Pii: s0278-6915(02)00096-0

Food and Chemical Toxicology 40 (2002) 1243–1255 Center for Occupational and Health Psychology, School of Psychology, Cardiff University, UK The literature suggests that the following effects on behavior of adult humans may occur when individuals consume moderate amounts of caffeine. (1) Caffeine increases alertness and reduces fatigue. This may be especially important in low arousal situations(e.g. working at night). (2) Caffeine improves performance on vigilance tasks and simple tasks that require sustained response.
Again, these effects are often clearest when alertness is reduced, although there is evidence that benefits may still occur when theperson is unimpaired. (3) Effects on more complex tasks are difficult to assess and probably involve interactions between the caf-feine and other variables which increase alertness (e.g. personality and time of day). (4) In contrast to the effects of caffeine con-sumption, withdrawal of caffeine has few effects on performance. There is often an increase in negative mood following withdrawalof caffeine, but such effects may largely reflect the expectancies of the volunteers and the failure to conduct ‘‘blind’’ studies. (5)Regular caffeine usage appears to be beneficial, with higher users having better mental functioning. (6) Most people are very goodat controlling their caffeine consumption to maximise the above positive effects. For example, the pattern of consumption over theday shows that caffeine is often consumed to increase alertness. Indeed, many people do not consume much caffeine later in the daysince it is important not to be alert when one goes to sleep. In contrast to effects found from normal caffeine intake, there arereports that have demonstrated negative effects when very large amounts are given or sensitive groups (e.g. patients with anxietydisorders) were studied. In this context caffeine has been shown to increase anxiety and impair sleep. There is also some evidencethat fine motor control may be impaired as a function of the increase in anxiety. Overall, the global picture that emerges depends onwhether one focuses on effects that are likely to be present when caffeine is consumed in moderation by the majority of the popu-lation or on the effects found in extreme conditions. The evidence clearly shows that levels of caffeine consumed by most peoplehave largely positive effects on behavior. Excessive consumption can lead to problems, especially in sensitive individuals. # 2002Published by Elsevier Science Ltd.
Keywords: Caffeine; Mental performance; Mood; Attention; Memory; Caffeine withdrawal sumed from food and drinks, and the very differenteffects observed with excessive amounts or in very sen- The aim of the present article is to review the effects sitive individuals. Unlike other areas of research (e.g.
of caffeine on human behavior. The main areas of studies of health effects), most studies of the behavioral behavior reviewed are mood, mental performance and effects of caffeine have examined acute changes follow- sleep. Certain areas, although related to behavior, such ing a single dose. Less is known about effects of regular as the underlying CNS mechanisms, are not reviewed in consumption, although there are now enough data on detail. This is because most of the research in this area this topic to draw tentative conclusions. In addition to has involved animal studies. These have provided plau- studying the effects of caffeine consumption, the research sible mechanisms for many of the effects observed in has also considered possible changes in behavior as a humans, but whether specific details (e.g. effects of dose) function of caffeine withdrawal. A critical appraisal of generalize across species is unknown. Other areas, such as claims that caffeine withdrawal influences performance the claims that caffeine is addictive, are not covered here but are discussed in a paper by Dews et al. in this issue.
It is important to note that in research reviewed in all In all areas it is important to make a distinction between subsequent sections, there is considerable variation in the effects of amounts of caffeine that are normally con- the methodology and measuring devices. This has thedisadvantage that it is often difficult to compare differentstudies. However, persistence of effects across a range of * Tel.: +44-2920-874757; +44-2920-874758.
methodologies enhances the validity of the reported E-mail address: smithap@cardiff.ac.uk (A. Smith).
0278-6915/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
P I I : S 0 2 7 8 - 6 9 1 5 ( 0 2 ) 0 0 0 9 6 - 0 A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 masked by increases in negative mood. Indeed, a num-ber of results suggest that caffeine may increase anxiety A large number of studies have shown that consump- Anecdotal evidence suggests that when individuals tion of caffeine leads to increased alertness (or reduced have an excessive amount of caffeine they may become fatigue). These effects have often been demonstrated anxious. Similarly, some psychiatric patients attribute using paradigms involving low alertness situations (e.g.
their problems to consumption of caffeine, which has following administration of benzodiazepines—Johnson led to a diagnosis of ‘‘caffeinism’’. Other patients, espe- et al., 1990; early morning—Smith et al., 1992; working at cially those with anxiety disorders, report that caffeine night—Smith et al., 1993a; when the person has a cold— may exacerbate their problems. The validity of these Smith et al., 1997a; and sleeploss—Bonnet et al., 1995).
questions will now be assessed by consideration of the However, beneficial effects of caffeine have been demon- strated in individuals in an alert state (e.g. Leathwood and Lieberman (1992) stated that ‘‘. . . it appears that caf- Pollet, 1982; Rusted, 1994, 1999; Smith et al., 1994a,b; feine can increase anxiety when administered in single Warburton, 1995). Many of these studies have used quite bolus doses of 300 mg or higher, which is many times high doses of caffeine (e.g. 250 mg—Johnson et al., greater than the amount present in a single serving of a 1990; 3 mg/kg—Smith et al., 1994a,b), which would not typical caffeine-containing beverage. However, in lower be consumed typically in a single drink in real-life doses it appears to have little effect on this mood-state situations. However, other studies have demonstrated or, under certain circumstances, it may even reduce similar effects with realistic doses (e.g. Leathwood and anxiety levels. It has also been observed that caffeine Pollet, 1982; Warburton, 1995). These results have reduces self-rated depression when administered in implications for many practical situations in that safety is often at risk when alertness is low. However, a better The literature supports Lieberman’s view since only a impression of this will be seen when the data from per- small proportion of the studies reviewed show increases formance tasks are considered. Many of the studies in anxiety following administration of caffeine. Stern et have administered caffeine in coffee and it is unclear al. (1989) found that individuals who choose a high dose whether it is the caffeine alone, or caffeine in combina- of caffeine reported positive mood changes whereas tion with other compounds in the coffee, which under- non-choosers reported anxiety and dysphoria. Loke lies the behavioral effects. Recent research (Smith et al., (1988) found that caffeine reduced fatigue but also led 1999b) has shown that it is the caffeine rather than the to increased tension and nervousness. Increased anxiety combination of the caffeine and the type of drink in was also reported following caffeine by Loke et al.
which it is presented that is important. Similar results (1985) where the doses were high (either 3 or 6 mg/kg).
have also been demonstrated with caffeine given as a Similarly, Sicard et al. (1996) found increased anxiety capsule and in a drink. Other research also shows few following 600 mg of caffeine. Green and Suls (1996) also differences between decaffeinated coffee and drinks such found that caffeine increased anxiety, and again the as fruit juice. All of these pieces of evidence point to volunteers were consuming very high amounts (125 mg caffeine as the main determinant of the behavioral caffeine per cup of coffee over the day). Overall, these results suggest that increases in anxiety following caf- One must now consider why some studies have failed feine are often only found following consumption of to find effects of caffeine on alertness. For example, amounts that would rarely be ingested by the majority Svensson et al. (1980) found no effect of 100 mg of caf- feine on the mood state of 23 volunteers. Similarly, It is now important to assess whether caffeine leads to Swift and Tiplady (1988) found no effect of 200 mg of mood problems when the person ingesting it already has caffeine on the mood of elderly volunteers. This lack of a high level of anxiety. It has been claimed that some effect could possibly reflect sample size or other details people abstain from caffeinated drinks because of the of the methodology. Lieberman (1992) suggests that accompanying jitteriness and nervousness (Goldstein et beneficial effects of caffeine on alertness are most easily al., 1969). Other authors have even gone as far as to demonstrated when circadian alertness is low and mood suggest that caffeine acts as a ‘‘fairly convincing model is measured in the context of doing demanding perfor- of generalised anxiety’’ (Lader and Bruce, 1986). Caf- mance tasks. Rusted (1999) also suggests that mood feinism refers to a constellation of symptoms associated effects occur after changes in performance, and this may with very high caffeine intake that are virtually indis- account for the absence of effects in certain studies.
tinguishable from severe chronic anxiety (Greden, 1974).
Another possible explanation of the failure to find Caffeinism is usually associated with daily intakes of positive mood changes in certain studies is that they are between 1000 and 1500 mg. However, it appears to be a A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 rather specific condition and there is little evidence for period of abstinence or the greater effects of caffeine correlations between caffeine intake and anxiety in when arousal is low. Finally, claims about the negative either non-clinical volunteers (Lynn, 1973; Hire, 1978) effects of caffeine withdrawal require closer examination or psychiatric outpatients (Eaton and Mcleod, 1984).
since they can often be interpreted in ways other than caf- Correlational studies of associations between caffeine feine dependence (e.g. expectancy—Smith, 1996a; Rubin and anxiety have a number of methodological pro- and Smith, 1999). Indeed, in most of the studies that have blems. Although it is better to examine the anxiety-pro- demonstrated increases in negative effect following caffeine ducing effects of a direct caffeine challenge, much of this withdrawal, the subjects have not been blind but have been research has used very high doses (e.g. 6.6 mg/kg— told or even instructed to abstain from caffeine. This is Veleber and Templer, 1984; 10 mg/kg—Charney et al., clearly very different from the double-blind methodology 1985), which makes it difficult to determine whether the typically used to study effects of caffeine challenge.
issue is relevant to doses more representative of normalconsumption. Other studies have used a caffeine chal- lenge with psychiatric patients with anxiety disorders(e.g. Bruce et al., 1992), although even with this group The preceding sections confirm the suggestions about there is little evidence that smaller doses exacerbate the effects of caffeine on mood that were made at the start.
existing anxiety (Mathew and Wilson, 1990).
Other research examined whether caffeine is capable of increasing the anxiety induced by other stressors.
 caffeine, even in low doses, may increase alert- Shanahan and Hughes (1986) found that 400 mg of ness and this is readily apparent in low arousal caffeine increased anxiety when paired with a stressful task. However, other research (e.g. Hasenfratz and  high doses of caffeine can lead to increased anxiety Battig, 1992; Smith et al., 1997b) has not been able to provide any evidence of interactive effects of caffeine  caffeine withdrawal increases negative affect but this may reflect expectancy effects. It is also Overall, the literature suggests that extremely high doses unlikely that this can account for the positive of caffeine may increase anxiety, but that this is rarely seen mood effects produced by caffeine when given to within the normal range of ingestive behavior.
non-consumers or to volunteers who have not The next section considers another area where caf- feine is claimed to be associated with adverse effects,namely when it is withdrawn.
The next section considers effects of caffeine on the Caffeine withdrawal has been widely studied because it is meant to provide crucial evidence on whether caf- feine is addictive or leads to some kind of dependence.
The most frequent outcome measure has been reporting of headache, but mood has been examined in otherstudies. Ratcliff-Crain et al. (1989) reported that caffeine There are already a number of extensive reviews (e.g.
deprivation led to increased reporting of stress by heavy Lieberman, 1992) which have summarized early studies coffee drinkers. This has recently been confirmed by of the effects of caffeine on human performance. This Schuh and Griffiths (1997), who found that caffeine section outlines the main points to emerge from such withdrawal was associated with feelings of fatigue and reviews, and is then followed by a critical evaluation of decreased feelings of alertness. Indeed, Silverman et al.
(1992) found that about 10% of volunteers with a moder-ate daily intake (235 mg per day) reported increased depression and anxiety when caffeine was withdrawn.
There is a great deal of evidence to shows that caffeine Other researchers (e.g. James, 1994) have argued that blocks the effects of the naturally occurring neuromo- caffeine has no beneficial behavioral effects but merely dulator adenosine (Snyder, 1984). This produces a net removes negative effects associated with caffeine with- increase in CNS activity because the inhibitory action of drawal. Smith (1995) has argued against this general view adenosine is blocked. The effects of caffeine on adenosine of caffeine effects on a number of grounds. First, it cannot activity, and the subsequent effects for neurotransmitters account for the behavioral effects seen in animals or non- such as norepinephrine, occur at concentrations found as consumers, where withdrawal cannot occur. Secondly, a function of dietary intake. Other mechanisms have been caffeine withdrawal cannot account for behavioral demonstrated (e.g. calcium mobilisation, prostaglandin changes following caffeine consumption after a short antagonism, phosphodiesterase inhibition), but these only A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 become relevant when caffeine is administered in doses vigilance, but there is an absence of data suggesting that that are at least 20–30 times higher than those found in the it may actually lead to impairments.
diet (Snyder, 1984). In humans peak plasma levels occur15–45 min after ingestion and its plasma half-life is 5–6 h.
Different mammals use different pathways to metabolise Although there have been a large number of early stud- caffeine, and many studies of animal behavior are, there- ies of effects of caffeine on more complex cognitive pro- fore, not relevant to its effects on humans.
cesses, it is hard to draw definitive conclusions. If, forexample, one considers verbal learning tasks then one 3.1.1.1. Early views of behavioral effect of caffeine. Most finds a large number of studies which have shown no effect consumers of caffeine regard it as a mild stimulant when of caffeine (e.g. Clubley et al., 1979; File et al., 1982; Battig consumed in moderate doses, as recognised over 400 et al., 1984; Loke et al., 1985). Many of these failures to years ago by Pietro della Vale (cited by Tannahill, detect effects on memory were carried out using a meth- 1989). Reviews written in the 1980s (e.g. Sawyer et al., odology that was sensitive enough to detect caffeine 1982; Dews, 1984) suggested that the effects were highly effects on psychomotor or sustained attention tasks (e.g.
variable and the subject of considerable controversy.
Lieberman et al., 1987; Roache and Griffiths, 1987).
Indeed, Dews (1984) has suggested that when caffeine is It would appear from these early studies that the administered in the doses found in foods, its effects are effects of caffeine on cognitive performance are often ‘‘so slight and subtle that the investigator is usually glad too small to detect as general groupeffects. Some posi- tive results have been obtained (e.g. Erikson et al., 1985) The above view probably reflects the numerous pro- but only in very specific conditions (when testing was blems associated with early studies of caffeine. First, conducted at a slow, not a fast rate). Again, however, many studies have used insensitive tests and designs. It there appears to be little evidence suggesting impair- is quite plausible that the effects of small doses of caf- ments following consumption of caffeine.
feine will be marginal and many of the negative resultsmay well reflect the absence of appropriate statistical power. This problem is often compounded by the use of Regina et al. (1974) examined the effects of caffeine on weak experimental designs (e.g. no baseline data) and a simulated driving task. The results showed beneficial failure to consider individual differences.
effects of caffeine and confirmed findings using labora-tory vigilance tasks. Studies conducted by the military (cited by Lieberman, 1992) have also shown that caf- 3.1.1.2.1. Sensory functions. Lieberman (1992) stated feine can improve a critical military task, namely sentry that ‘‘there is no evidence to suggest that moderate doses of caffeine have direct effects on sensory function,although well controlled studies using state-of-the-art 3.1.1.6. Caffeine and daytime sleepiness methods have not been conducted’’.
Beneficial effects of caffeine in low arousal situations will be considered more closely in a following section.
3.1.1.2.2. Simple and choice reaction time. There are a However, studies conducted in the 1980s clearly number of studies which have shown beneficial effects of demonstrated that caffeine can remove the sedative caffeine on simple reaction time (e.g. Clubley et al., effects of certain drugs (e.g. diazepam—File et al., 1982; 1979) and choice reaction time (Smith et al., 1977; Lie- berman et al., 1987; Roache and Griffiths, 1987). Otherstudies have demonstrated such effects in some groups 3.1.1.7. Caffeine, personality and time of day but not others (e.g. the elderly but not the young—Swift Research by Revelle and colleagues (see Revelle et al., and Tiplady, 1988) and with some doses but not others 1987, for a review) showed that caffeine facilitated the (beneficial effects with 200 and 400 mg but not 600 performance of impulsive individuals and impaired the performance of non-impulsive individuals taking com-plex cognitive tests in the morning. In the evening, the opposite pattern of results was observed. This has been Several papers suggest that caffeine, in moderate interpreted in terms of relationships between optimum doses, will improve vigilance (Regina et al., 1974; Club- levels of arousal and complex tasks performance. Such ley et al., 1979). Indeed, Lieberman (1992) suggests that effects do not appear with simple tasks, where even high beneficial effects can be observed with doses as low as 32 levels of alertness facilitate performance.
mg ‘‘when a sensitive test is employed in an appropriatetesting paradigm’’.
3.1.1.8. Adverse effects of caffeine on performance Other researchers (e.g. Loke and Meliska, 1984) have 3.1.1.8.1. Fine motor performance. Anecdotally, it has failed to demonstrate significant effects of caffeine on been suggested that the increased arousal induced by A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 consumption of caffeine impairs hand steadiness. How- of the population increase alertness and the ability ever, early studies failed to demonstrate such effects to sustain attention. Adverse effects occur when (e.g. Lieberman et al., 1987) or found them only in non- excessive doses are consumed or when caffeine is consumers (e.g. Kuznicki and Turner, 1986).
3.1.1.8.2. Caffeine withdrawal. Lieberman (1992) dis- cusses the effects of caffeine on headache and mood butcites no evidence to suggest that it influences performance.
Recent studies of effects of caffeine on performance This issue will be discussed again in a later section.
have confirmed many of the earlier results. For example,the beneficial effects of caffeine on psychomotor speed 3.1.1.9. A cost–benefit analysis of early studies of the and vigilance have been replicated (Frewer and Lader, 1991; Mitchell and Redman, 1992; Fine et al., 1994).
Lieberman (1992) reaches the following conclusions Similarly, the absence of groupeffects in more complex about the beneficial and adverse behavioral effects of cognitive tasks has also been observed (Loke, 1990, caffeine. ‘‘When caffeine is consumed in the range of 1992). Isolated reports of caffeine-induced impairments doses found in many foods, it improves the ability of are also to be found (e.g. decreased hand steadiness— individuals to perform tasks requiring sustained atten- Bovim et al., 1995). Research has continued to study the tion, including simulated automobile driving. In addi- effects of caffeine on attention tasks with one aim being tion, when administered in the same dose range, caffeine to identify mechanisms underlying the effects. For increases self-reported alertness and decreases sleepi- example, Smith et al. (1999a) and Smith (2001) have ness’’. ‘‘Adverse behavioral effects occur when caffeine shown that caffeine increases the speed of processing is consumed in excessive doses or by individuals who are new stimuli, confirming results reported by Streufert et overly sensitive to the substance’’.
al. (1997). Lorist and Snel (1997) have also shown that Overall, the early studies suggested that consumption target detection and response preparation are enhanced of caffeine may have beneficial effects which can by caffeine, and Ruijter et al. (1999) have demonstrated improve safety in automobile driving, other transport that the quantity of information processed is greater operations and industry. The next section reviews more after caffeine. In contrast, no effects of caffeine on out- recent research to see whether adjustments to this view put processes (e.g. movement time) have been demon- strated (Lorist, 1998), nor has caffeine been shown toreduce resistance to distraction (Kenemans and Verba- 3.1.2. A summary of results from studies of caffeine and Research has also sought to link performance effects seen with laboratory tasks with possible effects in real-  A number of different CNS mechanisms by life activities. Brice and Smith (2001a) found that caf- which caffeine could change performance were feine improved both performance on a driving simulator and laboratory tests of attention. Similarly, when per-  Early reviews of the performance effects sug- formance of reaction time tasks was measured before gested that effects were variable and slight and after a normal working day, caffeine consumption  There appeared to be no direct effects of caffeine during the day was seen to reduce the slowing of reac- tion times seen at the end of the day (suggesting that  A number of studies showed that caffeine caffeine may maintain performance levels at work).
A major research issue has been whether caffeine can  Sustained attention has also been shown to be remove impairments produced by fatigue or drugs.
These studies are now briefly reviewed.
 Effects of caffeine on memory have not been 3.2.1. Effects of caffeine in low arousal states  Simulations of real-life tasks (e.g. driving) have A number of studies from the late 1980s and early 1990s show that caffeine removes the performance  Caffeine has been shown to remove impairments impairments produced by sleep loss, fatigue, working at night or by sedative drugs (Rogers et al., 1989; Johnson  Complex interactions between caffeine, person- et al., 1990, 1991; Nicholson et al., 1990; Zwyghuizen- Doorenbos et al., 1990). These findings have important  Some studies have shown that caffeine impairs implications for safety-critical jobs and for maintaining operational efficiency in low alertness situations. Indeed,  A cost–benefit analysis suggested that doses of other research has shown that the decreased alertness caffeine similar to those consumed by the majority produced by consumption of lunch can be eliminated by A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 consumption of caffeinated coffee (Smith et al., 1991a; receptors, which in turn affects the release of a variety of Smith and Phillips, 1993). Furthermore, alertness is often neurotransmitters (e.g. noradrenaline, acetylcholine, reduced by minor illnesses such as the common cold, and dopamine and the GABA/benzodiazepine system). One recent research has shown that caffeine can remove the approach to understanding the CNS mechanisms impaired performance and negative mood associated with underlying the effects of caffeine has been to combine these illnesses (Smith et al., 1997a). The ability of caffeine pharmacological challenges with administration of caf- to counteract the effects of fatigue has been confirmed feine. Alertness can be reduced by changing a number of using simulations of driving (Horne and Reyner, 1996; the neurotransmitter systems. For example, by using Reyner and Horne, 1997) and also simulated assembly- clonidine, a drug which reduces the turnover of central line work (Muehlbach and Walsh, 1995). The latter study noradrenaline, it is possible to mimic sleep deprivation demonstrated significant improvements after caffeine on in a period of a few hours. Smith and Brice (2001) five consecutive nights and showed no decrements when report results from a study combining caffeine/placebo caffeine was withdrawn. Effects of caffeine on more com- and clonidine/placebo conditions. Caffeine was found to plex cognitive tasks are again less clear (Linde, 1995), reverse the effects of clonidine but produce few effects which may reflect the importance of individual differences.
when the volunteers were alert. However, some effects Some of the above studies allow one to assess the of caffeine (e.g. the beneficial effect on encoding of new magnitude of the beneficial effects of caffeine. For information; the cardiovascular effects) were not related example, Smith et al. (1993a) found that consumption to changes in the noradrenergic system. Indeed, Rusted of caffeine at night maintained individuals at the levels and Smith (unpublished) have shown that the encoding seen in the day. Another approach has been to compare of new information reflects the cholinergic system and the effects of caffeine with other approaches aimed at there is evidence from other studies (Riedel et al., 1995) counteracting sleepiness. Bonnet and Arand (1994a,b) that caffeine also influences this neurotransmitter.
report that the combination of a prophylactic nap and There are five other areas that need to be considered caffeine was more effective in maintaining nocturnal here. The first is concerned with the role of individual Other research has demonstrated that administration of caffeine can remove impairments produced by seda- tive drugs (e.g. alcohol—Kerr et al., 1991; Hasenfratz et A number of studies have confirmed the interactive al., 1993; Rush et al., 1993; Hasenfratz and Battig, 1994; effects of caffeine and the personality dimension of Scopolamine—Riedel et al., 1995; Lorazepam—Rush et impulsivity (Smith et al., 1991b, 1994b; Gupta, 1988a,b; al., 1994a; Triazolam—Rush et al., 1994b).
Gupta and Gupta, 1999). Some of these results fit pre- One issue is whether positive effects of caffeine are dictions from the Yerkes-Dodson law which states that largely restricted to low alertness situations. Battig and performance is an inverted-U function of arousal with a Buzzi (1986) argued that caffeine can improve perfor- negative relationshipbetween optimal arousal and task mance beyond a mere restoration of fatigue. Other difficulty. Anderson (1994) found that performance was studies have shown that fatigued subjects show larger an interactive function of task difficulty, caffeine and performance after caffeine than do well-rested volun- impulsivity. Performance on an easy letter cancellation teers (Linde, 1994; Lorist et al., 1994a,b). This pattern task was improved as caffeine dose increased, but on a of results fits with the conclusion of Smith (1994, 1996b, difficult task impulsive subjects (less aroused) improved 1998), namely that effects of small doses of caffeine are while non-impulsive subjects (more aroused) improved most easily demonstrated in low alertness paradigms then deteriorated. Other results do not fit this pattern whereas more global effects can be observed with doses and could reflect other individual differences such as of 200–300 mg. Indeed, very high doses may increase the risk of producing some of the adverse effects descri-bed earlier, as has been confirmed in a study of the dose–response relationships of caffeine (Hasenfratz and Fillmore and colleagues (Fillmore and Vogel-Sprott, Battig, 1994). The results of this last study suggested 1992, 1994, 1995; Fillmore, 1994, 1999) have clearly that beneficial behavioral effects were found with doses demonstrated that effects of caffeine depend on a per- comparable to those found in caffeine-containing bev- son’s expectations. Indeed, these expectations can gen- erages. Furthermore, these levels of caffeine do not eralize to placebo conditions if the individual is led to exacerbate negative effects produced by stressful condi- expect that they are consuming a caffeinated beverage.
tions (e.g. electrical shocks—Hasenfratz and Battig, In many experiments the role of expectations has not been assessed and these could account for at least some It is highly likely that many different CNS mechan- of the conflicting results in the caffeine literature.
isms underlie the effects of caffeine on behavior. Caf- Another possible confounding factor is the role of feine’s major effect is as an antagonist of the adenosine A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 3.2.4. Regular level of caffeine consumption examine performance, and that where they do, any There has been far less research on the effects of reg- effects are selective, not very pronounced, and largely ular caffeine consumption than on acute effects. How- unrelated to the beneficial effects of caffeine reported in ever, a number of papers suggest that high consumers demonstrate better performance (Loke, 1988, 1989; Rogers et al. (1995) have reviewed a number of studies Smith et al., 1993b). This view is confirmed by other of caffeine withdrawal and performance. They conclude studies which suggest that non-consumers of caffeine that ‘‘. . .in a review of recent studies we find no have the worst performance (Jacobsen and Thurman- unequivocal evidence of impaired psychomotor perfor- Lacey, 1992) especially when challenged with caffeine.
mance associated with caffeine withdrawal’’. Indeed, There are exceptions which have shown high users to be they found that caffeine improved performance in both impaired (Mitchell and Redman, 1992), although these deprived volunteers and non-consumers (Richardson et effects are often restricted to the performance of specific al., 1994). Furthermore, other studies which suggest that withdrawal may impair performance (e.g. Rizzo et The strongest evidence for beneficial effects of regular al., 1988; Bruce et al., 1991) can be interpreted in other caffeine consumption comes from a study by Jarvis ways than deprivation (e.g. changes in state—Overton, (1993). He examined the relationshipbetween habitual 1984). Other research (e.g. Griffiths et al., 1986) has coffee and tea consumption and cognitive performance failed to demonstrate negative effects of caffeine with- using data from a cross-sectional survey of a repre- drawal on performance. Another problem for the caf- sentative sample of over 9000 British adults. Subjects feine withdrawal explanation of improved performance completed tests of simple reaction time, choice reaction following caffeine is that it cannot account for the ben- time, incidental verbal memory and visuo-spatial rea- eficial effects of caffeine observed even when volunteers soning, in addition to providing self-reports of usual abstain for a short period of time (Smith et al., 1994b; coffee and tea intake. After controlling extensively for potential confounding variables, a dose–response trend The effects of caffeine withdrawal are still con- to improved performance with higher levels of coffee troversial. This can be seen when one looks at reviews of consumption (best performance associated with about the topic (Daly and Fredholm, 1999; Nehlig, 1999) and 400 mg caffeine per day) was found for all tests. Esti- at empirical studies. For example, James (1998) showed mated overall caffeine consumption showed a dose– that caffeine withdrawal impaired short-term memory response relationship to improved cognitive perfor- performance but caffeine ingestion had no effect. In mance that was strongest in those who had consumed contrast, Smith (1999) reported that caffeine improved high levels for the longest time period (the 55 years+age attention in both those who had been deprived of caf- group). Studies by Hogervorst et al. (1998) and Rogers feine for a short period and those who had no caffeine and Dernoncourt (1998) have failed to replicate these for 7 days. In this study caffeine withdrawal had no effects using acute caffeine challenges which suggests that effect on performance. Studies of children (Goldstein the above effects do reflect regular consumption patterns and Wallace, 1997; Bernstein et al., 1998) have sug- rather than recent intake of caffeine.
gested that caffeine withdrawal may have negative Overall, the previous sections confirm that the effects effects on mood and performance. Other studies (e.g.
of caffeine on performance are largely beneficial. How- Comer et al., 1997) suggest that effects of withdrawal ever, this view has recently been questioned by James are restricted to mood and that performance is unal- (1994), who argues that the beneficial effects of caffeine tered. Like many areas of caffeine research, some of the are really only removal of negative effects produced by effects that have been attributed to withdrawal are open caffeine withdrawal. The final section considers whether to other interpretations. For example, Lane (1997), caffeine withdrawal does actually impair performance Phillips-Bute and Lane (1997) and Lane and Phillips- Bute (1998) compared days when mid-morning coffeewas either caffeinated or decaffeinated. Caffeine con- 3.2.5. Effects of caffeine withdrawal on performance sumption was associated with better performance and The view that beneficial effects of caffeine reflect mood. The authors interpret this as a negative effect of degraded performance in the caffeine-free conditions caffeine withdrawal whereas one could interpret it as a (James, 1994) crucially depends on the strength of the positive effect of caffeine. Other studies of caffeine evidence for withdrawal effects. James states that ‘‘there withdrawal effects have methodological problems such is an extensive literature showing that caffeine with- as the lack of pre-drink baselines (e.g. James, 1998; drawal has significant adverse effects on human perfor- Robelin and Rogers, 1998) or the consideration of mance’’. If one examines the details of the studies cited asymmetric transfer when using within subject designs to support this view (Griffiths et al., 1990; Hughes et al., (e.g. James, 1998). Indeed, while there have been a great 1990, 1991; van Dusseldorpand Katan, 1990; Silverman number of studies of the behavioral effects of caffeine, et al., 1992) one finds that some of them do not even there are still a large number of methodological issues A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 that need further consideration and these are outlined  Caffeine, in doses that would be obtained in normal consumption of caffeinated products, improves One method that can resolve whether caffeine effects performance of vigilance tasks and simple tasks are due to removal of caffeine withdrawal is to study the which benefit from a high level of alertness.
effects on caffeine on the behavior of non-consumers.
 The beneficial effects of caffeine can be most Smith et al. (2001) describe results from three studies that, easily demonstrated in low arousal situations.
used this technique. All studies were able to show beneficial effects of caffeine in both regular consumers and shown when reduced alertness is not involved.
non-consumers, which raises serious doubts about the  Caffeine improves performance of artificial tasks and simulations of driving and industrial work.
These findings suggest that it will be of benefit in safety-critical situations and will improve opera- 3.2.6.1. Consumption regimen. Most studies of the effects of caffeine have administered a single large dose,  Effects on more complex cognitive tasks are less often equivalent to the person’s total daily consumption clear cut, probably because of the moderating level. Caffeine is usually ingested in a number of smaller influence of factors such as personality and time doses (Brice and Smith, 2002) and it is unclear whether of day. However, even this area shows few costs effects observed after a single large dose are the same as those produced by an identical level produced by con-  It has been claimed that the positive effects of suming several caffeine containing drinks over a longer caffeine really reflect removal of negative effects time period. Smith and Brice (2001a,b) describe a study of caffeine withdrawal. This view cannot account that examined this issue and found that the improved for effects observed in non-consumers or non- mood and enhanced performance found after a single deprived individuals. In addition, there is little dose of 200 mg were also observed following four doses evidence of caffeine withdrawal impairing tasks of 65 mg given at hourly intervals (which resulted in an which show improvements following ingestion of identical final level to the single 200 mg dose).
 More research is required on the effects of reg- 3.2.6.2. Metabolismof caffeine. Most of the beneficial ular levels of caffeine consumption on perfor- effects of caffeine show a linear dose–response relation- shipupto about 300 mg and this is then followed by evidence that high consumption is associated either a flattening of the curve or, sometimes, impaired with better performance, especially in the elderly.
performance at higher doses. Brice and Smith (2001b)  A cost–benefit analysis of the effects of caffeine examined the relationshipbetween metabolism of a on performance reveals that this is an area where fixed dose of caffeine (as indicated by saliva levels) and most of the evidence suggest benefits.
mood and performance changes and found that therewas no strong association between the two. This is nottoo surprising in that it is not caffeine levels in the per-iphery per se which produce the behavioral changes but secondary CNS mechanisms. The individual differencesin the metabolism of the caffeine may be very different Much of the research on caffeine and sleephas been from the individual differences in the CNS mechanisms concerned with removing unwanted sleepiness either which, plausibly, accounts for the lack of a strong when the person is working at night or when they are association between plasma (or saliva levels) and beha- sleepdeprived. The fact that caffeine can remove slee- piness means that it can, of course, interfere with nor-mal sleep. However, patterns of consumption suggest 3.2.6.3. Effects of low doses. A number of studies (e.g.
that individuals usually control their caffeine intake to Lieberman et al., 1987; Durlach, 1998; Smith et al., 1999b) prevent interference with sleep. If large amounts of caf- have shown that beneficial effects of doses of caffeine typi- feine are consumed shortly before trying to sleep, then it cally found in commercial products can now be demon- will undoubtedly disturb sleep. The experimental evi- strated in both measures of mood and performance.
dence for such effects is well established and is brieflysummarized below.
3.3. Caffeine and mental performance: a summary A number of studies have shown that caffeine increa- ses sleeplatency (e.g. Zwyghuizen-Doorenbos et al., A detailed review of recent studies confirmed views 1990), and reduces sleepduration (Hicks et al., 1983).
described in a review by Lieberman (1992). To reiterate, Caffeine often produces its effects by increasing latencies in the first half of the night (Bonnet and Webb, 1979) A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 which is different from the insomnia seen in hypnotic consumers being more resistant to effects of caffeine on withdrawal (Brezinova et al., 1975). It would not sleep. Secondly, the impact of caffeine-induced changes appear, therefore, that caffeine-induced insomnia acts as in sleepon behavior the next day and long-term health is a good general model of insomnia, as suggested by some not known. Finally, high levels of caffeine consumption do not appear to be strongly related to sleep parameters.
There are large individual differences in the effects of This again suggests that consumption is usually con- caffeine on sleep. For example, one study has shown trolled to avoid any potential adverse effects on sleep.
that even caffeine given in the early morning can influ-ence the subsequent night’s sleep(Landolt et al., 1995), whereas other individuals report that they can consumecaffeine-containing beverages before bedtime with noadverse impact on their sleep (Colton et al., 1967; Levy  It is quite clear that high doses of caffeine in the and Zylber-Katz, 1983). There are probably many rea- late evening will increase the time taken for some sons for these differences, but it appears to be estab- lished that high consumers appear less likely to report  The effects of smaller doses vary from individual to sleepdisturbance than those who only consume caffeine individual, and even when sleepis affected there is infrequently (Snyder and Sklar, 1984). Indeed, other no clear evidence that the effects are of a sufficient results suggest that tolerance develops to effects of caf- magnitude to influence health and well-being.
feine on sleep(Zwyghuizen-Doorenbos et al., 1990) but  Indeed, people are usually very good at control- that there are no withdrawal effects on sleepwhen caf- ling their caffeine intake, which means that there feine is no longer given (Searle, 1994). It is also unclear is not any strong evidence relating level of caffeine whether the sleepdisturbance produced by caffeine has an impact on behavior the next day, with one studyshowing no changes in mood and performance followingcaffeine disturbed sleep(Smith et al., 1993b).
While it is quite easy to demonstrate effects of late- night caffeine on sleep, it is much harder to find evidence The present article has reviewed the effects of caffeine that high levels of consumption per se will affect sleep.
on mood, mental performance and sleep. In all areas it Hicks et al. (1983) conducted a survey to examine the is apparent that there is a big difference between the associations between daily caffeine consumption, habi- effects of amounts of caffeine that are normally con- tual sleepduration and sleepsatisfaction. The results sumed and those observed when excessive amounts are showed an inverse relationshipbetween level of daily ingested, or when very sensitive individuals are studied.
consumption of caffeinated drinks and habitual sleep Most of the research has examined acute effects of single duration, but no significant association between caffeine doses, and further studies are needed to produce a more consumption and sleep satisfaction. Dekker et al. (1993) detailed profile of effects of regular levels of consump- examined the impact of caffeine consumption on the tion. However, the general picture to emerge is that sleepof locomotive engineers and their spouses. For the when caffeine is consumed in moderation by the major- engineers only, caffeine consumption was correlated ity of the population there are unlikely to be many with longer sleep latency. The effect was not apparent in negative effects. Indeed, the positive effects may be important in maintaining efficiency and safety in both Other surveys have found little evidence of associ- the workplace and other environments. Excessive con- ations between caffeine consumption and sleep. For sumption of caffeine will produce problems, and appro- example, Lee (1992) examined data from 760 nurses.
priate information should be given to minimise effects in The results showed that age and family factors con- psychiatric patients and other sensitive groups. It is tributed to differences in sleepmuch more than caffeine.
important to balance this with information on the ben- Similarly, Greenwood et al. (1995) found no effect of efits of caffeine, for most consumers can usually control caffeine consumption on the sleep of 72 rotating-shift their intake to maximise the beneficial effects and reduce workers. Finally, a study of sleepin elderly women or prevent adverse effects due to over-consumption or found no differences in level of caffeine consumption in good and poor sleepers (Bliwise, 1992).
Overall, the research on the effects of caffeine on sleep leads to three main conclusions. First, large amounts of caffeine (e.g. over 3 mg/kg in a single beverage) con- Alford, C., Bhatti, J., Leith, T., Jamieson, A., Hindmarch, I., 1996.
sumed in the late evening will prevent individuals from Caffeine-induced sleepdisruption: effects on waking the following going to sleepand reduce sleepduration. Effects of day and its reversal with an hypnotic. Human Psychopharmacology smaller doses show large individual variation, with high Clinical and Experimental 11, 185–198.
A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 Anderson, K.J., 1994. Impulsivity, caffeine, and task difficulty: a Durlach, P.J., 1998. The effects of a low dose of caffeine on cognitive within-subjects test of the Yerkes-Dodson law. Personality and performance. Psychopharmacology 140, 116–119.
Individual Differences 16, 813–829.
Eaton, W.W., McLeod, J., 1984. Consumption of coffee or tea and Battig, K., Buzzi, R., 1986. Effect of coffee on the speed of subject- symptoms of anxiety. American Journal of Public Health 74, 66–68.
paced information processing. Neuropsychobiology 16, 126–130.
Erikson, G.C., Hager, L., Houseworth, C., Duggan, J., Petros, T., Battig, K., Buzzi, R., Martin, J.R., Feierabend, J.M., 1984. The effects Beckwith, B., 1985. The effects of caffeine on memory for word lists.
of caffeine on psychological functions and mental performance.
Physiology and Behavior 35, 47–51.
File, S.A., Bond, A.J., Lister, R.G., 1982. Interaction between effects Bernstein, G.A., Carroll, M.E., Dean, N.W., Crosby, R.D., Perwien, of caffeine and lorazepam in performance tests and self-ratings.
A.R., Benowitz, N.L., 1998. Caffeine withdrawal in normal school- Journal of Clinical Psychopharmacology 2, 102–106.
age children. Journal of the American Academy of Child and Ado- Fillmore, M.T., 1994. Investigating the behavioral effects of caffeine: the contribution of drug-related expectancies. Special Issue: Caffeine Bliwise, N.G., 1992. Factors related to sleepquality in healthy elderly research. Pharmacopsychoecologia 7, 63–73.
women. Psychology and Aging 7, 83–88.
Fillmore, M.T., 1999. Behavioral effects of caffeine: the role of drug- Bonnet, M.H., Arand, D.L., 1994a. The use of prophylactic naps and related expectancies. In: Gupta, B.S., Gupta, U. (Eds.), Caffeine and caffeine to maintain performance during a continuous operation.
behavior: current views and research trends. CRC Press, Boca Bonnet, M.H., Arand, D.L., 1994b. Impact of naps and caffeine on Fillmore, M., Vogel-Sprott, M., 1992. Expected effect of caffeine on extended nocturnal performance. Physiology and Behavior 56, 103– motor performance predicts the type of response to placebo. Psy- Bonnet, M.H., Gomez, S., Wirth, O., Arand, D.L., 1995. The use of Fillmore, M.T., Vogel-Sprott, M., 1994. Psychomotor performance caffeine versus prophylactic naps in sustained performance. Sleep under alcohol and under caffeine: expectancy and pharmacological effects. Experimental and Clinical Psychopharmacology 2, 319–327.
Bonnet, M.H., Webb, W.B., 1979. The return to sleep. Biological Fillmore, M.T., Vogel-Sprott, M., 1995. Behavioral effects of com- bining alcohol and caffeine: contribution of drug-related expectan- Bovim, G., Naess, P., Helle, J., Sand, T., 1995. Caffeine influence on cies. Experimental and Clinical Psychopharmacology 3, 33–38.
the motor steadiness battery in neuropsychological tests. Journal of Fine, B.J., Kobrick, J.L., Lieberman, H.R., Marlowe, B., Riley, R.H., Clinical and Experimental Neuropsychology 17, 472–476.
Tharion, W.J., 1994. Effects of caffeine or diphenhydramine on Brezinova, V., Oswald, I., Loudon, J., 1975. Two types of insomnia: visual vigilance. Psychopharmacology 114, 233–238.
too much waking or not enough sleep. British Journal of Psychiatry Frewer, L.J., Lader, M., 1991. The effects of caffeine on two compu- terized tests of attention and vigilance. Human Psychopharmacol- Brice C., Smith A.P., 2001a. The effects of caffeine on simulated driving, ogy Clinical and Experimental 6, 119–128.
subjective alertness and sustained attention. Human Psychopharma- Goldstein, A., Kaizer, S., Whitby, O., 1969. Psychotropic effects of caffeine in man: iv. Quantitative and qualitative differences asso- Brice C., Smith A.P., 2001b. Caffeine levels in saliva: associations with ciated with habituation to coffee. Clinical Pharmacology and Ther- psychosocial factors and behavioural effects. Human Psycho- Goldstein, A., Wallace, M.E., 1997. Caffeine dependence in school- Brice, C.F., Smith, A.P., 2002. Factors associated with caffeine con- children? Experimental and Clinical Psychopharmacology 5, 388– sumption. International Journal of Food Sciences and Nutrition 53, Greden, J.F., 1974. Anxiety or caffeinism: a diagnostic dilemma.
Bruce, M., Scott, N., Shine, P., Lader, M., 1991. Caffeine withdrawal: American Journal of Psychiatry 131, 1089–1092.
A contrast of withdrawal symptoms in normal subjects who have Green, P.J., Suls, J., 1996. The effects of caffeine on ambulatory blood abstained from caffeine for 24 hours and for 7 days. Journal of pressure, heart rate, and mood in coffee drinkers. Journal of Beha- Bruce, M., Scott, N., Shine, P., Lader, M., 1992. Anxiogenic effects of Greenwood, K.M., Rich, W.J., James, J.E., 1995. Sleephygiene prac- caffeine in patients with anxiety disorders. Archives of General tices and sleepduration in rotating-shift shiftworkers. Work and Charney et al., 1985 Charney, D.S., Heninger, G.R., Jatlow, P.I., Griffiths, R.R., Bigelow, G.E., Liebson, I.A., 1986. Human coffee 1985. Increased anxiogenic effects of caffeine in panic disorders.
drinking. Reinforcing and physical dependence producing effects.
Archives of General Psychiatry 42, 233–243.
Journal of Pharmacology and Experimental Therapeutics 239, 416– Clubley, M., Bye, C.E., Henson, T.A., Peck, A.W., Riddington, C.J., 1979. Effects of caffeine and cyclizine alone and in combination on Griffiths, R.R., Evans, S.M., Heishman, S.J., Preston, K.L., Sannerud, human performance, subjective effects and EEG activity. British C.A., Wolf, B., Woodson, P.P., 1990. Low-dose caffeine physical Journal of Clinical Pharmacology 7, 157–163.
dependence in humans. Journal of Pharmacology and Experimental Colton, T., Gosselin, R.E., Smith, R.P., 1967. The tolerance of coffee drinkers to caffeine. Clinical Pharmacology and Therapeutics 9, 31–39.
Gupta, U., 1988a. Effects of impulsivity and caffeine on human cog- Comer, S.D., Haney, M., Foltin, R.W., Fischman, M.W., 1997. Effects nitive performance. Pharmacopsychoecologia 1, 33–41.
of caffeine withdrawal in humans living in a residential laboratory.
Gupta, U., 1988b. Personality, caffeine and human cognitive perfor- Experimental and Clinical Psychopharmacology 5, 399–403.
mance. Pharmacopsychoecologia 1, 79–84.
Daly, J.W., Fredholm, B.B., 1998. Caffeine—an atypical drug of Gupta, U., Gupta, B.S., 1999. Caffeine, impulsivity and performance.
dependence. Drug and Alcohol Dependence 51, 199–206.
In: Gup ta, B.S., Gup ta, U. (Eds.), Caffeine and Behavior: Current Dekker, D.K., Pakey, M.J., Popkin, S.M., Tepas, D.I., 1993. Loco- Views and Research Trends. CRC Press, Boca Raton, FL, pp. 191–205.
motive engineers and their spouses: coffee consumption, mood, and Hasenfratz, M., Battig, K., 1992. No psychophysiological interactions sleep reports. Special issue: Night and Shiftwork. Ergonomics 36, between caffeine and stress? Psychopharmacology 109, 283–290.
Hasenfratz, M., Battig, K., 1994. Acute dose-effect relationships of Dews, P.B., 1984. Behavioral effects of caffeine. In: Dews, P.B. (Ed.), caffeine and mental performance, EEG, cardiovascular and sub- jective parameters. Psychopharmacology 114, 281–287.
A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 Hasenfratz, M., Bunge, A., Dal Pra, D., Battig, K., 1993. Antagonistic Handbook of Human Performance, vol. 2. Academic Press, Lon- effects of caffeine and alcohol on mental performance parameters.
Pharmacology, Biochemistry and Behavior 46, 463–465.
Lieberman, H.R., Wurtman, R.J., Emde, G.G., Roberts, C., Covielle, Hicks, R.A., Hicks, G.J., Reyes, J.R., Cheers, Y., 1983. Daily caffeine I.L.G., 1987. The effects of low doses of caffeine on human perfor- use and the sleepof college students. Bulletin of the Psychonomic mance and mood. Psychopharmacology 92, 308–312.
Linde, L., 1994. An auditory attention task. A note on the processing Hire, J.N., 1978. Anxiety and caffeine. Psychological Reports 42, 833– of verbal information. Perceptual and Motor Skills 78, 563–570.
Linde, L., 1995. Mental effects of caffeine in fatigued and non-fatigued Horne, J.A., Reyner, L.A., 1996. Counteracting driver sleepiness: female and male subjects. Ergonomics 38, 864–885.
effects of napping, caffeine, and placebo. Psychophysiology 33, 306– Loke, W.H., 1988. Effects of caffeine on mood and memory. Physiol- Hogervorst, E., Riedel, W.J., Scmitt, J.A., 1998. J and Jolles J. Caf- Loke, W.H., 1989. Effects of caffeine on task difficulty. Psychologica- feine improves memory performance during distraction in middle- aged but not in young or old subjects. Human Psychopharmacology Loke, W.H., 1990. Effects of repeated caffeine administration on cog- nition and mood. Human Psychopharmacology Clinical and Hughes, J.R., Higgins, S.T., Bickel, W.K., Hunt, W.K., Fenwick, J.W., Gulliver, S.B., Mireault, G.C., 1991. Caffeine self-administra- Loke, W.H., 1992. The effects of caffeine and automaticity on a visual tion, withdrawal and adverse effects among coffee drinkers.
information processing task. Human Psychopharmacology Clinical Archives of General Psychiatry 48, 611–617.
Hughes, J.R., Oliveto, A.H., Helzer, J.E., Higgins, S.T., Bickel, W.K., Loke, W.H., Hinrichs, J.V., Ghoneim, M.M., 1985. Caffeine and dia- 1990. Should caffeine abuse, dependence or withdrawal be added to zepam: separate and combined effects on mood, memory, and psy- DSM-IV and ICD-10? American Journal of Psychiatry 149, 33–40.
chomotor performance. Psychopharmacology 87, 344–350.
Jacobson, B.H., Thurman-Lacey, S.R., 1992. Effect of caffeine on Loke, W.H., Meliska, C.J., 1984. Effects of caffeine use and indigestion motor performance by caffeine-naive and-familiar subjects. Percep- on a protracted visual vigilance task. Psychopharmacology 84, 54–57.
tual and Motor Skills 74, 151–157.
Lorist, M.M., 1998. Caffeine and information processing in man. In: James, J.E., 1994. Does caffeine enhance or merely restore degraded Snel, J., Lorist, M.M. (Eds.), Nicotine, Caffeine and Social Drink- psychomotor performance? Neuropsychobiology 30, 124–125.
ing: Behavior and Brain Function. Academic Press, Amsterdam, pp.
James, J.E., 1998. Acute and chronic effects of caffeine on perfor- mance, mood, headache and sleep. Neuropsychobiology 38, 32–41.
Lorist, M.M., Snel, J., 1997. Caffeine effects on perceptual and motor Jarvis, M.J., 1993. Does caffeine intake enhance absolute levels of processes. Electroencephalography and Clinical Neurophysiology cognitive performance? Psychopharmacology 110, 45–52.
Johnson, L.C., Freeman, C.R., Spinweber, C.L., Gomez, S.A., 1991.
Lorist, M.M., Snel, J., Kok, A., 1994a. Influence of caffeine on infor- Subjective and objective measures of sleepiness: effect of benzodia- mation processing stages in well rested and fatigued subjects. Psy- zepine and caffeine on their relationship. Psychophysiology 28, 65– Lorist, M.M., Snel, J., Kok, A., Mulder, G., 1994b. Influence of caf- Johnson, L.C., Spinweber, G.L., Gomez, S.Z., 1990. Benzodiazepines feine on selective attention in well-rested and fatigued subjects.
and caffeine: effect on daytime sleepiness, performance, and mood.
Lynn, R., 1973. National differences in anxiety and the consumption Kenemans, J.L., Verbaten, M., 1998. Caffeine and visuo-spatial of coffee. British Journal of Social Clinical Psychology 12, 92–93.
attention. Psychopharmacology 135, 353–360.
Mathew, R.J., Wilson, W.H., 1990. Behavioral and cerebrovascular Kerr, J.S., Sherwood, N., Hindmarch, I., 1991. Separate and com- effects of caffeine in patients with anxiety disorders. Acta Psychia- bined effects of the social drugs on psychomotor performance. Psy- Mitchell, P.J., Redman, J.R., 1992. Effects of caffeine, time of day and Kuznicki, J.T., Turner, L.S., 1986. The effects of caffeine on caffeine user history on study-related performance. Psychopharmacology users and non-users. Physiology and Behavior 37, 397–408.
Lader, M., Bruce, M., 1986. States of anxiety and their induction by Muehlbach, M.J., Walsh, J.K., 1995. The effects of caffeine on simulated drugs. British Journal of Clinical Pharmacology 22, 252–261.
night-shift work and subsequent daytime sleep. Sleep 18, 22–29.
Landolt, H.P., Werth, E., Borbely, A.A., Dijk, D.J., 1995. Caffeine Nehlig, A., 1999. Are we dependent upon coffee and caffeine? A review intake (200mg) in the morning affects human sleepand EEG power on human and animal data. Neuroscience and Biobehavioral spectra at night. Brain Research 675, 67–74.
Lane, J.D., 1997. Effects of brief caffeinated-beverage deprivation on Nicholson, A.N., Pascoe, P.A., Stone, B.M., 1990. The sleep-wakeful- mood, symptoms and psychomotor performance. Pharmacology, ness continuum: interactions with drugs which increase wakefulness Biochemistry and Behavior 58, 203–208.
and enhance alertness. International symposium: fatigue, sleep Lane, J.D., Phillips-Bute, B.G., 1998. Caffeine deprivation affects vig- deprivation, circadian rhythms and their interactions with alcohol ilance performance and mood. Physiology and Behavior 65, 171– and other drugs (1989, Santa Monica, California). Alcohol, Drugs Leathwood, P.D., Pollet, P., 1982. Diet-induced mood changes in Overton, D.A., 1984. State dependent learning and drug discrimina- normal populations. Journal of Psychiatric Research 17, 147–154.
tions. In: Iversen, L.L., Inersen, S.D., Snyder, S.H. (Eds.), Hand- Lee, K.A., 1992. Self-reported sleep disturbances in employed women.
book of Psychopharmacology, vol. 18. Plenum Press, New York, Levy, M., Zylber-Katz, E., 1983. Caffeine metabolism and coffee- Phillips-Bute, B.G., Lane, J.D., 1997. Caffeine withdrawal symptoms attributed sleepdisturbances. Clinical Pharmacology and Ther- following brief caffeine deprivation. Physiology and Behavior 63, Lieberman, H.R., 1988. Beneficial effects of caffeine. In: Twelfth Ratliff-Crain, J., O’Keeffe, M.K., Baum, A., 1989. Cardiovascular International Scientific Colloquiumon Coffee. Paris, ASIC.
reactivity, mood, and task performance in deprived and non- Lieberman, H.R., 1992. Caffeine. In: Smith, A.P., Jones, D.M. (Eds.), deprived coffee drinkers. Health Psychology 8, 427–447.
A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 Regina, E.G., Smith, G.M., Keiper, C.G., McKelvey, R.K., 1974.
Shanahan, M.P., Hughes, R.N., 1986. Potentiation of performance- Effects of caffeine on alertness in simulated automobile driving.
induced anxiety by caffeine in coffee. Psychological Reports 59, 83– Journal of Applied Psychology 59, 483–489.
Revelle, W., Anderson, J.K., Humphries, M.S., 1987. Empirical tests Sicard, B.A., Perault, M.C., Enslen, M., Chauffard, F., Vandel, B., and theoretical extensions of arousal-based theories of personality.
Tachon, P., 1996. The effects of 600 mg of slow release caffeine on In: Strelau, J., Eysenck, H.J. (Eds.), Personality Dimensions and mood and alertness. Aviation, Space, and Environmental Medicine Reyner, L.A., Horne, J.A., 1997. Suppression of sleepiness in drivers: Silverman, K., Evans, S.M., Strain, E.C., Griffiths, R.R., 1992. With- combination of caffeine with a short nap. Psychophysiology 34, drawal syndrome after the double-blind cessation of caffeine con- sumption. New England Journal of Medicine 327, 1109–1114.
Richardson, N.J., Rogers, P.J., Elliman, N.A., O’Dell, R.J., 1994.
Smith, A., 1994. Caffeine, performance, mood and states of reduced Mood and performance effects of caffeine in relation to acute and alertness. Special Issue: Caffeine research. Pharmacopsychoecologia chronic caffeine deprivation. Pharmacology, Biochemistry and Smith, A.P., 1995. Caffeine, caffeine withdrawal and psychomotor Riedel, W., Hogervorst, E., Leboux, R., Verhey, F., 1995. Caffeine performance: a reply to James. Neuropsychobiology 31, 200–201.
attenuates scopolamine-induced memory impairment in humans.
Smith, A.P., 1996a. Caffeine dependence: an alternative view. Nature Rizzo, A.A., Stamps, L.E., Lawrence, A.F., 1988. Effects of caffeine Smith, A.P., 1996b. Food, drink and mental performance. In: War- withdrawal on motor performance and heart rate changes. Interna- burton, D.M., Sherwood, N. (Eds.), Pleasure and Quality of Life.
tional Journal of Psychophysiology 6, 9–14.
Roache, J.O., Griffiths, R.R., 1987. Interactions of diazepam and caf- Smith, A.P., 1998. Effects of caffeine on attention. Low levels of feine: behavioral and subjective dose effects in humans. Pharmacol- arousal. In: Snel, J., Lorist, M. (Eds.), Nicotine, Caffeine and Social ogy, Biochemistry and Behavior 26, 801–812.
Drinking. Harwood Press, Amsterdam, pp. 215–228.
Robelin, M., Rogers, P.J., 1998. Mood and psychomotor performance Smith, A.P., 1999. Caffeine,caffeine withdrawal and performance effi- effects of the first but not subsequent, cup-of-coffee equivalent doses ciency. In: Gupta, B.S., Gupta, Uma (Eds.), Caffeine and Behavior: of caffeine consumed after overnight caffeine abstinence. Behavioral Current Views and Research Trends. CRC Press, Boca Raton, FL, Rogers, A.S., Spencer, M.B., Stone, B.M., Nicholson, A.N., 1989. The Smith, A.P., Brice, C., 2001a. The behavioral effects of caffeine. In: influence of a 1hr. napon performance overnight. Ergonomics 32, Proceedings of the 217th American Chemical Society National Rogers, P.J., Dernoncourt, C., 1998. Regular caffeine consumption: a Smith, A.P., Brice C., 2001b. Behavioural effects of caffeine in coffee.
balance of adverse and beneficial effects for mood and psychomotor In: Proceedings of ASIC ‘ 99, Helsinki.
performance. Pharmacology, Biochemistry and Behavior 59, 1039– Smith, A.P., Brockman, B., Flynn, R., Maben, A., Thomas, M., 1993a. Investigation of the effects of coffee on alertness and per- Rogers, P.J., Richardson, N.J., Dernoncourt, C., 1995. Caffeine use: is formance during the day and night. Neuropsychobiology 27, 217– there a net benefit for mood and psychomotor performance? Neu- Smith, A.P., Brice, C.F., Nguyen-van-Tam, D., 2001. Beneficial effects Rubin, G.J., Smith, A.P., 1999. Caffeine withdrawal and headaches.
of caffeinated coffee and effects of withdrawal. 19th International Nutritional Neuroscience 2, 123–126.
Scientific Colloquium on Coffee, Trieste, Association Scientifique Ruijter, J., Lorist, M.M., Snel, J., 1999. The influence of different doses of caffeine on visual task performance. Journal of Psycho- Smith, A.P, Clark, R., Gallagher, J., 1999a. Breakfast cereal and caf- feinated coffee: effects on working memory, attention, mood and Rush, C.R., Higgins, S.T., Bickel, W.K., Hughes, J.R., 1994a. Acute cardiovascular function. Physiology and Behavior 67, 9–17.
behavioral effects of lorazepam and caffeine, alone and in combina- Smith, A.P., Kendrick, A.M., Maben, A.L., 1992. Effects of breakfast tion, in humans. Behavioral Pharmacology 5, 245–254.
and caffeine on performance and mood in the late morning and Rush, C.R., Higgins, S.T., Hughes, J.R., Bickel, W.K., 1994b. Acute after lunch. Neuropsychobiology 26, 198–204.
behavioral effects of triazolam and caffeine, alone and in combina- Smith, A.P., Kendrick, A., Maben, A., Salmon, J., 1994a. Effects of tion, in humans. Experimental and Clinical Psychopharmacology 2, breakfast and caffeine on cognitive performance, mood and cardio- vascular functioning. Appetite 22, 39–55.
Rush, C.R., Higgins, S.T., Hughes, J.R., Bickel, W.K., Wiegner, M.S., Smith, A.P., Maben, A., Brockman, P., 1993b. The effects of caffeine 1993. Acute behavioral and cardiac effects of alcohol and caffeine, and evening meals on sleepand performance, mood and cardiovas- alone and in combination, in humans. Behavioral Pharmacology 22, cular functioning the following day. Journal of Psychopharmacol- Rusted, J., 1994. Caffeine and cognitive performance: effects on mood Smith, A.P., Maben, A., Brockman, P., 1994b. Effects of evening or mental processing? Pharmacopsychoecologia 7, 49–54.
meals and caffeine on cognitive performance, mood and cardiovas- Rusted, J., 1999. Caffeine and cognitive performance: effects on mood cular functioning. Appetite 22, 57–65.
or mental processing? In: Gupta, B.S., Gupta, U. (Eds.), Caffeine Smith, A. P., Phillips, W., 1993. Effects of low doses of caffeine in and Behavior: Current Views and Research Trends. CRC Press, coffee on human performance and mood. In: 15th International Scientific Colloquim on Coffee, vol. 2. Association Scientifique Sawyer, D.A., Harry, J.L., Turin, A.C., 1982. Caffeine and human Internationale de Cafe, Paris, pp. 461–469.
behavior: arousal, anxiety, and performance effects. Journal of Smith, A.P., Rusted, J.M., Eaton-Williams, P., Savory, M., Leath- wood, P., 1991a. Effects of caffeine given before and after lunch on Schuh, K.J., Griffiths, R.R., 1997. Caffeine reinforcement: the role of sustained attention. Neuropsychobiology 23, 160–163.
withdrawal. Psychopharmacology 130, 320–326.
Smith, A.P., Rusted, J.M., Savory, M., Eaton-Williams, P., Hall, S.R., Searle, G.F., 1994. The effect of dietary caffeine manipulation on 1991b. The effects of caffeine, impulsivity and time of day on per- blood caffeine, sleepand disturbed behaviour. Journal of Intellec- formance, mood and cardiovascular function. Journal of Psycho- tual Disability Research 38, 383–391.
A. Smith / Food and Chemical Toxicology 40 (2002) 1243–1255 Smith, A.P., Sturgess, W., Gallagher, J., 1999b. Effects of a low dose Streufert, S., Satish, U., Pogash, R., Gingrich, D., Landis, R., Roache, of caffeine given in different drinks on mood and performance.
J., Severs, W., 1997. Excess coffee consumption in simulated com- Human Psychopharmacology 14, 473–482.
plex work settings: detriment or facilitation of performance? Journal Smith, A.P., Thomas, M., Perry, K., Whitney, H., 1997a. Caffeine and of Applied Psychology 82, 774–782.
the common cold. Journal of Psychopharmacology 11, 319–324.
Svensson, E., Persson, L.O., Sjoberg, L., 1980. Mood effects of diaze- Smith, A.P., Whitney, H., Thomas, M., Perry, K., Brockman, P., pam and caffeine. Psychopharmacology 67, 73–80.
1997b. Effects of caffeine and noise on mood, performance and Swift, C.G., Tiplady, B., 1988. The effects of age on the response to cardiovascular functioning. Human Psychopharmacology 12, 27– caffeine. Psychopharmacology 94, 29–31.
Tannahill, R., 1989. Food in History. Crown Publishers, New York.
Smith, D.L., Tong, J.E., Leigh, G., 1977. Combined effects of tobacco van Dusseldorp, M., Katan, M.B., 1990. Headache caused by caffeine and caffeine on the components of choice reaction-time, heart rate, withdrawal among moderate coffee drinkers switched to ordinary and hand steadiness. Perceptual and Motor Skills 45, 635–639.
decaffeinated coffee: a 12 week double blind trial. British Medical Snyder, S.H., 1984. Adenosine as a mediator of the behavioral effects of xanthines. In: Dews, P.B. (Ed.), Caffeine. Springer, New York.
Veleber, D.M., Templer, D.I., 1984. Effects of caffeine on anxiety and Snyder, S.H., Sklar, P., 1984. Behavioral and molecular actions of depression. Journal of Abnormal Psychology 93, 120–122.
caffeine: focus on adenosine. Journal of Psychiatric Research 18, Warburton, D.M., 1995. Effects of caffeine on cognition and mood without caffeine abstinence. Psychopharmacology 119, 66–70.
Stern, K.N., Chait, L.D., Johanson, C.E., 1989. Reinforcing and sub- Zwyghuizen-Doorenbos, A., Roehrs, T.A., Lipschutz, L., Timms, V., jective effects of caffeine in normal human volunteers. Psycho- Roth, T., 1990. Effects of caffeine on alertness. Psychopharmacol-

Source: http://cedec.umfcv.ro/files/e/f/Effects%20of%20caffeine%20on%20human%20behavior.pdf