- Email: [email protected]
Effects of d-amphetamine on behavioral control in stimulant abusers: the role of prepotent response tendencies
Drug and Alcohol Dependence 71 (2003) 143 /152 www.elsevier.com/locate/drugalcdep
Effects of d-amphetamine on behavioral control in stimulant abusers: the role of prepotent response tendencies Mark T. Fillmore a,b,*, Craig R. Rush a,b,c, Cecile A. Marczinski a a
b
Department of Psychology, University of Kentucky, Lexington, KY 40506-0044, USA Department of Behavioral Science, University of Kentucky, Lexington, KY 40539-0086, USA c Department of Psychiatry, University of Kentucky, Lexington, KY 40539-0086, USA
Received 18 December 2002; received in revised form 6 March 2003; accepted 7 March 2003
Abstract The present study examined the effects of d -amphetamine on the ability to execute and inhibit behavior in a context where preliminary information signaled the likelihood that a response should be executed or suppressed. Eight adults (5 men and 3 women) with a history of stimulant abuse performed a cued go no-go task that required quick responses to go targets and suppression of responses to no-go targets. Performance was tested under four oral doses of d -amphetamine, 0 (placebo), 5, 10 and 20 mg, administered double-blind and in mixed order. d -Amphetamine produced a dose-dependent increase in inhibitory failures following invalid go cues and had no effect on inhibitory failures following valid no-go cues. d -Amphetamine had little effect on response execution as measured by reaction time. Subjective and physiological effects of d -amphetamine were also observed. The findings demonstrate that stimulant effects on fundamental aspects of behavioral control can be mediated by environmental cues that alter response tendencies. Identification of environmental conditions in which stimulants are likely to disinhibit behavior could provide insight into mechanisms that underlie the association between long-term stimulant use and poor impulse control. # 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: d -Amphetamine; Response inhibition; Human
1. Introduction Studies have shown that stimulant drugs can enhance human performance (e.g. Koelega, 1993; Weiss and Laties, 1962). Stimulants have been reported to allay fatigue, increase vigilance, speed performance, prolong effort, and generally increase work output (for review, see Koelega, 1993). Stimulants, such as methylphenidate (Ritalin) and d-amphetamine, are prescribed to treat behavioral and cognitive impairments associated with attention deficit/hyperactivity disorder (ADHD) and other disorders of self-control (Tannock, 1998). It also has been suggested that illicit use of stimulants, such as d -amphetamine and cocaine, might be motivated in part by a desire to self-medicate behavioral and cognitive deficits (e.g. Khantzian, 1985; Schiffer, 1988). However,
* Corresponding author. Tel.: /1-859-257-4728; fax: /1-859-3231979. E-mail address: [email protected] (M.T. Fillmore).
evidence for performance-enhancing effects of stimulants is not entirely consistent. Other studies have shown that stimulants, such as cocaine, can produce negative behavioral effects, such as impulsive responding (Fillmore et al., 2002). Although the reasons for the inconsistent findings are not clear, some researchers have pointed to cross-study differences in drug-use history, behavioral assessment methods, and experimental procedures (Ward et al., 1997). The complex nature of stimulant effects on human functioning highlights the need for model-based examinations and interpretation of their behavioral effects (Koelega, 1993). A number of theories postulate that behavioral control is governed by two distinct systems: one that activates behavior and one that inhibits behavior (Fowles, 1987; Gray, 1976, 1977; Logan and Cowan, 1984; Patterson and Newman, 1993; Quay, 1997). The relative strength of each of these systems is generally assumed to determine behavioral control, and impaired behavioral control might arise from either a
03765-8716/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0376-8716(03)00089-9
144
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
weakened inhibitory system or from a heightened activation system. Model-based cognitive assessments, such as stopsignal tasks, have been used to study how drugs alter inhibitory and activational response tendencies to either enhance or reduce behavioral control (deWit et al., 2002; Fillmore et al., 2002). Stop-signal tasks are choice reaction time (RT) measures that require individuals to quickly respond to go-signals and inhibit a response when a stop-signal occurs (Logan, 1994; Logan and Cowan, 1984). Studies of stop-signal performance have shown that CNS-depressants, such as alcohol and triazolam, impair behavioral control by reducing inhibitory tendencies at doses that do not affect the activation of responses (e.g. Fillmore et al., 2001; Fillmore and Vogel-Sprott, 1999, 2000; Mulvihill et al., 1997). The stop-signal paradigm also has been used to study the effects of stimulant use. Individuals with a history of stimulant abuse display reduced response inhibition compared with age-matched control subjects (Fillmore and Rush, 2002). Moreover, stimulant abusers displayed reduced behavioral inhibition in response to acute oral administration of cocaine (Fillmore et al., 2002). These findings are important because they implicate a druginduced impairment of inhibitory tendencies that could help explain the documented association between stimulant abuse and poor impulse control (Jentsch and Taylor, 1999). Although the stop-signal model has been useful in demonstrating that stimulants can affect response inhibition, the model does not provide information on environmental conditions that might exacerbate or ameliorate drug effects on response inhibition and response execution. The environment likely exerts some stimulus control over inhibitory and activational tendencies. Stimulus cues often precede signals to inhibit or respond and can provide preliminary information regarding the type of signal likely to follow (i.e. go or stop). Predictive cues can facilitate the activation and inhibition of behavior by initiating preparatory processes required for the activation or inhibition of an action (Duncan, 1981; Posner, 1980; Posner et al., 1980). Cue effects on the activation and inhibition of behavior have been studied in the laboratory using cued go no-go tasks (e.g. Miller et al., 1991). The tasks measure the effect of preliminary information on the ability to quickly execute and suddenly suppress responses to subsequent go and stop-signals. The tasks typically present a stimulus cue followed by a go or nogo target stimulus that requires a response to be either executed (go) or suppressed (no-go). The cue provides information concerning the probability that a go or nogo target will be presented. The cue-target relationship is manipulated so that cues have a high probability of correctly signaling a target and a low probability of
incorrectly signaling a target. Correct (i.e. valid) cues tend to facilitate response activation and response inhibition. For example, responses to go targets are faster when they are preceded by a go cue. The speeding effect is attributed to covert response preparation that occurs before the actual go target is presented (Posner, 1980). Response preparation can have detrimental effects on behavioral control when the response must be suppressed. Go cues are invalid (i.e. incorrect signals) when they precede a no-go target. Failures to suppress responses to no-go targets are more frequent following invalid go cues, suggesting that response preparation can decrease the likelihood that an action can be suddenly inhibited when necessary (Miller et al., 1991). Studies have used cued go no-go tasks to test alcohol effects on response activation and response inhibition (Abroms et al., 2002; Marczinski and Fillmore, 2003). Those studies showed stereotypic, dose-dependent impairing effects on the ability to inhibit and to execute responses that were preceded by invalid cues. By contrast, no impairing effects of alcohol were observed when cues correctly signaled the inhibition or activation of responses. The research highlights the potential contribution of preliminary information in determining drug effects on behavioral control. To date, no research has tested the effects of a stimulant drug on the ability to activate or inhibit behavior in the presence of cues that correctly or incorrectly signal these actions. Given that stimulants can impair inhibitory control in individuals with a history of stimulant abuse (Fillmore et al., 2002), it is important to determine if this effect is intensified or attenuated in contexts where cues correctly or incorrectly signal the preparation to execute or inhibit an action. The present research tested the effects of damphetamine (5, 10, 20 mg, and placebo) on behavioral control in contexts where preliminary information (i.e. cues) either facilitated or impeded the ability to quickly activate or suddenly inhibit a response.
2. Method 2.1. Subjects Eight adult volunteers (5 men and 3 woman) with histories of cocaine use participated in the study. Subjects’ ages ranged between 20 and 48 years (mean /37.0, S.D. /8.2). The racial make-up of the sample was African American (n/6), Asian (n/1) and Caucasian (n/1). Participants were recruited via notices posted on community bulletin boards and by word of mouth. The study was approved by the University of Kentucky Medical Institutional Review Board, and subjects provided their written, informed consent prior to participating.
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
Volunteers had to have: a minimum of grade 8 education, demonstrated reading ability, no uncorrected vision or auditory problems, and no self-reported major Axis I psychiatric disorder, head trauma, or other CNS injury. Volunteers also had to meet all of the following criteria: (1) a score of at least 4 on the Drug Abuse Screening Test (Skinner, 1982); (2) self-report of past month stimulant use of either d-amphetamine or cocaine. All subjects reported cocaine use and reported smoking cocaine in the form of crack. No volunteer was currently in, or seeking, treatment for their substance use. The sample reported an average frequency of cocaine use of 2.7 days in the past week (S.D. /2.4), and 11.2 days in the past month (S.D./10.7). The sample reported smoking a mean daily amount of 1.9 rocks of crack cocaine (S.D. /2.3), and spending an average of $87.80 per week on cocaine (S.D. /107.4). The sample reported using cocaine for an average period of 7.8 years (S.D. /5.8). All volunteers also reported using nicotine, alcohol, and caffeine. The mean total weekly alcohol consumption for this sample was 42.8 standard drinks (S.D. /45.1). The sample reported smoking an average of 15.8 (S.D. /9.3) cigarettes per day, and an average, total daily caffeine consumption of 95.7 mg (S.D. /98.1). Most of the sample also reported some past month use of marijuana (n/5). One individual reported using a hallucinogen once in the past month and two individuals reported using a benzodiazepine once in the past month. No other drug-use, including prescription medication, was reported by the sample. Prior to participation, all potential volunteers completed a comprehensive medical history questionnaire, drug-use questionnaire, mini-mental status examination (Folstein et al., 1975), and vital sign assessment. Routine, clinical laboratory blood chemistry tests and an electrocardiogram were conducted on all potential volunteers. Potential volunteers with histories of serious physical disease, current physical disease, impaired cardiovascular functioning, chronic obstructive pulmonary disease, seizure, head trauma, CNS tumors, or past histories of psychiatric disorder, (i.e. Axis I, DSM IV), other than substance use or dependence, were excluded from participation. Female volunteers had to have an effective form of birth control in use. All subjects were in good health with no contraindications to stimulant drugs. 2.2. Apparatus and materials 2.2.1. Cued go no-go task Participants performed a cued go no-go RT task that was operated using E-Prime experiment generation software (Schneider et al., 2002) and was performed on a PC computer. A trial involved the following sequence of events: (1) presentation of a fixation point
145
(/) for 800 ms, (2) a blank, white screen for 500 ms, (3) a cue, displayed for one of five stimulus onset asynchronies (SOAs/100, 200, 300, 400 and 500 ms), (4) a go or no-go target, that remained visible until a response occurred or 1000 ms had elapsed, and (5) an inter-trial interval of 700 ms. The cue was a rectangle (7.5 /2.5 cm2) framed in a 0.8 mm black outline that was presented in the center of the computer monitor against a white background. The cue was presented in either a horizontal (height /2.5 cm, width/7.5 cm) or vertical (height /7.5 cm, width/2.5 cm) orientation. The go and no-go targets were the colors, green and blue, respectively. They were displayed on the monitor as a solid hue that filled the interior of the rectangle cue. Subjects were instructed to press the forward slash (/) key on the keyboard as soon as a go (green) target appeared and to suppress the response when a no-go (blue) target was presented. Key presses were made with the index finger of the preferred hand. The go and no-go targets were presented in hues that were easily distinguishable by all participants. The orientation of the cue (horizontal or vertical) signaled the probability that a go or no-go target would be displayed. Cues that were presented vertically preceded the go target on 80% of the trials and preceded the no-go target on 20% of the trials. Cues that were presented horizontally preceded the no-go target on 80% of the trials and preceded the go target on 20% of the trials. Therefore, based on cue-target pairings, vertical and horizontal cues operated as go and no-go cues, respectively. The different SOAs (100, 200, 300, 400 and 500 ms) between cues and targets encouraged volunteers to pay attention to the cues and the variability and randomness of the SOAs prevented the participants from anticipating the exact onset of the targets. A test consisted of 500 trials that presented the four possible cue-target combinations (see Table 1). An equal
Table 1 Mean peak physiological effect and DEQ ratings under each dose condition Dose condition (mg) 0
5
10
20
Physiological effects Heart rate (BPM) Systolic BP (mmHg) Diastolic BP (mmHg)
76.9 (8.6) 138.9 (12.1) 86.9 (4.6)
78.8 (10.6) 141.5 (10.5) 84.0 (6.6)
77.4 (10.1) 142.7 (10.0) 88.8 (7.2)
79.0 (10.1) 152.2 (12.3) 91.9 (5.1)
DEQ effects Any effects Good effects
0.5 (0.5) 0.2 (0.5)
0.8 (0.7) 0.6 (0.7)
0.8 (0.7) 0.6 (0.7)
1.4 (0.9) 1.1 (1.0)
N/8. Parentheses indicate standard deviations.
146
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
number of vertical (250) and horizontal (250) cues were presented before an equal number of go (250) and no-go (250) target stimuli. Each cue-target combination was presented at each of the five SOAs, and an equal number of SOAs separated each cue-target combination. The presentation of cue-target combinations and SOAs was random. For each trial, the computer recorded whether or not a response occurred and if so, the RT in milliseconds was measured from the onset of the target until the key was pressed. To encourage fast and accurate responding, feedback was provided to the subject. Feedback was presented during the inter-trial interval on all trials by displaying the words ‘correct’ or ‘incorrect’ along with millisecond RTs on trials in which responses occurred. Subjects had a 10 min rest break after completing the first 250 trials. A test required approximately 30 min to complete. 2.2.2. Drug effect questionnaire This questionnaire, used in prior research (e.g. Fillmore et al., 2002; Rush et al., 1998), consisted of 15 items that were presented individually on the computer monitor. Subjects rated each item using the computer mouse to select among one of five response options: not at all, a little bit, moderately, quite a bit, and very much (scored numerically from 0 to 4, respectively). The items rated were: any effects, active/ alert/energetic, bad effects, good effects, high, irregular heart beat/racing, like, anxious/nervous, pay for this drug, rush, shaky/jittery, take this drug again, talkative/ friendly, nauseated/queasy, and sluggish/fatigued/lazy. 2.3. General procedures The study was conducted at the Laboratory of Human Behavioral Pharmacology at the University of Kentucky. All participants were tested individually. All sessions were scheduled from Monday to Friday between 8:30 am and 4:30 pm. Subjects were instructed not to drink alcohol or take any drug during their scheduled lab days. During all sessions, participants provided a breath sample using an Intoxilyzer Model 400 (CMI Inc. Owensborough, KY) to ensure a zero blood alcohol concentration (BAC). A urine sample also was obtained to test for the presence of cocaine/benzoylecgonine, benzodiazepines, barbiturates, tetrahydrocannabinol (THC), d -amphetamine, and opiates (On Trak TesTstiks, Roche Diagnostics Corporation, Indianapolis, IN). Females were also tested for pregnancy via urine analysis. With the exception of cocaine/benzoylecgonine and THC, no volunteers tested positive for the presence of any drug at the time of study. Subjects were informed that the study examined the effects of various drugs (e.g. stimulants, benzodiazepines, barbiturates) on mood and behavior. They were given no information about the specific drug examined in the study. Subjects attended a
familiarization/practice session and four dose administration test sessions. Sessions required 7.5 h to complete and were separated by a minimum of 48 h. 2.4. Familiarization/practice session Subjects were acquainted with the task requirements and general testing procedure. All testing occurred in a quiet test room that contained a chair, desk, and computer. The experimenter explained the requirements for the cued go no-go task. To verify that the participants could easily distinguish the colors of the stimuli in the task, each subject was presented with a color vision test that required them to discriminate between the green and blue colors that were used as go and no-go targets in the cued RT task. Subjects were told that they would see empty rectangular boxes become filled with green or blue ‘target’ colors. Subjects were instructed to press the forward slash key (/) on the keyboard as quickly as possible whenever a green target appeared and to make no response whenever a blue target appeared. The computer displayed how fast a subject responded to each go target by presenting the milliseconds required from target onset until the key press. Subjects were encouraged to make fast responses (i.e. in the fewest milliseconds). No information regarding cuetarget relationships was provided. Subjects then performed practice tests on the task. A single test is sufficient to produce cue-dependant responding (Marczinski and Fillmore, 2003). Subjects were also familiarized with the drug effect questionnaire (DEQ), and physiological monitoring equipment. Subjects completed the questionnaires and had their heart rate and blood pressure measured by a Dinamap XL Vital Signs Monitor (Johnson and Johnson Medical Inc., Arlington TX) at intermittent intervals over a 6-h period. 2.5. Dose administration sessions Prior to beginning each session, subjects ate a low-fat breakfast at 7:00 am. Subjects were allowed to smoke one cigarette between 7:30 and 8 am. No smoking was allowed during the sessions. 2.5.1. Baseline testing and dose administration The testing began at 9 am with baseline (i.e. precapsule) DEQ and physiological measures. Subjects then received one of four doses of d-amphetamine: 0 (placebo), 5, 10 and 20 mg. All doses were administered orally in a double-blind fashion. Drug doses were prepared by encapsulating 5 mg d -amphetamine in size 00 blue and white opaque capsules combined with lactose filler. Placebo capsules contained only lactose. Subjects ingested four capsules during each dose administration session. The combination of active and placebo
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
capsule determined the dose. Capsules were taken orally with 150 ml of water. Dose order across the four dose administration sessions was random without replacement and included a safety constraint that prevented any subject from receiving 20 mg d -amphetamine on the first session. Thus eight different dose orders were tested, with each order unique to a single subject. The highest dose (20 mg d-amphetamine) was administered to at least two subjects in each of the three sessions that followed the first session. The dose was received by two subjects in the second session, by four subjects in the third session, and by two subjects in the four session. Medications were prepared by the University of Kentucky Investigational Pharmacy. 2.5.2. Post-capsule testing Subjects performed the cued go no-go task at 2.5 h post-capsule. Because the test required 30 min to complete, the assessment spanned 2.5 /3.0 h postcapsule. This time interval was chosen to maximize the likelihood of observing a drug effect on inhibitory control. Time-course analyses of oral d -amphetamine dose effects on physiological and subjective measures typically show peak effects approximately 2.5 /3.0 h following administration (Rush et al., 1998; Vree and Henderson, 1980). Subjects completed the DEQ at 3.0 and 4.0 h post-capsule. Physiological measures were recorded at 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 5.5 h post-capsule. No volunteers experienced any adverse reaction to the drug or any procedure. Volunteers were paid $400 for their participation. 2.6. Criterion measures and data analyses of cued go nogo performance The two primary measures of interest were the participants’ failures to inhibit responses to no-go targets (failures of response inhibition) and their speed of responding to go targets (response execution). 2.6.1. Failure of response inhibition Failure of response inhibition was measured as the proportion of no-go targets in which a participant failed to inhibit a response. These p-inhibition failure scores were calculated for each cue condition (go and no-go) on each test. Scores were analyzed by a 4 Dose (0, 5, 10, 20 mg d -amphetamine) X 2 Cue (go cue vs. no-go cue) within-subjects ANOVA. 2.6.2. Response execution Response execution was measured by the RT to go targets. Shorter RTs indicated greater facilitation of response execution. A mean RT score for a participant was calculated for each cue. Responses with RTs less
147
than 100 ms and greater than 1000 ms were excluded. These outliers were infrequent, occurring on average less than 0.25% of the trials for which a response was observed (i.e. less than one trial per test). Scores were analyzed by a 4 (Dose) X 2 (Cue) within-subjects ANOVA. Omission errors were also recorded. These errors occurred when participants failed to respond to go targets. Omission errors were infrequent and occurred on less than 1% of go target trials (/2 trials per test).
2.6.3. Cue-dependency scores The degree to which a subject’s p-inhibition failures and RT was influenced by valid and invalid cues were also represented as single cue-dependency scores, based on the difference in performance between the invalid and valid cue conditions for a test. Greater cuedependency scores reflect greater reliance on the environmental context (i.e. the cues) to maintain efficient suppression and execution of responses (Neill, 1997; Posner and Snyder, 1975). The cue-dependency score for p-inhibition failures was calculated as the p-inhibition failure score in the go cue condition minus the pinhibition failure score in the no-go cue condition. A larger cue-dependency score for p-inhibition failures indicates greater reliance on valid cues (no-go cues) to successfully inhibit responses to no-go targets. The cuedependency score for RT was calculated as the mean RT in the no-go cue condition minus the mean RT in the go cue condition. A larger cue-dependency score for RT indicates greater reliance on valid cues (go cues) for fast response execution.
2.7. Criterion measures and data analyses of physiological and subjective effects Two sets of analyses examined physiological and selfreported subjective effects. First, peak effect data were calculated and analyzed by one-factor within-subjects ANOVA with Dose (0, 5 10 and 20 mg d -amphetamine) as a repeated measures factor. For physiological measures, peak effect was defined as the maximum value from 0.5 to 5.5 h after drug administration. For subjective measures, peak effect was the maximum value from 3 to 4 h after drug administration. Second, timecourse data were analyzed by two-factor, within-subjects ANOVA of Dose (0, 5, 10 and 20 mg) and Time (precapsule, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 5.5 h). Subjective effect ratings were analyzed for each DEQ item by a two-factor, within-subjects ANOVA of Dose (0, 5, 10 and 20 mg) and Time (pre-capsule, 3.0 and 4.0 h).
148
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
3. Results 3.1. Dose effects on inhibition failures (P-inhibition failures) A 4 (Dose) X 2 (Cue) within-subjects ANOVA of pinhibition failure scores revealed a significant Dose X Cue interaction (F3,21 /3.9, P /0.023). Fig. 1 illustrates the mean p-inhibition failure scores. The figure shows that p-inhibition failures following no-go cues were low and were not affected by dose. By contrast, p-inhibition failures were greater following go cues and increased as a function of d -amphetamine dose. These observations were confirmed by two simple effect one-way ANOVAs that compared p-inhibition failure scores across doses separately for each cue condition. There was no significant dose effect on p-inhibition failure scores following no-go cues (P /0.380). By contrast, a significant dose effect was observed following the go cues (F3,21 /4.1, P /0.020). Follow-up simple effects analyses of the go cue condition compared p-inhibition failure scores under each active dose to placebo. Fig. 1 shows that the 5 and 20 mg active doses significantly increased p-inhibition failure scores compared with placebo (Ps B/0.018). Although greater p-inhibition failures were observed under the 10 mg dose compared with placebo, the difference was not statistically significant (P /0.209). 3.2. Dose effects on response execution (reaction time) A 4 (Dose) X 2 (Cue) within-subjects ANOVA of RTs to go targets revealed significant main effects of Dose
(F3,21 /3.7, P /0.027) and Cue (F1,7 /159.9, P B/ 0.001). There was no significant interaction (P / 0.692). Fig. 2 shows that RTs were slower following no-go cues. The figure also shows that RTs decreased slightly as a function of dose and this speeding effect was similar across cue conditions. Follow-up simple effects analyses compared RT scores under each active dose to placebo in each cue condition. No significant differences were obtained (Ps /0.127). 3.3. Cue-dependency scores for P-inhibition failures and reaction time Dose effects on the suppression and execution of responses were also examined in terms of the magnitude of cue-dependency (the difference in performance between the invalid and valid cues). Fig. 3 plots the mean cue-dependency scores under each dose for p-inhibition failures and RT. The figure shows that cue-dependency of inhibition failures increased as a function of dose. By contrast, the cue-dependency of RT showed no orderly effect of dose. 3.4. Physiological and subjective effect measures ANOVAs of time-course effects obtained significant dose X time interactions for heart rate (F33,231 /1.7, P /0.009), and systolic blood pressure (F33,231 /1.6, P /0.018). No significant effects were obtained for diastolic blood pressure (Ps /0.095). ANOVAs of peak effects obtained significant dose effects for systolic (F3,21 /4.8, P /0.011), and diastolic blood pressure (F3,21 /3.1, P /0.048). No significant dose effect was
Fig. 1. Mean proportion of failures to inhibit responses to no-go targets following go and no-go cues under four doses of oral d -amphetamine: 0 (placebo), 5, 10, and 20 mg. Capped vertical lines show standard errors of the mean. *, indicates significant differences from placebo, P B/0.05.
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
149
Fig. 2. Mean RT to respond to go targets following go and no-go cues under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean.
obtained for peak heart rate (F3,21 /0.4, P /0.746). Table 1 presents the mean peak physiological effects for each measure under the four dose conditions. Peak effect DEQ ratings of items, Any Effect (F3,21 /6.0, P / 0.004) and Good Effect (F3,21 /3.2, P /0.042), increased under active dose conditions. Table 1 presents the mean peak rating of these two DEQ items under the four dose conditions. No other DEQ measures showed significant dose effects in either time-course or peak effect analyses (Ps /0.05).
4. Discussion This research used a cued go no-go RT task to test the acute effects of d -amphetamine on aspects of behavioral control in individuals with a history of stimulant abuse. The results showed that the effect of d-amphetamine on response suppression was dependent on cues that signaled the likelihood that a response should be executed or suppressed. Overall, inhibitory failures were more frequent following go cues compared with
Fig. 3. Left. Mean cue-dependency scores for proportion of failures to inhibit responses to no-go targets under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean. Right. Mean cue-dependency scores for RT to respond to go targets under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean.
150
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
no-go cues. d -Amphetamine produced a dose-dependent increase in inhibitory failures following go cues and had no effect on inhibitory failures following no-go cues. Inhibition failures following no-go cues were nearly absent in all dose conditions. A near-zero measure in this cue condition could impose a floor effect that would be a problem for studies of treatments expected to improve inhibitory control by reducing inhibitory failures. d -Amphetamine had little effect on response execution as measured by RT. Overall, volunteers displayed faster RTs to go targets that followed go cues compared with those that followed no-go cues. d Amphetamine produced a significant, but only slight speeding effect on RT that was evident in both cue conditions. The present findings are consistent with our prior research that showed long-term stimulant abusers displayed reduced behavioral inhibition on a stop-signal task in response to acute oral administration of cocaine (Fillmore et al., 2002). Like the stop-signal task, the cued go no-go task used in the present study measured behavioral control by the ability to quickly execute and suddenly suppress responses. However, unlike the stopsignal task, the cued task concerned responses to stimuli that were intended to be dependent on environmental cues. As such, the cued go no-go task did not include a no-cue, or neutral cue condition because those conditions reduce the overall proportion of valid cue trials and decrease participants’ reliance on cues for response preparation (Gottlob et al., 1999; Posner and Snyder, 1975). The large cue effects on RT and response suppression in the present study confirm that the 80% reliable cue-target association was effective in producing cue-dependent behavior. The dose-dependent increase in cue-dependency of inhibition failures suggests that subjects became more reliant on environment cues for inhibiting responses under the drug. Previous research shows that alcohol also increases the cue-dependency of inhibitory failures (Marczinski and Fillmore, 2003). However, that study found that alcohol also increased the cue-dependency of RT by making subjects more reliant on valid go cues for maintaining fast response execution. By contrast, d amphetamine had a slight speeding effect on RT, but did not increase the degree to which RT was cue-dependent. Evidence that d -amphetamine had different effects on cue-dependency of inhibitory and activational aspects of behavior supports the independence assumption regarding these two mechanisms of behavioral control (e.g. Logan and Cowan, 1984). The dissociative effect is also consistent with other studies of stimulants that find differences between inhibitory and activational effects. For example, Tannock et al. (1995) measured the ability to inhibit and execute responses following 0.0, 0.3, 0.6 and 0.9 mg/kg methylphenidate in children diagnosed with ADHD and found that response execution was
facilitated in a dose-dependent manner. However, measures of response inhibition revealed an inverted U-shaped dose response curve, suggesting that facilitative effects of psychostimulants on inhibitory control might be confined to a particular dose window. Demonstration of these types of dissociative drug effects are potentially important because they provide information about the specificity of a drug effect on a behavioral system and can identify levels of independence between behavioral mechanisms and their underlying neuroanatomical systems. Evidence that d -amphetamine effects on behavioral control can be environmentally-dependent might help explain reports of different effects of d -amphetamine on behavioral control. Some recent studies have found that d -amphetamine can improve inhibitory control in healthy subjects as measured by stop-signal tasks (deWit et al., 2000, 2002). Stimulant effects on behavioral control could differ depending on the degree to which inhibitory and activational aspects of behavior are under stimulus control. Impairing effects on inhibitory control might be more likely in contexts where there is a strong response tendency. Subject characteristics might also contribute to differences in stimulant effects. There is growing evidence that long-term habitual stimulant users display neuropsychological impairments and have neuroanatomical abnormalities, including deficits of inhibitory control (e.g. Fillmore and Rush, 2002; Lane et al., 1998). Some investigators suggest that repeated dopaminergic activation of prefrontal pathways by chronic stimulant use eventually impairs inhibitory functions, leading to a loss of control over behavioral impulses (e.g. Lyvers, 2000; Volkow et al., 1996). Neuronal changes owing to chronic stimulant use could alter the acute behavioral response to stimulants in those with a long history of stimulant use. The importance of these and other factors as determinants of stimulant effects on behavioral control await exploration. Cue-dependent models of behavioral control also have been applied to study deficits of behavioral control in relation to other conditions, such as age-related cognitive decline. A general finding from that research is that elderly subjects display greater reliance on cues in performance tests as compared with younger adults (Laver and Burke, 1993). Such age-related increases in cue-dependency for behavioral control parallel acute drug-induced increases in cue-dependancy, like that observed in the present research. Such inter-disciplinary comparisons might provide valuable insight into the potential interactions between drug effects and chronic conditions that are characterized by impaired control. The continued broad application of these models could allow for potentially informative comparisons between differences sources of impaired control (e.g. druginduced vs. age-related).
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
The present findings contribute to our growing understanding of how drugs affect behavioral control in humans. The findings suggest that moderate doses of d -amphetamine can be particularly detrimental to behavioral control when it depends on the sudden and complete suppression of responses. The present results are consistent with studies of cocaine that also demonstrate drug-associated impairments in the ability to inhibit behavior (Fillmore and Rush, 2002; Fillmore et al., 2002). The current study expands upon that research by demonstrating that environmental cues are an important factor in determining stimulant effects on fundamental aspects of behavioral control.
Acknowledgements This research was supported by grants DA14079 and DA10325 from the National Institute on Drug Abuse and by grant AA12895 from the National Institute on Alcohol Abuse and Alcoholism. Offprint requests should be sent to Mark T. Fillmore, Ph.D., Department of Psychology, University of Kentucky, Lexington, KY 40506-0044. E-mail: [email protected].
References Abroms, B.D., Fillmore, M.T., Marczinski, C.A., 2002. Alcoholinduced impairment of behavioral control: effects on alteration and suppression of prepotent responses, J. Stud. Alcohol, submitted for publication. deWit, H., Crean, J., Richards, J.B., 2000. Effects of d -amphetamine and ethanol on a measure of behavioral inhibition in humans. Behav. Neurosci. 114, 830 /837. deWit, H., Engasser, J.L., Richards, J.B., 2002. Acute administration of d -amphetamine decreases impulsivity in healthy volunteers. Neuropsychopharmacology 27, 813 /825. Duncan, J., 1981. Directing attention in the visual field. Percept. Psychophys. 30, 90 /93. Fillmore, M.T., Rush, C.R., 2002. Impaired inhibitory control in chronic cocaine abusers. Drug Alcohol Depend. 66, 265 /273. Fillmore, M.T., Rush, C.R., Kelly, H.K., Hays, L., 2001. Triazolam impairs inhibitory control of behavior in humans. Exp. Clin. Psychopharmacol. 9, 363 /371. Fillmore, M.T., Rush, C.R., Hays, L., 2002. Acute effects of oral cocaine on inhibitory control of behavior in humans. Drug Alcohol Depend. 67, 157 /167. Fillmore, M.T., Vogel-Sprott, M., 1999. An alcohol model of impaired inhibitory control and its treatment in humans. Exp. Clin. Psychopharmacol. 7, 49 /55. Fillmore, M.T., Vogel-Sprott, M., 2000. Response inhibition under alcohol: effects of cognitive and motivational conflict. J. Stud. Alcohol 61, 239 /246. Folstein, M., Folstein, S., Hugh, P., 1975. Mini-Mental State */a practical method for grading the cognitive state of patients for the clinician. J. Psychiat. Res. 12, 189 /198. Fowles, D.C., 1987. Application of a behavioral theory of motivation to the concepts of anxiety and impulsivity. J. Res. Pers. 21, 417 / 435.
151
Gottlob, L.R., Cheal, M., Lyon, D.R., 1999. Time course of locationcuing effects with a probability manipulation. J. Gen. Psychol. 126, 261 /270. Gray, J.A., 1976. The behavioral inhibition system: a possible substrate for anxiety. In: Feldman, M.P., Broadhurst, A. (Eds.), Theoretical and Experimental Bases of the Behavior Therapies. Wiley, London, pp. 3 /41. Gray, J.A., 1977. Drug effects on fear and frustration. Possible limbic site of action of minor tranquilizers. In: Iverson, L.L., Iverson, S.D., Snyder, S.H. (Eds.), Handbook of Psychopharmacology, vol. 8. Plenum, New York, pp. 433 /529. Jentsch, J.D., Taylor, J.R., 1999. Impulsivity resulting from frontostriatal dysfunction in drug abuse: implication for the control of behavior by reward-related stimuli. Psychopharmacology 146, 373 /390. Khantzian, E., 1985. The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. Am. J. Psychiat. 142, 1259 /1264. Koelega, H.S., 1993. Stimulant drugs and vigilance performance: a review. Psychopharmacology 111, 1 /16. Lane, S.D., Cherek, D.R., Dougherty, D.M., Moeller, F.G., 1998. Laboratory measurement of adaptive behavior change in humans with a history of substance dependence. Drug Alcohol Depend. 51, 239 /252. Laver, G.D., Burke, D.M., 1993. Why do semantic priming effects increase in old age? A meta-analysis. Psychol. Aging 8, 34 /43. Logan, G.D., 1994. On the ability to inhibit thought and action: a user’s guide to the stop-signal paradigm. In: Dagenbach, D., Carr, T.H. (Eds.), Inhibitory Processes in Attention, Memory, and Language. Academic Press, San Diego, CA, pp. 189 /239. Logan, G.D., Cowan, W.B., 1984. On the ability to inhibit thought and action: a theory of an act of control. Psychol. Rev. 91, 295 / 327. Lyvers, M., 2000. Loss of control in alcoholism and drug addiction: a neuroscientific interpretation. Exp. Clin. Psychopharmacol. 8, 225 /249. Marczinski, C.A., Fillmore, M.T., 2003. Pre-response cues reduce the impairing effects of alcohol on the execution and suppression of responses. Exp. Clin. Psychopharmacol. 9, 110 /117. Miller, J., Schaffer, R., Hackley, S.A., 1991. Effects of preliminary information in a Go versus No-go task. Acta Psychol. 76, 241 /292. Mulvihill, L.E., Skilling, T.A., Vogel-Sprott, M., 1997. Alcohol and the ability to inhibit behavior in men and women. J. Stud. Alcohol 58, 600 /605. Neill, W.T., 1997. Episodic retrieval in negative priming and repetition priming. J. Exp. Psychol: Learn. Mem. Cognit. 23, 1291 /1305. Patterson, C.M., Newman, J.P., 1993. Reflectivity and learning from aversive events: toward a psychological mechanism for the syndromes of disinhibition. Psychol. Rev. 100, 716 /736. Posner, M.I., 1980. Orienting of attention. Q. J. Exp. Psychol. 32, 3 / 25. Posner, M.I., Snyder, C.R., 1975. Facilitation and inhibition in the processing of signals. In: Rabbitt, P., Dornic, S. (Eds.), Attention and Performance, vol. V. Academic Press, New York. Posner, M.I., Snyder, C.R., Davidson, B.J., 1980. Attention and the detection of signals. J. Exper. Psychol. 109, 160 /174. Quay, H.C., 1997. Inhibition and attention deficit hyperactivity disorder. J. Abnorm. Child Psychol. 25, 7 /13. Rush, C.R., Kollins, S.H., Pazzaglia, P.J., 1998. Discriminativestimulus and participant-rated effects of methylphenidate, bupropion and triazolam in d -amphetamine-trained humans. Exp. Clin. Psychopharm. 6, 32 /44. Schiffer, F., 1988. Psychotherapy of nine successfully treated cocaine abusers: techniques and dynamics. J. Sub. Abuse Treat. 5, 131 / 137. Schneider, W., Eschman, A., Zuccolotto, A., 2002. E-Prime User’s Guide. Psychology Software Tools Inc, Pittsburgh, PA.
152
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
Skinner, H.A., 1982. The drug abuse screening test. Addict. Behav. 7, 363 /371. Tannock, R., 1998. Attention deficit hyperactivity disorder: advances in cognitive, neurobiological, and genetic research. J. Child Psychol. Psychiat. 39, 65 /99. Tannock, R., Schachar, R., Logan, G., 1995. Methylphenidate and cognitive flexibility: dissociated dose effects in hyperactive children. J. Abnorm. Child Psychol. 23, 235 /267. Volkow, N.D., Ding, Y., Fowler, J.S., Wang, G., 1996. Cocaine addiction: hypothesis derived from imaging studies with PET. J. Addict. Dis. 15, 55 /71.
Vree, T.B., Henderson, P.T., 1980. Pharmacokinetics of amphetamines: in vivo and in vitro studies of factors governing their elimination. In: Caldwell, J. (Ed.), Amphetamine and Related Stimulants: Chemical Biological Clinical and Sociological Aspects. CRC Press, Boca Raton, FL, pp. 47 /68. Ward, A.S., Kelly, T.H., Foltin, R.W., Fischman, M.W., 1997. Effects of d -amphetamine on task performance and social behavior of humans in a residential laboratory. Exp. Clin. Psychopharm. 5, 130 /136. Weiss, B., Laties, V.G., 1962. Enhancement of human performance by caffeine and the amphetamines. Pharmacol. Rev. 14, 1 /36.
Effects of d-amphetamine on behavioral control in stimulant abusers: the role of prepotent response tendencies Mark T. Fillmore a,b,*, Craig R. Rush a,b,c, Cecile A. Marczinski a a
b
Department of Psychology, University of Kentucky, Lexington, KY 40506-0044, USA Department of Behavioral Science, University of Kentucky, Lexington, KY 40539-0086, USA c Department of Psychiatry, University of Kentucky, Lexington, KY 40539-0086, USA
Received 18 December 2002; received in revised form 6 March 2003; accepted 7 March 2003
Abstract The present study examined the effects of d -amphetamine on the ability to execute and inhibit behavior in a context where preliminary information signaled the likelihood that a response should be executed or suppressed. Eight adults (5 men and 3 women) with a history of stimulant abuse performed a cued go no-go task that required quick responses to go targets and suppression of responses to no-go targets. Performance was tested under four oral doses of d -amphetamine, 0 (placebo), 5, 10 and 20 mg, administered double-blind and in mixed order. d -Amphetamine produced a dose-dependent increase in inhibitory failures following invalid go cues and had no effect on inhibitory failures following valid no-go cues. d -Amphetamine had little effect on response execution as measured by reaction time. Subjective and physiological effects of d -amphetamine were also observed. The findings demonstrate that stimulant effects on fundamental aspects of behavioral control can be mediated by environmental cues that alter response tendencies. Identification of environmental conditions in which stimulants are likely to disinhibit behavior could provide insight into mechanisms that underlie the association between long-term stimulant use and poor impulse control. # 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: d -Amphetamine; Response inhibition; Human
1. Introduction Studies have shown that stimulant drugs can enhance human performance (e.g. Koelega, 1993; Weiss and Laties, 1962). Stimulants have been reported to allay fatigue, increase vigilance, speed performance, prolong effort, and generally increase work output (for review, see Koelega, 1993). Stimulants, such as methylphenidate (Ritalin) and d-amphetamine, are prescribed to treat behavioral and cognitive impairments associated with attention deficit/hyperactivity disorder (ADHD) and other disorders of self-control (Tannock, 1998). It also has been suggested that illicit use of stimulants, such as d -amphetamine and cocaine, might be motivated in part by a desire to self-medicate behavioral and cognitive deficits (e.g. Khantzian, 1985; Schiffer, 1988). However,
* Corresponding author. Tel.: /1-859-257-4728; fax: /1-859-3231979. E-mail address: [email protected] (M.T. Fillmore).
evidence for performance-enhancing effects of stimulants is not entirely consistent. Other studies have shown that stimulants, such as cocaine, can produce negative behavioral effects, such as impulsive responding (Fillmore et al., 2002). Although the reasons for the inconsistent findings are not clear, some researchers have pointed to cross-study differences in drug-use history, behavioral assessment methods, and experimental procedures (Ward et al., 1997). The complex nature of stimulant effects on human functioning highlights the need for model-based examinations and interpretation of their behavioral effects (Koelega, 1993). A number of theories postulate that behavioral control is governed by two distinct systems: one that activates behavior and one that inhibits behavior (Fowles, 1987; Gray, 1976, 1977; Logan and Cowan, 1984; Patterson and Newman, 1993; Quay, 1997). The relative strength of each of these systems is generally assumed to determine behavioral control, and impaired behavioral control might arise from either a
03765-8716/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0376-8716(03)00089-9
144
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
weakened inhibitory system or from a heightened activation system. Model-based cognitive assessments, such as stopsignal tasks, have been used to study how drugs alter inhibitory and activational response tendencies to either enhance or reduce behavioral control (deWit et al., 2002; Fillmore et al., 2002). Stop-signal tasks are choice reaction time (RT) measures that require individuals to quickly respond to go-signals and inhibit a response when a stop-signal occurs (Logan, 1994; Logan and Cowan, 1984). Studies of stop-signal performance have shown that CNS-depressants, such as alcohol and triazolam, impair behavioral control by reducing inhibitory tendencies at doses that do not affect the activation of responses (e.g. Fillmore et al., 2001; Fillmore and Vogel-Sprott, 1999, 2000; Mulvihill et al., 1997). The stop-signal paradigm also has been used to study the effects of stimulant use. Individuals with a history of stimulant abuse display reduced response inhibition compared with age-matched control subjects (Fillmore and Rush, 2002). Moreover, stimulant abusers displayed reduced behavioral inhibition in response to acute oral administration of cocaine (Fillmore et al., 2002). These findings are important because they implicate a druginduced impairment of inhibitory tendencies that could help explain the documented association between stimulant abuse and poor impulse control (Jentsch and Taylor, 1999). Although the stop-signal model has been useful in demonstrating that stimulants can affect response inhibition, the model does not provide information on environmental conditions that might exacerbate or ameliorate drug effects on response inhibition and response execution. The environment likely exerts some stimulus control over inhibitory and activational tendencies. Stimulus cues often precede signals to inhibit or respond and can provide preliminary information regarding the type of signal likely to follow (i.e. go or stop). Predictive cues can facilitate the activation and inhibition of behavior by initiating preparatory processes required for the activation or inhibition of an action (Duncan, 1981; Posner, 1980; Posner et al., 1980). Cue effects on the activation and inhibition of behavior have been studied in the laboratory using cued go no-go tasks (e.g. Miller et al., 1991). The tasks measure the effect of preliminary information on the ability to quickly execute and suddenly suppress responses to subsequent go and stop-signals. The tasks typically present a stimulus cue followed by a go or nogo target stimulus that requires a response to be either executed (go) or suppressed (no-go). The cue provides information concerning the probability that a go or nogo target will be presented. The cue-target relationship is manipulated so that cues have a high probability of correctly signaling a target and a low probability of
incorrectly signaling a target. Correct (i.e. valid) cues tend to facilitate response activation and response inhibition. For example, responses to go targets are faster when they are preceded by a go cue. The speeding effect is attributed to covert response preparation that occurs before the actual go target is presented (Posner, 1980). Response preparation can have detrimental effects on behavioral control when the response must be suppressed. Go cues are invalid (i.e. incorrect signals) when they precede a no-go target. Failures to suppress responses to no-go targets are more frequent following invalid go cues, suggesting that response preparation can decrease the likelihood that an action can be suddenly inhibited when necessary (Miller et al., 1991). Studies have used cued go no-go tasks to test alcohol effects on response activation and response inhibition (Abroms et al., 2002; Marczinski and Fillmore, 2003). Those studies showed stereotypic, dose-dependent impairing effects on the ability to inhibit and to execute responses that were preceded by invalid cues. By contrast, no impairing effects of alcohol were observed when cues correctly signaled the inhibition or activation of responses. The research highlights the potential contribution of preliminary information in determining drug effects on behavioral control. To date, no research has tested the effects of a stimulant drug on the ability to activate or inhibit behavior in the presence of cues that correctly or incorrectly signal these actions. Given that stimulants can impair inhibitory control in individuals with a history of stimulant abuse (Fillmore et al., 2002), it is important to determine if this effect is intensified or attenuated in contexts where cues correctly or incorrectly signal the preparation to execute or inhibit an action. The present research tested the effects of damphetamine (5, 10, 20 mg, and placebo) on behavioral control in contexts where preliminary information (i.e. cues) either facilitated or impeded the ability to quickly activate or suddenly inhibit a response.
2. Method 2.1. Subjects Eight adult volunteers (5 men and 3 woman) with histories of cocaine use participated in the study. Subjects’ ages ranged between 20 and 48 years (mean /37.0, S.D. /8.2). The racial make-up of the sample was African American (n/6), Asian (n/1) and Caucasian (n/1). Participants were recruited via notices posted on community bulletin boards and by word of mouth. The study was approved by the University of Kentucky Medical Institutional Review Board, and subjects provided their written, informed consent prior to participating.
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
Volunteers had to have: a minimum of grade 8 education, demonstrated reading ability, no uncorrected vision or auditory problems, and no self-reported major Axis I psychiatric disorder, head trauma, or other CNS injury. Volunteers also had to meet all of the following criteria: (1) a score of at least 4 on the Drug Abuse Screening Test (Skinner, 1982); (2) self-report of past month stimulant use of either d-amphetamine or cocaine. All subjects reported cocaine use and reported smoking cocaine in the form of crack. No volunteer was currently in, or seeking, treatment for their substance use. The sample reported an average frequency of cocaine use of 2.7 days in the past week (S.D. /2.4), and 11.2 days in the past month (S.D./10.7). The sample reported smoking a mean daily amount of 1.9 rocks of crack cocaine (S.D. /2.3), and spending an average of $87.80 per week on cocaine (S.D. /107.4). The sample reported using cocaine for an average period of 7.8 years (S.D. /5.8). All volunteers also reported using nicotine, alcohol, and caffeine. The mean total weekly alcohol consumption for this sample was 42.8 standard drinks (S.D. /45.1). The sample reported smoking an average of 15.8 (S.D. /9.3) cigarettes per day, and an average, total daily caffeine consumption of 95.7 mg (S.D. /98.1). Most of the sample also reported some past month use of marijuana (n/5). One individual reported using a hallucinogen once in the past month and two individuals reported using a benzodiazepine once in the past month. No other drug-use, including prescription medication, was reported by the sample. Prior to participation, all potential volunteers completed a comprehensive medical history questionnaire, drug-use questionnaire, mini-mental status examination (Folstein et al., 1975), and vital sign assessment. Routine, clinical laboratory blood chemistry tests and an electrocardiogram were conducted on all potential volunteers. Potential volunteers with histories of serious physical disease, current physical disease, impaired cardiovascular functioning, chronic obstructive pulmonary disease, seizure, head trauma, CNS tumors, or past histories of psychiatric disorder, (i.e. Axis I, DSM IV), other than substance use or dependence, were excluded from participation. Female volunteers had to have an effective form of birth control in use. All subjects were in good health with no contraindications to stimulant drugs. 2.2. Apparatus and materials 2.2.1. Cued go no-go task Participants performed a cued go no-go RT task that was operated using E-Prime experiment generation software (Schneider et al., 2002) and was performed on a PC computer. A trial involved the following sequence of events: (1) presentation of a fixation point
145
(/) for 800 ms, (2) a blank, white screen for 500 ms, (3) a cue, displayed for one of five stimulus onset asynchronies (SOAs/100, 200, 300, 400 and 500 ms), (4) a go or no-go target, that remained visible until a response occurred or 1000 ms had elapsed, and (5) an inter-trial interval of 700 ms. The cue was a rectangle (7.5 /2.5 cm2) framed in a 0.8 mm black outline that was presented in the center of the computer monitor against a white background. The cue was presented in either a horizontal (height /2.5 cm, width/7.5 cm) or vertical (height /7.5 cm, width/2.5 cm) orientation. The go and no-go targets were the colors, green and blue, respectively. They were displayed on the monitor as a solid hue that filled the interior of the rectangle cue. Subjects were instructed to press the forward slash (/) key on the keyboard as soon as a go (green) target appeared and to suppress the response when a no-go (blue) target was presented. Key presses were made with the index finger of the preferred hand. The go and no-go targets were presented in hues that were easily distinguishable by all participants. The orientation of the cue (horizontal or vertical) signaled the probability that a go or no-go target would be displayed. Cues that were presented vertically preceded the go target on 80% of the trials and preceded the no-go target on 20% of the trials. Cues that were presented horizontally preceded the no-go target on 80% of the trials and preceded the go target on 20% of the trials. Therefore, based on cue-target pairings, vertical and horizontal cues operated as go and no-go cues, respectively. The different SOAs (100, 200, 300, 400 and 500 ms) between cues and targets encouraged volunteers to pay attention to the cues and the variability and randomness of the SOAs prevented the participants from anticipating the exact onset of the targets. A test consisted of 500 trials that presented the four possible cue-target combinations (see Table 1). An equal
Table 1 Mean peak physiological effect and DEQ ratings under each dose condition Dose condition (mg) 0
5
10
20
Physiological effects Heart rate (BPM) Systolic BP (mmHg) Diastolic BP (mmHg)
76.9 (8.6) 138.9 (12.1) 86.9 (4.6)
78.8 (10.6) 141.5 (10.5) 84.0 (6.6)
77.4 (10.1) 142.7 (10.0) 88.8 (7.2)
79.0 (10.1) 152.2 (12.3) 91.9 (5.1)
DEQ effects Any effects Good effects
0.5 (0.5) 0.2 (0.5)
0.8 (0.7) 0.6 (0.7)
0.8 (0.7) 0.6 (0.7)
1.4 (0.9) 1.1 (1.0)
N/8. Parentheses indicate standard deviations.
146
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
number of vertical (250) and horizontal (250) cues were presented before an equal number of go (250) and no-go (250) target stimuli. Each cue-target combination was presented at each of the five SOAs, and an equal number of SOAs separated each cue-target combination. The presentation of cue-target combinations and SOAs was random. For each trial, the computer recorded whether or not a response occurred and if so, the RT in milliseconds was measured from the onset of the target until the key was pressed. To encourage fast and accurate responding, feedback was provided to the subject. Feedback was presented during the inter-trial interval on all trials by displaying the words ‘correct’ or ‘incorrect’ along with millisecond RTs on trials in which responses occurred. Subjects had a 10 min rest break after completing the first 250 trials. A test required approximately 30 min to complete. 2.2.2. Drug effect questionnaire This questionnaire, used in prior research (e.g. Fillmore et al., 2002; Rush et al., 1998), consisted of 15 items that were presented individually on the computer monitor. Subjects rated each item using the computer mouse to select among one of five response options: not at all, a little bit, moderately, quite a bit, and very much (scored numerically from 0 to 4, respectively). The items rated were: any effects, active/ alert/energetic, bad effects, good effects, high, irregular heart beat/racing, like, anxious/nervous, pay for this drug, rush, shaky/jittery, take this drug again, talkative/ friendly, nauseated/queasy, and sluggish/fatigued/lazy. 2.3. General procedures The study was conducted at the Laboratory of Human Behavioral Pharmacology at the University of Kentucky. All participants were tested individually. All sessions were scheduled from Monday to Friday between 8:30 am and 4:30 pm. Subjects were instructed not to drink alcohol or take any drug during their scheduled lab days. During all sessions, participants provided a breath sample using an Intoxilyzer Model 400 (CMI Inc. Owensborough, KY) to ensure a zero blood alcohol concentration (BAC). A urine sample also was obtained to test for the presence of cocaine/benzoylecgonine, benzodiazepines, barbiturates, tetrahydrocannabinol (THC), d -amphetamine, and opiates (On Trak TesTstiks, Roche Diagnostics Corporation, Indianapolis, IN). Females were also tested for pregnancy via urine analysis. With the exception of cocaine/benzoylecgonine and THC, no volunteers tested positive for the presence of any drug at the time of study. Subjects were informed that the study examined the effects of various drugs (e.g. stimulants, benzodiazepines, barbiturates) on mood and behavior. They were given no information about the specific drug examined in the study. Subjects attended a
familiarization/practice session and four dose administration test sessions. Sessions required 7.5 h to complete and were separated by a minimum of 48 h. 2.4. Familiarization/practice session Subjects were acquainted with the task requirements and general testing procedure. All testing occurred in a quiet test room that contained a chair, desk, and computer. The experimenter explained the requirements for the cued go no-go task. To verify that the participants could easily distinguish the colors of the stimuli in the task, each subject was presented with a color vision test that required them to discriminate between the green and blue colors that were used as go and no-go targets in the cued RT task. Subjects were told that they would see empty rectangular boxes become filled with green or blue ‘target’ colors. Subjects were instructed to press the forward slash key (/) on the keyboard as quickly as possible whenever a green target appeared and to make no response whenever a blue target appeared. The computer displayed how fast a subject responded to each go target by presenting the milliseconds required from target onset until the key press. Subjects were encouraged to make fast responses (i.e. in the fewest milliseconds). No information regarding cuetarget relationships was provided. Subjects then performed practice tests on the task. A single test is sufficient to produce cue-dependant responding (Marczinski and Fillmore, 2003). Subjects were also familiarized with the drug effect questionnaire (DEQ), and physiological monitoring equipment. Subjects completed the questionnaires and had their heart rate and blood pressure measured by a Dinamap XL Vital Signs Monitor (Johnson and Johnson Medical Inc., Arlington TX) at intermittent intervals over a 6-h period. 2.5. Dose administration sessions Prior to beginning each session, subjects ate a low-fat breakfast at 7:00 am. Subjects were allowed to smoke one cigarette between 7:30 and 8 am. No smoking was allowed during the sessions. 2.5.1. Baseline testing and dose administration The testing began at 9 am with baseline (i.e. precapsule) DEQ and physiological measures. Subjects then received one of four doses of d-amphetamine: 0 (placebo), 5, 10 and 20 mg. All doses were administered orally in a double-blind fashion. Drug doses were prepared by encapsulating 5 mg d -amphetamine in size 00 blue and white opaque capsules combined with lactose filler. Placebo capsules contained only lactose. Subjects ingested four capsules during each dose administration session. The combination of active and placebo
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
capsule determined the dose. Capsules were taken orally with 150 ml of water. Dose order across the four dose administration sessions was random without replacement and included a safety constraint that prevented any subject from receiving 20 mg d -amphetamine on the first session. Thus eight different dose orders were tested, with each order unique to a single subject. The highest dose (20 mg d-amphetamine) was administered to at least two subjects in each of the three sessions that followed the first session. The dose was received by two subjects in the second session, by four subjects in the third session, and by two subjects in the four session. Medications were prepared by the University of Kentucky Investigational Pharmacy. 2.5.2. Post-capsule testing Subjects performed the cued go no-go task at 2.5 h post-capsule. Because the test required 30 min to complete, the assessment spanned 2.5 /3.0 h postcapsule. This time interval was chosen to maximize the likelihood of observing a drug effect on inhibitory control. Time-course analyses of oral d -amphetamine dose effects on physiological and subjective measures typically show peak effects approximately 2.5 /3.0 h following administration (Rush et al., 1998; Vree and Henderson, 1980). Subjects completed the DEQ at 3.0 and 4.0 h post-capsule. Physiological measures were recorded at 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 5.5 h post-capsule. No volunteers experienced any adverse reaction to the drug or any procedure. Volunteers were paid $400 for their participation. 2.6. Criterion measures and data analyses of cued go nogo performance The two primary measures of interest were the participants’ failures to inhibit responses to no-go targets (failures of response inhibition) and their speed of responding to go targets (response execution). 2.6.1. Failure of response inhibition Failure of response inhibition was measured as the proportion of no-go targets in which a participant failed to inhibit a response. These p-inhibition failure scores were calculated for each cue condition (go and no-go) on each test. Scores were analyzed by a 4 Dose (0, 5, 10, 20 mg d -amphetamine) X 2 Cue (go cue vs. no-go cue) within-subjects ANOVA. 2.6.2. Response execution Response execution was measured by the RT to go targets. Shorter RTs indicated greater facilitation of response execution. A mean RT score for a participant was calculated for each cue. Responses with RTs less
147
than 100 ms and greater than 1000 ms were excluded. These outliers were infrequent, occurring on average less than 0.25% of the trials for which a response was observed (i.e. less than one trial per test). Scores were analyzed by a 4 (Dose) X 2 (Cue) within-subjects ANOVA. Omission errors were also recorded. These errors occurred when participants failed to respond to go targets. Omission errors were infrequent and occurred on less than 1% of go target trials (/2 trials per test).
2.6.3. Cue-dependency scores The degree to which a subject’s p-inhibition failures and RT was influenced by valid and invalid cues were also represented as single cue-dependency scores, based on the difference in performance between the invalid and valid cue conditions for a test. Greater cuedependency scores reflect greater reliance on the environmental context (i.e. the cues) to maintain efficient suppression and execution of responses (Neill, 1997; Posner and Snyder, 1975). The cue-dependency score for p-inhibition failures was calculated as the p-inhibition failure score in the go cue condition minus the pinhibition failure score in the no-go cue condition. A larger cue-dependency score for p-inhibition failures indicates greater reliance on valid cues (no-go cues) to successfully inhibit responses to no-go targets. The cuedependency score for RT was calculated as the mean RT in the no-go cue condition minus the mean RT in the go cue condition. A larger cue-dependency score for RT indicates greater reliance on valid cues (go cues) for fast response execution.
2.7. Criterion measures and data analyses of physiological and subjective effects Two sets of analyses examined physiological and selfreported subjective effects. First, peak effect data were calculated and analyzed by one-factor within-subjects ANOVA with Dose (0, 5 10 and 20 mg d -amphetamine) as a repeated measures factor. For physiological measures, peak effect was defined as the maximum value from 0.5 to 5.5 h after drug administration. For subjective measures, peak effect was the maximum value from 3 to 4 h after drug administration. Second, timecourse data were analyzed by two-factor, within-subjects ANOVA of Dose (0, 5, 10 and 20 mg) and Time (precapsule, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 5.5 h). Subjective effect ratings were analyzed for each DEQ item by a two-factor, within-subjects ANOVA of Dose (0, 5, 10 and 20 mg) and Time (pre-capsule, 3.0 and 4.0 h).
148
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
3. Results 3.1. Dose effects on inhibition failures (P-inhibition failures) A 4 (Dose) X 2 (Cue) within-subjects ANOVA of pinhibition failure scores revealed a significant Dose X Cue interaction (F3,21 /3.9, P /0.023). Fig. 1 illustrates the mean p-inhibition failure scores. The figure shows that p-inhibition failures following no-go cues were low and were not affected by dose. By contrast, p-inhibition failures were greater following go cues and increased as a function of d -amphetamine dose. These observations were confirmed by two simple effect one-way ANOVAs that compared p-inhibition failure scores across doses separately for each cue condition. There was no significant dose effect on p-inhibition failure scores following no-go cues (P /0.380). By contrast, a significant dose effect was observed following the go cues (F3,21 /4.1, P /0.020). Follow-up simple effects analyses of the go cue condition compared p-inhibition failure scores under each active dose to placebo. Fig. 1 shows that the 5 and 20 mg active doses significantly increased p-inhibition failure scores compared with placebo (Ps B/0.018). Although greater p-inhibition failures were observed under the 10 mg dose compared with placebo, the difference was not statistically significant (P /0.209). 3.2. Dose effects on response execution (reaction time) A 4 (Dose) X 2 (Cue) within-subjects ANOVA of RTs to go targets revealed significant main effects of Dose
(F3,21 /3.7, P /0.027) and Cue (F1,7 /159.9, P B/ 0.001). There was no significant interaction (P / 0.692). Fig. 2 shows that RTs were slower following no-go cues. The figure also shows that RTs decreased slightly as a function of dose and this speeding effect was similar across cue conditions. Follow-up simple effects analyses compared RT scores under each active dose to placebo in each cue condition. No significant differences were obtained (Ps /0.127). 3.3. Cue-dependency scores for P-inhibition failures and reaction time Dose effects on the suppression and execution of responses were also examined in terms of the magnitude of cue-dependency (the difference in performance between the invalid and valid cues). Fig. 3 plots the mean cue-dependency scores under each dose for p-inhibition failures and RT. The figure shows that cue-dependency of inhibition failures increased as a function of dose. By contrast, the cue-dependency of RT showed no orderly effect of dose. 3.4. Physiological and subjective effect measures ANOVAs of time-course effects obtained significant dose X time interactions for heart rate (F33,231 /1.7, P /0.009), and systolic blood pressure (F33,231 /1.6, P /0.018). No significant effects were obtained for diastolic blood pressure (Ps /0.095). ANOVAs of peak effects obtained significant dose effects for systolic (F3,21 /4.8, P /0.011), and diastolic blood pressure (F3,21 /3.1, P /0.048). No significant dose effect was
Fig. 1. Mean proportion of failures to inhibit responses to no-go targets following go and no-go cues under four doses of oral d -amphetamine: 0 (placebo), 5, 10, and 20 mg. Capped vertical lines show standard errors of the mean. *, indicates significant differences from placebo, P B/0.05.
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
149
Fig. 2. Mean RT to respond to go targets following go and no-go cues under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean.
obtained for peak heart rate (F3,21 /0.4, P /0.746). Table 1 presents the mean peak physiological effects for each measure under the four dose conditions. Peak effect DEQ ratings of items, Any Effect (F3,21 /6.0, P / 0.004) and Good Effect (F3,21 /3.2, P /0.042), increased under active dose conditions. Table 1 presents the mean peak rating of these two DEQ items under the four dose conditions. No other DEQ measures showed significant dose effects in either time-course or peak effect analyses (Ps /0.05).
4. Discussion This research used a cued go no-go RT task to test the acute effects of d -amphetamine on aspects of behavioral control in individuals with a history of stimulant abuse. The results showed that the effect of d-amphetamine on response suppression was dependent on cues that signaled the likelihood that a response should be executed or suppressed. Overall, inhibitory failures were more frequent following go cues compared with
Fig. 3. Left. Mean cue-dependency scores for proportion of failures to inhibit responses to no-go targets under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean. Right. Mean cue-dependency scores for RT to respond to go targets under four doses of oral d -amphetamine: 0 (placebo), 5, 10 and 20 mg. Capped vertical lines show standard errors of the mean.
150
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
no-go cues. d -Amphetamine produced a dose-dependent increase in inhibitory failures following go cues and had no effect on inhibitory failures following no-go cues. Inhibition failures following no-go cues were nearly absent in all dose conditions. A near-zero measure in this cue condition could impose a floor effect that would be a problem for studies of treatments expected to improve inhibitory control by reducing inhibitory failures. d -Amphetamine had little effect on response execution as measured by RT. Overall, volunteers displayed faster RTs to go targets that followed go cues compared with those that followed no-go cues. d Amphetamine produced a significant, but only slight speeding effect on RT that was evident in both cue conditions. The present findings are consistent with our prior research that showed long-term stimulant abusers displayed reduced behavioral inhibition on a stop-signal task in response to acute oral administration of cocaine (Fillmore et al., 2002). Like the stop-signal task, the cued go no-go task used in the present study measured behavioral control by the ability to quickly execute and suddenly suppress responses. However, unlike the stopsignal task, the cued task concerned responses to stimuli that were intended to be dependent on environmental cues. As such, the cued go no-go task did not include a no-cue, or neutral cue condition because those conditions reduce the overall proportion of valid cue trials and decrease participants’ reliance on cues for response preparation (Gottlob et al., 1999; Posner and Snyder, 1975). The large cue effects on RT and response suppression in the present study confirm that the 80% reliable cue-target association was effective in producing cue-dependent behavior. The dose-dependent increase in cue-dependency of inhibition failures suggests that subjects became more reliant on environment cues for inhibiting responses under the drug. Previous research shows that alcohol also increases the cue-dependency of inhibitory failures (Marczinski and Fillmore, 2003). However, that study found that alcohol also increased the cue-dependency of RT by making subjects more reliant on valid go cues for maintaining fast response execution. By contrast, d amphetamine had a slight speeding effect on RT, but did not increase the degree to which RT was cue-dependent. Evidence that d -amphetamine had different effects on cue-dependency of inhibitory and activational aspects of behavior supports the independence assumption regarding these two mechanisms of behavioral control (e.g. Logan and Cowan, 1984). The dissociative effect is also consistent with other studies of stimulants that find differences between inhibitory and activational effects. For example, Tannock et al. (1995) measured the ability to inhibit and execute responses following 0.0, 0.3, 0.6 and 0.9 mg/kg methylphenidate in children diagnosed with ADHD and found that response execution was
facilitated in a dose-dependent manner. However, measures of response inhibition revealed an inverted U-shaped dose response curve, suggesting that facilitative effects of psychostimulants on inhibitory control might be confined to a particular dose window. Demonstration of these types of dissociative drug effects are potentially important because they provide information about the specificity of a drug effect on a behavioral system and can identify levels of independence between behavioral mechanisms and their underlying neuroanatomical systems. Evidence that d -amphetamine effects on behavioral control can be environmentally-dependent might help explain reports of different effects of d -amphetamine on behavioral control. Some recent studies have found that d -amphetamine can improve inhibitory control in healthy subjects as measured by stop-signal tasks (deWit et al., 2000, 2002). Stimulant effects on behavioral control could differ depending on the degree to which inhibitory and activational aspects of behavior are under stimulus control. Impairing effects on inhibitory control might be more likely in contexts where there is a strong response tendency. Subject characteristics might also contribute to differences in stimulant effects. There is growing evidence that long-term habitual stimulant users display neuropsychological impairments and have neuroanatomical abnormalities, including deficits of inhibitory control (e.g. Fillmore and Rush, 2002; Lane et al., 1998). Some investigators suggest that repeated dopaminergic activation of prefrontal pathways by chronic stimulant use eventually impairs inhibitory functions, leading to a loss of control over behavioral impulses (e.g. Lyvers, 2000; Volkow et al., 1996). Neuronal changes owing to chronic stimulant use could alter the acute behavioral response to stimulants in those with a long history of stimulant use. The importance of these and other factors as determinants of stimulant effects on behavioral control await exploration. Cue-dependent models of behavioral control also have been applied to study deficits of behavioral control in relation to other conditions, such as age-related cognitive decline. A general finding from that research is that elderly subjects display greater reliance on cues in performance tests as compared with younger adults (Laver and Burke, 1993). Such age-related increases in cue-dependency for behavioral control parallel acute drug-induced increases in cue-dependancy, like that observed in the present research. Such inter-disciplinary comparisons might provide valuable insight into the potential interactions between drug effects and chronic conditions that are characterized by impaired control. The continued broad application of these models could allow for potentially informative comparisons between differences sources of impaired control (e.g. druginduced vs. age-related).
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
The present findings contribute to our growing understanding of how drugs affect behavioral control in humans. The findings suggest that moderate doses of d -amphetamine can be particularly detrimental to behavioral control when it depends on the sudden and complete suppression of responses. The present results are consistent with studies of cocaine that also demonstrate drug-associated impairments in the ability to inhibit behavior (Fillmore and Rush, 2002; Fillmore et al., 2002). The current study expands upon that research by demonstrating that environmental cues are an important factor in determining stimulant effects on fundamental aspects of behavioral control.
Acknowledgements This research was supported by grants DA14079 and DA10325 from the National Institute on Drug Abuse and by grant AA12895 from the National Institute on Alcohol Abuse and Alcoholism. Offprint requests should be sent to Mark T. Fillmore, Ph.D., Department of Psychology, University of Kentucky, Lexington, KY 40506-0044. E-mail: [email protected].
References Abroms, B.D., Fillmore, M.T., Marczinski, C.A., 2002. Alcoholinduced impairment of behavioral control: effects on alteration and suppression of prepotent responses, J. Stud. Alcohol, submitted for publication. deWit, H., Crean, J., Richards, J.B., 2000. Effects of d -amphetamine and ethanol on a measure of behavioral inhibition in humans. Behav. Neurosci. 114, 830 /837. deWit, H., Engasser, J.L., Richards, J.B., 2002. Acute administration of d -amphetamine decreases impulsivity in healthy volunteers. Neuropsychopharmacology 27, 813 /825. Duncan, J., 1981. Directing attention in the visual field. Percept. Psychophys. 30, 90 /93. Fillmore, M.T., Rush, C.R., 2002. Impaired inhibitory control in chronic cocaine abusers. Drug Alcohol Depend. 66, 265 /273. Fillmore, M.T., Rush, C.R., Kelly, H.K., Hays, L., 2001. Triazolam impairs inhibitory control of behavior in humans. Exp. Clin. Psychopharmacol. 9, 363 /371. Fillmore, M.T., Rush, C.R., Hays, L., 2002. Acute effects of oral cocaine on inhibitory control of behavior in humans. Drug Alcohol Depend. 67, 157 /167. Fillmore, M.T., Vogel-Sprott, M., 1999. An alcohol model of impaired inhibitory control and its treatment in humans. Exp. Clin. Psychopharmacol. 7, 49 /55. Fillmore, M.T., Vogel-Sprott, M., 2000. Response inhibition under alcohol: effects of cognitive and motivational conflict. J. Stud. Alcohol 61, 239 /246. Folstein, M., Folstein, S., Hugh, P., 1975. Mini-Mental State */a practical method for grading the cognitive state of patients for the clinician. J. Psychiat. Res. 12, 189 /198. Fowles, D.C., 1987. Application of a behavioral theory of motivation to the concepts of anxiety and impulsivity. J. Res. Pers. 21, 417 / 435.
151
Gottlob, L.R., Cheal, M., Lyon, D.R., 1999. Time course of locationcuing effects with a probability manipulation. J. Gen. Psychol. 126, 261 /270. Gray, J.A., 1976. The behavioral inhibition system: a possible substrate for anxiety. In: Feldman, M.P., Broadhurst, A. (Eds.), Theoretical and Experimental Bases of the Behavior Therapies. Wiley, London, pp. 3 /41. Gray, J.A., 1977. Drug effects on fear and frustration. Possible limbic site of action of minor tranquilizers. In: Iverson, L.L., Iverson, S.D., Snyder, S.H. (Eds.), Handbook of Psychopharmacology, vol. 8. Plenum, New York, pp. 433 /529. Jentsch, J.D., Taylor, J.R., 1999. Impulsivity resulting from frontostriatal dysfunction in drug abuse: implication for the control of behavior by reward-related stimuli. Psychopharmacology 146, 373 /390. Khantzian, E., 1985. The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. Am. J. Psychiat. 142, 1259 /1264. Koelega, H.S., 1993. Stimulant drugs and vigilance performance: a review. Psychopharmacology 111, 1 /16. Lane, S.D., Cherek, D.R., Dougherty, D.M., Moeller, F.G., 1998. Laboratory measurement of adaptive behavior change in humans with a history of substance dependence. Drug Alcohol Depend. 51, 239 /252. Laver, G.D., Burke, D.M., 1993. Why do semantic priming effects increase in old age? A meta-analysis. Psychol. Aging 8, 34 /43. Logan, G.D., 1994. On the ability to inhibit thought and action: a user’s guide to the stop-signal paradigm. In: Dagenbach, D., Carr, T.H. (Eds.), Inhibitory Processes in Attention, Memory, and Language. Academic Press, San Diego, CA, pp. 189 /239. Logan, G.D., Cowan, W.B., 1984. On the ability to inhibit thought and action: a theory of an act of control. Psychol. Rev. 91, 295 / 327. Lyvers, M., 2000. Loss of control in alcoholism and drug addiction: a neuroscientific interpretation. Exp. Clin. Psychopharmacol. 8, 225 /249. Marczinski, C.A., Fillmore, M.T., 2003. Pre-response cues reduce the impairing effects of alcohol on the execution and suppression of responses. Exp. Clin. Psychopharmacol. 9, 110 /117. Miller, J., Schaffer, R., Hackley, S.A., 1991. Effects of preliminary information in a Go versus No-go task. Acta Psychol. 76, 241 /292. Mulvihill, L.E., Skilling, T.A., Vogel-Sprott, M., 1997. Alcohol and the ability to inhibit behavior in men and women. J. Stud. Alcohol 58, 600 /605. Neill, W.T., 1997. Episodic retrieval in negative priming and repetition priming. J. Exp. Psychol: Learn. Mem. Cognit. 23, 1291 /1305. Patterson, C.M., Newman, J.P., 1993. Reflectivity and learning from aversive events: toward a psychological mechanism for the syndromes of disinhibition. Psychol. Rev. 100, 716 /736. Posner, M.I., 1980. Orienting of attention. Q. J. Exp. Psychol. 32, 3 / 25. Posner, M.I., Snyder, C.R., 1975. Facilitation and inhibition in the processing of signals. In: Rabbitt, P., Dornic, S. (Eds.), Attention and Performance, vol. V. Academic Press, New York. Posner, M.I., Snyder, C.R., Davidson, B.J., 1980. Attention and the detection of signals. J. Exper. Psychol. 109, 160 /174. Quay, H.C., 1997. Inhibition and attention deficit hyperactivity disorder. J. Abnorm. Child Psychol. 25, 7 /13. Rush, C.R., Kollins, S.H., Pazzaglia, P.J., 1998. Discriminativestimulus and participant-rated effects of methylphenidate, bupropion and triazolam in d -amphetamine-trained humans. Exp. Clin. Psychopharm. 6, 32 /44. Schiffer, F., 1988. Psychotherapy of nine successfully treated cocaine abusers: techniques and dynamics. J. Sub. Abuse Treat. 5, 131 / 137. Schneider, W., Eschman, A., Zuccolotto, A., 2002. E-Prime User’s Guide. Psychology Software Tools Inc, Pittsburgh, PA.
152
M.T. Fillmore et al. / Drug and Alcohol Dependence 71 (2003) 143 /152
Skinner, H.A., 1982. The drug abuse screening test. Addict. Behav. 7, 363 /371. Tannock, R., 1998. Attention deficit hyperactivity disorder: advances in cognitive, neurobiological, and genetic research. J. Child Psychol. Psychiat. 39, 65 /99. Tannock, R., Schachar, R., Logan, G., 1995. Methylphenidate and cognitive flexibility: dissociated dose effects in hyperactive children. J. Abnorm. Child Psychol. 23, 235 /267. Volkow, N.D., Ding, Y., Fowler, J.S., Wang, G., 1996. Cocaine addiction: hypothesis derived from imaging studies with PET. J. Addict. Dis. 15, 55 /71.
Vree, T.B., Henderson, P.T., 1980. Pharmacokinetics of amphetamines: in vivo and in vitro studies of factors governing their elimination. In: Caldwell, J. (Ed.), Amphetamine and Related Stimulants: Chemical Biological Clinical and Sociological Aspects. CRC Press, Boca Raton, FL, pp. 47 /68. Ward, A.S., Kelly, T.H., Foltin, R.W., Fischman, M.W., 1997. Effects of d -amphetamine on task performance and social behavior of humans in a residential laboratory. Exp. Clin. Psychopharm. 5, 130 /136. Weiss, B., Laties, V.G., 1962. Enhancement of human performance by caffeine and the amphetamines. Pharmacol. Rev. 14, 1 /36.