleep inertia, Ferrara et al. (2000) reported that sleep inertia impairs sensory motor and cognitive tasks, is characterized by confusion and disorientation, worsens when individuals are sleep deprived, impairs accuracy more than speed, and varies in duration and severity with circadian phase.
Reports have been inconsistent with respect to the length of sleep inertia following awakening [e.g., Dinges et al. (1981)]. It is important to have an estimate of sleep inertia duration because it must be controlled when testing how performance, arousal, and mood are affected following a nap; that is, these effects must be tested after sleep inertia has dissipated. Ferrara and De Gennaro (2000) found that sleep inertia impaired accuracy more than speed, while Hofer-Tinguely et al. (2005) observed the opposite, that speed was impaired more than accuracy. These differences may be related to the specific tasks measured and/or differences in the timing of measurement. A number of factors likely influence the degree to which sleep inertia is experienced, including subject and test characteristics and homeostatic and circadian factors.
The benefits of napping in older adults
Napping has been shown to benefit young adults. As older adults may be sleepier and have more opportunity to nap due to retirement, it is likely that napping is both more frequent in older adults, and perhaps more beneficial to this group. In a previous review paper, Takahashi (2003) focused on the role of napping in individuals who typically experience sleep loss such as shift workers and those with sleep and medical disorders. While a subset of older adults certainly have sleep disorders and poor health, even healthy older adults who experience age-typical changes to their sleep can benefit from appetitive naps.
Research on napping in older adults is quite limited and is typically confined to survey data used to describe the frequency of napping in such a population. For instance, Webb and Swinburne (1971) reported that almost all individuals in a small sample of 19 elderly adults took naps; they suggested that some napping was appetitive. Beh (1994) and Ohayon and Zulley (1999) found that the prevalence of napping increased linearly with age which was also shown with the 2003 ‘Sleep in America’ poll (National Sleep Foundation, 2003). For adults aged 65 and over, Beh reported that nap frequency increased with each 5-year age range, and that a significantly greater proportion of the people aged 75 and above took a daily nap compared with those 70 years old and younger. Ohayon and Zulley similarly found that 22% of individuals aged 15 to 99 years napped at least twice a week, but this number increased to 53% for those aged just 75 and above.
Several studies have investigated the benefits of napping for older adults in particular. Tanaka et al. (2001, 2002) investigated the impact of a nap on the well-being of older adults and found that the combination of a 30-min afternoon nap and moderate intensity exercise in the evening improved sleep quality (i.e., reduced wake after sleep onset, increased sleep efficiency) and reduced subjective daytime sleepiness. The authors also showed that mental health was improved after this intervention. Creighton (1995) demonstrated that some cognitive improvements were apparent when naps were taken by older hospital patients. Participants completed measures of alertness, concentration, strength, coordination, and reaction time. Patients improved on all but reaction time, which is contrary to the napping literature in young adults where response time is typically improved following a nap, e.g., Song et al. (2002). Tanaka and Shirakawa (2004) reported that short naps and moderate exercise produce not only improvements in sleep and well-being but also improved performance on a verbal memory test. Campbell et al. (2005) also confirmed that napping improved performance on a logical reasoning task, two-letter search task, Stroop congruency task, and Wilkinson four-choice reaction time task, in 55 to 85 year olds, with benefits emerging immediately after the nap and throughout the next day. Tamaki et al. (1999, 2000) employed a repeated measures design to assess the benefits of napping in older adults. They demonstrated that with nap and resting conditions 1 week apart, a daytime nap improved performance on a visual detection task and decreased subjective and objective sleepiness and fatigue in older individuals.
A recent study by Milner and Cote (2007) investigated whether naps of different lengths benefited younger, middle-aged, and older adults in different ways. Given that sleep is typically less restorative with advancing age, they hypothesized that older adults might need a longer nap to accrue the same benefits as younger adults. The authors confirmed that naps are beneficial for mood and cognitive performance but found that older adults are able to garner as much benefit from a nap as their younger counterparts. Interestingly, however, quantitative EEG measures collected during wakefulness suggested that older adults had to put forth more effort to achieve the same results on performance tests.
In summary, research examining napping in older adults has shown that they nap more frequently than their younger counterparts, and that napping is beneficial to older adults who nap for both restorative and appetitive reasons. Specifically, research has confirmed that napping leads to improvements in well-being and cognitive performance for older adults.
Experience with napping
Very little research has been conducted to investigate individuals’ levels of experience with napping (i.e., whether they habitually nap or not). Most studies such as those by Taub et al. (1976, 1977) and Taub (1979) utilized only habitual nappers simply to ensure that participants slept in the laboratory environment. Anecdotally, it seems that there are individuals who avoid napping and those who must have a daily nap. Adults may choose to nap on a regular basis for many reasons, including perception of sleepiness, opportunity (Cottrell and Hildebrandt Karraker, 2002), and quality and amount of nocturnal sleep (Chan et al., 1989; Cottrell and Hildebrandt Karraker, 2002). A standard definition of habitual napping remains lacking but several operational definitions based on frequency of napping have been put forth (Table 2).
Table 2. Definitions of habitual and non-habitual nappers
|Lawrence and Shurley (1970)|
| Collected information on frequency of napping, typical nap duration, and typical nap time from 505 male and female undergraduates, aged 16–45|
| 78.6% of students napped when not working|
| 4.5% of students napped 4 or more days per week; 6.5% napped 3 days per week; 15.6% napped 2 days per week; 24.9% napped 1 day per week; all others had variable napping frequencies|
| Approximately one-third of students napped for 1 h, one-third for 2–3 h, one-third had no consistent nap duration|
| Over half the students napped in the afternoon|
|Taub et al. (1976)|
| Habitual nappers were those who napped in the afternoon one or more times per week, for 0.5–2 h, for at least 2 years|
|Evans et al. (1977)|
| Nappers were those who sometimes, usually, or always took naps during the day, and they also found napping to have restorative effects (less sleepy, more satisfied, and more subjective benefit after a 60-min nap than non-nappers)|
| Non-nappers were those who rarely or never napped|
| 61 to 70 year olds|
| Habitual nappers defined as those who took a daily daytime nap (about half the sample)|
| Only habitual nappers awoke refreshed from a nap|
|Bliwise and Swan (2005)|
| Habitual nappers were characterized by older age, higher depression scores, and poorer performance on the trail making test|
|Milner et al. (2006)|
| 70.8% of 137 first-year undergraduates were habitual nappers (napped every day or once or twice a week)|
| 29.2% of respondents were non-habitual nappers (napped at most once or twice a month)|
An interesting line of research is to examine the structure of the nap in these two groups. Individuals who nap frequently and those who do not nap may differ in their ability to sleep during a daytime nap, the quality of their sleep, and/or the benefits derived from napping. Habitual nappers might be predisposed to be good daytime nappers, or they may have learned to become skilled nappers through practice, perhaps being rewarded by increased well-being following the nap. Non-habitual nappers might self-select out of daytime napping because they experience no benefit. They may be unable to fall asleep, awaken too often, or sleep lightly; they may also sleep too deeply, thus receiving negative feedback from sleep inertia following the nap.
Dinges (1992) compared nap sleep architecture of habitual and non-habitual nappers. Habitual nappers were found to have more Stage 1 and more stage shifts, indicating that non-habitual nappers had more consolidated sleep. Those who do not regularly nap may therefore experience greater sleep inertia, perhaps explaining why some people choose not to nap. Furthermore, it may be the case that habitual nappers are more sensitive to ultradian rhythms (i.e., biological rhythms with a periodicity shorter than 24 h), which allows them to fall asleep more easily in the daytime (Schulz, 1993). Dinges (1992) also reported that habitual nappers had lower prenap body temperatures on days when they napped compared with days when they did not nap, and lower prenap body temperatures compared with non-habitual nappers on nap days. Although Dinges suggested that this effect reflected the preparatory response of habitual nappers to take a nap, it seems equally likely that habitual nappers take naps in response to this lowered temperature.
Differences between habitual and non-habitual nappers may also be due to or reflected in underlying EEG differences. For instance, EEG power values that reflect depth or character of sleep may differ between these groups. In addition, differences in sleep phasic events such as k-complexes and sleep spindles may represent individual differences in sleep-related information processing such as consolidation of newly learned skills. In a study investigating these issues, Milner et al. (2006) compared habitual and non-habitual nappers on a measure of motor procedural learning. Power spectral analysis performed on the EEG during the nap revealed that habitual nappers had more alpha EEG frequencies. The fact that they were able to maintain a tonic level of alertness throughout the nap suggests that habitual nappers may be able to disperse sleep inertia more quickly and thereby perform task in a shorter time frame. Although previous research has confirmed that naps result in benefits to mood, alertness, and cognitive processing in habitual nappers, other studies have documented no performance differences between habitual and non-habitual nappers [e.g., Daiss et al. (1986); Keyes (1989)]. Milner et al., however, demonstrated that motor procedural learning was impaired for non-habitual nappers who took a nap in lab. Postnap performance improvement on the motor procedural task was predicted by the number of sleep spindles and sigma power (13.5–15 Hz) during the nap; importantly, this finding held true for habitual nappers only. Additional research of this type would help to elucidate the effect of previous napping experience on the benefits that can be gained from a nap.
In conclusion, the existing literature shows that certain variables, such as the timing and duration of a nap, age, and experience with napping are important moderators of the benefits of naps. Information gained from the study of napping impacts our understanding of the functions of sleep itself. As researchers are beginning to employ practical daytime napping paradigms to investigate the role of sleep in learning and memory [e.g., Mednick et al. (2003); Tucker et al. (2006)], it will be important for the validity of these results to consider the influence of timing and duration of the nap, the timing of pre- and postnap testing sessions, and the participant characteristics. There is still a great deal of basic research to be carried out in the area of napping, all of which will better inform the application of prescribed napping in sleepy and applied populations. The following is a research agenda for the investigation of benefits of napping:
- • Identifying the ideal time of day to nap.
- • Assess the time course of benefits following naturalistically longer naps.
- • Investigate factors that contribute to why healthy, sleep-satiated individuals become habitual or non-habitual nappers.
- • Determine additional individual difference variables that influence the benefits derived from a nap.
- • Obtain more specific information on the types of cognitive domains and performance tasks that are improved following napping.
- • Gain more insight into the mechanisms responsible for the robust improvement in mood and affect regulation that consistently follows a nap.
- • Gain a better understanding of the significance of subjective perception of improved alertness and well-being following a nap.
- • Apply more quantitative electrophysiological methods such as EEG and ERPs to understand the neurophysiological basis of nap-related improvements.
- • Investigate the contribution of sleep phasic events such as k-complexes and spindles to the sleep-dependent changes in performance.
- • Use multiple channel EEG/ERPs and imaging techniques to localize neural sources driving the immediate benefits in arousal, mood, and performance that can be achieved from such a short bout of sleep.
The Brock University Sleep Research Laboratory is funded by the Natural Science and Engineering Research Council (NSERC) of Canada.