- Email: [email protected]
Chronic ecstasy (MDMA) use is associated with deficits in task-switching but not inhibition or memory updating executive functions
Drug and Alcohol Dependence 83 (2006) 181–184
Short communication
Chronic ecstasy (MDMA) use is associated with deficits in task-switching but not inhibition or memory updating executive functions Richard Ian Dafters ∗ Psychology Department, 58 Hillhead Street, Glasgow University, Glasgow, Scotland G12 8QB, United Kingdom Received 26 May 2005; received in revised form 29 September 2005; accepted 5 November 2005
Abstract The study aimed to establish whether prior ecstasy (MDMA) use caused impairments in tasks which targeted specific executive functions and whether the magnitude of impairments were correlated with the level of drug use. Subjects who had used ecstasy and cannabis (N = 33), subjects who had used cannabis but not ecstasy (N = 17) and subjects who had used neither drug (N = 18) were tested on a standard and modified form of the Stroop Colour-Word test to look for impairments in inhibition and in task-switching, respectively. A Keep Track Task was used to assess working memory updating ability. Ecstasy users, but neither of the other groups were significantly impaired (slower reaction times) on the task-switching version of the Stroop test (p < .05), but not on the standard version or on the Keep Track Task. Regression analysis showed a significant correlation between level of lifetime ecstasy use and magnitude of the switching-impairment. The findings are interpreted in line with recent theoretical views which regard executive function as composed of multiple independent cognitive components such as inhibition, switching and memory updating, only some of which may be affected by drug use or other factors. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Ecstasy (MDMA); Executive functions; Switching; Inhibition; Updating
1. Introduction Despite increasing research interest, a clear picture of Ecstasy’s (MDMA’s) long-term effects (if any) on executive cognitive function has yet to emerge. Thus, while there have been reports of deficits in standard executive tasks such as the Wisconsin Card Sorting Test (WCST) and on the Tower of Hanoi (TOH) task (Fox et al., 2001) as well as tests of cognitive flexibility and working memory (Bolla et al., 1998; Wareing et al., 2000), other studies using the same or similar tests have failed to observe deficits (Thomasius et al., 2003; Morgan et al., 2002). A contributing factor to the lack of agreement may be that many of the standard tests of executive function are complex tasks incorporating independent cognitive functions only some of which may be affected by prior drug use. In an influential paper, Miyake et al. (2000) proposed that executive functions comprised three generally-agreed components: monitoring and updating of short-term memory, inhibition of pre-potent responses and task-shifting. Using factor analysis techniques, they were able to
∗
Tel.: +44 141 330 4559. E-mail address: [email protected].
0376-8716/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.drugalcdep.2005.11.006
show that these three components are clearly separable and that standard executive function tests such as the WCST, the Tower of Hanoi (TOH) and the Random Letter Generation tasks tended to “load” heavily on particular components. The complex and varied nature of the executive tests also means that they may not be comparable in terms of the confounding non-executive functions they involve. Thus, deficits may appear in some tasks, but not others because of differences in the “lower-level” perceptual, attentional and motor tasks they involve. Drug related deficits may be due to impairment of these lower-level subcomponents. A more effective strategy to examine effects of ecstasy use on executive functions may be to concentrate on tasks which focus on individual executive components and tasks which are matched as far as possible in terms of lower-level processes. To this end, the present study used a modified Stroop Colour-Word paradigm to look for deficits in inhibition and/or task-switching in ecstasy users and a Keep Track Task (adapted from Miyake et al., 2000) to assess updating capacity. The Stroop ColourWord task in its normal form is regarded as a prototypical and relatively pure measure of inhibition of a pre-potent response tendency. Subjects are required to report the ink colour in which a word is printed while ignoring the incompatible word name. For example, the word green presented in red ink requires a
182
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
response of red. Typically, responses are slower on incompatible (incongruent) trials relative to congruent trials in which the ink colour and word name match. A nice feature of this paradigm, which was exploited in the present study, is that with a simple modification a task-switching-component can be introduced. On a proportion of trials the stimulus may be marked in some way, for example by underlining, indicating that the word name not the ink colour should be identified. Impairment on both versions of the Stroop task may indicate an inhibition deficit, while impairment limited to the modified form of the task is indicative of a task-switching-deficit. As a control for the lower-level confounding processes referred to earlier, both versions of the Stroop task may be accompanied by exposure to colour- and word-only versions in which the task is to identify the colour of a simple coloured rectangle or the name of a colour-word in standard black typeface. Impairment in these tasks would indicate low-level perceptual, attentional or reading/naming deficits which might contribute to Stroop performance. A Keep Track Task (adapted from Miyake et al., 2000) was used to assess subjects’ ability to update working memory. Subjects viewed serially a list of nouns drawn from several categories displayed at the bottom of the screen. At the end of the list the task was to report the last word that occurred in each of the categories. Accurate performance on this task requires constant updating of the contents of short-term memory. Since the majority of regular ecstasy users are also heavy users of cannabis it is important to control for possible executive deficits arising from cannabis use. Thus, as in previous studies we compared three-groups of subjects: ecstasy–cannabis users, cannabis users (who used equivalent levels of cannabis but no ecstasy) and non-drug users (Gouzoulis-Mayfrank et al., 2000; Dafters et al., 2004). Levels of other drugs used were entered as covariates during analysis. 2. Methods 2.1. Participants Subjects were university students or their friends contacted via posters and word of mouth who had taken part in an unrelated
experiment three-weeks previously when the subject demographics were taken (see Table 1). All subjects signed a written consent form and the experiment was ratified by local ethical review. Group assignment was made on the basis of a selfreported drug use history questionnaire which asks subjects to recall their level of consumption of various substances over the previous week, month, year and date of first use as well as recording age and educational background. Subjects with a lifetime use of 50 or more ecstasy tablets and who had smoked cannabis a minimum of 50 times were assigned to the ecstasy–cannabis (ECST-CANN) group (N = 18); subjects who had taken fewer than 50 ecstasy tablets, but had smoked cannabis on 50 or more occasions were assigned to the CANNABIS group (N = 17); subjects meeting neither criteria were assigned to the non-user group (N = 18). Subjects were excluded on stated history of neurological disorder, brain injury, colour-blindness and if they claimed to have taken ecstasy within the previous 5 days, or cannabis within the previous 48 h. 2.2. Procedure The Stroop tasks were presented in a sound-attenuated cubicle on a Macintosh G3 computer fitted with a three-key response box providing millisecond resolution timing and programmed using Psyscope. There were four phases in the experiment. In phase 1(128 trials), stimuli were small rectangular blocks of colour (red, blue or green) presented serially and randomly in the centre of the screen. Subjects indicated the colour by pressing one of three-keys with the index and second finger of the dominant hand and the index finger of the non-dominant hand. The response terminated the stimulus and initiated a 500 ms inter-stimulus interval. In phase 2 (128 trials) the colours were replaced by the words red, green and blue printed in black font – the subject indicated the colour named as quickly as possible. In phase 3 (256 trials), subjects experienced Stroop Colour-Word interference—the words red, blue and green were presented, but in incongruent ink colour, for example, the word red in green ink. The subjects were required to identify the ink colour as quickly as possible using the keys. In phase 4 (256 trials), the stimuli and procedure were the same, but on a randomised half of the trials
Table 1 Subject demographics Measure Age (years) Gender (number of males)
Non-users (N = 18) 22.67 (2.56) 10
Estimated lifetime consumption Ecstasy (pills) 0.00 (0.00) Cannabis (joints) 0.00 (0.00) Cocaine (grams) 0.00 (0.00) Amphetamine (grams) 0.11 (0.47) Alcohol (units) 22243.0 (2049.3) Tobacco (cigarettes) 504.3 (1171.1) Heroin (times used) 0.00 (0.00) LSD (pills) 0.00 (0.00) a b
Different from non-user group. Different from cannabis group (p < 0.05).
Cannabis users (N = 17) 23.19 (1.15) 13 4.23 (7.01) 6180.5 (10549)a 1.06 (1.39)a 1.47 (5.81) 11528 (10232)a 4454.7 (4436)a 0.12 (0.33) 0.23 (0.56)
Ecstasy–cannabis users (N = 18)
Ecstasy–cannabis users (task-order reversed) (N = 15)
23.24 (2.33) 12
22.93 (2.34) 9
522.33 (936.71)a,b 5570.9 (9364.9)a 38.72 (94.2) 132.61 (389.2) 10955.1 (9100.7)a 18291.1 (21851)a,b 8.33 (35.3) 28.56 (93.47)
475.87 (406.41)a,b 5200 (7792.9)a 4.93 (7.36)a,b 22 (64.4) 10220 (7052)a 13190 (9763)a,b 0.00 (0.00) 1.83 (3.87)
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
the word was underlined in black – on these trials the subject was to report the word name and ignore the ink colour. Each phase was preceded by 24 practice trials, the data from which was discarded. Although it was considered unlikely as all subjects had completed the sequence of four phases within 30 min and the first two phases were very easy, the use of a fixed phase sequence means that any deficits found in the ecstasy users in the crucial later phases could result from fatigue, loss of motivation or proactive interference effects. For this reason, a second group of ecstasy–cannabis users selected as before (N = 15) carried out the sequence of tasks in reverse order, i.e. switching, inhibition, word naming, then colour-naming. In all groups the Keep Track Task followed the Stroop tasks. In the Keep Track Task, a measure of working memory updating ability, subjects were presented with four target categories written on a piece of card which remained visible during the test. Eighteen words were presented including two or threeexemplars from each of the four displayed categories and from the other two categories in the total pool of six used in the experiment (animals, fruits, furniture, tools, vehicles and musical instruments). The words printed centrally on cards were placed by the experimenter one at a time in front of the subject for two seconds. After the last card had been displayed the subject was to recall the last word seen from each of the four target categories. The test was repeated three-times with a different subset of target categories and exemplars giving a total possible correct score of twelve. All differences between groups and tasks were analysed using Anova and post-hoc Tukey HSD tests. 3. Results No subject made more than 24 errors (out of 256 trials) in the two most difficult conditions of the Stroop test (inhibition and inhibition + switching) and error rates were generally too low for meaningful statistical analysis. Analysis was carried out on the reaction time (RT) data shown in Table 2. One-way Anovas were used to examine the effect of the druggroup on the RT’s for each phase of the Stroop experiment and on the total number of words correctly recalled in the Keep Track Task. Levels of drug use were similar in the two ECST–CANN groups given opposite phase sequences in the Stroop tasks (p > .5) and no significant differences were found between them on any response measure (p > .5 in each case). Thus, the possibility of the outcome being affected by task-sequencing effects such as fatigue or proactive interference may be ruled out and the subjects in the two groups were combined for subsequent analysis. Since there were differences between the drug using
183
Fig. 1. Scatter plot and fitted regression line of estimated lifetime consumption of ecstasy (pills) plotted against RT on the switching-task version of the Stroop test. One outlier with consumption of >4000 pills has been omitted to preserve scale.
subjects and the non-users in level of other drug use (see Table 1) lifetime consumption of cocaine, amphetamines, alcohol and tobacco were included as covariates in the analysis. Comparing non-users, cannabis users and combined ecstasy–cannabis users the only significant group effect occurred in the taskswitch version of the Stroop test (F(2, 61) = 3.529, p < .05). A post-hoc Tukey’s HSD test revealed that the ecstasy–cannabis group was significantly impaired on the switching-task relative to the cannabis and non-user groups (p < .05 in each case) and that the cannabis and non-user groups did not differ from each other. Previous studies have attempted to correlate the level of prior ecstasy use with the magnitude of cognitive impairment measured (Bolla et al., 1998; Gouzoulis-Mayfrank et al., 2000; Fox et al., 2001) and it has been argued that the demonstration of a relationship between severity of drug use and magnitude of cognitive deficit is more indicative of a causal relationship than an overall difference between different user groups (VerdejoGarcia et al., 2005). For this reason, in the present study multiple regression analysis was used to look for a correlation between lifetime ecstasy (and other drug) consumption and the degree of impairment (slowing) on the switching-task. Regression showed that mean-RT on the switching-task was correlated with lifetime ecstasy use (beta = 0.590, p < 0.01), but not with lifetime use of cannabis, amphetamine, alcohol, cocaine or tobacco (see Fig. 1).
Table 2 Mean reaction times (S.D.) on the four versions of the Stroop task and total word recall (max = 12) on the Keep Track Task Group
Colour-RT
Word-RT
Inhibition-RT
Switching-RT
Keep track task (words)
Non-users (N = 18) Cannabis (N = 17) ECST–CANN (N = 33)
699 (129) 724 (120) 713 (134)
682 (117) 684 (124) 683 (133)
838 (163) 829 (150) 834 (85)
848 (131) 856 (114) 966 (137)*
6.1 (1.9) 5.8 (1.7) 6.3 (2.5)
*
Significant difference from non-users and cannabis groups (p < 0.05).
184
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
4. Discussion Subjects with a history of recreational ecstasy use showed an impairment on a specific task-switching version of the Stroop task relative to subjects with a similar level of cannabis use and subjects who used neither drug. The deficit was significant when use of alcohol, cocaine, amphetamine and tobacco were controlled for using covariate analysis and occurred irrespective of whether the crucial switching-phase was at the beginning or the end of the task-sequence. Regression analysis showed that within the ecstasy-using group past ecstasy use, but not use of cannabis or other drugs was correlated with the degree of impairment in task-switching. The absence of impairment in the other versions of the task and in the Keep Track Task indicates that the deficit is limited to the switching-component and is not due to problems with inhibition of pre-potent responses, to impaired lower-level perceptual/attentional factors, or to a deficit in the ability to update working memory. The specificity of the deficit vindicates the approach adopted here of attempting to isolate for study the individual cognitive components which make up the executive functions. Ecstasyrelated deficits restricted to task-switching were recently reported elsewhere. Ecstasy-users were impaired on a mixedtask condition involving switching between judgements of relative numerosity of dots or triangles compared to task-repetition (non-switch) conditions and made more repetition errors on the WCST, a complex executive function task which loads primarily on the switching-component, while no drug related impairments were found in tasks designed to tap inhibition processes or working memory function (von Gesau et al., 2004). There is strong evidence linking ecstasy use with metabolic reductions of both dopamine and serotonin in frontal lobe regions (Semple et al., 1999; Cohen et al., 1996) so impairment of executive function is a likely outcome of prior ecstasy use. The evidence here suggests that concentration on tasks which target specific cognitive components may be more likely to reveal deficits than using standard batteries of complex executive tasks.
Acknowledgements Research supported by internal funds. References Bolla, K.I., McCann, U.D., Ricaurte, G.A., 1998. Memory impairment in abstinent MDMA (“ecstasy”) users. Neurology 5, 1532–1537. Cohen, Z., Bonvento, G., Lacombe, P., Hamel, E., 1996. Serotonin in the regulation of brain microcirculation. Prog. Neurobiol. 50, 335–362. Dafters, R.I., Hoshi, R., Talbot, A.C., 2004. Contribution of cannabis and MDMA (“ecstasy”) to cognitive changes in long-term polydrug users. Psychopharmacology 173, 405–410. Fox, H.C., Parrott, A.C., Turner, J.J.D., 2001. Ecstasy use: cognitive deficits related to dosage rather than self-reported problematic use of the drug. J. Psychopharmacol. 15, 273–281. Gouzoulis-Mayfrank, E., Daumann, J., Tuchtenhagen, G., Pelz, S., Becker, S., Kunert, H.K., Fimm, B., Sass, H., 2000. Impaired cognitive performance in drug-free users of recreational ecstasy (MDMA). J. Neurol. Neurosurg. 68, 719–725. Miyake, A., Friedman, N.P., Emerson, M.J., Witzki, A.H., Howerter, A., 2000. The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: a latent variable analysis. Cognit. Psychol. 41, 49–100. Morgan, M.J., McFie, L., Fleetwood, L.H., Robinson, J.A., 2002. Ecstasy (MDMA): are the psychological problems associated with its use reversed by prolonged absence ? Psychopharmacology 159, 294–303. Semple, D.M., Ebmeier, K.P., Glubus, M.F., O’Carroll, R.F., Johnsone, F.C., 1999. Reduced in vivo binding to the serotonin transporter in the cerebral cortex of MDMA (“ecstasy”) users. Br. J. Psychiatry. 175, 63–69. Thomasius, R., Peterson, K., Buchert, R., Andresen, B., Zapleova, P., Wartberg, L., Nebeling, B., Schmoldt, A., 2003. Mood, cognition and serotonin transporter availability in current and former ecstasy users. Psychopharmacology 167, 85–96. Verdejo-Garcia, J., Lopez-Torrecillas, F., Aguilar de Arcos, F., Perez-Garcia, M., 2005. Differential effects of MDMA, cocaine and cannabis use severity on distinctive components of the executive functions in polysubstance users; a multiple regression analysis. Addict. Behav. 30, 89–101. von Gesau, A.N., Stalenhoef, P., Huizinga, M., Snel, J., Ridderinkhof, K.R., 2004. Impaired executive function in male MDMA (“ecstasy”) users. Psychopharmacology 175, 331–341. Wareing, M., Fisk, J.E., Murphy, P.N., 2000. Working memory deficits in current and previous users of MDMA (“ecstasy”). Br. J. Psychol. 91, 181–188.
Short communication
Chronic ecstasy (MDMA) use is associated with deficits in task-switching but not inhibition or memory updating executive functions Richard Ian Dafters ∗ Psychology Department, 58 Hillhead Street, Glasgow University, Glasgow, Scotland G12 8QB, United Kingdom Received 26 May 2005; received in revised form 29 September 2005; accepted 5 November 2005
Abstract The study aimed to establish whether prior ecstasy (MDMA) use caused impairments in tasks which targeted specific executive functions and whether the magnitude of impairments were correlated with the level of drug use. Subjects who had used ecstasy and cannabis (N = 33), subjects who had used cannabis but not ecstasy (N = 17) and subjects who had used neither drug (N = 18) were tested on a standard and modified form of the Stroop Colour-Word test to look for impairments in inhibition and in task-switching, respectively. A Keep Track Task was used to assess working memory updating ability. Ecstasy users, but neither of the other groups were significantly impaired (slower reaction times) on the task-switching version of the Stroop test (p < .05), but not on the standard version or on the Keep Track Task. Regression analysis showed a significant correlation between level of lifetime ecstasy use and magnitude of the switching-impairment. The findings are interpreted in line with recent theoretical views which regard executive function as composed of multiple independent cognitive components such as inhibition, switching and memory updating, only some of which may be affected by drug use or other factors. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Ecstasy (MDMA); Executive functions; Switching; Inhibition; Updating
1. Introduction Despite increasing research interest, a clear picture of Ecstasy’s (MDMA’s) long-term effects (if any) on executive cognitive function has yet to emerge. Thus, while there have been reports of deficits in standard executive tasks such as the Wisconsin Card Sorting Test (WCST) and on the Tower of Hanoi (TOH) task (Fox et al., 2001) as well as tests of cognitive flexibility and working memory (Bolla et al., 1998; Wareing et al., 2000), other studies using the same or similar tests have failed to observe deficits (Thomasius et al., 2003; Morgan et al., 2002). A contributing factor to the lack of agreement may be that many of the standard tests of executive function are complex tasks incorporating independent cognitive functions only some of which may be affected by prior drug use. In an influential paper, Miyake et al. (2000) proposed that executive functions comprised three generally-agreed components: monitoring and updating of short-term memory, inhibition of pre-potent responses and task-shifting. Using factor analysis techniques, they were able to
∗
Tel.: +44 141 330 4559. E-mail address: [email protected].
0376-8716/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.drugalcdep.2005.11.006
show that these three components are clearly separable and that standard executive function tests such as the WCST, the Tower of Hanoi (TOH) and the Random Letter Generation tasks tended to “load” heavily on particular components. The complex and varied nature of the executive tests also means that they may not be comparable in terms of the confounding non-executive functions they involve. Thus, deficits may appear in some tasks, but not others because of differences in the “lower-level” perceptual, attentional and motor tasks they involve. Drug related deficits may be due to impairment of these lower-level subcomponents. A more effective strategy to examine effects of ecstasy use on executive functions may be to concentrate on tasks which focus on individual executive components and tasks which are matched as far as possible in terms of lower-level processes. To this end, the present study used a modified Stroop Colour-Word paradigm to look for deficits in inhibition and/or task-switching in ecstasy users and a Keep Track Task (adapted from Miyake et al., 2000) to assess updating capacity. The Stroop ColourWord task in its normal form is regarded as a prototypical and relatively pure measure of inhibition of a pre-potent response tendency. Subjects are required to report the ink colour in which a word is printed while ignoring the incompatible word name. For example, the word green presented in red ink requires a
182
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
response of red. Typically, responses are slower on incompatible (incongruent) trials relative to congruent trials in which the ink colour and word name match. A nice feature of this paradigm, which was exploited in the present study, is that with a simple modification a task-switching-component can be introduced. On a proportion of trials the stimulus may be marked in some way, for example by underlining, indicating that the word name not the ink colour should be identified. Impairment on both versions of the Stroop task may indicate an inhibition deficit, while impairment limited to the modified form of the task is indicative of a task-switching-deficit. As a control for the lower-level confounding processes referred to earlier, both versions of the Stroop task may be accompanied by exposure to colour- and word-only versions in which the task is to identify the colour of a simple coloured rectangle or the name of a colour-word in standard black typeface. Impairment in these tasks would indicate low-level perceptual, attentional or reading/naming deficits which might contribute to Stroop performance. A Keep Track Task (adapted from Miyake et al., 2000) was used to assess subjects’ ability to update working memory. Subjects viewed serially a list of nouns drawn from several categories displayed at the bottom of the screen. At the end of the list the task was to report the last word that occurred in each of the categories. Accurate performance on this task requires constant updating of the contents of short-term memory. Since the majority of regular ecstasy users are also heavy users of cannabis it is important to control for possible executive deficits arising from cannabis use. Thus, as in previous studies we compared three-groups of subjects: ecstasy–cannabis users, cannabis users (who used equivalent levels of cannabis but no ecstasy) and non-drug users (Gouzoulis-Mayfrank et al., 2000; Dafters et al., 2004). Levels of other drugs used were entered as covariates during analysis. 2. Methods 2.1. Participants Subjects were university students or their friends contacted via posters and word of mouth who had taken part in an unrelated
experiment three-weeks previously when the subject demographics were taken (see Table 1). All subjects signed a written consent form and the experiment was ratified by local ethical review. Group assignment was made on the basis of a selfreported drug use history questionnaire which asks subjects to recall their level of consumption of various substances over the previous week, month, year and date of first use as well as recording age and educational background. Subjects with a lifetime use of 50 or more ecstasy tablets and who had smoked cannabis a minimum of 50 times were assigned to the ecstasy–cannabis (ECST-CANN) group (N = 18); subjects who had taken fewer than 50 ecstasy tablets, but had smoked cannabis on 50 or more occasions were assigned to the CANNABIS group (N = 17); subjects meeting neither criteria were assigned to the non-user group (N = 18). Subjects were excluded on stated history of neurological disorder, brain injury, colour-blindness and if they claimed to have taken ecstasy within the previous 5 days, or cannabis within the previous 48 h. 2.2. Procedure The Stroop tasks were presented in a sound-attenuated cubicle on a Macintosh G3 computer fitted with a three-key response box providing millisecond resolution timing and programmed using Psyscope. There were four phases in the experiment. In phase 1(128 trials), stimuli were small rectangular blocks of colour (red, blue or green) presented serially and randomly in the centre of the screen. Subjects indicated the colour by pressing one of three-keys with the index and second finger of the dominant hand and the index finger of the non-dominant hand. The response terminated the stimulus and initiated a 500 ms inter-stimulus interval. In phase 2 (128 trials) the colours were replaced by the words red, green and blue printed in black font – the subject indicated the colour named as quickly as possible. In phase 3 (256 trials), subjects experienced Stroop Colour-Word interference—the words red, blue and green were presented, but in incongruent ink colour, for example, the word red in green ink. The subjects were required to identify the ink colour as quickly as possible using the keys. In phase 4 (256 trials), the stimuli and procedure were the same, but on a randomised half of the trials
Table 1 Subject demographics Measure Age (years) Gender (number of males)
Non-users (N = 18) 22.67 (2.56) 10
Estimated lifetime consumption Ecstasy (pills) 0.00 (0.00) Cannabis (joints) 0.00 (0.00) Cocaine (grams) 0.00 (0.00) Amphetamine (grams) 0.11 (0.47) Alcohol (units) 22243.0 (2049.3) Tobacco (cigarettes) 504.3 (1171.1) Heroin (times used) 0.00 (0.00) LSD (pills) 0.00 (0.00) a b
Different from non-user group. Different from cannabis group (p < 0.05).
Cannabis users (N = 17) 23.19 (1.15) 13 4.23 (7.01) 6180.5 (10549)a 1.06 (1.39)a 1.47 (5.81) 11528 (10232)a 4454.7 (4436)a 0.12 (0.33) 0.23 (0.56)
Ecstasy–cannabis users (N = 18)
Ecstasy–cannabis users (task-order reversed) (N = 15)
23.24 (2.33) 12
22.93 (2.34) 9
522.33 (936.71)a,b 5570.9 (9364.9)a 38.72 (94.2) 132.61 (389.2) 10955.1 (9100.7)a 18291.1 (21851)a,b 8.33 (35.3) 28.56 (93.47)
475.87 (406.41)a,b 5200 (7792.9)a 4.93 (7.36)a,b 22 (64.4) 10220 (7052)a 13190 (9763)a,b 0.00 (0.00) 1.83 (3.87)
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
the word was underlined in black – on these trials the subject was to report the word name and ignore the ink colour. Each phase was preceded by 24 practice trials, the data from which was discarded. Although it was considered unlikely as all subjects had completed the sequence of four phases within 30 min and the first two phases were very easy, the use of a fixed phase sequence means that any deficits found in the ecstasy users in the crucial later phases could result from fatigue, loss of motivation or proactive interference effects. For this reason, a second group of ecstasy–cannabis users selected as before (N = 15) carried out the sequence of tasks in reverse order, i.e. switching, inhibition, word naming, then colour-naming. In all groups the Keep Track Task followed the Stroop tasks. In the Keep Track Task, a measure of working memory updating ability, subjects were presented with four target categories written on a piece of card which remained visible during the test. Eighteen words were presented including two or threeexemplars from each of the four displayed categories and from the other two categories in the total pool of six used in the experiment (animals, fruits, furniture, tools, vehicles and musical instruments). The words printed centrally on cards were placed by the experimenter one at a time in front of the subject for two seconds. After the last card had been displayed the subject was to recall the last word seen from each of the four target categories. The test was repeated three-times with a different subset of target categories and exemplars giving a total possible correct score of twelve. All differences between groups and tasks were analysed using Anova and post-hoc Tukey HSD tests. 3. Results No subject made more than 24 errors (out of 256 trials) in the two most difficult conditions of the Stroop test (inhibition and inhibition + switching) and error rates were generally too low for meaningful statistical analysis. Analysis was carried out on the reaction time (RT) data shown in Table 2. One-way Anovas were used to examine the effect of the druggroup on the RT’s for each phase of the Stroop experiment and on the total number of words correctly recalled in the Keep Track Task. Levels of drug use were similar in the two ECST–CANN groups given opposite phase sequences in the Stroop tasks (p > .5) and no significant differences were found between them on any response measure (p > .5 in each case). Thus, the possibility of the outcome being affected by task-sequencing effects such as fatigue or proactive interference may be ruled out and the subjects in the two groups were combined for subsequent analysis. Since there were differences between the drug using
183
Fig. 1. Scatter plot and fitted regression line of estimated lifetime consumption of ecstasy (pills) plotted against RT on the switching-task version of the Stroop test. One outlier with consumption of >4000 pills has been omitted to preserve scale.
subjects and the non-users in level of other drug use (see Table 1) lifetime consumption of cocaine, amphetamines, alcohol and tobacco were included as covariates in the analysis. Comparing non-users, cannabis users and combined ecstasy–cannabis users the only significant group effect occurred in the taskswitch version of the Stroop test (F(2, 61) = 3.529, p < .05). A post-hoc Tukey’s HSD test revealed that the ecstasy–cannabis group was significantly impaired on the switching-task relative to the cannabis and non-user groups (p < .05 in each case) and that the cannabis and non-user groups did not differ from each other. Previous studies have attempted to correlate the level of prior ecstasy use with the magnitude of cognitive impairment measured (Bolla et al., 1998; Gouzoulis-Mayfrank et al., 2000; Fox et al., 2001) and it has been argued that the demonstration of a relationship between severity of drug use and magnitude of cognitive deficit is more indicative of a causal relationship than an overall difference between different user groups (VerdejoGarcia et al., 2005). For this reason, in the present study multiple regression analysis was used to look for a correlation between lifetime ecstasy (and other drug) consumption and the degree of impairment (slowing) on the switching-task. Regression showed that mean-RT on the switching-task was correlated with lifetime ecstasy use (beta = 0.590, p < 0.01), but not with lifetime use of cannabis, amphetamine, alcohol, cocaine or tobacco (see Fig. 1).
Table 2 Mean reaction times (S.D.) on the four versions of the Stroop task and total word recall (max = 12) on the Keep Track Task Group
Colour-RT
Word-RT
Inhibition-RT
Switching-RT
Keep track task (words)
Non-users (N = 18) Cannabis (N = 17) ECST–CANN (N = 33)
699 (129) 724 (120) 713 (134)
682 (117) 684 (124) 683 (133)
838 (163) 829 (150) 834 (85)
848 (131) 856 (114) 966 (137)*
6.1 (1.9) 5.8 (1.7) 6.3 (2.5)
*
Significant difference from non-users and cannabis groups (p < 0.05).
184
R.I. Dafters / Drug and Alcohol Dependence 83 (2006) 181–184
4. Discussion Subjects with a history of recreational ecstasy use showed an impairment on a specific task-switching version of the Stroop task relative to subjects with a similar level of cannabis use and subjects who used neither drug. The deficit was significant when use of alcohol, cocaine, amphetamine and tobacco were controlled for using covariate analysis and occurred irrespective of whether the crucial switching-phase was at the beginning or the end of the task-sequence. Regression analysis showed that within the ecstasy-using group past ecstasy use, but not use of cannabis or other drugs was correlated with the degree of impairment in task-switching. The absence of impairment in the other versions of the task and in the Keep Track Task indicates that the deficit is limited to the switching-component and is not due to problems with inhibition of pre-potent responses, to impaired lower-level perceptual/attentional factors, or to a deficit in the ability to update working memory. The specificity of the deficit vindicates the approach adopted here of attempting to isolate for study the individual cognitive components which make up the executive functions. Ecstasyrelated deficits restricted to task-switching were recently reported elsewhere. Ecstasy-users were impaired on a mixedtask condition involving switching between judgements of relative numerosity of dots or triangles compared to task-repetition (non-switch) conditions and made more repetition errors on the WCST, a complex executive function task which loads primarily on the switching-component, while no drug related impairments were found in tasks designed to tap inhibition processes or working memory function (von Gesau et al., 2004). There is strong evidence linking ecstasy use with metabolic reductions of both dopamine and serotonin in frontal lobe regions (Semple et al., 1999; Cohen et al., 1996) so impairment of executive function is a likely outcome of prior ecstasy use. The evidence here suggests that concentration on tasks which target specific cognitive components may be more likely to reveal deficits than using standard batteries of complex executive tasks.
Acknowledgements Research supported by internal funds. References Bolla, K.I., McCann, U.D., Ricaurte, G.A., 1998. Memory impairment in abstinent MDMA (“ecstasy”) users. Neurology 5, 1532–1537. Cohen, Z., Bonvento, G., Lacombe, P., Hamel, E., 1996. Serotonin in the regulation of brain microcirculation. Prog. Neurobiol. 50, 335–362. Dafters, R.I., Hoshi, R., Talbot, A.C., 2004. Contribution of cannabis and MDMA (“ecstasy”) to cognitive changes in long-term polydrug users. Psychopharmacology 173, 405–410. Fox, H.C., Parrott, A.C., Turner, J.J.D., 2001. Ecstasy use: cognitive deficits related to dosage rather than self-reported problematic use of the drug. J. Psychopharmacol. 15, 273–281. Gouzoulis-Mayfrank, E., Daumann, J., Tuchtenhagen, G., Pelz, S., Becker, S., Kunert, H.K., Fimm, B., Sass, H., 2000. Impaired cognitive performance in drug-free users of recreational ecstasy (MDMA). J. Neurol. Neurosurg. 68, 719–725. Miyake, A., Friedman, N.P., Emerson, M.J., Witzki, A.H., Howerter, A., 2000. The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: a latent variable analysis. Cognit. Psychol. 41, 49–100. Morgan, M.J., McFie, L., Fleetwood, L.H., Robinson, J.A., 2002. Ecstasy (MDMA): are the psychological problems associated with its use reversed by prolonged absence ? Psychopharmacology 159, 294–303. Semple, D.M., Ebmeier, K.P., Glubus, M.F., O’Carroll, R.F., Johnsone, F.C., 1999. Reduced in vivo binding to the serotonin transporter in the cerebral cortex of MDMA (“ecstasy”) users. Br. J. Psychiatry. 175, 63–69. Thomasius, R., Peterson, K., Buchert, R., Andresen, B., Zapleova, P., Wartberg, L., Nebeling, B., Schmoldt, A., 2003. Mood, cognition and serotonin transporter availability in current and former ecstasy users. Psychopharmacology 167, 85–96. Verdejo-Garcia, J., Lopez-Torrecillas, F., Aguilar de Arcos, F., Perez-Garcia, M., 2005. Differential effects of MDMA, cocaine and cannabis use severity on distinctive components of the executive functions in polysubstance users; a multiple regression analysis. Addict. Behav. 30, 89–101. von Gesau, A.N., Stalenhoef, P., Huizinga, M., Snel, J., Ridderinkhof, K.R., 2004. Impaired executive function in male MDMA (“ecstasy”) users. Psychopharmacology 175, 331–341. Wareing, M., Fisk, J.E., Murphy, P.N., 2000. Working memory deficits in current and previous users of MDMA (“ecstasy”). Br. J. Psychol. 91, 181–188.