1 Department of Psychology, University of Toronto, Toronto, Ontario M5S 3G3 and
2 Department of Psychology, Queens University, Kingston, Ontario K7L 3N6, Canada
Received 3 July 2002; in revised form 1 October 2002; accepted 13 October 2002
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Alcohol increases cognitive impulsivity in rats (Poulos et al., 1998), but does not influence discounting of delayed monetary rewards in humans (Richards et al., 1999
). A possible explanation for this discrepancy is that alcohol-induced impulsivity is moderated by alcohol myopia (Steele and Josephs, 1990
). According to this theory, alcohol intoxication restricts information processing, causing an individual to extract meaning solely from the most salient cues in their environment. If cues provoking an impulsive choice are more salient than cues inhibiting that choice, alcohol increases impulsivity. In contrast, if either or both of the impelling and inhibiting cues are weak, alcohol myopia does not influence responding. Alcohol myopia theory may explain why Richards et al.(1999)
failed to find an effect of alcohol on delay discounting because the two response choices (e.g. $5.00 now or $10.00 in 365 days) were both weak cues. Practice effects may also have masked any treatment effects, because participants received five exposures to the delay-discounting task (practice, pre-placebo, post-placebo, pre-alcohol and post-alcohol). Finally, trait impulsivity affects responding in that personality measures of impulsivity correlate with performance on the delay-discounting task (Richards et al., 1999
), with alcohol use and with alcohol-related problems (Nagoshi et al., 1991
; Grau and Ortet, 1999
). Even rats exhibit individual differences in responsiveness to alcohol in tests of cognitive impulsivity (Poulos et al., 1998
).
The purpose of the present study was to examine the factors affecting cognitive impulsivity in intoxicated individuals. We hypothesized that intoxicated participants would make more impulsive choices only when the cues impelling impulsive behaviour were salient. Cognitive impulsivity was measured in sober, intoxicated and control (i.e. no drink) participants using the delay-discounting task; cue salience was manipulated by presenting one of the response choices in bold, as this manipulation successfully changes the direction of responding in intoxicated participants (MacDonald et al., 2000). To test whether trait impulsivity and cognitive impulsivity were correlated, participants completed a questionnaire measure of impulsivity prior to completing the task. We also assessed the relations among cognitive impulsivity, blood alcohol level (BAL) and self-reported desire to consume more alcohol, because cognitive impulsively predicts voluntary consumption of alcohol in rats (Poulos et al., 1998
), and alcohol dose-dependently increases subjective reports of drug liking and the desire to ingest more alcohol in humans (Kirk and de Wit, 2000
). Finally, we measured motor impulsivity using a go/no-go task in order to compare the effects of alcohol on two different types of impulsivity (i.e. cognitive and motor).
![]() |
SUBJECTS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Dependent measures
ZuckermanKuhlman Personality Questionnaire (ZKPQ). The impulsiveness sensation seeking scale (ImpSS) of the ZKPQ (Zuckerman et al., 1993) consists of 19 items regarding a lack of planning and the tendency to act impulsively. The ImpSS was chosen because it is a significant predictor of risk behaviours (e.g. drug use and gambling) in college students (Zukerman and Kuhlman, 2000), but does not contain items relating to alcohol consumption, thereby reducing the possibility that alcohol use will act as a confound.
Delay discounting. A computerized version of the delay-discounting task (Mitchell, 1999) was presented on an IBM ThinkPad. Participants were presented with a series of hypothetical choices between the standard and the alternative item. The standard item was $10.00 available after one of six delays (0, 7, 30, 90, 180 or 365 days). The alternative item was an amount of money ($0.01, $0.25, $0.50, and further amounts increasing in 0.50 increments up to $10.50) available after 0 days (i.e. at the end of the session). Standard and alternative items were presented, without replacement, in a random order. The order in which the standard and alternative items were presented within each question was also randomized. For each participant, an indifference point was determined for each of the six delays. The indifference point is the value of the alternative (the immediate reward) at which the preference switches from the delayed to the immediate reward. Usually, the indifference point is discrete, but when it was not, it was defined as the point midway between the lowest value of the alternative at which the participant chose the alternative for two consecutive, descending values and the highest value of the standard for which the participant chose the standard for two consecutive ascending values.
Participants received the following verbal instructions: You will have the opportunity to choose between different amounts of money available after different delays. The test consists of about 140 questions. Just indicate the alternative you prefer by pressing the appropriate key on the keyboard and then press enter. If you press the wrong key by mistake, simply press backspace and retype your answer. At the end of the session, one of the choices you made will be selected at random and you will receive whatever choice you made on that question and at the time stated in the question. So, if on that trial, you selected an immediate amount of money, you will receive the money in cash at the end of the session. If you selected the delayed money, the money will be placed in an envelope with your name on it and will be available to you when the time has elapsed. Participants were also informed that this money was in addition to the $10 they would receive for their participation in the study.
In the impelling cue condition, the alternative item was presented in bold and the instructions were altered to further impel an impulsive choice. When informing participants that one of their choices would be selected at random at the end of the session, the experimenter only described the outcome of selecting the alternative item on that choice, by saying: So, for example: if on that trial, you selected an immediate amount of money, you will receive the money in cash at the end of the session.
Go/No-Go. The go/no-go task (Owen et al., 1991) is a measure of motor impulsivity. A series of 18 characters (letters and digits) is flashed in the centre of a screen for 300 ms per character, at 900 ms intervals. Participants respond by pressing the space bar as quickly as they can when they see a specified stimulus (letters or digits) and withholding responses when presented with the other stimulus. Midway through the task, the target stimulus is switched and the participant must respond to the other stimulus. Measures include response reaction time, errors of commission (responding to an incorrect stimulus) and errors of omission (failing to respond to the specified stimulus).
Drug Effects Questionnaire (DEQ). The DEQ (Kirk and De Wit, 2000) is a self-report measure of alcohols effects and the desire to consume more alcohol. A modified version of the DEQ was employed, using three of the four scales. Each scale is a visual analogue scale 100 mm in length. The scales consist of ratings of feel effects, like effects and want more. The left ends of these scales are labelled not at all (or dislike a lot for ratings of like effects). The right ends of these scales are labelled a lot (or very much for ratings of want more).
Design and procedure
Participants first completed the ZKPQ. Participants were randomly assigned to the intoxicated, placebo or sober condition. Within each condition, half the participants received the original version of the delay-discounting task and half received the impelling cue version. Sober participants completed the dependent measures and were debriefed.
Participants in the intoxicated condition were weighed and then received 0.7 g of alcohol/kg body weight divided into three alcoholic drinks. The drinks contained one part alcohol to two parts Wink soda. Participants in the placebo condition were also weighed and then consumed three non-alcoholic beverages containing one part soda water and two parts Wink soda. The rims of the glasses were dipped in alcohol and 5 ml of alcohol was placed on the surface of each placebo drink (MacDonald et al., 2000). Intoxicated and placebo participants consumed each drink within 2 min, with an 18 min rest period between drinks. Twenty minutes after the final beverage, participants completed the dependent measures. BAL measures were obtained using a breathalyser and participants were then fully debriefed.
Data analysis
Delay discounting is best supported by a hyperbolic function (Green et al., 1999; Richards et al., 1999
). Therefore, to assess the rate of discounting of delayed rewards, the curvilinear equation fitting function of SigmaPlot was used to fit each participants indifference points to the hyperbolic equation: V = A / (1 + kD), where V is the current, subjective value of the delayed reward, A is the amount of the delayed reward, D is the delay to the reward and k is a free parameter representing the rate of devaluation of the delayed reward. The fit of the hyperbolic model to individual participants data was evaluated based on values of t (the ratio of the parameter estimate to its standard error), with (n p) degrees of freedom, where n is the number of data points (6) and p is the number of free parameters (1). A value of t less than the critical value of t for P < 0.05 indicates that the parameter estimate is not an adequate fit for the data. Fit to the hyperbolic model across conditions was compared by means of a chi-squared test and between specific conditions using MannWhitney U tests, owing to the skewness of the r2 values and the unequal group sizes.
The distribution of k values was extremely positively skewed and leptokurtic and the group variances were significantly heterogeneous. Therefore, a log-transformation [log10(k + 0.0001] was computed before performing the analyses. Median indifference points with interquartile ranges are presented, because distributions of some of the indifference points were not normal. There were no significant differences between sober and placebo participants in mean log-transformed k values in either the standard or the impelling cue condition, or in errors of commission on the go/no-go. Therefore, subsequent analyses were conducted by collapsing across the placebo and sober groups to form a no-alcohol group.
A two-way ANOVA was conducted to compare the effects of drink (alcohol or no-alcohol) and the cue salience manipulation (standard or impelling cue) in the delay-discounting task on log-transformed k values. A univariate ANOVA was computed to examine the effect of the delay-discounting condition on log-transformed k values, with BAL as a covariate. The effect of alcohol on go/no-go performance was examined by way of a two-tailed independent samples t-test. Finally, Pearson correlations were conducted to assess the relations among different measures of impulsivity.
Two participants in the alcohol, standard cue condition were inadvertently not administered the ZKPQ. Additionally, delay-discounting data were lost for one participant in the placebo, standard delay-discounting condition, leaving 10 participants in this group.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
Figure 1 shows that the indifference points were higher in the alcohol than the no-alcohol condition (binomial test, P = 0.063). This finding provides further support for the notion that alcohol intoxication reduced the tendency to discount delayed rewards. The difference between median indifference points for the two groups approached significance at the 30-day delay only (KruskalWallis,
2 = 3.254, P = 0.071). In order to partial out any effects of BAL, a univariate ANOVA of the log-transformed k values was computed, with the delay-discounting condition (standard or impelling cue) as the fixed factor and BAL as a covariate. In this analysis, the delay-discounting condition had no significant effect on log-transformed k values.
|
Correlations among measures of impulsivity
There was no correlation between ImpSS and delay discounting in either the alcohol or no-alcohol conditions (Pearson correlations, r = -0.170 and r = -0.163, respectively). However, in the alcohol group, there was a significant positive correlation between log-transformed k values and errors of commission on the go/no-go task (Pearson correlation, r = 0.420, P < 0.01). A median split was performed to divide participants into those who made few (low) or many (high) errors of commission. A two-way ANOVA (errors of commission: low or high x drink, alcohol or no-alcohol) was computed, with log-transformed k values as the dependent variable. Participants who were high on errors of commission discounted delayed rewards at a higher rate than participants who were low on errors of commission [F(1,69) = 9.75, P < 0.05] and there was an interaction between errors of commission and drink [F(1,69) = 6.90, P < 0.05]. The effect of drink approached significance [F(1,69) = 3.31, P = 0.07]. These findings are illustrated in Fig. 2. t-tests revealed that, in the group low on errors of commission, rates of discounting were higher in the no-alcohol than in the alcohol group [t(27.72) = -2.93, P < 0.01].
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The reduction in cognitive impulsivity when intoxicated could be explained by the alcohol myopia theory in that aspects of the experimental environment (e.g. presence of the experimenter, knowledge that the experimenter would later view the data), may have acted as salient inhibiting cues, reducing impulsive behaviour in intoxicated individuals. This notion is supported by the finding that intoxicated participants who made few errors of commission on the go/no-go task also responded less impulsively on the delay-discounting task, only when intoxicated. That is, perhaps intoxicated participants who were most aware of inhibiting cues in the environment were more cautious on both measures of impulsivity.
Contrary to the initial prediction, the impelling cue manipulation did not moderate the effects of alcohol on delay discounting. It is possible that the impelling cues were not salient enough to influence participants responses, particularly because both possible outcomes were presented simultaneously. In contrast, in the real world, as well as in the study by MacDonald et al.(2000), cues regarding the delayed reward or negative outcome are not available during decision-making, unless the individual removes his attention from the current (i.e. more salient) environmental cues.
BALs correlated significantly with want more ratings, confirming that alcohol dose-dependently increases subjective reports of drug liking and the desire to consume more alcohol (Kirk and de Witt, 2000). Given this, further assessment of the effects of alcohol intoxication on preference for alcohol rewards would further our understanding of how alcohol consumption is perpetuated within a drinking session, and would have important implications for theories of how alcohol misuse is initiated and maintained.
Finally, trait impulsivity, as assessed by the ImpSS, was not associated with cognitive impulsivity on the delay-discounting task, even though it predicts alcohol use and other risk-taking behaviours in college students (Zuckerman and Kuhlman, 2000). The relation between personality measures of impulsivity and performance on behavioural tasks, however, depends on the specific questionnaire (Mitchell, 1999
; Richards et al., 1999
). In the current study, high scores on the personality measure were associated with slow responses on the go/no-go task, suggesting that impulsiveness sensation-seeking reflects motor impulsivity, rather than cognitive impulsivity.
In conclusion, the current study found that, contrary to common assumptions, alcohol intoxication in humans was associated with more cautious or less impulsive responding. Additionally, intoxicated participants were more likely to show lack of fit to the hyperbolic model, suggesting that they responded less consistently on the delay-discounting task. The delay-discounting paradigm could be further modified to test the predictions of alcohol myopia more effectively. For example, manipulations that increase inhibition conflict by using larger sums of money (Steele and Southwick, 1985) would be expected to increase the likelihood that alcohol will induce cognitive impulsivity. To enhance cue salience, participants attention could be drawn to the consequences of winning the immediate reward prior to completing the delay-discounting task. A further test of the alcohol myopia theory would involve assessing the effect of inhibiting cues on delay discounting. The theory would predict that, when faced with strong, salient, inhibiting cues, intoxicated individuals will discount delayed rewards less steeply than sober participants.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
FOOTNOTES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bigelow, G. E. (2001) An operant behavioural perspective on alcohol abuse and dependence. In International Handbook of Alcohol Dependence and Problems, Heather, N., Peters, T. J. and Stockwell, T. eds, pp. 299315. John Wiley & Sons Ltd, Chichester, UK.
Brunner, D. and Hen, R. (1997) Insights into the neurobiology of impulsive behavior from serotonin receptor knockout mice. Annals of the New York Academy of Sciences 836, 81103.[Abstract]
Cherpitel, C. J. (1993) Alcohol and injuries: A review of international emergency room studies. Addiction 88, 923937.[ISI][Medline]
Cherpitel, C. J. (1999) Substance use, injury and risk-taking dispositions in the general population. Alcoholism: Clinical and Experimental Research 23, 121126.[ISI][Medline]
Critchlow, B. (1986) The powers of John Barleycorn: Beliefs about the effects of alcohol on social behavior. American Psychologist 41, 751764.[CrossRef][ISI][Medline]
de Wit, H., Crean, J. and Richards, J. B. (2000) Effects of d-amphetamine and ethanol on a measure of behavioural inhibition in humans. Behavioral Neuroscience 114, 830837.[CrossRef][ISI][Medline]
Ericksen, K. P. and Trocki, K. F. (1992) Behavioral risk factors for sexually transmitted diseases in American households. Social Sciences Medicine 34, 843853.[CrossRef]
Finn, P. R., Justus, A., Mazas, C. and Steinmetz, J. E. (1999) Working memory, executive processes and the effects of alcohol on Go/No-Go learning: testing a model of behavioural regulation and impulsivity. Psychopharmacology 146, 465472.[ISI][Medline]
Grau, E. and Ortet, G. (1999) Personality traits and alcohol consumption in a sample of non-alcoholic women. Personality and Individual Differences 27, 10571066.[CrossRef][ISI]
Green, L., Myerson, J. and Ostaszewski, P. (1999) Amount of reward has opposite effects on the discounting of delayed and probabilistic outcomes. Journal of Experimental Psychology: Learning, Memory and Cognition 25, 418427.[CrossRef][ISI][Medline]
Kirby, K. N., Petry, N. M. and Bickel, W. K. (1999) Heroin addicts have higher discount rates for delayed rewards than non-drugusing controls. Journal of Experimental Psychology: General 128, 7887.[CrossRef][ISI][Medline]
Kirk, J. M. and de Wit, H. (2000) Individual differences in the priming effect of ethanol in social drinkers. Journal of Studies on Alcohol 61, 6471.[ISI][Medline]
MacDonald, T. K., Fong, G. T., Zanna, M. P. and Martineau, A. M. (2000) Alcohol myopia and condom use: Can alcohol intoxication be associated with more prudent behavior? Journal of Personality and Social Psychology 78, 605619.[CrossRef][ISI][Medline]
Mitchell, S. H. (1999) Measures of impulsivity in cigarette smokers and non-smokers. Psychopharmacology 146, 455464.[ISI][Medline]
Nagoshi, C. T., Wilson, J. R. and Rodriguez, L. A. (1991) Impulsivity, sensation seeking, and behavioural and emotional responses to alcohol. Alcoholism: Clinical and Experimental Research 15, 661667.[ISI][Medline]
Owen, A. M., Roberts, A. C., Polkey, C. E., Sahakain, B. J. and Robbins, T. W. (1991) Extra-dimensional versus intra-dimensional set shifting performance following frontal lobe excisions, temporal lobe excisions or amygdalo-hippocampectomy in man. Neuropsychologia 29, 9931006.[CrossRef][ISI][Medline]
Pernanen, K. (1976) Alcohol and crimes of violence. In The Biology of Alcoholism, Volume 4: Social Aspects of Alcoholism, Kissin, B. and Begleiter, H. eds, pp. 351444. Plenum Press, New York.
Petry, N. M. (2001) Delay discounting of money and alcohol in actively using alcoholics, currently abstinent alcoholics, and controls. Psychopharmacology 154, 243250.[CrossRef][ISI][Medline]
Poulos, C. X., Parker, J. L. and Lê, D. A. (1998) Increased impulsivity after injected alcohol predicts later alcohol consumption in rats: evidence for loss-of-control drinking and marked individual differences. Behavioral Neuroscience 112, 12471257.[CrossRef][ISI][Medline]
Richards, J. B., Zhang, L., Mitchell, S. H. and de Wit, H. (1999) Delay or probability discounting in a model of impulsive behavior: effect of alcohol. Journal of the Experimental Analysis of Behavior 71, 121143.[ISI][Medline]
Steele, C. M. and Josephs, R. A. (1990) Alcohol myopia: its prized and dangerous effects. American Psychologist 45, 921933.[CrossRef][ISI][Medline]
Steele, C. M. and Southwick, L. (1985) Alcohol and social behavior. I: The psychology of drunken excess. Journal of Personality and Social Psychology 48, 1834.[CrossRef][ISI][Medline]
Vuchinich, R. E. and Tucker, J. A. (1998). Choice, behavioral economics, and addictive behavior patterns. In Treating Addictive Behaviors, Miller, W. R. and Heather, N. eds, pp. 93104. Plenum Press, New York.
Zuckerman, M. and Kuhlman, D. M. (2000) Personality and risk-taking: common biosocial factors. Journal of Personality 68, 9991029.[CrossRef][ISI][Medline]
Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P. and Kraft, M. (1993) A comparison of three structural models for personality: the big three, the big five, and the alternative five. Journal of Personality and Social Psychology 65, 757768.[CrossRef][ISI]