Decision-making and inhibitory control after smoking-related priming in nicotine dependent smokers and never-smokers

Accepted Manuscript Decision-making and inhibitory control after smoking-related priming in nicotine dependent smokers and never-smokers

Anja Kräplin, Stefan Scherbaum, Gerhard Bühringer, Thomas Goschke PII: DOI: Reference:

S0306-4603(18)30327-7 doi:10.1016/j.addbeh.2018.08.020 AB 5662

To appear in:

Addictive Behaviors

Received date: Revised date: Accepted date:

18 April 2018 27 July 2018 14 August 2018

Please cite this article as: Anja Kräplin, Stefan Scherbaum, Gerhard Bühringer, Thomas Goschke , Decision-making and inhibitory control after smoking-related priming in nicotine dependent smokers and never-smokers. Ab (2018), doi:10.1016/ j.addbeh.2018.08.020

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Decision-making and inhibitory control after smoking-related priming in nicotine dependent smokers and never-smokers

Anja Kräplin1 , Stefan Scherbaum1 , Gerhard Bühringer1,2 , Thomas Goschke1

PT

1 Faculty of Psychology, Technische Universität Dresden, Germany

SC

RI

2 IFT Institut für Therapieforschung, Munich, Germany

Correspondence concerning this article should be addressed to: Anja Kräplin

NU

Work group Addictive Behaviors, Risk Analysis and Risk Management, Faculty of

01187 Dresden, Germany E-mail: [email protected]

Stefan Scherbaum

EP T

Telefax: +49 351 463-39830

ED

Telephone: +49 351 463-39848

MA

Psychology, School of Science, Technische Universität Dresden, Chemnitzer Straße 46, D-

AC C

Chair of Research Methods and Computational Cognitive Modelling, Faculty of Psychology, School of Science, Technische Universität Dresden, Zellescher Weg 17, D-01062 Dresden, Germany, E-mail: [email protected]

Gerhard Bühringer Chair of Work group Addictive Behaviors, Risk Analysis and Risk Management, Faculty of Psychology, School of Science, Technische Universität Dresden, Chemnitzer Straße 46, D01187 Dresden, Germany, E-mail: [email protected] Thomas Goschke

ACCEPTED MANUSCRIPT Chair of General Psychology, Faculty of Psychology, School of Science, Technische Universität

Dresden,

Zellescher

Weg

17,

D-01062

Dresden,

Germany,

E-mail:

[email protected]

Ethical approval

PT

The study protocol was approved by the local Ethics Committee at the Technische Universität

RI

Dresden, Germany (reference number: EK 170062012) and was in accordance with the 1964

SC

Helsinki declaration and its later amendments.

NU

Funding

This research was supported by the German Research Foundation (DFG) under Grant GO -

AC C

EP T

ED

940/1 2016 and SFB 940/2 2017).

MA

720/8-1 and within the Collaborative Research Centre “Volition and Cognitive Control” (SFB

ACCEPTED MANUSCRIPT Abstract Impaired decision-making and inhibitory control are important characteristics of nicotine dependence (ND). We aimed to test 1) the effects of smoking-related priming cues on subsequent decision-making and inhibitory control in ND and 2) how these priming effects are related to valence ratings, nicotine deprivation and craving. A sample of 27 smokers with

PT

ND according to DSM-IV and a control group of 33 never-smokers performed an

RI

intertemporal choice task and a go/no-go task. Before each trial of the tasks, a priming cue

SC

appeared that was either smoking-related or neutral. Valence ratings, nicotine deprivation and craving were assessed with self-reports. After smoking-related compared to neutral primes,

NU

the ND group exhibited increased delay discounting (β=0.07, 95% confidence interval (CI): 0.01− 0.14) and shorter go reaction times (β=-0.13, CI: -0.32− -0.01) compared to the never-

MA

smoker group. The speed-up in go trials after smoking-related compared to neutral cues was significantly related to more pleasant valence ratings (β=0.07, CI:0.01– 0.13), a longer time

ED

since last cigarette (β=-0.17, CI:-0.30– -0.03), and increased craving (β=-0.19, CI: -0.33– 0.06) within the ND group. We found evidence for small group effects indicating that

EP T

individuals with ND compared to never-smokers decide more dysfunctional and react faster after smoking-related compared to neutral cues. Faster reactions after smoking-related cues

AC C

within the ND group, especially in states of increased nicotine deprivation and craving, without more errors could be explained by an increased attentional focus. Cue-induced alterations in decision-making and inhibitory control in ND highly depend on the temporal sequence of cue presentation.

Keywords Decision- making, inhibitory control, priming, nicotine dependence, craving

ACCEPTED MANUSCRIPT 1 Introduction Nicotine dependence (ND) is highly prevalent in Western societies (Andlin-Sobocki & Rehm, 2005) and associated with detrimental health consequences and societal costs (Effertz & Mann, 2013). Therefore, progress in understanding the processes related to the onset and course of ND merits high priority. Dysfunctional changes of valuation systems and cognitive

PT

control networks have been suggested as possible underlying processes in the development

RI

and course of substance use disorders (Goschke, 2014). Supporting these assumptions,

SC

previous studies in ND revealed impaired performance (for an overview, see MacKillop et al., 2011; Smith, Mattick, Jamadar, & Iredale, 2014) and corresponding altered brain reactivity

NU

during tasks requiring decision-making and inhibitory control (for an overview, see Moeller & Paulus, 2018; Peters & Büchel, 2011). Moreover, impaired decision-making and inhibitory

MA

control have been shown to be associated with relapse (Froeliger, McConnell, Bell, & et al., 2017; Krönke, Wolff, Benz, & Goschke, 2015; for an overview, see Moeller & Paulus, 2018).

ED

Adding to the assumption that impaired cognitive processes are antecedent risk factors of ND, acute and chronic effects of nicotine consumption may also impair decision-making and

EP T

inhibitory control, which in turn facilitate further nicotine consumption (Dawkins, Powell, West, Powell, & Pickering, 2007; Field, Santarcangelo, Sumnall, Goudie, & Cole, 2006;

AC C

Mitchell, 2004).

Despite this substantial body of research, it remains an open question whether these decision-making and inhibitory dysfunctions are general characteristics of ND or whether they are exclusively or particularly impaired in disorder-related contexts, i.e. triggered by smoking-related cues (Goschke, 2014). This open question arises in the light of empirical evidence from neuroimaging and intervention studies supporting that lasting substanceinduced changes in dopaminergic systems involved in reward-based learning are important mechanism underlying ND and other substance use disorders (e.g. Froeliger, Mathew, et al., 2017; Neuhofer & Kalivas, 2018; for an overview, see Volkow, Wang, Tomasi, & Baler,

ACCEPTED MANUSCRIPT 2013). These changes result in markedly increased salience of and responsivity to substancerelated cues which thus receive priority in cognitive processing (Redish, Jensen, & Johnson, 2008; Robinson & Berridge, 2008). Various neuroimaging studies found evidence that elevated frontal, limbic and striatal brain activation in response to substance-related cues and weakened functional connectivity of these same regions are transdiagnostic mechanisms in

PT

substance use disorders and important predictors of relapse (Kearney-Ramos et al., 2018;

RI

Moeller & Paulus, 2018).

SC

Based on these findings, it has been assumed that decision-making and inhibitory control are not only generally impaired in ND given a neutral context but would be further impaired

NU

by smoking-related cues (Dawe & Loxton, 2004). Behavioral studies in smokers testing this assumption yielded heterogeneous results including evidence for increased delay discounting

MA

and impaired inhibitory control in response to smoking-related compared to neutral stimuli (Cox, Fadardi, & Pothos, 2006; Field, Rush, Cole, & Goudie, 2007; Field et al., 2006; Odum,

ED

2011; Yi et al., 2008) and no evidence for such effects (Field et al., 2007; Luijten, Littel, & Franken, 2011).

EP T

Considering previous study designs, the causal effect of substance-related cues on decision-making and inhibitory control performance remains unclear. In the above-mentioned

AC C

behavioral studies, the substance-related cues were simultaneously applied with or as response-relevant stimuli. With these task sets, it remains an open research question if substance-related cues 1) elicit subsequent over-valuation of immediate rewards and habitual responding (Redish et al., 2008) or 2) conflict with the task instructions requiring, for instance, inhibition of substance-related stimuli instead of approach behavior (Veling, Holland, & van Knippenberg, 2008). To address this open research question, it is important to study the effects of substance-related priming cues on subsequent task performance in neutral tasks. A second open research question concerns the impact of cue valence or the deprivation

ACCEPTED MANUSCRIPT and craving levels on the priming effect which are factors known to be related to the preferential cognitive processing of smoking-related cues (Field & Cox, 2008). In the current study, we aimed to find out whether impairments in decision-making and inhibitory control in ND are specifically prominent after smoking-related cues addressing the above specified open research questions. Our first research question was whether task-

PT

irrelevant smoking-related priming cues affect subsequent decision-making and inhibitory

RI

control in a sample clinically diagnosed with ND. We predicted poorer performance in

SC

participants with ND compared to control participants, especially on task trials immediately following smoking-related compared to neutral priming cues. Our second research question

NU

was whether these priming effects are related to cue valence, time since last cigarette and subjective craving. We predicted a stronger priming effect in individuals with more positive

MA

valence ratings, increased nicotine deprivation and craving. By disentangling the effects of smoking-related cues from task performance in ND, our results will contribute to a better

ED

understanding of the causal relation between substance-related cues and impaired decision-

EP T

making and inhibitory control in ND.

2 Material and methods

AC C

2.1 Design and participants

In a cross-sectional study, we applied a two-by-two design with the observed (nonrandomized) between factor group (ND vs. never-smoking control group) and the within factor condition (smoking-related vs. neutral). We conducted a priori power analysis with G*Power 3.1.9.2 (Faul, Erdfelder, Lang, & Buchner, 2007) for linear multiple regression with two predictors (group and performance in the neutral condition, see 2.4 Data analyses). Based on previous studies (Field et al., 2007; Field et al., 2006; Kreusch, Vilenne, & Quertemont, 2013; Monk, Sunley, Qureshi, & Heim, 2016), we assumed a medium effect size. The power analysis revealed that a total of 64 participants, i.e. 32 per group, would be required to achieve

ACCEPTED MANUSCRIPT a power of 0.8 with alpha of 0.05. The two groups were recruited through advertisements and postings on community boards. We screened whether the interested smokers fulfilled ND according to DSM-IV-TR within the last 12 months (American Psychiatric Association (APA), 2000) and whether the nonsmokers fulfilled the criteria of never-smoking defined by having smoked less than 20 times in their life and having not smoked in the last 12 month

PT

(Pomerleau, Pomerleau, Snedecor, & Mehringer, 2004). Furthermore, we screened both

RI

groups for the following exclusion criteria: (1) age under 18 or over 55 years (based on

SC

evidence for general decisions-making deficits in adults >55 compared to <55 years; see Denburg, Tranel, & Bechara, 2005; Fein, McGillivray, & Finn, 2007; and based on findings

NU

showing that cognitive task performance declines below average in the mid-50s, see Murman, 2015), (2) current use of psychotropic substances or medications, (3) disorders that might

MA

influence test performance (e.g., craniocerebral injury), (4) mental disorders according to DSM-IV in the last 12 months, and (5) first language other than German. From 96 screened

ED

individuals, 60 included participants were invited to a test session where a last exclusion criterion was (6) a positive urine drug screening. No participant had to be excluded at the test

EP T

session. The final sample consisted of 27 participants in the ND group and 33 in the control group (see Table 1). The control group included only participants that had never smoked in

AC C

their lives. All variables except the smoking-related did not significantly differ between the two groups (see Table 1).

----Please insert Table 1 about here----

2.2 Procedure At a first personal appointment, participants signed informed consent. We administered the Munich Composite International Diagnostic Interview (DIA-X/M-CIDI; Wittchen & Pfister, 1997) to assess other mental disorders according to DSM-IV and we screened intelligence

ACCEPTED MANUSCRIPT with the Multiple-choice Vocabulary Intelligence Test version B (MWT-B; Lehrl, 2005). All included participants were invited for an individual two hours test session in a quiet room at the university. Tasks in the test session were administered in random order using Octave 3.2.4 (Eaton, Bateman, & Hauberg, 2009). Each task included a training of 10 trials with trial-bytrial feedback on performance. If participants had no further questions, they started the

PT

experiment. Before the test session, we assessed breath carbon monoxide concentration using

RI

Neomed CO-Check (NEOMED GmbH, 2015) to verify the participants’ statements regarding

SC

their smoking behavior. Furthermore, we assessed the time since the last cigarette in minutes and subjective craving ratings before, during and after the test session on a visual analog scale

NU

from 0 to 10 (see Table 1). In the ND group, none of the participants used the breaks during the session for smoking. The study protocol was approved by the Ethics Committee at the

MA

Technische Universität Dresden, Germany.

ED

2.3 Measurements

EP T

Priming cues

We administered an intertemporal choice (ITC) task and a go/no-go task containing either neutral or smoking-related pictures before each trial. Participants were instructed to passively

AC C

look at the pictures and concentrate on the task. After the tasks, participants had to rate the pictures with respect to valence and arousal. The 77 neutral pictures were derived from the International Affective Picture System (IAPS) (Lang, Bradley, & Cuthbert, 1998). The 77 smoking-related pictures (e.g., cigarette, ashtray) were derived from non-copyrighted Internet sources. All priming pictures were color photographs with a size of 800 x 600 pixels and with 72 dpi resolution (see Figure 1). We used the smoking-related pictures from a pilot study that best matched the neutral pictures with respect to their content (one-third scenes with people and two-third with objects) and arousal ratings.

ACCEPTED MANUSCRIPT ----Please insert Figure 1 about here----

Inhibitory control We applied an already implemented go/no-go task from our lab for which basic task effects have been already confirmed to facilitate the interpretation of results (Beck et al., 2016). In

PT

300 trials, participants responded as fast as possible to the letter ‘M’ (go trials), but had to

RI

withhold their response to the letter ‘W’ (no-go trials). No-go stimuli were presented on 20%

SC

randomly chosen trials. After a fixation cross of 300 milliseconds (ms), the letters were presented for 150 ms at the center of the screen. Before each trial, either a neutral or a

NU

smoking-related picture was randomly selected and presented for 500 ms. To avoid slowing down of the task, we presented the pictures within the original inter-stimulus interval which

MA

randomly varied between 1250 and 1750ms. The dependent measures were the mean reaction

ED

times for go trials and the proportion of errors in no-go trials.

Decision-making

EP T

We used an already validated ITC task from our lab to facilitate the interpretation of results (Kräplin et al., 2014). In 192 trials, participants were instructed to decide between a smaller

AC C

monetary gain delivered sooner and a larger monetary gain delivered later. The sooner/smaller reward was randomly selected from a pool of items with a mean value of 20 euro, a standard deviation of two euro, and a range between 17 and 23 euro. The value of the later/larger reward was generated by increasing the sooner/smaller reward by 1, 3.5, 7, 12, 18, 27, 40, or 80%. The time delay for the sooner/smaller reward was either “now”, in 7, or in 14 days and the later/larger reward was additionally delayed by 1, 2, 3, 5, 7, 9, 12, or 15 days. Either a neutral or a smoking-related picture was randomly selected and presented for 500 ms before each trial. Our dependent variable was the k-value derived from participants’ choices using the procedure described first in Kirby and Maraković (1996).

ACCEPTED MANUSCRIPT

2.4 Data analyses Two participants (one from the ND group) were excluded from the analyses of the go/no-go task as high error rates indicated a low adherence to the task instruction.

PT

For all statistical analyses, we used linear regression analyses because, unlike ANOVA and t-tests (which are based on the same underlying statistical model), this allows

RI

for an alternative robust estimation method. Robust regression accounts for residuals with

SC

different variances and extreme values that would otherwise have a strong impact on results (narrowing the scope of a mean effect results). If the results differed between ordinary and

NU

robust linear regression, we report the latter result (no matter what the result is) because this is

MA

based on much weaker assumptions on the data. This follows the general recommendation of Field and Wilcox (2017) how to deal with the usual situation of violated assumptions in clinical and experimental data that easily distorts the results calculated with standard methods.

ED

Two pre-analyses were conducted to validate the experiment: 1. Manipulation check:

EP T

Do the ND group and the control group differ in their ratings of smoking-related and neutral pictures concerning pleasantness and arousal. 2. Are there generalized decision-making and inhibitory control impairments in the neutral condition?

AC C

Then we tested the following hypotheses: 1. Is the difference (delta) between the smoking-related and neutral condition larger in the ND than in the control group? We tested this hypothesis regarding between group effects in two separate regression analyses for decision-making and inhibitory control performance. The predictor of our regression models was the dummy-coded group (0, 1). Our outcomes were the performance contrasts (delta) of trials after smoking-related minus those after neutral pictures (e.g., k-value in the smoking-related minus k-value in the neutral condition). We adjusted this comparison (difference of difference) by participants’ values in the neutral condition. This takes into account the regression to the mean and group differences yet at the

ACCEPTED MANUSCRIPT outset of measurement. In case of significant between group effects, we analyzed within group effects as post-hoc tests, i.e. to test whether performance within one group differs between the smoking-related and neutral condition. 2. Is the ‘priming effect’ in individuals with a) more positive picture ratings, b) increased nicotine deprivation and c) increased craving higher? (Separate analyses for decision-making

PT

and inhibitory control performance). The outcome of the regression models was again the

RI

‘priming effect’ operationalized as difference (delta) between smoking-related and neutral

SC

pictures. The predictors were the pictures valence ratings (a) of all participants as well as the time since last cigarette (b) and the subjective craving ratings (c) of the ND group only.

NU

To facilitate the interpretation of results, we additionally calculated the probability of our hypotheses to be true, i.e. a true difference (between groups) falling into a certain range

MA

(according to the z-standardized values between 0 and 1 for delay discounting, between -1 and 0 for go reaction times). Note that this interpretation does not apply to any conventional,

ED

frequentist analysis, but the frequentist results can be interpreted from the Bayesian perspective if one assumes that all information on the parameter of interest comes from the

EP T

data (‘flat prior’). Then the ‘posterior distribution’ (distribution of a difference = regression coefficient, given the data) equals the frequentist likelihood distribution (distribution of data,

AC C

given the unknown regression coefficient). Here, a (maximum-likelihood) estimate of a regression coefficient is non-centrally t-distributed with expectation = point estimate and variance = squared standard error of the estimate (and n-2 degrees of freedom, n = sample size). All statistical analyses were conducted using Stata 15.0 (Stata Corp., 2017).

3. Results 3.1 Pre-analyses Please note the, except for descriptive data, we report regression analyses results of the zstandardized values for a better interpretation of results and figures.

ACCEPTED MANUSCRIPT

Manipulation check Between groups, the ND group rated the smoking-related pictures significantly more pleasant compared to the control group (β =-1.50, p<0.001, CI: -1.99–-1.01, see Table 2). In contrast, groups did neither differ significantly in the arousal ratings of the smoking-related pictures (β

SC

RI

(β=-0.09, p=0.70, CI:-0.58– 0.39) ratings of the neutral pictures.

PT

=0.00, p=0.99, CI: -0.58– 0.58) nor in valence (β=0.19, p=0.13, CI:-0.06– 0.43) and arousal

General impairments in the neutral condition

NU

----Please insert Table 2 about here----

MA

A significant between group effect in the neutral priming condition indicated general decision-making impairments in the ND group, i.e. that the ND group displayed increased

ED

discounting of delayed rewards compared to the control group (β=0.11, p=0.01, CI: 0.02– 0.19). In contrast, there was no significant between group effects on the indicators of

EP T

inhibitory control in the neutral condition, i.e. go reaction times (β=0.13, p=0.69, CI:-0.04–

AC C

0.67) and error rates (β=-0.04, p=0.56, CI: -0.58–0.49).

3.2 First hypothesis: Priming effects Decision-making

The ND group exhibited a significantly increased priming effect on decision-making compared to the control group, i.e. a higher discounting of delayed rewards after smokingrelated compared to neutral pictures (see Table 3 and Figure 2). From the likelihood function of the beta estimate, we calculated that the probability of beta falling into the range between 0 and 1 is 99%. However, given that the confidence interval is very close to zero (0.01− 0.14), the revealed effect seem to be small. In our post-hoc analyses, we found no significant within-

ACCEPTED MANUSCRIPT group effects between the priming conditions for the ND group (β=0.15, p=0.22, CI: -0.25– 0.56) or the control group (β=0.07, p=0.71, CI: -0.29– 0.42).

----Please insert Table 3 about here----

RI

PT

----Please insert Figure 2 about here----

SC

Inhibitory control

We found a significant between group effect concerning go reaction times (p=0.02; CI: -0.32−

NU

-0.01; see Table 3 and Figure 2). From the likelihood function of the beta estimate, we calculated that the probability of beta falling into the range between -1 and 0 is 98%. Like for

MA

decision-making, the confidence interval of our effect is very close to zero which indicated a small effect. Our post-hoc analyses within groups revealed a significant effect of the priming

ED

condition on reaction times in the ND group (β=-0.13, p=0.03, CI: -0.26– 0.00) while we found no significant effect of condition on reaction times within the control group (β=0.04,

AC C

effect.

EP T

p=0.46, CI: -0.63– 0.17). Concerning error rates, found no evidence for a between group

3.3 Second hypothesis: Relation of the priming effects with picture ratings, nicotine deprivation, and craving

There was no significant association of the revealed priming effect on decision-making with the valence ratings of the smoking-related pictures in both groups (β=-0.11, p=0.16, CI:-0.32– 0.10). In contrast, we found that the more pleasant smoking-related pictures were rated compared to neutral pictures, the shorter were the reaction times in go trials following these pictures compared to neutral pictures (β=0.07, p=0.02, CI:0.01– 0.13, see Figure 3). From the

ACCEPTED MANUSCRIPT likelihood function of the beta estimate, we calculated that the probability of beta falling into the range between 0 and 1 is 98%. The confidence interval indicated a small effect.

----Please insert Figure 3 about here----

PT

Within the ND group, there was no significant associations of the revealed priming

RI

effect on decision-making with the time since last cigarette (β=-0.07, p=0.61, CI: -0.57– 0.43)

SC

or the subjective craving ratings (β=0.35, p=0.06, CI: -0.09– -0.77). In contrast, we found a significant relation between the priming effect on go reaction times and time since last

NU

cigarette (β=-0.17, p=0.01, CI:-0.30– -0.03) as well as subjective craving ratings (β=-0.19, p=0.004, CI: -0.33– -0.06). The probability of beta falling into the range between -1 and 0

MA

was 99 % for time since last cigarette and for craving. The confidence intervals indicated small to medium effects. The longer the last cigarette was ago and the higher the craving

ED

ratings during the test session, the shorter were the reaction times of participants in the ND

EP T

group in go trials after smoking-related in contrast to neutral pictures (see Figure 4).

AC C

----Please insert Figure 4 about here----

4 Discussion

We tested, to our best knowledge, for the first time the effects of smoking-related priming cues on subsequent decision-making and inhibitory control in ND. After smoking-related compared to neutral cues, the ND group exhibited increased delay discounting and shorter go reaction times compared to the never-smoker group. Faster go reaction times after smokingrelated cues were associated with increased pleasantness ratings and with increased time since last cigarette and subjective craving in the ND group.

ACCEPTED MANUSCRIPT 4.1 General discussion As first hypothesis, we assumed that ND is characterized by generally impaired decisionmaking and inhibitory control after neutral cues and that smoking-related cues elicit (further) impairments of these functions (Dawe & Loxton, 2004; Goschke, 2014; Redish et al., 2008; Robinson & Berridge, 2008). Our results revealed evidence for a general dysfunctional

PT

decision-making in ND but not for generally impaired inhibitory control which is in line with

RI

previous studies (MacKillop et al., 2011; Smith et al., 2014). Regarding the hypothesized

SC

additional impairing effect of substance-related cue on these cognitive functions, we found evidence for small between group effects indicating that individuals with ND compared to

NU

never-smokers decide and react differently after smoking-related compared to neutral cues. However, post-hoc comparisons revealed no evidence for within-group effects of condition on

MA

subsequent decision-making. This is in contrast to previous studies using hypothetical cigarette purchases (Field et al., 2006; Odum, 2011; Yi et al., 2008) while one study applying

ED

a decision-making task in a smoking-related context reported also no within-group effects (Field et al., 2007). In sum, the increased discounting of delayed rewards (money, cigarettes)

performance failure.

EP T

in ND seems to reflect a generalized competence impairment instead of a context-specific

AC C

Regarding inhibitory control, post-hoc comparisons revealed evidence that individuals with ND were faster in go trials after smoking-related cues compared to neutral cues whereas there was no evidence for such an effect on false alarm rates. Smoking-related cues may trigger habitual behavior but simultaneously increase attention by positive affect (Chiew & Braver, 2011; Goschke & Bolte, 2014) and approach motivation (Harmon-Jones, Gable, & Price, 2012; Hicks, Friedman, Gable, & Davis, 2012). This assumption is underpinned by the revealed small to medium effect that the priming effect on go reaction times was positively related to pleasantness ratings, time since the last cigarette, and craving ratings which are known to increase positive approach behavior and allocation of attentional resources in

ACCEPTED MANUSCRIPT individuals with substance use disorders (Berridge, Robinson, & Aldridge, 2009; Field & Cox, 2008; Robinson & Berridge, 2008). The missing evidence for smoking-related priming effects on false alarm rates in ND is contrary to previous results that have found such evidence in substance use disorders applying substance-related cues with or as the response-relevant stimuli (Cox et al., 2006; Kreusch et

PT

al., 2013; Monk et al., 2016; Noël et al., 2007; Petit, Kornreich, Noël, Verbanck, &

RI

Campanella, 2012). This contrasting evidence underpins the assumption that smoking-related

SC

cues which are applied with or as response-relevant stimuli conflict with processes required by the task instruction while a task-irrelevant antecedent cue presentation might facilitate

MA

NU

attentional processes.

4.2 Limitations and strength

In previous studies, very different populations of smokers have been assessed including

ED

participants with light, regular, or dependent smoking. It is therefore valuable that we

EP T

assessed the priming effects in clinically diagnosed ND to specifically contribute to disorder models. With respect to our strict exclusion criteria, our results on ND may have limited generalizability because ND is related to several comorbid mental disorders (John, Meyer,

AC C

Rumpf, & Hapke, 2004). Nevertheless, the internal validity of our results is high and we can draw conclusions specifically for ND. Concerning our sample, another limitation may be that the ND group had a lower proportion of male participants. It has been summarized that there were greater differences between female problem and non-problem substance users in inhibitory control compared to men (Weafer & de Wit, 2014). Hence, gender differences in inhibitory control processes may have confounded our results. However, one previous study has shown that gender effects were independent from the substance-relatedness of the applied stimuli in an inhibitory control task (Nederkoorn, Baltus, Guerrieri, & Wiers, 2009). Furthermore, the ND group comprised more female participants so that we would have

ACCEPTED MANUSCRIPT expected a lower inhibitory control in the ND group which we did not confirm. Appling randomized trial-by-trial priming with substance-related cues may not have been sufficient to produce a pronounced effect of the smoking-related cues. A future study may use blocks of smoking-related and neutral pictures or more realistic cues to alleviate this problem.

PT

5 Conclusions and future research

RI

The cue-induced alterations in decision-making and inhibitory control in ND depend on

SC

the temporal sequence of cue presentation. Smoking-related cues seem to elicit subsequent changes in valuation signals and inhibitory and attentional processes in ND which are

NU

different from never smokers. Applied with or as response-relevant stimuli as has been done in previous studies, smoking-related cues conflict with processes required by the task

MA

instruction and result in more performance failures.

Future research may assess performance during and after the presentation of

ED

substance-related cues applying neuroimaging methods to further clarify the temporal dynamics of their effects on decision-making and inhibitory control processes and their role in

EP T

the course of ND. One recent study, for instance, has shown that diminished neural activity in the inhibitory control circuitry during an inhibition tasks with smoking-related cues was a risk

AC C

factor for relapse (Gilman et al., 2018). Further studies in this direction would allow deepening our understanding of self-control failures of individuals with substance use disorders in substance-related contexts (e.g., relapse) even after control competences have been strengthened with therapeutic strategies.

ACCEPTED MANUSCRIPT References

AC C

EP T

ED

MA

NU

SC

RI

PT

American Psychiatric Association (APA). (2000). Diagnostic and statistical manual of mental disorder, text revision (DSM-IV-TR) (4th, text revision ed.). Washington, DC: American Psychiatric Association. Andlin-Sobocki, P., & Rehm, J. (2005). Cost of addiction in Europe. European Journal of Neurology, 12, 28-33. doi:10.1111/j.1468-1331.2005.01194.x Beck, S. M., Ruge, H., Schindler, C., Burkart, M., Miller, R., Kirschbaum, C., & Goschke, T. (2016). Effects of Ginkgo biloba extract EGb 761® on cognitive control functions, mental activity of the prefrontal cortex and stress reactivity in elderly adults with subjective memory impairment – a randomized double-blind placebo-controlled trial. Human Psychopharmacology: Clinical and Experimental, 31(3), 227-242. doi:10.1002/hup.2534 Berridge, K. C., Robinson, T. E., & Aldridge, J. W. (2009). Dissecting components of reward: ‘Liking’, ‘wanting’, and learning. Current Opinion in Pharmacology, 9(1), 65-73. doi:10.1016/j.coph.2008.12.014 Chiew, K. S., & Braver, T. S. (2011). Positive affect versus reward: emotional and motivational influences on cognitive control. Frontiers in Psychology, 2, 279. doi:10.3389/fpsyg.2011.00279 Cox, W. M., Fadardi, J. S., & Pothos, E. M. (2006). The Addiction-Stroop Test: Theoretical considerations and procedural recommendations. Psychological Bulletin, 132(3), 443476. doi:10.1037/0033-2909.132.3.443 Dawe, S., & Loxton, N. J. (2004). The role of impulsivity in the development of substance use and eating disorders. Neuroscience & Biobehavioral Reviews, 28(3), 343-351. doi:10.1016/j.neubiorev.2004.03.007 Dawkins, L., Powell, J. H., West, R., Powell, J., & Pickering, A. (2007). A double-blind placebo-controlled experimental study of nicotine: II—Effects on response inhibition and executive functioning. Psychopharmacology, 190(4), 457-467. doi:10.1007/s00213006-0634-6 Denburg, N. L., Tranel, D., & Bechara, A. (2005). The ability to decide advantageously declines prematurely in some normal older persons. Neuropsychologia, 43(7), 10991106. doi:10.1016/j.neuropsychologia.2004.09.012 Eaton, J. W., Bateman, D., & Hauberg, S. (2009). GNU Octave version 3.2.4 manual: A highlevel interactive language for numerical computations. Retrieved from https://octave.org/doc/v4.2.1/. Effertz, T., & Mann, K. (2013). The burden and cost of disorders of the brain in Europe with the inclusion of harmful alcohol use and nicotine addiction. European Neuropsychopharmacology, 23(7), 742-748. doi:10.1016/j.euroneuro.2012.07.010 Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175-191. doi:10.3758/bf03193146 Fein, G., McGillivray, S., & Finn, P. (2007). Older adults make less advantageous decisions than younger adults: cognitive and psychological correlates. Journal of the International Neuropsychological Society, 13(3), 480-489. doi:10.1017/S135561770707052X Field, M., & Cox, W. M. (2008). Attentional bias in addictive behaviors: A review of its development, causes, and consequences. Drug and Alcohol Dependence, 97(1–2), 1-20. doi:10.1016/j.drugalcdep.2008.03.030 Field, M., Rush, M., Cole, J., & Goudie, A. (2007). The smoking Stroop and delay discounting in smokers: Effects of environmental smoking cues. Journal of Psychopharmacology, 21(6), 603-610. doi:10.1177/0269881106070995

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

Field, M., Santarcangelo, M., Sumnall, H., Goudie, A., & Cole, J. (2006). Delay discounting and the behavioural economics of cigarette purchases in smokers: The effects of nicotine deprivation. Psychopharmacology, 186(2), 255-263. doi:10.1007/s00213-0060385-4 Froeliger, B., Mathew, A. R., McConnell, P. A., Eichberg, C., Saladin, M. E., Carpenter, M. J., & Garland, E. L. (2017). Restructuring reward mechanisms in nicotine addiction: A pilot fMRI study of mindfulness-oriented recovery enhancement for cigarette smokers. Evidence-Based Complementary and Alternative Medicine, 2017, 1-10. doi:10.1155/2017/7018014 Froeliger, B., McConnell, P. A., Bell, S., & et al. (2017). Association between baseline corticothalamic-mediated inhibitory control and smoking relapse vulnerability. JAMA Psychiatry, 74(4), 379-386. doi:10.1001/jamapsychiatry.2017.0017 Gilman, J. M., Radoman, M., Schuster, R. M., Pachas, G., Azzouz, N., Fava, M., & Evins, A. E. (2018). Anterior insula activation during inhibition to smoking cues is associated with ability to maintain tobacco abstinence. Addictive Behaviors Reports, 7, 40-46. doi:10.1016/j.abrep.2018.01.002 Goschke, T. (2014). Dysfunctions of decision-making and cognitive control as transdiagnostic mechanisms of mental disorders: Advances, gaps, and needs in current research. International Journal of Methods in Psychiatric Research, 23(S1), 41-57. doi:10.1002/mpr.1410 Goschke, T., & Bolte, A. (2014). Emotional modulation of control dilemmas: The role of positive affect, reward, and dopamine in cognitive stability and flexibility. Neuropsychologia, 62, 403-423. doi:10.1016/j.neuropsychologia.2014.07.015 Harmon-Jones, E., Gable, P., & Price, T. F. (2012). The influence of affective states varying in motivational intensity on cognitive scope. Frontiers in Integrative Neuroscience, 6(73), 1-5. doi:10.3389/fnint.2012.00073 Hicks, J. A., Friedman, R. S., Gable, P. A., & Davis, W. E. (2012). Interactive effects of approach motivational intensity and alcohol cues on the scope of perceptual attention. Addiction, 107(6), 1074–1080. doi:10.1111/j.1360-0443.2012.03781.x John, U., Meyer, C., Rumpf, H.-J., & Hapke, U. (2004). Smoking, nicotine dependence and psychiatric comorbidity-a population-based study including smoking cessation after three years. Drug and Alcohol Dependence, 76(3), 287-295. doi:10.1016/j.drugalcdep.2004.06.004 Kearney-Ramos, T. E., Dowdle, L. T., Lench, D. H., Mithoefer, O. J., Devries, W. H., George, M. S., . . . Hanlon, C. A. (2018). Transdiagnostic effects of ventromedial prefrontal cortex transcranial magnetic stimulation on cue reactivity. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 3(7), 599-609. doi:10.1016/j.bpsc.2018.03.016 Kirby, K. N., & Maraković, N. N. (1996). Delay-discounting probabilistic rewards: Rates decrease as amounts increase. Psychonomic Bulletin & Review, 3(1), 100-104. doi:10.3758/bf03210748 Kräplin, A., Dshemuchadse, M., Behrendt, S., Scherbaum, S., Goschke, T., & Bühringer, G. (2014). Dysfunctional decision- making in pathological gambling: pattern specificity and the role of impulsivity. Psychiatry Research, 215(3), 675–682. doi:10.1016/j.psychres.2013.12.041 Kreusch, F., Vilenne, A., & Quertemont, E. (2013). Response inhibition toward alcoholrelated cues using an alcohol go/no-go task in problem and non-problem drinkers. Addictive Behaviors, 38(10), 2520-2528. doi:10.1016/j.addbeh.2013.04.007 Krönke, K.-M., Wolff, M., Benz, A., & Goschke, T. (2015). Successful smoking cessation is associated with prefrontal cortical function during a Stroop task: A preliminary study.

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

Psychiatry Research: Neuroimaging, 234(1), 52-56. doi:10.1016/j.pscychresns.2015.08.005 Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1998). International affective picture system (IAPS): Technical manual and affective ratings. Gainsville, FL: University of Florida Center for Research in Psychophysiology. Lehrl, S. (2005). Mehrfachwahl-Wortschatz-Intelligenztest MWT-B. [Multiple-choice Vocabulary Intelligence Test - Version B] (5. ed.). Balingen: Spitta Verlag. Luijten, M., Littel, M., & Franken, I. H. A. (2011). Deficits in inhibitory control in smokers during a go/nogo task: an investigation using event-related brain potentials. PloS One, 6(4), e18898. doi:10.1371/journal.pone.0018898 MacKillop, J., Amlung, M. T., Few, L. R., Ray, L. A., Sweet, L. H., & Munafo, M. R. (2011). Delayed reward discounting and addictive behavior: A meta-analysis. Psychopharmacology, 216(3), 305–321. doi:10.1007/s00213-011-2229-0 Mitchell, S. H. (2004). Effects of short-term nicotine deprivation on decision-making: Delay, uncertainty and effort discounting. Nicotine & Tobacco Research, 6(5), 819-828. doi:10.1080/14622200412331296002 Moeller, S. J., & Paulus, M. P. (2018). Toward biomarkers of the addicted human brain: Using neuroimaging to predict relapse and sustained abstinence in substance use disorder. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 80, 143154. doi:10.1016/j.pnpbp.2017.03.003 Monk, R. L., Sunley, J., Qureshi, A. W., & Heim, D. (2016). Smells like inhibition: The effects of olfactory and visual alcohol cues on inhibitory control. Psychopharmacology, 233(8), 1331-1337. doi:10.1007/s00213-016-4221-1 Murman, D. L. (2015). The impact of age on cognition. Seminars in Hearing, 36(3), 111-121. doi:10.1055/s-0035-1555115 Nederkoorn, C., Baltus, M., Guerrieri, R., & Wiers, R. W. (2009). Heavy drinking is associated with deficient response inhibition in women but not in men. Pharmacology Biochemistry & Behavior, 93, 331–336. Neuhofer, D., & Kalivas, P. (2018). Metaplasticity at the addicted tetrapartite synapse: A common denominator of drug induced adaptations and potential treatment target for addiction. Neurobiology of Learning and Memory. doi:10.1016/j.nlm.2018.02.007 Noël, 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, 192(2), 291-298. doi:10.1007/s00213-006-0695-6 Odum, A. L. (2011). Delay discounting: Trait variable? Behavioural Processes, 87(1), 1-9. doi:10.1016/j.beproc.2011.02.007 Peters, J., & Büchel, C. (2011). The neural mechanisms of inter-temporal decision- making: understanding variability. Trends in Cognitive Sciences, 15(5), 227-239. doi:10.1016/j.tics.2011.03.002 Petit, G., Kornreich, C., Noël, X., Verbanck, P., & Campanella, S. (2012). Alcohol-related context modulates performance of social drinkers in a visual go/no-go task: a preliminary assessment of event-related potentials. PloS One, 7(5), e37466. doi:10.1371/journal.pone.0037466 Pomerleau, C. S., Pomerleau, O. F., Snedecor, S. M., & Mehringer, A. M. (2004). Defining a never-smoker: results from the nonsmokers survey. Addictive Behaviors, 29(6), 11491154. doi:10.1016/j.addbeh.2004.03.008 Redish, A. D., Jensen, S., & Johnson, A. (2008). A unified framework for addiction: Vulnerabilities in the decision process. Behavioral and Brain Sciences, 31, 415–487. doi:10.1017/S0140525X0800472X Robinson, T. E., & Berridge, K. C. (2008). Review: The incentive sensitization theory of addiction: Some current issues. Philosophical transactions of the Royal Society of

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

London. Series B, Biological sciences, 363(1507), 3137-3146. doi:10.1098/rstb.2008.0093 Smith, J. L., Mattick, R. P., Jamadar, S. D., & Iredale, J. M. (2014). Deficits in behavioural inhibition in substance abuse and addiction: A meta-analysis. Drug and Alcohol Dependence, 145, 1-33. doi:10.1016/j.drugalcdep.2014.08.009 Stata Corp. (2017). Stata statistical software: Release 15.0 Stata Corporation. Veling, H., Holland, R. W., & van Knippenberg, A. (2008). When approach motivation and behavioral inhibition collide: Behavior regulation through stimulus devaluation. Journal of Experimental Social Psychology, 44(4), 1013-1019. Volkow, N. D., Wang, G.-J., Tomasi, D., & Baler, R. D. (2013). Unbalanced neuronal circuits in addiction. Current Opinion in Neurobiology, 23(4), 639-648. doi:10.1016/j.conb.2013.01.002 Weafer, J., & de Wit, H. (2014). Sex differences in impulsive action and impulsive choice. Addictive Behaviors, 39(11), 1573-1579. doi:10.1016/j.addbeh.2013.10.033 Wittchen, H.-U., & Pfister, H. (1997). DIA-X Interview. Instruktionsmanual zur Durchführung von DIA-X-Interviews [Instruction manual for the DIA-X-Interview]. Frankfurt: Swets & Zeitlinger. Yi, R., Johnson, M. W., Giordano, L. A., Landes, R. D., Badger, G. J., & Bickel, W. K. (2008). The effects of reduced cigarette smoking on discounting future rewards: An initial evaluation. The Psychological Record, 58(2), 163-174. doi:10.1007/bf03395609

PT

ACCEPTED MANUSCRIPT

Fig.1. Modified go/no-go procedure with examples of neutral and smoking-related priming

AC C

EP T

ED

MA

NU

SC

RI

cues.

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig.2. Descriptive results for the z-standardized outcomes of decision-making (k-values of an intertemporal choice task) and inhibitory control performance (go-reaction times in a go/nogo task) separately for trials after smoking-related and after neutral pictures and separately for the predictor group (control group=CG, nicotine dependence group=ND). Regression analyses revealed evidence for a significant between group effect for performance differences after smoking-related compared to neutral pictures for the k-values (β=0.07, p=0.02; CI: 0.01− 0.14) and go reaction times (β=-0.16, p=0.02; CI: -0.32− -0.01).

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

ED

Fig.3. Positive relation (standardized β=0.07, p=0.02, CI: 0.01– 0.13) between the outcome

EP T

priming effect (differences in go reaction times in the go/no-go task between the smokingrelated and neutral priming condition) and the predictor valence ratings of the smoking-

AC C

related pictures (1=very pleasant to 9=very unpleasant) over the nicotine dependence (ND) and the control group (CG).

Note: ms=milliseconds; Differences below zero indicate faster reaction times in the go trials of the go/no -go task following smoking-related compared to neutral priming.

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

EP T

Fig.4. Negative relation between the outcome priming effect (differences in go reaction times in the go/no-go task between the smoking-related and neutral priming condition) and the z-

AC C

standardized predictors time since last cigarette (β=-0.17, p=0.01, CI:-0.30– -0.03) and subjective craving ratings during the test session (β=-0.19, p=0.004, CI: -0.33– -0.06) within the nicotine dependence group.

Note: ms=milliseconds; Differences below zero indicate faster reaction times in the go trials of the go/no -go task following smoking-related compared to neutral priming.

ACCEPTED MANUSCRIPT Table 1 Demographic and smoking-related characteristics with either means (M) and standard

Test statistics of ND

CG

AC C

EP T

ED

MA

NU

SC

RI

PT

group differences

deviations (SD), numbers (n) and percentages, or median and range for the nicotine dependence (ND) group and the control group of never-smokers (CG)

ACCEPTED MANUSCRIPT n Demographic

27

33

M (SD)

M (SD)

26.26 (6.93)

25.81 (7.14)

t-test CI: -4.10−3.22

Age

p=.81 CI: -2.34−5.22 100.89 (8.33)

102.33 (6.30)

Income < 900 Euro per 21 (77.8%)

Age of first cigarette

smoking

AC C

Attempts to quit

squared test χ2 =2.17 p=.07 χ2 =3.75

27 (81.8%) p=.06

Median (range)

Median (range)

t-test

17 (14-24)

-

-

4 (3-7)

0

-

1 (0-6)

-

-

10 (3-30)

0

-

7 (6-28)

2 (1-5)

EP T

DSM-IV criteria

Pearson's chi-

ED

Smoking-related

p=.45

18 (54.6%)

MA

month

for ND

SC

8 (29.6%)

NU

Male participants

n (%)

RI

n (%)

PT

Estimated IQ

Number of daily cigarettes Breath CO

CI: -10.14− -4.83

concentration (ppm)1)

p<.001

Last cigarette before 20 (5-240) test session (minutes)

-

-

ACCEPTED MANUSCRIPT Sum score of craving 8 (0-24)

-

-

ratings (0 to 30)

1) Breath carbon monoxide (CO) concentration in parts per million (ppm) was assessed before the test session using Neomed CO-Check (NEOMED GmbH, 2015). For non-smokers, values typically range

PT

from 0 to 5. Note. IQ=Intelligence quotient; DSM-IV=Diagnostic and Statistical Manual of Mental Disorders, 4th

AC C

EP T

ED

MA

NU

SC

RI

Edition (APA, 2000); CI= 95% confidence interval of group differences

ACCEPTED MANUSCRIPT Table 2 Means (M) and standard deviations (SD) of the study outcomes decision-making and inhibitory control performance and the valence and arousal ratings separately for the ND

CG

M (SD)

M (SD)

AC C

EP T

ED

MA

NU

SC

RI

PT

Variables

smoking-related and the neutral priming condition in the nicotine dependence (ND) group and the control group of never-smokers (CG)

ACCEPTED MANUSCRIPT Decision-making k-values ITC task Neutral condition

0.08 (0.16)

0.04 (0.09)

Smoking-related condition

0.10 (0.21)

0.05 (0.14)

PT

Inhibitory control

Smoking-related condition

333.33 (26.42)

333.28 (36.74)

SC

337.43 (24.40)

NU

Neutral condition

RI

Go reaction times (ms)

Nogo error rates (%)

334.44 (35.56)

11.86 (6.42)

12.19 (9.03)

Smoking-related condition

11.35 (5.74)

12.03 (8.24)

ED EP T

Valence ratings 1) Neutral pictures

MA

Neutral condition

4.06 (0.67)

3.83 (0.41)

4.10 (1.09)

5.91 (1.11)

Neutral pictures

3.18 (1.43)

3.30 (1.49)

Smoking-related pictures

4.10 (1.81)

4.26 (2.26)

AC C

Smoking-related pictures

Arousal ratings 1)

Note. ITC=inter-temporal choice, ms=milliseconds 1) Higher values indicate more negative valence ratings and increased arousal ratings of the pictures.

ACCEPTED MANUSCRIPT Table 3 Results of the regression analyses with the priming effects as outcomes (z-standardized performance differences of the smoking-related minus the neutral priming condition), with group as predictor (dummy coded, nicotine dependence versus never-smoker group as

Beta

Standard

t

p-value

RI

df

PT

reference), and adjusted for performance in the neutral priming condition

Priming effects 59

NU

on decision-

interval

SC

error

95% confidence

making 0.07

0.03

2.24

0.02

0.01− 0.14

Decision-making

-0.61

0.03

-20.79

<0.001

-0.67− -0.55

control

57

AC C

on inhibitory

EP T

Priming effects

ED

neutral condition

MA

Group

Reaction times Group

-0.16

0.08

-2.04

0.02

-0.32− -0.01

Inhibitory

-0.05

0.04

-1.13

0.26

-0.12− 0.03

control neutral condition Error rates

57

ACCEPTED MANUSCRIPT Group

-0.06

0.19

-0.34

0.63

-0.44− 0.31

Inhibitory

-0.39

0.09

-4.30

<0.001

-0.57− -0.21

control neutral condition

PT

Note. df= degrees of freedom

AC C

EP T

ED

MA

NU

SC

RI

We applied standard and robust regression which provides an alternative robust estimation method. If the results differed between ordinary and robust linear regression, we report the latter result (no matter what the result is) because this is based on much weaker assumptions on the data.

ACCEPTED MANUSCRIPT Highlights

- Nicotine dependence (ND) is related to generally increased delay discounting - No evidence for generally decreased inhibitory control in ND - Previously presented smoking-related primes elicit faster go reaction times in ND

PT

- Priming effect in ND increases with time since last cigarette and subjective craving

AC C

EP T

ED

MA

NU

SC

RI

- Effects of substance-related cues on cognitive processes depend on the timing of cues