Caffeine Expectancy Questionnaire (CaffEQ): construction, psychometric properties, and associations with caffeine use, caffeine dependence, and other related variables

Expectancies for drug effects predict drug initiation, use, cessation, and relapse, and may play a causal role in drug effects (i.e., placebo effects). Surprisingly little is known about expectancies for caffeine even though it is the most widely used psychoactive drug in the world. In a series of independent studies, the nature and scope of caffeine expectancies among caffeine consumers and nonconsumers were assessed, and a comprehensive and psychometrically sound Caffeine Expectancy Questionnaire (CaffEQ) was developed. After 2 preliminary studies, the CaffEQ was administered to 1,046 individuals from the general population along with other measures of interest (e.g., caffeine use history, anxiety). Exploratory factor analysis of the CaffEQ yielded a 7-factor solution. Subsequently, an independent sample of 665 individuals completed the CaffEQ and other measures, and a subset (n = 440) completed the CaffEQ again approximately 2 weeks later. Confirmatory factor analysis revealed good model fit, and test-retest reliability was very good. The frequency and quantity of caffeine use were associated with greater expectancies for withdrawal/dependence, energy/work enhancement, appetite suppression, social/mood enhancement, and physical performance enhancement and lower expectancies for anxiety/negative physical effects and sleep disturbance. Caffeine expectancies predicted various caffeine- associated features of substance dependence (e.g., use despite harm, withdrawal incidence and severity, perceived difficulty stopping use, tolerance). Expectancies for caffeine consumed via coffee were stronger than for caffeine consumed via soft drinks or tea. The CaffEQ should facilitate the advancement of our knowledge of caffeine and drug use in general.     

Repeated diazepam administration: effects on the acquisition and performance of response chains in humans.

The effects of repeated diazepam administration (80 mg) were assessed across a 12-hr time course with humans responding under a two-component multiple schedule of repeated acquisition and performance of response sequences. Subjects resided in an inpatient clinical research ward for the duration of the study. In each component of the multiple schedule, subjects completed sequences of 10 responses in a predetermined order using three keys of a numeric keypad. In the acquisition component, a new response sequence was to be acquired each session. In the performance component, the response sequence always remained the same. After stable responding was obtained and the effects of the placebo assessed, diazepam was administered for 3 consecutive days. The effects of repeated diazepam administration on overall percentage of errors across the two components of the multiple schedule were selective. In the acquisition component, the first dose of diazepam increased percentage errors with the magnitude of effects decreasing across the second and third days of diazepam administration. In the performance component, the percentage of errors was either minimally affected across all 3 days of diazepam administration or substantively increased on Day 1 with subsequent diazepam administrations having minimal effects. Effects on response rate were not selective. Diazepam decreased rates of responding in both schedule components, with the magnitude of effects decreasing across successive administrations. These results replicate previous findings in humans and nonhumans on the selective effects of diazepam on acquisition versus performance baselines. Also, the results suggest that the selective effects do not result from differences in reinforcement rate. Finally, the present results demonstrate that the selective recovery from repeated drug administration previously demonstrated in nonhumans using a repeated acquisition arrangement has generality to human behavior.     

Can caffeine antagonize alcohol-induced performance decrements in humans?

Design aspects and results of experiments that investigated the effects of caffeine on alcohol-induced performance decrements in humans are discussed. Simple conclusions concerning those outcomes are unwarranted because they seem to depend on the tasks used and the dosages of both drugs, with both antagonism and potentiation of alcohol-induced effects having been reported. Results indicate that legally intoxicated individuals cannot antagonize alcohol-induced, driving-related decrements with caffeine prior to driving an automobile, thought to be the major behavior for which caffeine is used in attempts to antagonize alcohol-induced decrements. We offer suggestions for research concerning subjects' habitual use or nonuse of caffeine and typical alcohol consumption levels, the interval between alcohol and caffeine ingestion, and the effects of caffeine and alcohol alone on performance tasks. We also suggest that statistical analyses should allow for a differentiation of results in which caffeine partially offsets an alcohol-induced decrement from more positive results in which caffeine returns functioning to its normal level.     

Effect of drugs on the cardiac conditioned response of dogs

Dogs were conditioned to an anxiety response characterized by an increase in heart rate upon presentation of a musical tone. Meprobamate, chlordiazepoxide and diazepam effectively prevented the cardiac conditioned response. Phenobarbital and chlorpromazine were less effective. Amphetamine, 1.5 orally, also blocked the cardiac conditioned response. Atropine, meperidine, caffeine, imipramine and -methyl-dopa had no effect on the cardiac conditioned response. Morphine and methyl phenidate were clearly active only in doses producing marked side effects. Ethanol in relatively small doses was mildly effective.     

Interactive Effects of Caffeine Consumption and Stressful Circumstances on Components of Stress: Caffeine Makes Men Less,But Women More Effective as Partners Under Stress

We tested whether increased caffeine consumption exacerbates stress and disrupts team performance, and we explored whether “tend and befriend” characterizes women's coping. We gave decaffeinated coffees, half of which contained added caffeine, to coffee drinkers in same‐sex, same‐aged dyads. We measured individual cognitive appraisals, emotional feelings, bodily symptoms, coping, and performance evaluations, together with dyad memory, psychomotor performance, and negotiation skills under higher or lower stressful conditions. Evidence consistent with the first hypothesis was weak, but we found that women performed better than did men on collaborative tasks under stress, provided caffeine had been consumed. The usefulness of multi component, cognitive‐relational approaches to studying the effects of caffeine on stress is discussed, together with special implications of the effects for men.     

Effect of caffeine on motor performance by caffeine-naive and -familiar subjects.

The purpose of this investigation was to assess the effect of caffeine on selected manipulation skills by caffeine-naive and caffeine-familiar subjects. The subjects were 20 caffeine-naive (less than 90 mg/d) and 20 caffeine-familiar (greater than 750 mg/d) college-age (21 +/- 1.7 yr.) women. Measurements included steadiness error time and frequency, duration of tracing, error time and frequency, and dexterity. Doses of 2.5, 5.0 mg.kg-1 body weight caffeine or a placebo (200 mg. methylcellulose) were administered randomly to all subjects on three separate occasions. A 2 x 3 repeated-measures analysis of variance yielded a significant group difference for steadiness error time between the 5 mg.kg-1 and 2.5 mg.kg-1 dose and between 5 mg.kg-1 and the placebo. For frequency of steadiness errors, the nonuser group posted significant gains for both 5.0 and 2.5 mg.kg-1 over the placebo control. On tracing error time and error frequency, 5.0 mg.kg-1 resulted in significant increases from both 2.5 mg.kg-1 and the placebo group. In the caffeine-naive group, both doses of caffeine led to significant increases in dexterity time from the placebo, and the 5.0 mg.kg-1 dose was significantly different from the 2.5 mg.kg-1 trial. It was concluded that caffeine had detrimental effects on selected performance skills of caffeine-naive women but not in caffeine-familiar women.     

The effect of caffeine on handwriting movements in skilled writers

In laboratory tasks, caffeine has been shown to improve psychomotor performance. The aim of the present experiment was to assess the effects of caffeine on a skilled everyday life task in habitual caffeine consumers. The assessment of handwriting movements of 20 adults was performed following the administration of 0mg/kg (placebo), 1.5mg/kg, 3.0mg/kg or 4.5mg/kg of caffeine. A digitising tablet was used for the assessment of fine motor movements. Participants were asked to perform a simple writing task. Kinematic analysis of handwriting movements showed that, in comparison to placebo administration, high doses of caffeine (i.e., 4.5mg/kg) can produce improvements in handwriting as indicated by more fluent handwriting movements as well as an increase in maximum velocity and maximum positive and negative accelerations. The results suggest that higher doses of caffeine can enhance psychomotor performance.     

Effects of caffeine on performance of low intensity tasks

31 college age men and women who consume less than three caffeinated beverages per week agreed to participate as subjects in research on the effects of acute caffeine intake on low intensity task performance. All subjects performed two randomly administered test conditions: (1) caffeine (5 mg/kg) and (2) placebo on separate visits following an initial 1-hr. orientation visit. Subjects were administered the beverage 30 min. prior to performing 12 separate tests assessing basic mathematics, simple response, logical reasoning, hand-eye coordination, and spatial and assembly skills. The Spielberger State Anxiety test was administered immediately after consuming the test beverage and once again at posttest. Analysis showed that caffeine did not significantly affect performance on all tests with the exception of the peripheral awareness (hand-eye coordination) test on which performance was higher after ingesting caffeine. The placebo treatment produced no effect on state anxiety, which contrasted with a significant rise in anxiety after caffeine consumption. State anxiety values were significantly greater after caffeine treatment relative to the placebo at pretest, and this difference persisted at posttest. These results demonstrated that the dose of caffeine increased scores on state anxiety for individuals who consumed less than three caffeinated beverages weekly but had very little effect on performance of low intensity tasks, except for a hand-eye coordination test involving peripheral awareness. Perhaps longer continuous performance of more demanding tasks would be more sensitive.     

Caffeine reduces time-of-day effects on memory performance in older adults

For most older adults, memory performance depends on the time of day, with performance being optimal early in the morning and declining during the late afternoon hours. In the present study, we asked whether this decline could be ameliorated by a simple stimulant, caffeine. Adults over the age of 65 who considered themselves "morning types" were tested twice over an interval of 5 to 11 days, once in the morning and once in the late afternoon. Participants ingested either coffee with caffeine or decaffeinated coffee at both sessions. Participants who ingested decaffeinated coffee showed a significant decline in memory performance from morning to afternoon. In contrast, those who ingested caffeine showed no decline in performance from morning to afternoon. The results suggest that time-of-day effects may be mediated by nonspecific changes in level of arousal.     

The effects of intravenous diazepam and hyoscine upon human memory

Diazepam and hyoscine are known to have amnesic effects when administered intravenously. The two drugs are pharmacologically quite different from each other and might be expected to produce qualitatively distinct patterns of impairment in formal memory tasks. Groups of normal volunteers received intravenous administrations of diazepam, hyoscine and saline following a double-blind procedure and were then tested on immediate serial recall. Diazepam and hyoscine produced similar deficits on concrete and abstract words whether scored for ordered recall or item recall. In terms of ordered recall, phonemic similarity produced impaired performance under all three administrations, but semantic similarity did not. In terms of item recall, diazepam and hyoscine produced impaired performance on unrelated words, but the impairment was reduced under conditions of either phonemic or semantic similarity. There were also some interesting differences between diazepam and hyoscine in terms of their effects upon the shape of the serial-position curve and upon the types of intrusion error. The results confirm that both diazepam and hyoscine impair acquisition processes but fail to distinguish the effects of the two drugs upon different categories of encoding operations.     

Discriminative stimulus effects of diazepam and buspirone in normal volunteers.

A within-subject design was used to characterize the effects of dose manipulations on discriminative and self-reported effects of oral diazepam and buspirone. Subjects were trained to discriminate diazepam (10 mg) versus placebo (n = 10), or buspirone (10 or 15 mg) versus placebo (n = 9). The compounds were identified to subjects by letter code before discrimination training began. In later sessions, correct identifications at 2 hr after the oral administration of drug earned money. All subjects showed accurate discrimination performance during the test-of-acquisition phase. In a low-dose generalization phase, diazepam and buspirone produced dose-related increases in drug identifications across a four-fold range of doses. In a subsequent low-dose training phase, in which subjects were trained to discriminate progressively lower drug doses, the median lowest discriminable dose of diazepam and buspirone was 2.5 and 7.5 mg, respectively. Dose-response functions for drug identifications were shifted leftward in the low-dose training phase relative to the low-dose generalization phase, suggesting that reinforcement of progressively lower doses enhances drug discriminability. The self-reported effects of diazepam and buspirone were similar (e.g., both drugs increased ratings of drug strength and clumsy/uncoordinated) and different (e.g., diazepam but not buspirone increased ratings of drowsy/sleepy; buspirone but not diazepam increased ratings of tense/nervous). This study demonstrates discriminative and self-reported effects of diazepam and buspirone at doses lower than previously shown to be behaviorally active, and suggests that at commonly used clinical doses, diazepam is relatively more discriminable than buspirone.     

Reaction time and diazepam plasma concentration in subjects with normal gastric pH

The present study investigated the relationship of reaction time with the plasma concentration of four different formulations of diazepam (liquid, original, new, and generic). On four separate days, over a 4-month period, subjects with normal gastric pH (N=7) were administered each of the diazepam formulations. Blood samples were obtained prior to each dose and at subsequent intervals. Likewise, reaction times were evaluated prior to dosing and then at ten different intervals post-dosing. As hypothesized, reaction time performance was associated with plasma diazepam concentration, regardless of formulation type. This suggests that reaction time is a sensitive measure of cerebral functioning for individuals who are treated with diazepam. In addition to providing a measure of drug effects on cognitive functioning, the results of this study have implications concerning possible hazards of operating machinery or driving motor vehicles for individuals being treated with diazepam.     

Reaction time and diazepam plasma concentration in subjects with normal gastric pH

The present study investigated the relationship of reaction time with the plasma concentration of four different formulations of diazepam (liquid, original, new, and generic). On four separate days, over a 4-month period, subjects with normal gastric pH (N=7) were administered each of the diazepam formulations. Blood samples were obtained prior to each dose and at subsequent intervals. Likewise, reaction times were evaluated prior to dosing and then at ten different intervals post-dosing. As hypothesized, reaction time performance was associated with plasma diazepam concentration, regardless of formulation type. This suggests that reaction time is a sensitive measure of cerebral functioning for individuals who are treated with diazepam. In addition to providing a measure of drug effects on cognitive functioning, the results of this study have implications concerning possible hazards of operating machinery or driving motor vehicles for individuals being treated with diazepam.     

Caffeine does not modulate inhibitory control

The effects of a 3mg/kg body weight (BW) dose of caffeine were assessed on behavioral indices of response inhibition. To meet these aims, we selected a modified AX version of the Continuous Performance Test (CPT), the stop task, and the flanker task. In three double-blind, placebo-controlled, within-subjects experiments, these tasks were administered to healthy participants. While the results for the AX-CPT were indicative of improved response inhibition after caffeine, they might also reflect caffeine-induced changes in mechanisms other than response inhibition (e.g., attentional processes). The results for the stop task and flanker task were more straightforward. That is, the effects of caffeine on overall flanker performance and selective response suppression as revealed by distribution-analytical techniques were negligible. In the stop task a global effect of caffeine on processing speed was seen, rather than specific effects on response inhibition. Taken together, these experiments showed that both active and reactive inhibition were not significantly modulated by caffeine. The present results are linked to neural circuits that underlie inhibitory control and the role of caffeine-induced strategic changes.     

Trait-anger enhances effects of caffeine on psychomotor vigilance performance

This study examined the combined effects of caffeine and the personality attribute of trait-anger on the speed of psychomotor vigilance performance during sleep deprivation. 23 young adult soldiers (19 male) were administered the State-Trait Anger Expression Inventory-2 when well-rested. Participants were then sleep deprived for three consecutive nights (77 hours total) during which they completed repeated psychomotor vigilance testing. Half of the participants received four doses of oral caffeine (200 mg every 2 hr.; 800 mg total) each night, while the other half were administered a placebo. For the first night, higher scores on trait-anger, outward anger expression, and intensity of anger expression predicted better sustained overnight vigilance performance, but only for those volunteers receiving caffeine. These correlations were not significant for the subsequent nights. Findings suggest a possible synergistic effect between personality traits associated with arousal of the central nervous system and vigilance-promoting effects of caffeine.     

Sociability/impulsivity and caffeine-induced arousal: Critical flicker/fusion frequency and procedural learning

The influence of sociability/impulsivity in caffeine-induced arousal effects was investigated in two separate experiments: Experiment 1 examined critical flicker/fusion frequency change scores (ΔCFFT) in 60 subjects; and Experiment 2 investigated procedural learning in 30 subjects. In the two experiments, subjects received either caffeine citrate (500 mg) or placebo. The pattern of results was consistent across both studies:(1) a strong interactive effect of sociability (as measured by the EPQ extraversion scale) by caffeine/placebo which showed that (a) subjects low in sociability showed the greatest increase in ACFFT and learned most under placebo, while the reverse was true under caffeine; (b) subjects high in sociability, showed no increase in ΔCFFT and learned least under placebo, while the reverse was true under caffeine; and (2) in neither experiment did impulsivity (as measured by the EPS impulsiveness scale) significantly interact with caffeine/placebo. The results are consistent with Eysenck's (The Biological Basis of Personality, 1967) theory of personality in suggesting that subjects low in sociability are highly arousable under low-arousal (placebo) but over-aroused under high-arousal (caffeine), with the reverse pattern of effects holding for subjects high in sociability. The implications of these data for the respective roles of sociability and impulsivity components of extraversion in arousal-mediated performance are discussed.     

Sleep Loss and Temporal Memory

Historical evidence suggests that sleep deprivation affects temporal memory, but this has not been studied systematically. We explored the effects of 36 hr of sleep deprivation on a neuropsychological test of temporal memory. To promote optimal performance, the test was short, novel, and interesting, and caffeine was used to reduce 'sleepiness'. A total of 40 young adults were randomized into four groups: control + caffeine (Cc), control + placebo (Cp), sleep deprived + caffeine (SDc), and sleep deprived + placebo (SDp). Controls slept normally. Caffeine (350 mg) or placebo were given just prior to testing. The task comprised colour photographs of unknown faces and had two components: recognition memory (distinction between previously presented and novel faces), and recency discrimination (temporal memory), when a previously shown face was presented. An interpolated task, self-ordered pointing, acted as a distraction. Caffeine had no effects within control conditions, but significantly reduced subjective sleepiness in SDc. Recognition was unaffected by sleep deprivation, whereas for recency, sleep deprivation groups scored significantly lower than controls. There was no significant improvement of recency with caffeine in the SDc group. Both sleep deprivation groups had poorer insight into their performance with recency. Self-ordered pointing remained unchanged. In conclusion, sleep deprivation impairs temporal memory (i.e. recency) despite other conditions promoting optimal performance.     

Diazepam and memory: Support for a duplex model of memory

In three experiments, the hypothesis that diazepam (Valium) selectively impairs the transfer of information from short-term to long-term memory was supported by differences between placebo and drugged subjects in effects of item difficulty, serial position (primacy-recency), and list length. In Experiment 1, diazepam reduced recall in the primacy portion of the serial position function, but produced no performance difference in the recency component. Recall decreased in placebo subjects as item difficulty increased, but drugged subjects were relatively uaffected by manipulation of difficulty. In Experiments 2 and 3, subjects treated with diazepam exhibited smaller gains in immediate recall with increasing list length than placebo controls. Retrieval of words learned before drug administration was not impaired by diazepam; in fact, it was significantly enhanced relative to control performance. The results add further support to the distinction between short- and long-term memory.     

Three consecutive nights of sleep loss: Effects of morning caffeine consumption on subjective sleepiness/alertness,reaction time and simulated driving performance

Caffeinated products are often consumed as a popular countermeasure to the effects of sleep loss. However, the efficacy of caffeine to exert these effects after consecutive nights of sleep loss is poorly understood. The aim of this study was to investigate the effects of three consecutive nights of restricted sleep and morning caffeine consumption on subjective ratings of sleepiness/alertness, reaction time, and simulated driving performance. Twenty healthy, habitual caffeine consumers (11 females; age: 23.3 ± 5.7 y; BMI: 22.3 ± 3.5 kg⋅m⁻²; caffeine intake: 204 ± 89 mg⋅day⁻¹; Mean ± SD) who had normal sleeping patterns (≥8 h⋅night⁻¹) participated in this double-blind, placebo-controlled, randomised study. Following one night of normal sleep (≥8 h time in bed (TIB)), participants underwent three consecutive nights of restricted sleep (5 h TIB). Participants received caffeine (200 mg; n = 10) or placebo (n = 10) capsules each morning and all participants received caffeine (100 mg) capsules each afternoon. Subjective ratings of alertness, concentration and tiredness were measured before and 1 h after morning capsule administration. Choice Reaction Time (CRT) was examined 1 h after morning capsule administration, with response speed and accuracy as outcome variables. Driving performance was assessed using a 30 min simulated driving task, with lateral (standard deviation of lane position [SDLP]; total number of line crossings [LC]) and longitudinal (standard deviation of speed [SDSP]) measures of vehicle control as outcome variables. Alertness and concentration significantly decreased, and tiredness increased across the three days of sleep loss. Caffeine only marginally alleviated these effects. No differences were observed between treatments or across trial days for response speed and accuracy on the CRT task. Likewise, no significant differences were observed between groups or across trial days for any measures of simulated driving performance. Overall, results from this study indicate that three consecutive days of sleep loss influence subjective ratings of alertness, concentration and tiredness, but does not alter CRT or simulated driving performance. Caffeine may alleviate some of the negative subjective effects imposed by restricted sleep, but the efficacy of caffeine to attenuate performance changes in CRT and driving performance were unable to be observed.     

Pavlovian Conditioning to a Diazepam Cue with Yohimbine as the Unconditional Stimulus

On multiple occasions, rats were administered diazepam (2.0 mg/kg, ip) followed 30 min thereafter by yohimbine hydrochloride (2.5 or 5.0 mg/kg) or isotonic saline (forward conditioning groups). Three additional groups (backward conditioning controls) were given equivalent injections, but in reverse order. After eight such pairings, the effects of a single injection of diazepam on motor performance (balancing on a rotating drum) was assessed. Rats that had received either dose of yohimbine during forward conditioning trials maintained their balance longer than the saline controls. After four additional conditioning trials, the animals’ activity patterns in a plus-maze screening test for anxiolytics were examined. Placed into the maze after a single test injection of isotonic saline, the behavior of all groups was virtually identical: less than 16% of total entries into or time spent in the four arms of the maze was spent in the two “open” arms (unprotected by surrounding walls). When tested in the maze again, but 35 min after a single injection of diazepam, the groups that had received diazepam but not yohimbine during the conditioning phase exhibited the expected increase in open-arm activity, and equivalent increases were found in backward conditioning groups. However, the group previously conditioned with 2.5 mg/kg of yohimbine following diazepam also showed an increased open-arm activity when tested with diazepam alone, but it was significantly greater than that seen in the control group. In contrast, the group conditioned with 5.0 mg/kg yohimbine following diazepam exhibited no effect of diazepam upon their plus maze activity; consequently, these animals spent less time in the open arms than either of the other groups. Yohimbine alone normally decreases open-arm activity (a putative “anxiogenic” effect) in a linear dose-dependent fashion. The fact that it had a bidirectional conditional effect on the diazepam cue drug demonstrates that a conditional response in drug → drug conditioning cannot always be predicted on the basis of the behavioral response to the signaled drug. Consideration is given to possible reasons for these effects of diazepam → yohimbine pairings in terms of the known neuropharmacological properties of yohimbine.