- Email: [email protected]
Effects of alcohol on disinhibition towards alcohol-related cues
Drug and Alcohol Dependence 127 (2013) 137–142
Contents lists available at SciVerse ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Effects of alcohol on disinhibition towards alcohol-related cues Sally Adams ∗ , Alia F. Ataya, Angela S. Attwood, Marcus R. Munafò School of Experimental Psychology, University of Bristol, 12a Priory Road, Bristol BS8 1TU, UK
a r t i c l e
i n f o
Article history: Received 30 March 2012 Received in revised form 21 June 2012 Accepted 23 June 2012 Available online 26 July 2012 Keywords: Alcohol Inhibitory control Alcohol-shifting task
a b s t r a c t Background: We investigated (1) the effects of acute alcohol on inhibition of alcohol-related versus neutral cues, (2) the effects of drinking status on inhibition of alcohol-related versus neutral cues, and (3) the similarity of any effects of alcohol or drinking status across two different cue types (lexical versus pictorial). Methods: Participants received 0.0 g/kg, 0.4 g/kg or 0.6 g/kg of alcohol in a between-subjects design. Healthy, heavy and light social alcohol users (n = 96) completed both lexical and pictorial cue versions of an alcohol-shifting task. Participants were instructed to respond to target stimuli by pressing the spacebar, but to ignore distracter stimuli. Errors towards distracter stimuli were analysed using a series of mixedmodel ANOVAs, with between-subjects factors of challenge and drinking status and within-subjects factors of distracter type (alcohol, neutral) and block (shift, non-shift). Results: Lexical commission error data indicated a main effect of distracter (F [1,90] = 43.25, p < 0.001, 2 = 0.33), which was qualified by a marginal interaction with challenge condition (F [2,90] = 2.77, p = 0.068, 2 = 0.06). Following an acute high dose of alcohol participants made more errors towards alcohol distracters. Pictorial commission error data indicated a significant main effect of distracter (F [1,90] = 67.40, p < 0.001, 2 = 0.43), such that all participants made more errors towards neutral image distracters versus alcohol distracter images. Conclusions: Our results reveal acute alcohol’s impairment of inhibitory control may be enhanced when a response towards alcohol-related lexical stimuli is required to be withheld. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Alcohol’s priming effects have been attributed to alcoholinduced impairment of cognitive mechanisms, including inhibitory control and response towards alcohol-related cues (Field et al., 2010). Alcohol is widely regarded to diminish inhibitory control of behaviour (Fillmore, 2007) that is, the ability to suppress a prepotent response and/or behavioural impulse and more recently goal-directed control (Hogarth et al., 2012). The stop-signal and go/no-go tasks have been developed to assess impairment of inhibitory control, and these indicate that alcohol specifically disrupts response inhibition at doses (0.4–0.6 g/kg) that are not associated with impairment of activational processes (de Wit et al., 2000; Easdon and Vogel-Sprott, 2000; Marczinski et al., 2005). These data are consistent with disinhibitory models of drug use, which suggest that alcohol intoxication may lead to a loss of intentional control over alcohol-seeking behaviours. Alcohol intoxication is also thought to increase the incentive value of alcohol-related stimuli. Following intoxication, alcohol cues gain strong motivational properties, which may be viewed as
∗ Corresponding author. Tel.: +44 117 9288547; fax: +44 117 9288588. E-mail address: [email protected] (S. Adams). 0376-8716/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.drugalcdep.2012.06.025
“hijacking” and influencing automatic alcohol cognitions. For example, studies have reliably indicated that moderate doses of alcohol (0.3–0.4 g/kg) enhance cognitive bias towards alcoholrelated cues. Moreover, acute alcohol intake may enhance positive automatic associations with alcohol-related cues (Palfai and Ostafin, 2003) and strengthen associations between alcohol and approach behaviours (Farris and Ostafin, 2008). These findings provide support for two distinct factors involved in acute alcohol consumption: inhibitory control and cognitive biases for alcohol-related cues; however, they do not indicate whether these mechanisms may interact under alcohol. Several models of addiction (Field et al., 2010; Goldstein and Volkow, 2002; Jentsch and Taylor, 1999) suggest that acute intoxication may simultaneously reduce response inhibition whilst increasing automatised salience attribution for drug-cues. It is proposed that top-down inhibitory processes are lowered, releasing automatic appetitive behaviours that would typically be under close regulation. In this state, response inhibition is reduced and response to salient alcohol-related cues is accentuated (Field et al., 2010). Two studies have examined inhibitory control and cognitive bias for alcohol-related cues in heavier drinking adolescents (Field et al., 2007) and alcohol-dependent individuals (Noel et al., 2007). However, only one study (Rose and Duka, 2008) has examined the acute effects of alcohol on inhibition of alcohol-related cues, using a
138
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
modified go/no-go task. This indicated no effects of alcohol (0.6 g/kg) on disinhibition towards alcohol-related cues. However, participants in this study also made no inhibition errors, suggesting ceiling effects on performance. To address this limitation we investigated the acute effects of alcohol on an alcohol-shifting task, designed to examine different components of inhibitory control (Noel et al., 2007). This task is capable of assessing (1) overall inhibition, (2) inhibition in reverse stimulus-reward associations (e.g., shifting attention to a different stimulus category) and (3) inhibition in the presence of alcohol-related cognitive biases. In the present study we were specifically interested in the third type of inhibition, i.e., the effects of alcohol on inhibition towards alcoholrelated stimuli. We examined the effects of two doses of alcohol on inhibitory control towards alcohol-related cues in light and heavy social drinkers. Two doses of alcohol (0.4 g/kg, 0.6 g/kg) were included to assess the extent to which inhibitory control towards alcoholrelated cues may be differentially sensitive to different priming doses. Additionally, light and heavy drinkers were included to examine the influence of individual differences in alcohol use on inhibition towards alcohol-related cues. These factors were important to the study design since dose effects and history of alcohol use have been suggested to moderate alcohol’s effects on inhibitory control and cognitive bias for alcohol-related cues (Field et al., 2010). Inhibitory control towards alcohol-related stimuli was measured using both lexical and pictorial cues, to directly compare the sensitivity of these cue-types and assess the extent to which these cues operate according to a common underlying mechanism. Cognitive bias studies have produced mixed findings according to cue type used (lexical vs. pictorial), suggesting pictorial images may serve as more ecologically valid cues for laboratory studies (Bruce and Jones, 2004; Townshend and Duka, 2001). We hypothesised, first, that participants primed with an acute dose of alcohol would demonstrate increased disinhibition towards alcohol-related cues, second, that heavier social drinkers compared to lighter drinkers would exhibit increased disinhibition towards alcohol-related cues, and, third, that the effects of acute alcohol and heaviness of drinking on disinhibition towards alcohol-related cues would differ according to cue type. 2. Methods 2.1. Design The study employed a double-blind between-subjects placebocontrolled design, comprising two between-subjects factors of challenge condition (0.0 g/kg alcohol, 0.4 g/kg alcohol, 0.6 g/kg alcohol) and drinking status (light drinkers, heavy drinkers).
2.3. Measures 2.3.1. Questionnaire measures. Questionnaire measures used included self-report measures of drinking behaviour, Alcohol Use Disorders Identification Test (AUDIT; Bohn et al., 1995), impulsivity, Barratt Impulsivity Scale (BIS-II; Patton et al., 1995), sensation seeking, Zuckerman–Kuhlman Personality Questionnaire Impulsive Sensation Seeking subscale (ImpSS; Zuckerman et al., 1993), mood, Profile of Mood States (POMS; McNair et al., 1992) and craving, Visual Analogue Scales (VAS). The AUDIT was completed once at baseline along with the ImpSS and BIS v11. The POMS and VAS measures were administered at baseline, 5 min post-challenge and 25 min post-challenge. 2.3.2. Alcohol-shifting tasks. The alcohol-shifting task is capable of assessing several types of inhibitory control. The task requires; selection of attention and response, shifting of attention and processing of emotionally relevant cues (i.e., alcohol-related stimuli). In the pictorial alcohol-shifting task (Noel et al., 2007), each trial began with a stimulus image, presented for 500 ms, followed by an inter-stimulus interval of 900 ms. Participants were instructed to respond as quickly and accurately as possible to target stimuli by pressing the spacebar, but to ignore distracter stimuli by pressing nothing. A 500 ms/900 Hz tone sounded for commission errors (pressing the spacebar following a distracter) but not for omission errors (failing to respond to a target). The task comprised 2 practice blocks, followed by 8 experimental blocks. Data from the practice blocks were not analysed. For the pictorial task stimuli consisted of 24 front-facing photographic images of bottled beverages (Adams et al., 2011). Images were 12 neutral (soft-drink images) and 12 alcohol-related (alcohol-beverage images), matched for colour, complexity, and content. Lexical stimuli were 20 alcoholrelated and 20 neutral words (Cox et al., 1999). For the pictorial task, each block contained 24 stimulus trials, consisting of 12 alcohol-related images and 12 neutral images. Alcohol-related and neutral images were randomised within each block. In each block of the task either alcohol-related (A) or neutral (N) images were identified as targets. Each block contained an equal number of target and distracter stimuli. Targets were presented in an order of either AANNAANN or NNAANNAA. Arrangement of the blocks in this order creates 4 shift and 4 non-shift blocks. In shift blocks, participants must respond to a different target than in the previous block, however in non-shift blocks, the target remains the same as the previous block. The lexical alcohol-shifting task was identical in design; however, each block contained 20 stimulus trials (10 alcohol-related and 10 neutral words). All tasks were programmed using E-Prime version 1.2 (Psychology Software Tools Inc., Pittsburgh, USA) and all stimuli were presented on a 15-in. monitor.
2.2. Participants
2.4. Procedure
Social drinkers were recruited from students and staff at the University of Bristol and members of the general public (n = 96; 50% male). Participants were assigned to drinking status groups based upon number of alcohol units consumed per/week, according to UK government limits for safe alcohol use. Participants were asked to report alcohol consumption on a typical drinking week. Light drinkers (n = 48; 50% male) were defined as individuals who consume ≥10 and ≤20 units of alcohol per week for males and ≥5 and ≤14 units of alcohol per week for females. Heavy drinkers (n = 48; 50% male) were defined as individuals who consume ≥21 and ≤50 units per week for males and ≥15 and ≤35 units per week for females. A unit is equivalent to 8 g of ethanol. Participants received £7 each for participation. The study was approved by the Faculty of Science Research Ethics Committee.
All participants were tested between noon and 6 pm in a laboratory in the School of Experimental Psychology. On the test day, after providing informed consent, all participants completed a screening process. Exclusion criteria included: current use of medication and illicit substances; family history of alcoholism; and alcohol consumed 12 h prior to the study. Recent alcohol consumption was assessed using a breath alcohol test. Alcohol consumption (units per week) was recorded to establish allocation to drinking status group (light drinkers, heavy drinkers) and weight was recorded for drink preparation. Following the completion of baseline measures (AUDIT, BIS-II, ImpSS, POMS and VAS), participants were given 5–10 min to consume the drink. Participants were randomly allocated to receive either vodka at 37.5% alcohol by volume, at a dose of 0.4 g/kg or 0.6 g/kg alcohol with tonic water (up to a maximum of
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
125 ml vodka), or a placebo consisting of only tonic water. All drinks were flavoured with lime and chilled. Following drink consumption, participants completed the first awareness check to determine whether they were aware if they had received alcohol or placebo. Next, participants were given 15 min to complete post-challenge measures (POMS and VAS). At the end of this time participants then completed the pictorial and lexical alcohol-shifting tasks, with task order counterbalanced across participants. On completion of the tasks, all participants completed a second awareness check and questionnaire measures of mood and craving (POMS and VAS). Participants were informed of their drink condition. At the end of the test day participants were paid and received a full debrief. 2.5. Data analysis For alcohol-shifting data, mixed-model ANOVAs of commission errors (failures to inhibit) with two within-subjects factors of distracter type (alcohol, neutral) and block (shift, non-shift) and two between-subjects factors of challenge condition (0.0 g/kg alcohol, 0.4 g/kg alcohol, 0.6 g/kg alcohol) and drinking status (light drinkers, heavy drinkers) were conducted. Additionally, mixedmodel ANOVAs of correct response RTs were conducted with a within-subject factor of target (alcohol, neutral) and block (shift, non-shift). All initial analyses also included between-subjects factors of task order (lexical first, pictorial first) and target order (alcohol first, neutral first). For all task data analyses factors of task order and target order did not interact significantly with other factors. Skewness tests for normality indicated that lexical and pictorial commission error and RT data were non-normal but could be corrected by a square root transformation. For craving and mood data mixed-model ANOVAs were conducted of mean VAS and POMS scores with a within-subjects factor of time (baseline, 5 min post-challenge, and 25 min post-challenge) and challenge condition and drinking status as between-subjects factors. 3. Results 3.1. Characteristics of participants Descriptive data were analysed using a series of univariate ANOVAs, with two between-subject factors of challenge condition and drinking status. Group interactions between challenge condition × drinking status did not significantly differ with respect to any demographic or baseline variables (ps > 0.07). Table 1 shows characteristics of light and heavy drinking participants by challenge condition. Exhaled breath alcohol level was 0.00 g/l at baseline for all participants and, at end of testing, 0.00 g/l in the 0.00 g/kg condition, 0.14 g/l in the 0.4 g/kg condition and 0.27 g/l in the 0.6 g/kg condition. At the end of testing breath alcohol level did not significantly differ between light and heavy drinkers (p > 0.99). For our 0.6 g/kg dose it has been shown that peak Blood Alcohol Concentration (BAC) occurs at 60 min post administration (Fillmore and Vogel-Sprott, 1998; Ostling and Fillmore, 2010) and all tasks were completed within 30 min. For our 0.4 g/kg dose it is estimated that that peak BAC occurs at 40 min post administration and all tasks were completed within 30 min. 3.2. Subjective alcohol craving A mixed-model ANOVA of VAS total craving scores indicated a marginal main effect of time (F [2,171] = 2.55, p = 0.08,
139
2 = 0.03) indicating an increase in craving scores from baseline (M = 16, SD = 8) to 5 min post-challenge (M = 17, SD = 9) and a decrease in craving scores from 5 min post-challenge to 25 min post-challenge (M = 16, SD = 9). All other interactions were nonsignificant (ps > 0.21). 3.3. Subjective mood A mixed-model ANOVA of POMS total scores indicated main effects of time for tension, vigour, fatigue, depression and anger (ps < 0.019). The main effect of fatigue was qualified by a significant time × challenge condition interaction (F [4,170] = 3.71, p = 0.007, 2 = 0.08). Further main effects of challenge condition and a challenge condition × drinking status interaction on depression were significant (ps < 0.024). All other main effects and interactions were non-significant (ps > 0.08). 3.4. Perceived alcohol content Awareness checks were performed immediately following drinking consumption (time 1) and at the end of testing (time 2). At time 1, 38% of participants in the 0.0 g/kg, 72% in the 0.4 g/kg and 88% in the 0.6 g/kg condition reported that their drink contained alcohol. At time 2 these figures were 44% in the 0.0 g/kg, 88% in the 0.4 g/kg and 94% in the 0.6 g/kg condition. A chi-square test indicated a relationship between perceived alcohol content and challenge condition at baseline (2 [2, n = 96] = 18.56, p < 0.001), and after testing, (2 [2, n = 96] = 25.33, p < 0.001). 3.5. Lexical alcohol-shifting task 3.5.1. Commission errors. A mixed-model ANOVA of transformed commission errors indicated a main effect of distracter type (F [1,90] = 43.25, p < 0.001, 2 = 0.33), with participants making more errors towards alcohol versus neutral distracters. A further main effect of block (F [1,90] = 59.37, p < 0.001, 2 = 0.40) was found, where participants made more errors in shift blocks compared to non-shift blocks. A main effect of challenge × drinking status on error rate was also significant (F [2,90] = 3.32, p = 0.040, 2 = 0.07). For the interaction between challenge condition × drinking status, simple effects analyses were conducted for challenge conditions separately. The difference between drinking status groups was only significant for participants in the 0.6 g/kg challenge condition group (F [1,30] = 4.23, p = 0.046, 2 = 0.13). Further post hoc tests indicated that heavier drinkers made more errors of commission than (M = 1.1, SD = 0.6) than lighter drinkers (M = 0.8, SD = 0.7) in the 0.6 g/kg challenge condition group. A significant interaction between distracter type × block (F [1,90] = 6.15, p = 0.015, 2 = 0.06) and a marginal interaction between distracter type × challenge condition (F [2,90] = 2.77, p = 0.068, 2 = 0.06) were observed. The distracter type × block interaction was qualified by a significant interaction between distracter type × block × drinking status (F [1,90] = 4.22, p = 0.043, 2 = 0.05). All other main effects and interactions were non-significant (ps > 0.11). Mean lexical and pictorial commission error rate data are presented in Table 2. For the marginal interaction of distracter × challenge condition, simple effects analyses were conducted for challenge conditions separately. For all challenge conditions the main effect of distracter was significant; 0.0 g/kg alcohol (F [1,30] = 4.73, p = 0.038, 2 = 0.14), 0.4 g/kg alcohol (F [1,30] = 11.78, p = 0.002, 2 = 0.28) and 0.6 g/kg alcohol (F [1,30] = 37.00, p < 0.001, 2 = 0.55). For the 0.0 g/kg condition, participants made more errors towards alcohol distracters (M = 0.9, SD = 0.6) compared to neutral distracters (M = 0.6, SD = 0.6). For the 0.4 g/kg condition, participants made more errors towards
140
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
Table 1 Participant demographic characteristics. Alcohol 0.4 g/kg (n = 32)
Alcohol 0.0 g/kg (n = 32) Light (n = 16) Age (years) Alcohol (units/week) Weight (kg) BIS-II (Total score) ImpSS (Total score) AUDIT (Total score) POMS Depression POMS Vigour POMS Tension POMS Fatigue POMS Confusion POMS Anger
20 ±2 12 ±6 67 ±11 69 ±8 9 ±3 11 ±4 17 ±4 21 ±4 15 ±3 12 ±3 10 ±4 14 ±3
Heavy (n = 16) 21 ±5 26 ±9 70 ±16 68 ±5 11 ±3 15 ±6 17 ±3 22 ±5 16 ±3 16 ±4 11 ±3 15 ±3
Light (n = 16) 22 ±4 12 ±4 68 ±8 72 ±6 12 ±4 12 ±4 23 ±9 20 ±6 16 ±4 16 ±8 13 ±4 16 ±5
Alcohol 0.6 g/kg (n = 32) Heavy (n = 16) 19 ±1 28 ±8 69 ±10 70 ±7 12 ±4 17 ±6 18 ±4 21 ±7 15 ±2 13 ±5 11 ±4 14 ±2
Light (n = 16) 22 ±4 12 ±5 72 ±6 70 ±6 11 ±3 12 ±5 18 ±4 23 ±7 15 ±4 13 ±5 10 ±2 15 ±3
Heavy (n = 16) 19 ±2 26 ±8 72 ±11 70 ±6 13 ±3 17 ±6 18 ±5 21 ±4 15 ±3 14 ±5 11 ±3 16 ±4
Participant demographic characteristics including age, alcohol consumption, weight, BIS-II, ImpSS, AUDIT and POMS (baseline) by drink challenge conditions and for light and heavy drinkers. Values are expressed as mean ± SD. Table 2 Mean number of commission errors for lexical and pictorial alcohol go/no-go data. Alcohol 0.4 g/kg (n = 32)
Alcohol 0.0 g/kg (n = 32)
Lexical Alcohol errors Neutral errors Pictorial Alcohol errors Neutral errors
Alcohol 0.6 g/kg (n = 32)
Light (n = 16)
Heavy (n = 16)
Light(n = 16)
Heavy (n = 16)
Light (n = 16)
Heavy (n = 16)
0.7 ±0.7 0.7 ±0.6
1 ±0.6 0.6 ±0.7
1 ±0.7 0.7 ±0.6
0.8 ±0.7 0.4 ±0.6
1 ±0.8 0.5 ±0.6
1 ±0.6 0.8 ±0.7
0.5 ±0.8 1.2 ±0.7
1 ±0.7 1.2 ±0.6
0.9 ±0.6 1.3 ±0.7
0.6 ±0.6 1 ±0.7
0.9 ±0.7 1.3 ±0.6
1 ±0.7 1.5 ±0.7
Mean number of commission errors for lexical and pictorial alcohol go/no-go data by drink challenge conditions and for light and heavy drinkers. Values are expressed as mean ± SD. Values refer to transformed data.
alcohol distracters (M = 0.9, SD = 0.7) compared to neutral distracters (M = 0.6, SD = 0.6). For the 0.6 g/kg condition, participants made more errors towards alcohol distracters (M = 1.2, SD = 0.7) compared to neutral distracters (M = 0.7, SD = 0.6). The difference between alcohol and neutral distracters was most pronounced in the 0.6 g/kg alcohol condition. These data are represented graphically in Fig. 1. 3.5.2. Reaction times. A mixed-model ANOVA of transformed RTs indicated a main effect of target (F [1,90] = 24.87, p < 0.001, 2 = 0.22), with all participants responding faster towards alcohol targets compared to neutral targets. A further main effect of block (F [1,90] = 30.03, p < 0.001, 2 = 0.25) was found, where participants were slower to respond in shift versus non-shift blocks. Significant interactions between target type × block (F [1,90] = 5.11, p = 0.026, 2 = 0.05) and block × drinking status × challenge condition (F [2,90] = 4.13, p = 0.019, 2 = 0.08) were also observed. A marginal interaction between target × block × drinking status × challenge
Fig. 1. Mean commission error rate towards alcohol and neutral distracter words by challenge condition groups (0.00 g/kg, 0.40 g/kg, 0.60 g/kg). Values are mean ± SE. Values refer to square root transformed data. Significant differences are marked with an asterisk. For all challenge conditions the main effect of distracter was significant; 0.0 g/kg alcohol (p = 0.038), 0.4 g/kg alcohol (p = 0.002) and 0.6 g/kg alcohol (p < 0.001).
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
condition was also observed (F [2,90] = 2.48, p = 0.089, 2 = 0.05). All other main effects and interactions were non-significant (ps > 0.13). These main effects and interactions were not analysed further. 3.6. Pictorial alcohol-shifting task 3.6.1. Commission errors. A mixed-model ANOVA of transformed commission errors indicated a main effect of distracter (F [1,90] = 67.40, p < 0.001, 2 = 0.43), with all participants making a greater number of errors towards neutral distracters compared to alcohol distracters. A further main effect of block (F [1,90] = 23.80, p < 0.001, 2 = 0.21) was found, where participants made more errors in shift versus non-shift blocks. A main effect of challenge × drinking status on error rate was significant (F [2,90] = 3.11, p = 0.049, 2 = 0.07). For the interaction between challenge condition × drinking status, simple effects analyses were conducted for challenge conditions separately. A significant difference between drinking status groups was only significant for participants in the 0.4 g/kg challenge condition group (F [1,30] = 4.25, p = 0.048, 2 = 0.12). Further post hoc tests indicated that light drinkers made more errors of commission than (M = 1.1, SD = 0.7) than heavy drinkers (M = 0.8, SD = 0.7) in the 0.4 g/kg challenge condition group. All other main effects and interactions were non-significant (ps > 0.10). 3.6.2. Reaction times. A mixed-model ANOVA of transformed RTs indicated a main effect of target (F [1,90] = 25.41, p < 0.001, 2 = 0.22), with all participants responding faster towards alcohol targets compared to neutral targets. A further main effect of block (F [1,90] = 54.33, p < 0.001, 2 = 0.38) was found, where participants were slower to respond in shift versus non-shift blocks. All other main effects and interactions were non-significant (ps > 0.13). These main effects and interactions were not analysed further. 4. Discussion Our data provide partial support for our first hypothesis, indicating a trend for acute alcohol consumption increasing disinhibition towards alcohol-related cues, but did not clearly support our second hypothesis that heavier drinkers would demonstrate increased disinhibition towards alcohol-related cues. Our findings did, however, suggest that the effects of alcohol on disinhibition were cue-specific and restricted to inhibition of responses towards lexical alcohol stimuli. Previous studies have indicated alcohol’s disinhibiting effects (de Wit et al., 2000; Easdon and Vogel-Sprott, 2000; Marczinski et al., 2005), and also alcohol’s capability to increase incentive value for alcohol cues (Adams et al., 2011; Duka and Townshend, 2004; Schoenmakers et al., 2008). However, our data are the first to suggest an impairing effect of alcohol on the ability to inhibit a response in the face of alcohol-related cues. Our data suggest that, following acute alcohol intoxication, social drinkers may exhibit decreased inhibitory control over responses towards alcohol-related stimuli. This result is consistent with previous evidence of selective processing towards affective stimuli on a task that involves selection of attention and inhibition of a response (Murphy et al., 1999; Noel et al., 2007). In contrast to previous evidence (Field et al., 2007), we did not observe any clear evidence for the influence of drinking status on inhibitory control towards alcohol-related cues. However, the study by Field and colleagues targeted a different component of impulsive behaviour (i.e., delay discounting) to the present study, suggesting that the influence of drinking status on inhibitory control may be selective to different mechanisms of impulsivity.
141
Studies of cognitive biases for alcohol-related cues have produced mixed findings according to cue type used (i.e., lexical versus pictorial). Consistent with our third hypothesis, the effects of alcohol on disinhibition towards alcohol-related cues differed according to cue type. Pictorial images are generally regarded as more ecologically valid cues for examining cognitive biases than lexical stimuli (Bruce and Jones, 2004; Townshend and Duka, 2001). However, we observed impairing effects of alcohol consumption on inhibition towards alcohol-related lexical stimuli only. Given the complex demands of the alcohol-shifting task, participants may have found lexical stimuli less ambiguous to interpret following alcohol intake, increasing the capability of these stimuli over pictorial cues to produce disinhibition under intoxication. Interestingly, we also observed cue-specific, main effects of distracter type (alcohol versus neutral) on disinhibition. For lexical cues, all participants show reduced inhibition towards alcohol-related cues, whilst for pictorial stimuli all participants showed reduced inhibition towards neutral cues. Together these cue-dependent findings may reflect differing mechanisms for response inhibition according to cue type. For example, lexical and visual encoding take place in different neural regions (Grady et al., 1998). We therefore tentatively suggest that lexical and pictorial cues may assay different cue-specific mechanisms of reward on inhibitory control. Alternatively, the novel content of our neutral pictorial stimuli (i.e., lesser known brands of soft drinks) may have lead to greater processing of these “unusual” cues, in turn producing greater disinhibition to neutral cues. Pictorial and lexical data indicated several further discrepant findings in overall response inhibition. Whilst previous evidence has indicated that binge drinkers and those who generally consume more alcohol may be more sensitive to the effects of acute alcohol consumption on response inhibition (Marczinski et al., 2007), our data indicate cue specific effects of drinking status on alcohol’s effects on response inhibition. For lexical cues, heavier drinkers receiving the highest dose of alcohol (0.6 g/kg) made the most commission errors. However, for pictorial cues lighter drinkers receiving a moderate dose of alcohol (0.4 g/kg) made the most commission errors. Whilst these data are broadly consistent with the notion that heavier drinkers are more hypersensitive to the disinhibiting effects of alcohol intoxication, they also suggest that this effect is restricted to inhibition towards lexical cues. Some other findings were also noteworthy. Both lexical and pictorial data reveal that all participants were more disinhibited in shift blocks. This finding may reflect increased difficultly in withholding a response whilst performing an attentional shift (i.e., switching to a different response target). For lexical data, participants also showed increased disinhibition towards alcohol-related distracters in shift blocks. This may suggest that disinhibition towards alcohol-related cues is more pronounced when an attentional shift to a different stimulus category is required. Consistent with previous work (Duka and Townshend, 2004), we do not observe any clear effects of alcohol on craving. We also observed faster RTs to alcohol versus neutral targets for lexical and pictorial cues, suggesting that all participants were quicker to approach alcohol-related cues. Given that our main hypotheses examined the acute effects of alcohol on inhibition, for all task data analyses baseline perceived alcohol content was initially added as a covariate, and then removed if it did not interact significantly with other factors. These data on perceived alcohol content indicate that alcohol expectancy did not influence the effects of acute alcohol consumption on response inhibition. There are some limitations to our study which should be considered when interpreting these results. First, we did not assess baseline performance on the alcohol-shifting tasks, which would have enabled comparisons with impulsivity and sensation-seeking
142
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
measurements before and after alcohol consumption. However, groups did not differ at baseline for impulsivity, sensation-seeking and mood. Second, response inhibition of salience attribution was assessed using a single measure. Given previous concerns with the measurement and complexity of both response inhibition and incentive salience, future studies may benefit from the use of two well-defined task paradigms for the assessment of these two distinct constructs. Our data indicate that, following an acute dose of alcohol, disinhibition is increased towards alcohol-related, lexical stimuli in social drinkers. These data provide initial support for alcoholinduced impairment of behavioural response to alcohol relevant stimuli. Additionally, our results suggest that all participants showed increased disinhibition towards lexical alcohol cues, but reduced disinhibition towards pictorial alcohol stimuli. It remains unclear whether this discrepancy in response inhibition towards alcohol stimuli reflects distinct cue-specific mechanisms of reward on inhibitory control. We also observed that under the influence of acute alcohol, heavier drinkers showed difficulty in shifting attention in the face of alcohol-related cues. This finding is consistent with previous evidence of enhanced biases towards alcohol-related cues in heavy drinkers compared with lighter social drinkers. Finally, we observed that acute alcohol produced overall deficits in inhibitory control by drinking status, where alcohol led to greater disinhibition to lexical cues in heavier drinkers. Together, our results indicate alcohol’s effects on different components of inhibitory control. Moreover, it appears that alcohol’s impairment of inhibitory control may be enhanced when a response towards alcohol-related stimuli is required to be withheld. Role of funding source Funding for this study was provided by a University of Bristol Postgraduate Scholarship, the University of Bristol had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors SA and MRM were responsible for the study design. SA was responsible for the collection and analysis of the data. SA drafted the manuscript. AFA, ASA, and MRM provided critical revision of the manuscript for important intellectual content. All authors critically reviewed content and approved final version for publication. Conflict of interest There are no conflicts of interest. References Adams, S., Ataya, A.F., Attwood, A.S., Munafò, M.R., 2011. Effects of acute alcohol consumption on alcohol-related cognitive biases in light and heavy drinkers are task-dependent. J. Psychopharmacol. 26, 245–253.
Bohn, M.J., Krahn, D.D., Staehler, B.A., 1995. Development and initial validation of a measure of drinking urges in abstinent alcoholics. Alcohol Clin. Exp. Res. 19, 600–606. Bruce, G., Jones, B.T., 2004. A pictorial Stroop paradigm reveals an alcohol attentional bias in heavier compared to lighter social drinkers. J. Psychopharmacol. 18, 527–533. Cox, W.M., Yeates, G.N., Regan, C.M., 1999. Effects of alcohol cues on cognitive processing in heavy and light drinkers. Drug Alcohol Depend. 55, 85–89. de Wit, 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. Duka, T., Townshend, J.M., 2004. The priming effect of alcohol pre-load on attentional bias to alcohol-related stimuli. Psychopharmacology (Berl) 176, 353–356. Easdon, C.M., Vogel-Sprott, M., 2000. Alcohol and behavioral control: impaired response inhibition and flexibility in social drinkers. Exp. Clin. Psychopharmacol. 8, 387–394. Farris, S.R., Ostafin, B.D., 2008. Alcohol consumption primes automatic alcoholapproach associations. Am. J. Drug Alcohol Abuse 34, 703–711. Field, M., Christiansen, P., Cole, J., Goudie, A., 2007. Delay discounting and the alcohol Stroop in heavy drinking adolescents. Addiction 102, 579–586. Field, M., Wiers, R.W., Christiansen, P., Fillmore, M.T., Verster, J.C., 2010. Acute alcohol effects on inhibitory control and implicit cognition: implications for loss of control over drinking. Alcohol Clin. Exp. Res. 34, 1346–1352. Fillmore, M.T., 2007. Acute alcohol-induced impairment of cognitive functions: past and present findings. Int. J. Disabil. Hum. Dev. 6, 115–125. Fillmore, M.T., Vogel-Sprott, M., 1998. Behavioral impairment under alcohol: cognitive and pharmacokinetic factors. Alcohol Clin. Exp. Res. 22, 1476–1482. Goldstein, R.Z., Volkow, N.D., 2002. Drug addiction and its underlying neurobiological basis: neuroimaging evidence for the involvement of the frontal cortex. Am. J. Psychiatry 159, 1642–1652. Grady, C.L., McIntosh, A.R., Rajah, M.N., Craik, F.I.M., 1998. Neural correlates of the episodic encoding of pictures and words. Proc. Natl. Acad. Sci. U.S.A. 95, 2703–2708. Hogarth, L., Attwood, A.S., Bate, H.A., Munafo, M.R., 2012. Acute alcohol impairs human goal-directed action. Biol. Psychol. 90, 154–160. Jentsch, J.D., Taylor, J.R., 1999. Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli. Psychopharmacology (Berl) 146, 373–390. Marczinski, C.A., Abroms, B.D., Van Selst, M., Fillmore, M.T., 2005. Alcohol-induced impairment of behavioral control: differential effects on engaging vs. disengaging responses. Psychopharmacology (Berl) 182, 452–459. Marczinski, C.A., Combs, S.W., Fillmore, M.T., 2007. Increased sensitivity to the disinhibiting effects of alcohol in binge drinkers. Psychol. Addict. Behav. 21, 346–354. McNair, D.M., Lorr, M., Droppleman, L.F., 1992. Profile of Mood States (POMS). Educational and Industrial Testing Service, San Diego, CA. Murphy, F.C., Sahakian, B.J., Rubinsztein, J.S., Michael, A., Rogers, R.D., Robbins, T.W., Paykel, E.S., 1999. Emotional bias and inhibitory control processes in mania and depression. Psychol. Med. 29, 1307–1321. Noel, X., Van der Linden, M., d’Acremont, M., Bechara, A., Dan, B., Hanak, C., Verbanck, P., 2007. Alcohol cues increase cognitive impulsivity in individuals with alcoholism. Psychopharmacology (Berl) 192, 291–298. Ostling, E.W., Fillmore, M.T., 2010. Tolerance to the impairing effects of alcohol on the inhibition and activation of behavior. Psychopharmacology (Berl) 212, 465–473. Palfai, T.P., Ostafin, B.D., 2003. The influence of alcohol on the activation of outcome expectancies: the role of evaluative expectancy activation in drinking behavior. J. Stud. Alcohol 64, 111–119. Patton, J.H., Stanford, M.S., Barratt, E.S., 1995. Factor structure of the Barratt Impulsiveness Scale. J. Clin. Psychol. 51, 768–774. Rose, A.K., Duka, T., 2008. Effects of alcohol on inhibitory processes. Behav. Pharmacol. 19, 284–291. Schoenmakers, T., Wiers, R.W., Field, M., 2008. Effects of a low dose of alcohol on cognitive biases and craving in heavy drinkers. Psychopharmacology (Berl) 197, 169–178. Townshend, J.M., Duka, T., 2001. Attentional bias associated with alcohol cues: differences between heavy and occasional social drinkers. Psychopharmacology (Berl) 157, 67–74. Zuckerman, M., Kuhlman, D.M., Joireman, J., Teta, P., Kraft, M., 1993. A comparison of 3 structural models for personality – the big 3, the big 5, and the alternative 5. J. Pers. Soc. Psychol. 65, 757–768.
Contents lists available at SciVerse ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Effects of alcohol on disinhibition towards alcohol-related cues Sally Adams ∗ , Alia F. Ataya, Angela S. Attwood, Marcus R. Munafò School of Experimental Psychology, University of Bristol, 12a Priory Road, Bristol BS8 1TU, UK
a r t i c l e
i n f o
Article history: Received 30 March 2012 Received in revised form 21 June 2012 Accepted 23 June 2012 Available online 26 July 2012 Keywords: Alcohol Inhibitory control Alcohol-shifting task
a b s t r a c t Background: We investigated (1) the effects of acute alcohol on inhibition of alcohol-related versus neutral cues, (2) the effects of drinking status on inhibition of alcohol-related versus neutral cues, and (3) the similarity of any effects of alcohol or drinking status across two different cue types (lexical versus pictorial). Methods: Participants received 0.0 g/kg, 0.4 g/kg or 0.6 g/kg of alcohol in a between-subjects design. Healthy, heavy and light social alcohol users (n = 96) completed both lexical and pictorial cue versions of an alcohol-shifting task. Participants were instructed to respond to target stimuli by pressing the spacebar, but to ignore distracter stimuli. Errors towards distracter stimuli were analysed using a series of mixedmodel ANOVAs, with between-subjects factors of challenge and drinking status and within-subjects factors of distracter type (alcohol, neutral) and block (shift, non-shift). Results: Lexical commission error data indicated a main effect of distracter (F [1,90] = 43.25, p < 0.001, 2 = 0.33), which was qualified by a marginal interaction with challenge condition (F [2,90] = 2.77, p = 0.068, 2 = 0.06). Following an acute high dose of alcohol participants made more errors towards alcohol distracters. Pictorial commission error data indicated a significant main effect of distracter (F [1,90] = 67.40, p < 0.001, 2 = 0.43), such that all participants made more errors towards neutral image distracters versus alcohol distracter images. Conclusions: Our results reveal acute alcohol’s impairment of inhibitory control may be enhanced when a response towards alcohol-related lexical stimuli is required to be withheld. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Alcohol’s priming effects have been attributed to alcoholinduced impairment of cognitive mechanisms, including inhibitory control and response towards alcohol-related cues (Field et al., 2010). Alcohol is widely regarded to diminish inhibitory control of behaviour (Fillmore, 2007) that is, the ability to suppress a prepotent response and/or behavioural impulse and more recently goal-directed control (Hogarth et al., 2012). The stop-signal and go/no-go tasks have been developed to assess impairment of inhibitory control, and these indicate that alcohol specifically disrupts response inhibition at doses (0.4–0.6 g/kg) that are not associated with impairment of activational processes (de Wit et al., 2000; Easdon and Vogel-Sprott, 2000; Marczinski et al., 2005). These data are consistent with disinhibitory models of drug use, which suggest that alcohol intoxication may lead to a loss of intentional control over alcohol-seeking behaviours. Alcohol intoxication is also thought to increase the incentive value of alcohol-related stimuli. Following intoxication, alcohol cues gain strong motivational properties, which may be viewed as
∗ Corresponding author. Tel.: +44 117 9288547; fax: +44 117 9288588. E-mail address: [email protected] (S. Adams). 0376-8716/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.drugalcdep.2012.06.025
“hijacking” and influencing automatic alcohol cognitions. For example, studies have reliably indicated that moderate doses of alcohol (0.3–0.4 g/kg) enhance cognitive bias towards alcoholrelated cues. Moreover, acute alcohol intake may enhance positive automatic associations with alcohol-related cues (Palfai and Ostafin, 2003) and strengthen associations between alcohol and approach behaviours (Farris and Ostafin, 2008). These findings provide support for two distinct factors involved in acute alcohol consumption: inhibitory control and cognitive biases for alcohol-related cues; however, they do not indicate whether these mechanisms may interact under alcohol. Several models of addiction (Field et al., 2010; Goldstein and Volkow, 2002; Jentsch and Taylor, 1999) suggest that acute intoxication may simultaneously reduce response inhibition whilst increasing automatised salience attribution for drug-cues. It is proposed that top-down inhibitory processes are lowered, releasing automatic appetitive behaviours that would typically be under close regulation. In this state, response inhibition is reduced and response to salient alcohol-related cues is accentuated (Field et al., 2010). Two studies have examined inhibitory control and cognitive bias for alcohol-related cues in heavier drinking adolescents (Field et al., 2007) and alcohol-dependent individuals (Noel et al., 2007). However, only one study (Rose and Duka, 2008) has examined the acute effects of alcohol on inhibition of alcohol-related cues, using a
138
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
modified go/no-go task. This indicated no effects of alcohol (0.6 g/kg) on disinhibition towards alcohol-related cues. However, participants in this study also made no inhibition errors, suggesting ceiling effects on performance. To address this limitation we investigated the acute effects of alcohol on an alcohol-shifting task, designed to examine different components of inhibitory control (Noel et al., 2007). This task is capable of assessing (1) overall inhibition, (2) inhibition in reverse stimulus-reward associations (e.g., shifting attention to a different stimulus category) and (3) inhibition in the presence of alcohol-related cognitive biases. In the present study we were specifically interested in the third type of inhibition, i.e., the effects of alcohol on inhibition towards alcoholrelated stimuli. We examined the effects of two doses of alcohol on inhibitory control towards alcohol-related cues in light and heavy social drinkers. Two doses of alcohol (0.4 g/kg, 0.6 g/kg) were included to assess the extent to which inhibitory control towards alcoholrelated cues may be differentially sensitive to different priming doses. Additionally, light and heavy drinkers were included to examine the influence of individual differences in alcohol use on inhibition towards alcohol-related cues. These factors were important to the study design since dose effects and history of alcohol use have been suggested to moderate alcohol’s effects on inhibitory control and cognitive bias for alcohol-related cues (Field et al., 2010). Inhibitory control towards alcohol-related stimuli was measured using both lexical and pictorial cues, to directly compare the sensitivity of these cue-types and assess the extent to which these cues operate according to a common underlying mechanism. Cognitive bias studies have produced mixed findings according to cue type used (lexical vs. pictorial), suggesting pictorial images may serve as more ecologically valid cues for laboratory studies (Bruce and Jones, 2004; Townshend and Duka, 2001). We hypothesised, first, that participants primed with an acute dose of alcohol would demonstrate increased disinhibition towards alcohol-related cues, second, that heavier social drinkers compared to lighter drinkers would exhibit increased disinhibition towards alcohol-related cues, and, third, that the effects of acute alcohol and heaviness of drinking on disinhibition towards alcohol-related cues would differ according to cue type. 2. Methods 2.1. Design The study employed a double-blind between-subjects placebocontrolled design, comprising two between-subjects factors of challenge condition (0.0 g/kg alcohol, 0.4 g/kg alcohol, 0.6 g/kg alcohol) and drinking status (light drinkers, heavy drinkers).
2.3. Measures 2.3.1. Questionnaire measures. Questionnaire measures used included self-report measures of drinking behaviour, Alcohol Use Disorders Identification Test (AUDIT; Bohn et al., 1995), impulsivity, Barratt Impulsivity Scale (BIS-II; Patton et al., 1995), sensation seeking, Zuckerman–Kuhlman Personality Questionnaire Impulsive Sensation Seeking subscale (ImpSS; Zuckerman et al., 1993), mood, Profile of Mood States (POMS; McNair et al., 1992) and craving, Visual Analogue Scales (VAS). The AUDIT was completed once at baseline along with the ImpSS and BIS v11. The POMS and VAS measures were administered at baseline, 5 min post-challenge and 25 min post-challenge. 2.3.2. Alcohol-shifting tasks. The alcohol-shifting task is capable of assessing several types of inhibitory control. The task requires; selection of attention and response, shifting of attention and processing of emotionally relevant cues (i.e., alcohol-related stimuli). In the pictorial alcohol-shifting task (Noel et al., 2007), each trial began with a stimulus image, presented for 500 ms, followed by an inter-stimulus interval of 900 ms. Participants were instructed to respond as quickly and accurately as possible to target stimuli by pressing the spacebar, but to ignore distracter stimuli by pressing nothing. A 500 ms/900 Hz tone sounded for commission errors (pressing the spacebar following a distracter) but not for omission errors (failing to respond to a target). The task comprised 2 practice blocks, followed by 8 experimental blocks. Data from the practice blocks were not analysed. For the pictorial task stimuli consisted of 24 front-facing photographic images of bottled beverages (Adams et al., 2011). Images were 12 neutral (soft-drink images) and 12 alcohol-related (alcohol-beverage images), matched for colour, complexity, and content. Lexical stimuli were 20 alcoholrelated and 20 neutral words (Cox et al., 1999). For the pictorial task, each block contained 24 stimulus trials, consisting of 12 alcohol-related images and 12 neutral images. Alcohol-related and neutral images were randomised within each block. In each block of the task either alcohol-related (A) or neutral (N) images were identified as targets. Each block contained an equal number of target and distracter stimuli. Targets were presented in an order of either AANNAANN or NNAANNAA. Arrangement of the blocks in this order creates 4 shift and 4 non-shift blocks. In shift blocks, participants must respond to a different target than in the previous block, however in non-shift blocks, the target remains the same as the previous block. The lexical alcohol-shifting task was identical in design; however, each block contained 20 stimulus trials (10 alcohol-related and 10 neutral words). All tasks were programmed using E-Prime version 1.2 (Psychology Software Tools Inc., Pittsburgh, USA) and all stimuli were presented on a 15-in. monitor.
2.2. Participants
2.4. Procedure
Social drinkers were recruited from students and staff at the University of Bristol and members of the general public (n = 96; 50% male). Participants were assigned to drinking status groups based upon number of alcohol units consumed per/week, according to UK government limits for safe alcohol use. Participants were asked to report alcohol consumption on a typical drinking week. Light drinkers (n = 48; 50% male) were defined as individuals who consume ≥10 and ≤20 units of alcohol per week for males and ≥5 and ≤14 units of alcohol per week for females. Heavy drinkers (n = 48; 50% male) were defined as individuals who consume ≥21 and ≤50 units per week for males and ≥15 and ≤35 units per week for females. A unit is equivalent to 8 g of ethanol. Participants received £7 each for participation. The study was approved by the Faculty of Science Research Ethics Committee.
All participants were tested between noon and 6 pm in a laboratory in the School of Experimental Psychology. On the test day, after providing informed consent, all participants completed a screening process. Exclusion criteria included: current use of medication and illicit substances; family history of alcoholism; and alcohol consumed 12 h prior to the study. Recent alcohol consumption was assessed using a breath alcohol test. Alcohol consumption (units per week) was recorded to establish allocation to drinking status group (light drinkers, heavy drinkers) and weight was recorded for drink preparation. Following the completion of baseline measures (AUDIT, BIS-II, ImpSS, POMS and VAS), participants were given 5–10 min to consume the drink. Participants were randomly allocated to receive either vodka at 37.5% alcohol by volume, at a dose of 0.4 g/kg or 0.6 g/kg alcohol with tonic water (up to a maximum of
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
125 ml vodka), or a placebo consisting of only tonic water. All drinks were flavoured with lime and chilled. Following drink consumption, participants completed the first awareness check to determine whether they were aware if they had received alcohol or placebo. Next, participants were given 15 min to complete post-challenge measures (POMS and VAS). At the end of this time participants then completed the pictorial and lexical alcohol-shifting tasks, with task order counterbalanced across participants. On completion of the tasks, all participants completed a second awareness check and questionnaire measures of mood and craving (POMS and VAS). Participants were informed of their drink condition. At the end of the test day participants were paid and received a full debrief. 2.5. Data analysis For alcohol-shifting data, mixed-model ANOVAs of commission errors (failures to inhibit) with two within-subjects factors of distracter type (alcohol, neutral) and block (shift, non-shift) and two between-subjects factors of challenge condition (0.0 g/kg alcohol, 0.4 g/kg alcohol, 0.6 g/kg alcohol) and drinking status (light drinkers, heavy drinkers) were conducted. Additionally, mixedmodel ANOVAs of correct response RTs were conducted with a within-subject factor of target (alcohol, neutral) and block (shift, non-shift). All initial analyses also included between-subjects factors of task order (lexical first, pictorial first) and target order (alcohol first, neutral first). For all task data analyses factors of task order and target order did not interact significantly with other factors. Skewness tests for normality indicated that lexical and pictorial commission error and RT data were non-normal but could be corrected by a square root transformation. For craving and mood data mixed-model ANOVAs were conducted of mean VAS and POMS scores with a within-subjects factor of time (baseline, 5 min post-challenge, and 25 min post-challenge) and challenge condition and drinking status as between-subjects factors. 3. Results 3.1. Characteristics of participants Descriptive data were analysed using a series of univariate ANOVAs, with two between-subject factors of challenge condition and drinking status. Group interactions between challenge condition × drinking status did not significantly differ with respect to any demographic or baseline variables (ps > 0.07). Table 1 shows characteristics of light and heavy drinking participants by challenge condition. Exhaled breath alcohol level was 0.00 g/l at baseline for all participants and, at end of testing, 0.00 g/l in the 0.00 g/kg condition, 0.14 g/l in the 0.4 g/kg condition and 0.27 g/l in the 0.6 g/kg condition. At the end of testing breath alcohol level did not significantly differ between light and heavy drinkers (p > 0.99). For our 0.6 g/kg dose it has been shown that peak Blood Alcohol Concentration (BAC) occurs at 60 min post administration (Fillmore and Vogel-Sprott, 1998; Ostling and Fillmore, 2010) and all tasks were completed within 30 min. For our 0.4 g/kg dose it is estimated that that peak BAC occurs at 40 min post administration and all tasks were completed within 30 min. 3.2. Subjective alcohol craving A mixed-model ANOVA of VAS total craving scores indicated a marginal main effect of time (F [2,171] = 2.55, p = 0.08,
139
2 = 0.03) indicating an increase in craving scores from baseline (M = 16, SD = 8) to 5 min post-challenge (M = 17, SD = 9) and a decrease in craving scores from 5 min post-challenge to 25 min post-challenge (M = 16, SD = 9). All other interactions were nonsignificant (ps > 0.21). 3.3. Subjective mood A mixed-model ANOVA of POMS total scores indicated main effects of time for tension, vigour, fatigue, depression and anger (ps < 0.019). The main effect of fatigue was qualified by a significant time × challenge condition interaction (F [4,170] = 3.71, p = 0.007, 2 = 0.08). Further main effects of challenge condition and a challenge condition × drinking status interaction on depression were significant (ps < 0.024). All other main effects and interactions were non-significant (ps > 0.08). 3.4. Perceived alcohol content Awareness checks were performed immediately following drinking consumption (time 1) and at the end of testing (time 2). At time 1, 38% of participants in the 0.0 g/kg, 72% in the 0.4 g/kg and 88% in the 0.6 g/kg condition reported that their drink contained alcohol. At time 2 these figures were 44% in the 0.0 g/kg, 88% in the 0.4 g/kg and 94% in the 0.6 g/kg condition. A chi-square test indicated a relationship between perceived alcohol content and challenge condition at baseline (2 [2, n = 96] = 18.56, p < 0.001), and after testing, (2 [2, n = 96] = 25.33, p < 0.001). 3.5. Lexical alcohol-shifting task 3.5.1. Commission errors. A mixed-model ANOVA of transformed commission errors indicated a main effect of distracter type (F [1,90] = 43.25, p < 0.001, 2 = 0.33), with participants making more errors towards alcohol versus neutral distracters. A further main effect of block (F [1,90] = 59.37, p < 0.001, 2 = 0.40) was found, where participants made more errors in shift blocks compared to non-shift blocks. A main effect of challenge × drinking status on error rate was also significant (F [2,90] = 3.32, p = 0.040, 2 = 0.07). For the interaction between challenge condition × drinking status, simple effects analyses were conducted for challenge conditions separately. The difference between drinking status groups was only significant for participants in the 0.6 g/kg challenge condition group (F [1,30] = 4.23, p = 0.046, 2 = 0.13). Further post hoc tests indicated that heavier drinkers made more errors of commission than (M = 1.1, SD = 0.6) than lighter drinkers (M = 0.8, SD = 0.7) in the 0.6 g/kg challenge condition group. A significant interaction between distracter type × block (F [1,90] = 6.15, p = 0.015, 2 = 0.06) and a marginal interaction between distracter type × challenge condition (F [2,90] = 2.77, p = 0.068, 2 = 0.06) were observed. The distracter type × block interaction was qualified by a significant interaction between distracter type × block × drinking status (F [1,90] = 4.22, p = 0.043, 2 = 0.05). All other main effects and interactions were non-significant (ps > 0.11). Mean lexical and pictorial commission error rate data are presented in Table 2. For the marginal interaction of distracter × challenge condition, simple effects analyses were conducted for challenge conditions separately. For all challenge conditions the main effect of distracter was significant; 0.0 g/kg alcohol (F [1,30] = 4.73, p = 0.038, 2 = 0.14), 0.4 g/kg alcohol (F [1,30] = 11.78, p = 0.002, 2 = 0.28) and 0.6 g/kg alcohol (F [1,30] = 37.00, p < 0.001, 2 = 0.55). For the 0.0 g/kg condition, participants made more errors towards alcohol distracters (M = 0.9, SD = 0.6) compared to neutral distracters (M = 0.6, SD = 0.6). For the 0.4 g/kg condition, participants made more errors towards
140
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
Table 1 Participant demographic characteristics. Alcohol 0.4 g/kg (n = 32)
Alcohol 0.0 g/kg (n = 32) Light (n = 16) Age (years) Alcohol (units/week) Weight (kg) BIS-II (Total score) ImpSS (Total score) AUDIT (Total score) POMS Depression POMS Vigour POMS Tension POMS Fatigue POMS Confusion POMS Anger
20 ±2 12 ±6 67 ±11 69 ±8 9 ±3 11 ±4 17 ±4 21 ±4 15 ±3 12 ±3 10 ±4 14 ±3
Heavy (n = 16) 21 ±5 26 ±9 70 ±16 68 ±5 11 ±3 15 ±6 17 ±3 22 ±5 16 ±3 16 ±4 11 ±3 15 ±3
Light (n = 16) 22 ±4 12 ±4 68 ±8 72 ±6 12 ±4 12 ±4 23 ±9 20 ±6 16 ±4 16 ±8 13 ±4 16 ±5
Alcohol 0.6 g/kg (n = 32) Heavy (n = 16) 19 ±1 28 ±8 69 ±10 70 ±7 12 ±4 17 ±6 18 ±4 21 ±7 15 ±2 13 ±5 11 ±4 14 ±2
Light (n = 16) 22 ±4 12 ±5 72 ±6 70 ±6 11 ±3 12 ±5 18 ±4 23 ±7 15 ±4 13 ±5 10 ±2 15 ±3
Heavy (n = 16) 19 ±2 26 ±8 72 ±11 70 ±6 13 ±3 17 ±6 18 ±5 21 ±4 15 ±3 14 ±5 11 ±3 16 ±4
Participant demographic characteristics including age, alcohol consumption, weight, BIS-II, ImpSS, AUDIT and POMS (baseline) by drink challenge conditions and for light and heavy drinkers. Values are expressed as mean ± SD. Table 2 Mean number of commission errors for lexical and pictorial alcohol go/no-go data. Alcohol 0.4 g/kg (n = 32)
Alcohol 0.0 g/kg (n = 32)
Lexical Alcohol errors Neutral errors Pictorial Alcohol errors Neutral errors
Alcohol 0.6 g/kg (n = 32)
Light (n = 16)
Heavy (n = 16)
Light(n = 16)
Heavy (n = 16)
Light (n = 16)
Heavy (n = 16)
0.7 ±0.7 0.7 ±0.6
1 ±0.6 0.6 ±0.7
1 ±0.7 0.7 ±0.6
0.8 ±0.7 0.4 ±0.6
1 ±0.8 0.5 ±0.6
1 ±0.6 0.8 ±0.7
0.5 ±0.8 1.2 ±0.7
1 ±0.7 1.2 ±0.6
0.9 ±0.6 1.3 ±0.7
0.6 ±0.6 1 ±0.7
0.9 ±0.7 1.3 ±0.6
1 ±0.7 1.5 ±0.7
Mean number of commission errors for lexical and pictorial alcohol go/no-go data by drink challenge conditions and for light and heavy drinkers. Values are expressed as mean ± SD. Values refer to transformed data.
alcohol distracters (M = 0.9, SD = 0.7) compared to neutral distracters (M = 0.6, SD = 0.6). For the 0.6 g/kg condition, participants made more errors towards alcohol distracters (M = 1.2, SD = 0.7) compared to neutral distracters (M = 0.7, SD = 0.6). The difference between alcohol and neutral distracters was most pronounced in the 0.6 g/kg alcohol condition. These data are represented graphically in Fig. 1. 3.5.2. Reaction times. A mixed-model ANOVA of transformed RTs indicated a main effect of target (F [1,90] = 24.87, p < 0.001, 2 = 0.22), with all participants responding faster towards alcohol targets compared to neutral targets. A further main effect of block (F [1,90] = 30.03, p < 0.001, 2 = 0.25) was found, where participants were slower to respond in shift versus non-shift blocks. Significant interactions between target type × block (F [1,90] = 5.11, p = 0.026, 2 = 0.05) and block × drinking status × challenge condition (F [2,90] = 4.13, p = 0.019, 2 = 0.08) were also observed. A marginal interaction between target × block × drinking status × challenge
Fig. 1. Mean commission error rate towards alcohol and neutral distracter words by challenge condition groups (0.00 g/kg, 0.40 g/kg, 0.60 g/kg). Values are mean ± SE. Values refer to square root transformed data. Significant differences are marked with an asterisk. For all challenge conditions the main effect of distracter was significant; 0.0 g/kg alcohol (p = 0.038), 0.4 g/kg alcohol (p = 0.002) and 0.6 g/kg alcohol (p < 0.001).
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
condition was also observed (F [2,90] = 2.48, p = 0.089, 2 = 0.05). All other main effects and interactions were non-significant (ps > 0.13). These main effects and interactions were not analysed further. 3.6. Pictorial alcohol-shifting task 3.6.1. Commission errors. A mixed-model ANOVA of transformed commission errors indicated a main effect of distracter (F [1,90] = 67.40, p < 0.001, 2 = 0.43), with all participants making a greater number of errors towards neutral distracters compared to alcohol distracters. A further main effect of block (F [1,90] = 23.80, p < 0.001, 2 = 0.21) was found, where participants made more errors in shift versus non-shift blocks. A main effect of challenge × drinking status on error rate was significant (F [2,90] = 3.11, p = 0.049, 2 = 0.07). For the interaction between challenge condition × drinking status, simple effects analyses were conducted for challenge conditions separately. A significant difference between drinking status groups was only significant for participants in the 0.4 g/kg challenge condition group (F [1,30] = 4.25, p = 0.048, 2 = 0.12). Further post hoc tests indicated that light drinkers made more errors of commission than (M = 1.1, SD = 0.7) than heavy drinkers (M = 0.8, SD = 0.7) in the 0.4 g/kg challenge condition group. All other main effects and interactions were non-significant (ps > 0.10). 3.6.2. Reaction times. A mixed-model ANOVA of transformed RTs indicated a main effect of target (F [1,90] = 25.41, p < 0.001, 2 = 0.22), with all participants responding faster towards alcohol targets compared to neutral targets. A further main effect of block (F [1,90] = 54.33, p < 0.001, 2 = 0.38) was found, where participants were slower to respond in shift versus non-shift blocks. All other main effects and interactions were non-significant (ps > 0.13). These main effects and interactions were not analysed further. 4. Discussion Our data provide partial support for our first hypothesis, indicating a trend for acute alcohol consumption increasing disinhibition towards alcohol-related cues, but did not clearly support our second hypothesis that heavier drinkers would demonstrate increased disinhibition towards alcohol-related cues. Our findings did, however, suggest that the effects of alcohol on disinhibition were cue-specific and restricted to inhibition of responses towards lexical alcohol stimuli. Previous studies have indicated alcohol’s disinhibiting effects (de Wit et al., 2000; Easdon and Vogel-Sprott, 2000; Marczinski et al., 2005), and also alcohol’s capability to increase incentive value for alcohol cues (Adams et al., 2011; Duka and Townshend, 2004; Schoenmakers et al., 2008). However, our data are the first to suggest an impairing effect of alcohol on the ability to inhibit a response in the face of alcohol-related cues. Our data suggest that, following acute alcohol intoxication, social drinkers may exhibit decreased inhibitory control over responses towards alcohol-related stimuli. This result is consistent with previous evidence of selective processing towards affective stimuli on a task that involves selection of attention and inhibition of a response (Murphy et al., 1999; Noel et al., 2007). In contrast to previous evidence (Field et al., 2007), we did not observe any clear evidence for the influence of drinking status on inhibitory control towards alcohol-related cues. However, the study by Field and colleagues targeted a different component of impulsive behaviour (i.e., delay discounting) to the present study, suggesting that the influence of drinking status on inhibitory control may be selective to different mechanisms of impulsivity.
141
Studies of cognitive biases for alcohol-related cues have produced mixed findings according to cue type used (i.e., lexical versus pictorial). Consistent with our third hypothesis, the effects of alcohol on disinhibition towards alcohol-related cues differed according to cue type. Pictorial images are generally regarded as more ecologically valid cues for examining cognitive biases than lexical stimuli (Bruce and Jones, 2004; Townshend and Duka, 2001). However, we observed impairing effects of alcohol consumption on inhibition towards alcohol-related lexical stimuli only. Given the complex demands of the alcohol-shifting task, participants may have found lexical stimuli less ambiguous to interpret following alcohol intake, increasing the capability of these stimuli over pictorial cues to produce disinhibition under intoxication. Interestingly, we also observed cue-specific, main effects of distracter type (alcohol versus neutral) on disinhibition. For lexical cues, all participants show reduced inhibition towards alcohol-related cues, whilst for pictorial stimuli all participants showed reduced inhibition towards neutral cues. Together these cue-dependent findings may reflect differing mechanisms for response inhibition according to cue type. For example, lexical and visual encoding take place in different neural regions (Grady et al., 1998). We therefore tentatively suggest that lexical and pictorial cues may assay different cue-specific mechanisms of reward on inhibitory control. Alternatively, the novel content of our neutral pictorial stimuli (i.e., lesser known brands of soft drinks) may have lead to greater processing of these “unusual” cues, in turn producing greater disinhibition to neutral cues. Pictorial and lexical data indicated several further discrepant findings in overall response inhibition. Whilst previous evidence has indicated that binge drinkers and those who generally consume more alcohol may be more sensitive to the effects of acute alcohol consumption on response inhibition (Marczinski et al., 2007), our data indicate cue specific effects of drinking status on alcohol’s effects on response inhibition. For lexical cues, heavier drinkers receiving the highest dose of alcohol (0.6 g/kg) made the most commission errors. However, for pictorial cues lighter drinkers receiving a moderate dose of alcohol (0.4 g/kg) made the most commission errors. Whilst these data are broadly consistent with the notion that heavier drinkers are more hypersensitive to the disinhibiting effects of alcohol intoxication, they also suggest that this effect is restricted to inhibition towards lexical cues. Some other findings were also noteworthy. Both lexical and pictorial data reveal that all participants were more disinhibited in shift blocks. This finding may reflect increased difficultly in withholding a response whilst performing an attentional shift (i.e., switching to a different response target). For lexical data, participants also showed increased disinhibition towards alcohol-related distracters in shift blocks. This may suggest that disinhibition towards alcohol-related cues is more pronounced when an attentional shift to a different stimulus category is required. Consistent with previous work (Duka and Townshend, 2004), we do not observe any clear effects of alcohol on craving. We also observed faster RTs to alcohol versus neutral targets for lexical and pictorial cues, suggesting that all participants were quicker to approach alcohol-related cues. Given that our main hypotheses examined the acute effects of alcohol on inhibition, for all task data analyses baseline perceived alcohol content was initially added as a covariate, and then removed if it did not interact significantly with other factors. These data on perceived alcohol content indicate that alcohol expectancy did not influence the effects of acute alcohol consumption on response inhibition. There are some limitations to our study which should be considered when interpreting these results. First, we did not assess baseline performance on the alcohol-shifting tasks, which would have enabled comparisons with impulsivity and sensation-seeking
142
S. Adams et al. / Drug and Alcohol Dependence 127 (2013) 137–142
measurements before and after alcohol consumption. However, groups did not differ at baseline for impulsivity, sensation-seeking and mood. Second, response inhibition of salience attribution was assessed using a single measure. Given previous concerns with the measurement and complexity of both response inhibition and incentive salience, future studies may benefit from the use of two well-defined task paradigms for the assessment of these two distinct constructs. Our data indicate that, following an acute dose of alcohol, disinhibition is increased towards alcohol-related, lexical stimuli in social drinkers. These data provide initial support for alcoholinduced impairment of behavioural response to alcohol relevant stimuli. Additionally, our results suggest that all participants showed increased disinhibition towards lexical alcohol cues, but reduced disinhibition towards pictorial alcohol stimuli. It remains unclear whether this discrepancy in response inhibition towards alcohol stimuli reflects distinct cue-specific mechanisms of reward on inhibitory control. We also observed that under the influence of acute alcohol, heavier drinkers showed difficulty in shifting attention in the face of alcohol-related cues. This finding is consistent with previous evidence of enhanced biases towards alcohol-related cues in heavy drinkers compared with lighter social drinkers. Finally, we observed that acute alcohol produced overall deficits in inhibitory control by drinking status, where alcohol led to greater disinhibition to lexical cues in heavier drinkers. Together, our results indicate alcohol’s effects on different components of inhibitory control. Moreover, it appears that alcohol’s impairment of inhibitory control may be enhanced when a response towards alcohol-related stimuli is required to be withheld. Role of funding source Funding for this study was provided by a University of Bristol Postgraduate Scholarship, the University of Bristol had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors SA and MRM were responsible for the study design. SA was responsible for the collection and analysis of the data. SA drafted the manuscript. AFA, ASA, and MRM provided critical revision of the manuscript for important intellectual content. All authors critically reviewed content and approved final version for publication. Conflict of interest There are no conflicts of interest. References Adams, S., Ataya, A.F., Attwood, A.S., Munafò, M.R., 2011. Effects of acute alcohol consumption on alcohol-related cognitive biases in light and heavy drinkers are task-dependent. J. Psychopharmacol. 26, 245–253.
Bohn, M.J., Krahn, D.D., Staehler, B.A., 1995. Development and initial validation of a measure of drinking urges in abstinent alcoholics. Alcohol Clin. Exp. Res. 19, 600–606. Bruce, G., Jones, B.T., 2004. A pictorial Stroop paradigm reveals an alcohol attentional bias in heavier compared to lighter social drinkers. J. Psychopharmacol. 18, 527–533. Cox, W.M., Yeates, G.N., Regan, C.M., 1999. Effects of alcohol cues on cognitive processing in heavy and light drinkers. Drug Alcohol Depend. 55, 85–89. de Wit, 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. Duka, T., Townshend, J.M., 2004. The priming effect of alcohol pre-load on attentional bias to alcohol-related stimuli. Psychopharmacology (Berl) 176, 353–356. Easdon, C.M., Vogel-Sprott, M., 2000. Alcohol and behavioral control: impaired response inhibition and flexibility in social drinkers. Exp. Clin. Psychopharmacol. 8, 387–394. Farris, S.R., Ostafin, B.D., 2008. Alcohol consumption primes automatic alcoholapproach associations. Am. J. Drug Alcohol Abuse 34, 703–711. Field, M., Christiansen, P., Cole, J., Goudie, A., 2007. Delay discounting and the alcohol Stroop in heavy drinking adolescents. Addiction 102, 579–586. Field, M., Wiers, R.W., Christiansen, P., Fillmore, M.T., Verster, J.C., 2010. Acute alcohol effects on inhibitory control and implicit cognition: implications for loss of control over drinking. Alcohol Clin. Exp. Res. 34, 1346–1352. Fillmore, M.T., 2007. Acute alcohol-induced impairment of cognitive functions: past and present findings. Int. J. Disabil. Hum. Dev. 6, 115–125. Fillmore, M.T., Vogel-Sprott, M., 1998. Behavioral impairment under alcohol: cognitive and pharmacokinetic factors. Alcohol Clin. Exp. Res. 22, 1476–1482. Goldstein, R.Z., Volkow, N.D., 2002. Drug addiction and its underlying neurobiological basis: neuroimaging evidence for the involvement of the frontal cortex. Am. J. Psychiatry 159, 1642–1652. Grady, C.L., McIntosh, A.R., Rajah, M.N., Craik, F.I.M., 1998. Neural correlates of the episodic encoding of pictures and words. Proc. Natl. Acad. Sci. U.S.A. 95, 2703–2708. Hogarth, L., Attwood, A.S., Bate, H.A., Munafo, M.R., 2012. Acute alcohol impairs human goal-directed action. Biol. Psychol. 90, 154–160. Jentsch, J.D., Taylor, J.R., 1999. Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli. Psychopharmacology (Berl) 146, 373–390. Marczinski, C.A., Abroms, B.D., Van Selst, M., Fillmore, M.T., 2005. Alcohol-induced impairment of behavioral control: differential effects on engaging vs. disengaging responses. Psychopharmacology (Berl) 182, 452–459. Marczinski, C.A., Combs, S.W., Fillmore, M.T., 2007. Increased sensitivity to the disinhibiting effects of alcohol in binge drinkers. Psychol. Addict. Behav. 21, 346–354. McNair, D.M., Lorr, M., Droppleman, L.F., 1992. Profile of Mood States (POMS). Educational and Industrial Testing Service, San Diego, CA. Murphy, F.C., Sahakian, B.J., Rubinsztein, J.S., Michael, A., Rogers, R.D., Robbins, T.W., Paykel, E.S., 1999. Emotional bias and inhibitory control processes in mania and depression. Psychol. Med. 29, 1307–1321. Noel, X., Van der Linden, M., d’Acremont, M., Bechara, A., Dan, B., Hanak, C., Verbanck, P., 2007. Alcohol cues increase cognitive impulsivity in individuals with alcoholism. Psychopharmacology (Berl) 192, 291–298. Ostling, E.W., Fillmore, M.T., 2010. Tolerance to the impairing effects of alcohol on the inhibition and activation of behavior. Psychopharmacology (Berl) 212, 465–473. Palfai, T.P., Ostafin, B.D., 2003. The influence of alcohol on the activation of outcome expectancies: the role of evaluative expectancy activation in drinking behavior. J. Stud. Alcohol 64, 111–119. Patton, J.H., Stanford, M.S., Barratt, E.S., 1995. Factor structure of the Barratt Impulsiveness Scale. J. Clin. Psychol. 51, 768–774. Rose, A.K., Duka, T., 2008. Effects of alcohol on inhibitory processes. Behav. Pharmacol. 19, 284–291. Schoenmakers, T., Wiers, R.W., Field, M., 2008. Effects of a low dose of alcohol on cognitive biases and craving in heavy drinkers. Psychopharmacology (Berl) 197, 169–178. Townshend, J.M., Duka, T., 2001. Attentional bias associated with alcohol cues: differences between heavy and occasional social drinkers. Psychopharmacology (Berl) 157, 67–74. Zuckerman, M., Kuhlman, D.M., Joireman, J., Teta, P., Kraft, M., 1993. A comparison of 3 structural models for personality – the big 3, the big 5, and the alternative 5. J. Pers. Soc. Psychol. 65, 757–768.