Attentional bias modification in smokers trying to quit: A longitudinal study about the effects of number of sessions

Journal of Substance Abuse Treatment 47 (2014) 50–57

Contents lists available at ScienceDirect

Journal of Substance Abuse Treatment

Attentional bias modification in smokers trying to quit: A longitudinal study about the effects of number of sessions☆ Fernanda Machado Lopes, Ph.D. ⁎, Augusto Viana Pires, Lisiane Bizarro, Ph.D. Department of Psychology, Laboratory of Experimental Psychology, Neurosciences and Behavior, Universidade Federal do Rio Grande do Sul (UFRGS), Rua Ramiro Barcellos 2600, Porto Alegre-RS, Brazil, 90035-003

a r t i c l e

i n f o

Article history: Received 13 October 2013 Received in revised form 27 January 2014 Accepted 3 March 2014 Keywords: Smoking Attentional bias modification Smoking cessation program Treatment Cognitive–Behavioral therapy

a b s t r a c t Attentional bias modification (ABM) to avoid smoking-related cues is a potentially new intervention in addition to existing therapy to stop smoking. We examined immediate and long-term changes in attentional bias and treatment outcomes from multiple ABM sessions in 67 smokers trying to quit. After assessing attentional bias baseline, participants were randomly allocated to one of three training groups: three sessions of ABM (avoid 3); two sessions of placebo-ABM and one session of ABM (avoid 1); and three sessions of placebo-ABM (avoid 0). At baseline, all groups had similar positive attentional bias, which became negative at 24 h post-training. After 1 month, avoid 1 and avoid 3 still exhibited negative attentional biases. Only avoid 3 maintained this effect at 6-month, but not at 12-month assessments. ABM produced a long-lasting automatic and maintained avoidance to smoking-related cues which depended on number of sessions; however its effects on treatment outcomes are uncertain. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Evidence suggests that implicit cognitive mechanisms such as attentional bias and cues reactivity influence the decision and behavior of drug use, playing an important role in maintaining this addiction (Field & Cox, 2008). The cues reactivity refers to the variety of responses (physiological or behavioral) that are observed when drug addicts, former addicts or frequent users are exposed to some stimuli that were previously associated with the drug effects (Rooke, Hine, & Thorsteinsson, 2008). Drug-related stimuli produce responses associated with its effects, including craving, excitement and difficulty sustaining abstinence (Robbins & Ehrman, 2004). Although cognitive–behavioral therapy is effective for smoking cessation (Focchi & Braun, 2005), it is aimed at explicit processes (e.g. motivation for treatment, relapse prevention), and not implicit automatic processes. Thus, investigation of new techniques focused on implicit cognition as complementary to traditional interventions is preeminent (Schoenmakers et al., 2010). Attentional bias modification (ABM) has been widely studied as an implicit training strategy to reduce cue reactivity in anxiety disorders (Amir, Beard, Burns, & Bomyea, 2009; MacLeod, Rutherford, Campbell, Ebsworthy, & Holker, 2002; Schmidt, Richey, Buckner, & Timpano, 2009) and a few studies in addiction (Attwood, O'Sullivan, Leonards,

☆ Support: The research was supported by a grant from Conselho Nacional de Desenvolvimento Científico e Tecnológico - Brazil - CNPq 401035/2007-7. ⁎ Corresponding author. Tel.: +55 51 33085363; fax: +55 51 93322233. E-mail addresses: [email protected] (F.M. Lopes), [email protected] (A.V. Pires), [email protected] (L. Bizarro). 0740-5472/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jsat.2014.03.002

Mackintosh, & Munafo, 2008; Field, Duka, Tyler, & Schoenmakers, 2009; Field & Eastwood, 2005; Schoenmakers et al., 2010). Since individuals affected by emotional disorders such as anxiety, depression, and addiction have increased attention toward events (words or images) related to their pathologies (Amir et al., 2009; Peuker, Lopes, & Bizarro, 2009), ABM could be a novel and promising approach with a potential clinical utility as an additional intervention. Attentional bias is the tendency for a person to drive or maintain attention to stimuli due to the value attributed to them. Theoretical models suggest that attentional bias results from repeated pairing of smoking cues with direct effects of nicotine, leading to a sensitized reaction to smokingrelated cues which become salient (Field & Cox, 2008; Franken, 2003; Lopes, Peuker, & Bizarro, 2008; Robinson & Berridge, 1993). Smokingrelated stimuli tend to capture the attention of smokers, and this is considered relevant to drug seeking and smoking cessation outcomes (Waters, Shiffman, Bradley, & Mogg, 2003). Smokers have positive bias for a cigarette compared to nonsmokers (Bradley, Field, Mogg, & De Houver, 2004; Lopes et al., 2008; Moog, Bradley, Field, & De Houwer, 2003) when deprived of nicotine (Field, Mogg, Zetteler, & Bradley, 2004) and even when motivated to quit smoking (Waters et al., 2003). On the other hand, former smokers showed long-lasting negative attentional bias, i.e. an avoidance to smoking-related cues, which might be a successful outcome of a smoking cessation attempt (Peuker & Bizarro, 2013). Thus, strategies that help to reduce and/or make this bias negative may contribute to a higher success rate in smoking cessation treatment. The visual-probe task is one of the most widely used tasks to investigate attentional bias (MacLeod, Mathews, & Tata, 1986), and a modified version is employed in ABM (MacLeod et al., 2002) which

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

has the advantage of assessing and training implicit cognition using equivalent procedures. In this computerized task, the participant must locate and identify a probe that appears to the left or right visual field. The appearance of the probe is preceded by the appearance of a pair of images (smoking-related and matched control). The difference between reaction times to the probe when it replaces each image indicates the interference of the content of the image on attentional processes. In a standard visual probe task, the probe replaces both images in equal frequency while in ABM, the probe always replaces the control images when training aims to develop avoidance to smoking-related images. Thus, in ABM one learns the implicit rule to disengage attention from smoking-related images and instead attend to the control image. It is expected that this change in attentional bias generalizes to real exposure to drug cues, reduces craving and helps to acquire and maintain abstinence. In order to study all attentional processes involved in attentional bias, it is necessary to target both initial, automatic detection of stimuli and later, maintained engagement stages of attention (Field, Mogg, Zetteler, & Bradley, 2004; Robbins & Ehrman, 2004). Initial orienting of attention is a relatively rapid process, which can be assessed when the stimuli are presented for brief exposure durations (e.g. 50–200 ms), while biases in maintained attention are more likely to be revealed when stimuli are presented for longer stimulus durations (e.g. 2000 ms) (Field, Mogg, Zetteler, & Bradley, 2004). Manipulation of the presentation duration of the stimuli, that is, the stimulus onset asynchrony (SOA), in the visual probe task indicated that attentional bias occurs in all stages of attention in smokers (Ehrman et al., 2002; Field, Mogg, & Bradley, 2004b). Similarly, former smokers exhibited negative attentional bias in short and long SOAs (200 ms, 500 ms and 2000 ms), but as the SOA became longer, their avoidance became stronger (Peuker & Bizarro, 2013). However, it is not clear how ABM changes automatic and maintained attention. Field et al. (2009) showed that after ABM, attentional bias in smokers was stronger at longer SOA (500 ms) than shorter (50 ms), regardless of whether ABM was employed to attend or to avoid smoking-related pictures. Schoenmakers et al. (2010) conducted a study using ABM to avoid alcohol-related pictures with abstinent alcoholic patients who showed a reduction in attentional bias for longer SOA (500 ms) but not for a shorter one (200 ms). These differences might be attributed to motivation for treatment (only alcoholic patients were in treatment). Nevertheless, the influence of AMB on attention processes requires further investigation. Studies on ABM in smokers have employed a single session of training which did not produce robust changes in attentional bias and other smoking-related outcomes (Attwood et al., 2008; Field et al., 2009; McHugh, Murray, Hearon, Calkins, & Otto, 2010). Five sessions of ABM had no effect on attentional bias, subjective craving and abstinence outcomes in smokers who set a quit day, had 7 sessions of behavioral support and used nicotine patches (Begh et al., 2013). In abstinent alcoholic patients, 5-session ABM reduced attentional bias, produced clinically relevant effects and generalized to new stimuli, but did not reduce craving (Schoenmakers et al., 2010). Craving correlates positively with attentional bias in smokers (Mogg & Bradley, 2002; Moog et al., 2003; Zack, Belsito, Scher, Eissenberg, & Corrigal, 2001), but it is still not clear how ABM changes craving or other variables related to smoking behavior. Positive correlations between craving and attentional bias scores were found just in males trained to attend to smoking-related cues (Attwood et al., 2008). A single session of ABM did not change subjective craving or other clinically relevant variables despite a short-lasting reduction on attentional bias after AMB (Field et al., 2009). Thus, it is not clear if ABM targets implicit automatic attentional process and, if it does, in which conditions (e.g. number of trials) it can help smokers to get better outcomes from smoking cessation therapy. In the present randomized controlled experimental study, we investigated the immediate (24 h) and long-term (1, 6 and

51

12 months) effects of different numbers of sessions of ABM on attentional bias and other smoking-related variables (number of cigarettes smoked per day, carbon monoxide in exhaled air, level of nicotine dependence and urge to smoke) in smokers enrolled in a smoking cessation program. We hypothesize that 3 sessions but not 1 session of ABM would produce negative attentional bias to smokingrelated pictures in all processes of attention, that is, we had a dose response effect expectation. 2. Materials and methods 2.1. Participants Smokers (n = 67) were recruited from 97 participants enrolled in a smoking cessation program (SCP) available to staff and students of a university campaigning for a smoke-free environment. To be included, participants had to meet the following criteria: a) be at least 18 years-old, b) have normal or corrected vision, c) smoke at least 5 cigarettes for more than 30 days (same criteria as Attwood et al., 2008), d) did not undergo any other treatment to quit smoking during the study period, e) did not score as dependent on any drug other than nicotine according to the Alcohol, Smoking and Substance Involvement Screening Test (see Material), f) and did not fulfill criteria for mental disorders on the Self Report Questionnaire (see Material). Participants were randomly allocated to one of three conditions defined according to the number of sessions of ABM: Group avoid 3 (n = 22), performed three sessions of ABM; Group avoid 1 (n = 22) performed two sessions of placebo condition (visual probe task with neutral pictures) and one session of ABM; and group avoid 0 (n = 23) performed two sessions of placebo condition and one session of a standard visual probe task with a different set of smokingrelated pictures to ensure that this group was exposed to a set of smoking-related pictures during ABM. Another group of 22 smokers from the same population who were not participants in the SCP and did not wish to quit smoking had their attentional bias assessed but did not take part in the ABM study. Other criteria to participate were the same. The attentional bias and other smoking-related variables of this group were compared to ABM groups at baseline only. 2.2. Overview of experimental design A mixed experimental design was employed. Between-subjects variables were groups defined by the number of trials of ABM. Within subjects variables were SOAs (50, 500 and 2000 ms) and assessments. All dependent measures were measured in 5 assessments: once before ABM and on follow-ups after the last session of ABM (24 h, 1, 6 and 12 months). 2.3. Materials Scores above seven in the Self Report Questionnaire (SRQ-20; Harding et al., 1980; validated in Brazil by Mari & Willians, 1986) and scores above 16 in the Brazilian version of the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST; Henrique, De Micheli, Lacerda, Lacerda, & Formigoni, 2004) were used as exclusion criteria. Number of cigarettes smoked per day, scores in Fagerström Test for Nicotine Dependence (FTND; Heatherton, Kozlowski, Frecker, & Fagerström, 1991) and level of carbon monoxide (CO) in breath using Smokerlyser (Bedfont, 1993) indicated severity of dependence and cigarette consumption for all assessments. The Questionnaire of Smoking Urges-Brief (Cox, Tiffany, & Christen, 2001; validated in Brazil by Araújo et al., 2007) monitored the urge to smoke for all assessments. The standard visual probe task (MacLeod et al., 1986) was used to measure the attentional bias before and after ABM in all 5 assessments

52

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

(baseline, 24 h, 1, 6 and 12 months after the last ABM session), and a modified visual probe task (MacLeod et al., 2002) was used for ABM. Pictorial stimuli were presented in both tasks in three SOAs (50 ms with 3 frames, 500 ms with 30 frames and 2000 ms with120 frames) to assess automatic and maintained attention. Both tasks were adapted for smoking-related stimuli, programmed and presented using E-Prime version 2.0 software [Psychological Software Tools Inc, USA]. The program ran on a laptop [Toshiba Satellite 1135-s1553, Intel Celeron 2,40Ghz, 512 Mb RAM, HD 30Gb, 4200 rpm; graphics (BGA) Intel 82852, 15” TFT monitor (4:3); Windows XP Home Edition 5.1, build 2600]. Responses were typed in an attached numerical USB-keyboard. In the standard visual probe task, the pictorial stimuli consisted of twelve pairs of smoking-related and matched control pictures obtained from a validated set developed for studies on cue reactivity with smokers (Lopes et al., 2012). Each picture was 80 mm high × 122 mm wide when displayed on the screen, and the distance between the inside edges of each image was 30 mm. In ABM training, the pictorial stimuli comprised a set of 48 pairs of smoking-related and matched control pictures from a bank of pairs obtained by our laboratory. In the placebo condition, 24 pairs of neutral pictures from International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert, 1999) were used. ABM training in avoid 3 group consisted of 36 different pairs of smoking-related and matched-control pictures divided into 3 sessions. In avoid 1 group, training comprised 12 pairs of smoking-related and matched-control pictures in one session and 24 pairs of neutral pictures divided into 2 sessions. Avoid 0 group was trained with 24 pairs of neutral pictures divided into 2 sessions and 12 pairs of smoking-related and matched-control pictures in one session of the standard visual probe task. 2.4. Procedures During the week preceding the first session of the SCP, participants signed an informed consent and performed the baseline assessment. By the informed consent they were informed that they would perform five attention assessments sessions and three attentional training sessions (total of 8 sessions) that would include some questionnaires and an attentional task over 1 year. They were not informed about the aim of the training because we wanted to measure the awareness of the contingencies in ABM training. All procedures were conducted between 8 a.m. and 8 p.m. in a quiet environment. The SCP was a 4-week cognitive behavioral group psychotherapy in line with guidelines of Instituto Nacional do Câncer and Ministério da Saúde (National Institute of Cancer/Ministry of Health). The program was offered during an institutional campaign for a smokefree environment. Participants were referred to an occupational physician if they wished to take pharmacological and/or nicotinereplacement therapy (exclusion criteria). Assessments using standard visual probe task were conducted at baseline, 24 h, 1, 6 and 12 months after the last ABM session. Participants were contacted by telephone to book their assessment sessions and were required to abstain from smoking for at least 1 h before coming to the laboratory. In these sessions, participants provided samples of expired CO, completed the questionnaires individually and performed the standard visual probe task. Experimental procedures of the visual probe task were equivalent to those used in another study (Field, Mogg, & Bradley, 2004a). Instructions were presented on the screen, at a distance of 50 cm from the participant. Participants answered first a question about their craving status: “How strong is your wish to smoke right now?” (craving scale, 0 = none, 9 = extremely strong). In order to ensure the understanding of the task and habituate to the procedure, participants performed 16 training trials and 2 buffer trials of the visual probe task with 4 pairs of neutral stimuli. After that, all participants performed a standard visual probe task, which lasted 10 minutes. In this task, each trial

started with a fixation cross presented for 500 ms in the center of the screen. The cross was replaced by a pair of images presented side-by-side for one of three random SOAs (50, 500 and 2000 ms). After the pair offset, a small arrow (pointing either up or down) appeared in the same location previously occupied by one of the pictures until the participant made a response. Participants were instructed to indicate as fast and accurately as possible the direction of the arrow by pressing a key (8 or 2, covered by a sticker showing the respective arrows) on the numerical keyboard and rest the finger in the middle key (5, covered by a blank sticker) between trials. During the visual probe task, each of the 12 pairs of smokingrelated and matched control pictures were randomly presented four times in each SOA: control pictures on the left side of the monitor replacing the up arrow and down arrow and control pictures on the right side of the monitor replacing the up arrow and down arrow, in a total of 144 trials. The probe (arrow up or down) replaced smoking-related and control pictures with equal frequency and remained on the screen until the participant pressed a correct or incorrect key (see Fig. 1). A separate sample of 22 smokers, who were not participants of the SCP, had a single assessment session to provide a valid comparison to the baseline of the experimental groups enrolled in the SCP. ABM sessions were conducted during the first 2 weeks of the 4-week SCP within at least a 24 hour interval between sessions. In each ABM session, a new set of smoking-related and control pictures were used, and the arrow replaced the control images in 100% of the trials. In placebo sessions, the arrow replaced 100% of the neutral images. Each session of ABM and placebo lasted 30 minutes and had 576 trials, split into 4 blocks of 144 trials (2 positions of the arrow X 2 sides of the screen X 12 pairs X 3 SOAs). Blocks lasted about 6 minutes, and participants had 2 minutes to rest between blocks. Training in avoid 3 group had 1728 ABM trials divided into three sessions. Avoid 1 group had one session (576 trials) of ABM, and the remaining two sessions were placebos. Training in avoid 0 group had 1 session of a standard visual probe task (576 trials to present smoking-related pictures without ABM), and the remaining two sessions were placebo. The order of all sessions was randomized for all groups. After both standard and modified visual probe tasks, participants always rated the pleasantness ("How pleasant is this image for you?") of all images, and relevance ("How relevant is this image for your smoking behavior?") of smoking-related pictures used in the session. The pleasantness scale ranged from −3 (very unpleasant) to +3 (very nice), indicated on the laptop keyboard (letters Z, X, C, V, B, N, and M were replaced by numbers − 3, −2, − 1, 0, + 1, + 2 and + 3, respectively). The relevance scale ranged from 1 (not at all important) to 7 (extremely important), indicated on the laptop keyboard (1 to 7). Each picture was displayed individually (177 mm high × 114 mm wide) for 2 s, followed by 500 ms intervals, and then the pleasantness/ relevance scale remained on the screen until a response was made. At the end of the first post-training assessment (24 h after ABM training) participants of avoid 3 and avoid 1 groups answered questions about the awareness of the contingencies in ABM training (Field et al., 2009), presented on the computer screen. The first open question asked participants to type a description of the relationship between the smoking-related pictures and the location of the probe during ABM sessions. The second question assessed their recognition of these contingencies by asking participants to select the best description from five statements (choosing b indicated awareness): a) “arrows always pointed up if there was a smoking-related picture on the right of the screen”; b) “arrows always appeared on the same side of the screen as the pictures not related to smoking”; c) “arrows always pointed down if there was a smoking-related picture on the left of the screen”; d) “arrows always appeared on the same side of the screen as the smoking-related pictures”; e) “arrows always pointed down if there was a smoking-related picture on the right of the screen”.

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

Standard VPT

+

Modified VPT

SOA 500ms

Placebo VPT

SOA 500ms

+

SOA 50ms 500ms 2000ms

Arrow replaces: 50% SRC 50% non-SRC

53

SOA 50ms 500ms 2000ms

Arrow replaces: 100% non-SRC

+

SOA 500ms

SOA 50ms 500ms 2000ms

Arrow replaces: 100% neutral images

Fig. 1. Scheme of visual probe tasks (VPT).

2.5. Data analysis Attentional bias scores were calculated by subtracting mean reaction time of trials where probes replaced smoking-related pictures from those where probes replaced control pictures. Positive scores indicated attentional bias to smoking-related cues, scores not different from zero according to t-tests indicated absence of attentional bias and negative scores indicated avoidance for smoking-related pictures or attentional bias to control pictures. For the visual probe task, data from one male participant from avoid 3 group was excluded due to errors on more than 20% of the trials. Only trials with errors (1.3%) and outliers (reaction times longer than 2000 ms or shorter than 200 ms) were discarded before analysis. Chi-square was employed to compare groups regarding sex distributions and to compare avoid 1 and avoid 3 regarding awareness of contingencies. A t-test compared attentional bias of participants aware of the contingencies to those unaware. One-way analysis of variance (ANOVA) was used to compare groups at baseline regarding age of participants, number of cigarettes smoked/day, expired CO, duration of regular smoking, age of first cigarette, quitting trials, level of nicotine dependence (FTND) and urge to smoke. ANOVA for repeated measures (GLM/RM) was employed to compare attentional bias of ABM groups enrolled in the SCP with attentional bias of a separate sample of smokers not enrolled in this program within SOAs on baseline. To test the immediate effects of ABM, ANOVA GLM/RM was employed to investigate the effects of group and SOAs on attentional bias between baseline and first post-training assessment (24 h after the last ABM session). All participants attended these assessments (n = 66; 100% retention). Considering drop-outs along other assessments, after 1 month there was a loss of 16 participants (n = 50; 76% retention). In the following assessments, the number of participants remained the same, although the identity of the missing ones was different between assessments. For the analysis of long-term effects of ABM, complete data were obtained from 41 participants (64% retention) in visual probe task (2 participants did not answer the task, and 1 was excluded due to

errors on more than 20% of the trials), and complete data were obtained from 44 participants in remaining items of the assessments. ANOVA GLM/RM was employed to compare progression of a) attentional bias, b) pleasantness and relevance of smoking-related pictures, c) craving status, and d) SCP outcomes (urge to smoke, number cigarettes smoked/day, level of expired CO, level of nicotine dependence) in the 5 assessments. SOA was not taken as an additional within-subject factor in these analyses. Pearson's correlations between attentional bias and other smoking-related variables after ABM were performed. When sphericity was violated in GLM/RM, the Greenhouse–Geisser correction was used, and the degrees of freedom were adjusted. All statistical analyses show effect sizes (d of Cohen for t-test and η 2 for eta-squared), were performed in the Statistical Package for the Social Sciences (SPSS) version 18.0, and significance was set at 5%. 3. Results A summary of the results was previously presented in a conference (Lopes, Pires, & Bizarro, 2013). 3.1. Baseline Table 1 shows the summary data for the demographic (sex, age), smoking-related variables and attentional bias in all SOAs obtained at the baseline of ABM groups and the separate sample of smokers not enrolled in the SCP. Groups did not differ in age or any of the variables related to smoking. There were more women (n = 54) than men (n = 35) in SCP, but this was not a significant difference. Although more women were randomly allocated to avoid 0 than avoid 3 group, a t-test did not reveal significant difference in attentional bias between sexes. All groups showed attentional bias for smoking-related images in all SOAs, as shown by the t-test against zero (M = 27.53, SEM = 9.09; t = 3.54, p = 0.001; d = 0.88). There was no difference between groups, or other interactions between factors. All smokers, wishing or not to quit, showed attentional bias to stimuli related to smoking.

54

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

Table 1 Characteristics of participants and attentional bias scores at baseline.

Age (years) Gender ratio (M:F) Cigarettes/day CO (ppm) Duration of smoking Age of first cigarette Attempts to quit (yes:no) FTND QSU-B AB SOA 50 ms AB SOA 500 ms AB SOA 2000 ms Total AB

Avoid 3

Avoid 1

Avoid 0

Smokers

Total

n = 22

n = 22

n = 23

n = 22

n = 89

46.32 (2.80) 13:09 21.45 (1.62) 22.18 (2.40) 27.64 (2.98) 17.00 (0.59) 18:04 5.91 (0.45) 29.04 (2.68) 32.84 (13.71) 28.08 (19.10) 13.61 (19.61) 24.84 (17.47)

43.77 (2.52) 10:12 19.64 (2.32) 19.95 (1.98) 26.86 (2.84) 17.14 (0.85) 16:06 4.09 (0.45) 34.22 (4.03) 8.91 (11.75) 46.11 (19.73) 36.26 (18.79) 30.43 (16.83)

45.26 (2.51) 02:21 18.48 (1.65) 20.87 (2.05) 29.09 (2.56) 16.17 (0.53) 19:04 4.43 (0.53) 27.60 (3.24) 24.56 (14.07) 35.74 (15.48) 7.14 (18.36) 22.48 (15.97)

44.05 (2.47) 10:12 22.55 (2.23) 23.82 (2.27) 27.73 (2.61) 15.86 (0.89) 14:08 4.73 (0.59) 39.36 (3.85) 22.85 (18.48) 42.84 (5.40) 31.5 (22.29) 32.40 (22.50)

44.85 (1.27) 35:54 20.51 (0.98) 21.70 (1.08) 27.84 (1.35) 16.54 (0.36) 67:22 4.79 (0.26) 32.50 (1.78) 22.29 (7.25) 38.19 (7.49) 22.12 (9.88) 27.53 (9.09)

Note. Values are means ± SEM. Cigarettes/day, number of cigarettes smoker per day; CO (ppm), expired alveolar carbon monoxide; duration of smoking, duration of regular smoking (years); Attempts to quit are shown in proportions of participants who responded yes and no; FTND, Fagerström Test for Nicotine Dependence (higher scores indicate a higher level of nicotine dependence); QSU-B, Brief form of Questionnaire of Smoking Urges; AB, attentional bias; SOA, stimulus onset asynchronies; ms, milliseconds.

Attentional bias became negative (t-test against zero = −4.75; p = 0.0001; d = 1.18) from baseline to first assessment (24 h) after training (F (1,63) = 32.22; p = 0.0001; η2 = 0.36). There was no difference between groups or interaction between groups and the within-group variables. A main effect of SOA was observed (F (2,126) =5.04, p = 0.008; η2 = 0.01) and the interaction between SOA and the time of assessment (F (2,126) = 3.78, p = 0.025; η 2 = 0.006). As SOAs became longer, attentional bias was lower after ABM, according to the linear component of this interaction. In a separate ANOVA of the results after training, the difference among groups (F (1,63) = 5.20; p = 0.008; η2 = 0.01) was due to an avoidance to smoking images seen in avoid 3 group but not avoid 0 group (Bonferroni, p b 0.05). Avoid 1 group was not different from these groups although the negative bias is intermediate. As SOAs became longer, avoidance became stronger in all groups (F (2,126) = 6.51; p = 0.002; η2 = 0.02). There was no significant interaction between group and SOA in this analysis. Regarding contingency awareness, 72% were aware of the relationship between the location of smoking-related pictures and the location of probes during ABM sessions. In avoid 3, 17 (81%) and in avoid 1, 14 (63%) participants were classified as being aware of the experimental contingencies. The rate of awareness was not different between groups (X 2 = 1.60; gl = 1; p = 0.206). Participants who were aware of the contingencies were not different in their attentional bias from those who were unaware. 3.3. Long-term effects of ABM ABM produced long-term effects on attentional bias and this effect was dose-dependent. There was a main effect of time (F (4,152) = 10.14; p = 0.0001; η 2 = 0.05) given that all groups showed a positive attentional bias at baseline (M = 19.55; SEM = 7.90), which became immediately negative in the first post-training assessment (M = − 39.33; SEM = 8.65) and was progressively attenuated, reaching a near-zero level by the last assessment (M = −1.91, SEM = 6.54). Taking all assessments together, the mean attentional bias of avoid 3 group was − 31 ms (SEM = 13.82), a strong avoidance pattern, and different from both avoid 1 (M = − 4.03; SEM = 11.36) and avoid 0 (M = 1.79; SEM = 8.76; F (2.38) = 8.91; p = 0.001, Bonferroni p b 0.05; η 2 = 0.04). A long-lasting avoidance to smoking-related cues was observed in avoid 3 group. In the first assessment after ABM training (24 hs after),

the mean attentional bias of this group was − 76 ms (SEM = 18.86), significantly different from avoid 1 (M = − 27.67; SEM = 9.32) and avoid 0 (M = − 11.08; SEM = 8.41; F (8,152) = 2.41; p = 0.018, Bonferroni p b 0.05; η 2 = 0.004). This strong negative bias persisted in the second assessment (1 month after ABM; M = − 56.25; SEM = 13.41), different from avoid 0 but not from avoid 1. Six months later, although there was no difference between groups in ANOVA, a t-test against zero (M = − 34.90; t = 3.14, p = 0.03; d = 0.78) showed that avoid 3 group was still showing avoidance to smoking-related cues. One year later, bias was close to zero, and thus similar to avoid 1 and avoid 0 (Fig. 2). Pleasantness of all pictures and relevance of smoking-related pictures did not change over assessments. However, control images were considered more pleasant (M = 1.03) than smoking-related ones (−0.54; F (1,39) = 54.42; p = 0.0001; η2 = 0.61). Craving status before and after the visual probe task was not affected by the task and did not change over time. There were no correlations between attentional bias measures and other smoking-related variables after ABM.

60 40 20

Mean attentional bias (ms)

3.2. Immediate effects of ABM

0 -20 -40 -60

Avoid 3 (n=14)

-80

Avoid 1 (n=15) -100 -120

** Baseline

24h

Avoid 0 (n=12)

* 1mo

6mo

12mo

Assessments Fig. 2. Attentional bias in all assessments. ** = Avoid 3 different from both avoid 1 and avoid 0. * = Avoid 3 different from avoid 0.

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

The SCP was successful at reducing the number of cigarettes smoked per day (F (4,164) = 10.83; p b 0.01; η 2 = 0.06), expired CO (F (4,164) = 5.74; p = 0.0001; η 2 = 0.02), and nicotine dependence (F (4,160) = 9.87; p = 0.0001; η 2 = 0.05) in all groups and assessments (Table 2). There was no main effect of group or other significant interactions. Urge to smoke and reported craving were not affected in participants of this program or ABM conditions. The number of participants who had quit smoking was different in each assessment, meaning that abstainers were not the same in every encounter. In the first assessment after ABM, which occurred during the program 1 month after ABM, until 1 week after finishing the program (second assessment), the quitting rate was 12%. Quitting rates increased to 24% during the last two assessments. 4. Discussion The results provided several novel findings extending conclusions from earlier studies about this promising research field. This study was the first to show the long-term effectiveness of ABM in a clinical population of smokers. Before starting an SCP, the participants of ABM groups were not different from other smokers regarding smoking behavior or attentional bias. The assessment at baseline indicated that smokers wishing or not to quit showed positive attentional bias to smoking-related pictures. Engaging in the SCP produced an immediate decrease in attentional bias. In fact, all smokers in treatment developed a conscious avoidance to smoking-related cues. The avoidance was stronger in the longer SOA, indicating that engaging in treatment per se has a greater impact in maintained than in automatic attention. In another study with smokers who were not in treatment to quit but received a single session of ABM (control, trained to avoid, and trained to attend) showed larger positive attentional bias scores at the longer SOA (500 ms) compared to the shorter SOA (50 ms) 24 h after ABM (Field et al., 2009). Longer SOAs produced larger positive bias in smokers not in treatment (Field et al., 2009) and negative bias with smokers in treatment in our study. Engagement in treatment may explain the direction of the effect (either negative or positive) on maintained attention which is more susceptible to motivational influence. Supporting this view are the results of a study with abstinent alcoholic patients who performed 5 sessions of ABM to avoid alcohol-related pictures. Between 3 and 4 days after ABM, attentional bias became negative (about − 35 ms). ABM reduced the difficulty to disengage attention (i.e. produced avoidance in SOA 500 ms) from alcohol-related stimuli but did not influence speeded detection (SOA 200 ms) of alcohol cues (Schoenmakers et al., 2010). The lack of interaction between SOA and group on immediate assessment of attentional bias indicated that ABM had a broad effect on attentional processes. ABM produced a dose-dependent reduction on attentional bias regardless of SOA. Those who performed placebo ABM showed negative attentional bias 24 h after the last session, but this effect was far more pronounced in those who performed 3 sessions of ABM. The present study is the first to show that the number of sessions of ABM is important to produce immediate strong negative attentional bias in smokers in treatment. This is more encouraging than results found in previous studies on ABM in

55

smokers which reported differences when comparing groups trained to avoid and trained to attend to smoking-related cues (Attwood et al., 2008; Field et al., 2009). These differences were influenced by contingency awareness (Attwood et al., 2008) and without evidence of generalization to novel stimuli (Field et al., 2009). The results shown here were independent of contingency awareness and generalized to novel stimuli increasing the ecological validity of the study. The highlight of the present study is the long-lasting avoidance of smoking-related pictures after 3 sessions of ABM. It seems that the number of sessions of ABM is important in maintaining the desired effect, and this encourages the use of ABM as a complementary intervention for smoking cessation programs. Attentional bias in former smokers was shown to be persistently negative for many years of abstinence indicating that negative attentional bias might be a marker for successful quitting (Peuker & Bizarro, 2013). The trend of the three experimental groups over the five assessments (Fig. 2) suggests that ABM produced an efficient and durable change. This implicit training operated throughout attentional processing, including initial orienting (SOA 50 ms). ABM might be an efficient way of developing automatic and maintained avoidance in the context of cognitive–behavioral therapy. Results that were not so encouraging were the effects of ABM in other measures of smoking behavior. The hypothesis that the reduction of bias would produce a greater reduction in smoking-related variables was not confirmed, as number of cigarettes smoked/day, CO and level of nicotine dependence were reduced across assessments in all groups. This suggests that engaging in the program and/or being followed for one year was a stronger cause of this reduction than ABM. Large-scale clinical trials may show the advantage of incorporating ABM into treatment programs (Field, Marhe, & Franken, 2013). On the other hand, a greater number of ABM sessions, perhaps held daily, might be necessary to impact the reduction of other variables related to smoking because of the complexity and diversity involved in smoking habits as well as the number of pairings of environmental cues and the act of smoking. A previous study conducted five ABM sessions with alcoholic patients and achieved preliminary clinical effect (Schoenmakers et al., 2010). Patients underwent a total of 2640 training trials, whereas during the three sessions in this study smokers were exposed to a total of 1728 training trials. Therefore, intensive training sessions can be conducted concurrently and after formal treatment acting to reinforce the motivation to achieve abstinence between sessions, as well as reinforce the maintenance of smoking cessation after the program. Despite the observed reduction in smoking-related variables in all groups, neither reported craving nor urge to smoke changed across assessments or groups, thus they were not associated with ABM. Some studies found association between attentional bias and craving, but only in participants trained to attend to drug related cues (Attwood et al., 2008; Field & Eastwood, 2005), in male but not female smokers (Attwood et al., 2008) or in drinkers that were aware of the experimental contingencies (Field et al., 2007). Self-reported scales of craving and urge to smoke are susceptible to social desirability, whereas attentional bias is unlikely to be subject to this influence (Rooke et al., 2008).

Table 2 Smoking-related variables at baseline and follow-ups.

Cigarettes/day CO (ppm) FTND QSU-B

Baseline

24 hs after ABM

1 month after ABM

6 months after ABM

12 months after ABM

20.45 20.48 5.07 28.93

14.52 15.25 3.60 25.32

13.32 15.61 3.23 24.90

13.07 16.91 3.16 23.81

11.59 14.16 3.19 22.97

(1.25) (1.29) (0.32) (2.30)

(1.25) (1.35) (0.36) (2.14)

(1.18) (1.41) (0.36) (2.21)

(1.28) (1.55) (0.39) (1.95)

(1.37) (1.52) (0.44) (2.14)

Note. Values are means ± SEM (n = 44). Cigarettes/day, number of cigarettes smoker per day; CO (ppm), expired alveolar carbon monoxide; FTND, Fagerstrom Test for Nicotine Dependence (higher scores indicate a higher level of nicotine dependence); QSU-B, Questionnaire of Smoking Urges – Brief (measure of cigarette craving).

56

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57

Studies comparing groups trained to avoid smoking-related cues and their control (Field et al., 2009; Schoenmakers et al., 2010), including the present study, did not find an effect of ABM on the craving for cigarettes, suggesting that although craving and attentional bias have direct influences on smoking behavior, they are probably independent. Other factors including peculiarities of habit, psychological dependence and possible comorbidities are also involved in the decision to acquire abstinence and maintain that behavior change. Moreover, few training sessions are able to modify the bias and may be sufficient to increase but not decrease craving. Both studies on anxiety (Heeren, Reese, McNally, & Philippot, 2012; Krebs, Hirsch, & Mathews, 2010) and drug addiction (Attwood et al., 2008; Field & Eastwood, 2005; Field et al., 2007), found training to increase bias (attend groups) had much more success in modifying bias and other variables. The reason might be that avoidance training demands a greater cognitive effort as it goes against the addiction. However, post-training group differences are not enough to consider that ABM training procedures can lead to behavioral changes. For this objective, it is necessary that the effects of training generalize to real life situations, correlate with improvement in symptoms and maintain this pattern in the long term. The absence of clinical outcomes in the present study could be related, besides the factors mentioned above, to a predisposition to have attentional bias, i.e., a subgroup of individuals who are more easily influenced by external stimuli with particularly high attentional bias at pre-treatment. Some studies that showed no positive bias at baseline also had no effect from attention training (Cowart & Ollendick, 2011; McHugh et al., 2010; Sharpe et al., 2012), and other studies showed the effect of the training only among the participants who already had a bias at baseline (Amir, Taylor, & Donohue, 2011). The authors suggested that this technique may be directed only to people who can be benefited, since the intensity with which subjects are influenced by external stimuli may vary from individual to individual. In the present study, the presence of bias toward cigarette in A1 was not considered as an inclusion criterion for the training, what could be a limitation to this study. Nevertheless both groups of smokers, those who showed a bias close to zero and those who already showed an avoidance pattern to stimuli related to cigarette at baseline, seemed to be benefited from the situation of training, once both increased their avoidance pattern to this class of stimuli. An intensive training may help maintain the negative bias, and this result could have an impact on reducing behavioral dependence, but future studies are necessary to compare the effect of training between smokers with and without bias at baseline to better investigate this issue. Another aspect that is relevant to interpret the changes on attentional bias produced by ABM is the quality of the stimuli used in the task. Despite the wide use of the visual probe task to assess attentional bias in drug users, little is known about the effects of the properties of the images used in the task (Miller & Fillmore, 2010) and about the face and content validity of these images (Lopes et al., 2012). The evaluation of the images regarding relevance to smoking behavior and pleasantness made by participants of this study was equivalent to that of a validation study (Lopes et al., 2012). Smokers tend to judge the valence of these stimuli more positively than nonsmokers (Bradley, Field, Healy, & Mogg, 2008; Moog et al., 2003). In the present study, the control images were considered as more pleasant than smoking-related ones, which in turn were considered unpleasant in all assessments. This was previously observed in former smokers (Peuker & Bizarro, 2013) and young smokers vs. nonsmokers (Lopes et al., 2012). Therefore, national policy on tobacco use (e.g. health warnings in cigarette packs) and restriction on advertisement might be responsible for this socially-desirable self-reported negative evaluation in smokers. On the other hand, the subjective pleasantness (like) might not be as relevant as the motivational salience of cues (want) to continue smoking (Bradley et al., 2008; Robinson & Berridge, 1993).

Taken together, the present results indicate that smokers can develop avoidance to smoking-related pictures in automatic and maintained attention processes for a long period of time if they receive multiple sessions of ABM. Furthermore, it suggests that engaging in a formal smoking cessation program produces changes in motivation detected in long SOAs in the visual probe task. However, it is not known how this change in attentional bias impacts smoking behavior. Perhaps the lab-context of data collection and the number of assessments or participants required to detect changes in smoking behavior in a reliable way were limitations of our experimental design. The sample size for this study may have been small for the design, and would only be sufficiently powered to detect quite large effect sizes. Thus, the absence of outcomes findings have to be interpreted cautiously because it may be just a power issue. In other words, smoking behavior might have been affected by ABM, but we could have not detected it. In order to do so, the number of participants in future ABM studies should be higher, but the practicalities of conducting multiple lab-based interventions are a serious limitation. These limitations might be partially overcome by ecological momentary assessments (Shiffman, 2009) and the use of internet-based ABM to deliver multiple sessions of ABM. References Amir, N., Beard, C., Burns, M., & Bomyea, J. (2009). Attention modificationprogram in individuals with generalized anxiety disorder. Journal of Abnormal Psychology, 118, 28–33, http://dx.doi.org/10.1037/a0012589. Amir, N., Taylor, C. T., & Donohue, M. C. (2011). Predictors of response to an attention modification program in generalized social. Journal of Consulting and Clinical Psychology, 79, 533–541, http://dx.doi.org/10.1037/a0023808. Araújo, R., Margareth, S., Moraes, J., Pedroso, R., Port, F., & Castro, M. (2007). Validação da versão brasileira do Questionnaire of Smoking Urges-Brief. Revista de Psiquiatria Clínica, 34, 166–175, http://dx.doi.org/10.1590/S0101-60832007000400002. Attwood, A. S., O'Sullivan, H., Leonards, U., Mackintosh, B., & Munafo, M. R. (2008). Attentional bias training and cue reactivity in cigarette smokers. Addiction, 103, 1875–1882, http://dx.doi.org/10.1111/j.1360-0443.2008.02335.x. Bedfont, S. (1993). Operator´s manual for mini and micro smokerlyzers. Begh, R., Munafo, M. R., Shiffman, S., Ferguson, S., Nichols, L., Mohammed, M., et al. (2013). A double-blind randomized controlled trial evaluating the efficacy of attentional retraining on attentional bias and craving in smokers attempting cessation. Proceedings of 2013 International Meeting of Society for Research on Nicotine and Tobacco, Boston, Massachusetts, USA (pp. 233). Bradley, B., Field, M., Healy, H., & Mogg, K. (2008). Do the affective properties of smoking-related cues influence attentional and approach bias in cigarette smokers? Journal of Psychopharmacology, 21, 1–9, http://dx.doi.org/10.1177/ 0269881107083844. Bradley, B., Field, M., Mogg, K., & De Houver, J. (2004). Attentional and evaluative biases for smoking cues in nicotine dependence: Component processes of biases in visual orienting. Behavioral Pharmacology, 15, 29–36, http://dx.doi.org/ 10.1097/01.fbp.0000113331.49506.b5. Cowart, M. J. W., & Ollendick, T. H. (2011). Attention training in socially anxious children: A multiple baseline design analysis. Journal of Anxiety Disorders, 25, 972–977, http://dx.doi.org/10.1016/j.janxdis.2011.06.005. Cox, L. S., Tiffany, A. G., & Christen, S. T. (2001). Evaluation of the brief questionnaire of smoking urges (QSU-brief) in laboratory and clinical settings. Nicotine & Tobacco Research, 3, 7–16, http://dx.doi.org/10.1080/14622200020032051. Ehrman, R., Robbins, S., Bromwell, M., Lankford, M., Monterosso, J., & O’Brien, C. (2002). Comparing attentional bias to smoking cues in current smokers, former smokers, and non-smokers using a dot-probe task. Drug and Alcohol Dependence, 67, 185–191, http://dx.doi.org/10.1016/S0376-8716(02)00065-0. Field, M., & Cox, W. M. (2008). Attentional bias in addictive behaviors: A review of its development, causes and consequences. Drug and Alcohol Dependence, 97, 1–20, http://dx.doi.org/10.1016/j.drugalcdep.2008.03.03. Field, M., Duka, T., Eastwood, B., Child, R., Santarcangelo, M., & Gayton, M. (2007). Experimental manipulation of attentional biases in heavy drinkers: do the effects generalise? Psychopharmacology, 192, http://dx.doi.org/10.1007/s00213-0070760-9. Field, M., Duka, T., Tyler, E., & Schoenmakers, T. (2009). Attentional bias modification in tobacco smokers. Nicotine & Tobacco Research, 11, http://dx.doi.org/10.1093/ntr/ ntp067. Field, M., & Eastwood, B. (2005). Experimental manipulation of attentional bias increases the motivation to drink alcohol. Psychopharmacology, 183, 350–357, http://dx.doi.org/10.1007/s00213-005-0202-5. Field, M., Marhe, R., & Franken, I. (2013). The clinical relevance of attentional bias in substance use disorders. CNS Spectrums, 1–6, http://dx.doi.org/10.1017/ S1092852913000321. Field, M., Mogg, K., & Bradley, B. (2004). Alcohol increases cognitive biases for smoking cues in smokers. Psychopharmacology, 180, 63–72, http://dx.doi.org/10.1007/ s00213-005-2251-1.

F.M. Lopes et al. / Journal of Substance Abuse Treatment 47 (2014) 50–57 Field, M., Mogg, K., & Bradley, B. (2004). Eye movements to smoking cues: Effects of nicotine deprivation. Psychopharmacology, 173, 116–123, http://dx.doi.org/10.1007/ s00213-003-1689-2. Field, M., Mogg, K., Zetteler, J., & Bradley, B. (2004). Attentional biases for alcohol cues in heavy and light social drinkers: The roles of initial orienting and maintained attention. Psychopharmacology, 176, 88–93, http://dx.doi.org/10.1007/s00213-0041855-1. Focchi, G., & Braun, I. (2005). Tratamento farmacológico do tabagismo. Revista de Psiquiatria Clínica, 32, 259–266. Franken, I. H. (2003). Drug craving and addiction: Integrating psychological and neuropsychopharmacological approaches. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 27, 563–579, http://dx.doi.org/10.1016/S0278-5846(03) 00081-2. Harding, T. W., Arango, M. V., Baltazar, J., Climent, C. E., Ibrahim, H. H., & Ignacio, L. L. (1980). Mental Disorders in primary health care: A study of their frequency and diagnosis in four development countries. Psychological Medicine, 10, 231–241, http://dx.doi.org/10.1017/S0033291700043993. Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerström, K. O. (1991). The Fagerström test for nicotine dependence: A revision of the Fagerström Tolerance Questionnaire. British Journal of Addiction, 86, 1119–1127, http://dx.doi.org/10.1111/j. 1360-0443.1991.tb01879.x. Heeren, A., Reese, H., McNally, R., & Philippot, P. (2012). Attention training toward and away from threat in social phobia: Effects on subjective, behavioral, and physiological measures of anxiety. Behaviour Research and Therapy, 50, 30–39, http://dx.doi.org/10.1016/j.brat.2011.10.005. Henrique, I. F., De Micheli, D., Lacerda, R. B., Lacerda, L. A., & Formigoni, M. L. (2004). Validação da versão brasileira do teste de triagem do envolvimento do álcool, cigarro e outras substâncias (ASSIST). Revista da Associação Médica Brasileira, 50, 199–206, http://dx.doi.org/10.1590/S0104-42302004000200039. Krebs, G., Hirsch, C. R., & Mathews, A. (2010). The effect of attention modification with explicit vs. minimal instructions on worry. Behaviour Research and Therapy, 48, 251–256, http://dx.doi.org/10.1016/j.brat.2009.10.009. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1999). International affective picture system (IAPS): Technical manual and affective ratings. Gainesville, Florida: The Center for Research in Psychophysiology. Lopes, F. M., Peuker, A. C., & Bizarro, L. (2008). Viés atencional em fumantes. Psico, 39, 280–288 (Retrieved from http://revistaseletronicas.pucrs.br/ojs/index.php/ revistapsico/article/view/4462). Lopes, F. M., Pires, A., & Bizarro, L. (2013). Attentional bias modification in smokers trying to quit: a longitudinal study. Proceedings of 2013 International Meeting of Society for Research on Nicotine and Tobacco, Boston, Massachusetts, USA (pp. 222). Lopes, F. M., Wagner, F., Peuker, A. C., Cunha, S., Trentini, C., & Bizarro, L. (2012). Face and content validity of smoking-related and matched control pictures. Avances en Psicologia Latinoamericana, 30, 209–220, http://dx.doi.org/10.1177/ 0269881107083844. MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal of Abnormal Psychology, 95, 15–20, http://dx.doi.org/10.1037/0021-843X.95.1.15. MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, L. (2002). Selective attention and emotional vulnerability: Assessing the causal basis of their association through the experimental manipulation of attentional bias. Journal of Abnormal Psychology, 111, 107–123, http://dx.doi.org/10.1037//0021-843x.111.1.107.

57

Mari, J., & Willians, P. A. (1986). A validity study of a psychiatric screening questionnaire (SRQ-20) in primary care in the city of São Paulo. British Journal of Psychiatry, 148, 23–26, http://dx.doi.org/10.1192/bjp.148.1.23. McHugh, R. K., Murray, H. W., Hearon, B. A., Calkins, A. W., & Otto, M. W. (2010). Attentional bias and craving in smokers: The impact of a single attentional training session. Nicotine & Tobacco Research, 12, 1261–1264, http://dx.doi.org/10.1093/ntr/ ntq171. Miller, M. A., & Fillmore, M. T. (2010). The effect of image complexity on attentional bias towards alcohol-related images in adult drinkers. Addiction, 105, 883–890, http://dx. doi.org/10.1111/j.1360-0443.2009.02860.x. Mogg, K., & Bradley, B. (2002). Selective processing of smoking- related cues in smokers: Manipulation of deprivation level and comparison of three measures of processing bias. Journal of Psychopharmacology, 16, 385–392, http://dx.doi.org/ 10.1177/026988110201600416. Moog, K., Bradley, B., Field, M., & De Houwer, J. (2003). Eye movements to smoking related pictures in smokers: Relationship between attentional biases and implicit and explicit measures of stimulus valence. Addiction, 98, 825–836, http://dx.doi.org/10.1046/j.1360-0443.2003.00392.x. Peuker, A. C., & Bizarro, L. (2013). Attentional avoidance of smoking cues in former smokers. Journal of Substance Abuse Treatment, 45, 517–522. Peuker, A. C., Lopes, F., & Bizarro, L. (2009). Viés atencional no abuso de drogas: teoria e método. Psicologia: Teoria e Pesquisa, 25, 603–609. Robbins, S., & Ehrman, R. (2004). The role of attentional bias in substance abuse. Behavioral and Cognitive Neuroscience Reviews, 3, 243–260, http://dx.doi.org/10.1177/ 1534582305275423. Robinson, T. E., & Berridge, K. C. (1993). The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Research Reviews, 18, 247–291, http://dx.doi.org/10.1016/0165-0173(93)90013-P. Rooke, S., Hine, D., & Thorsteinsson, E. (2008). Implicit cognition and substance use: A meta-analysis. Addictive Behaviors, 33, 1314–1328, http://dx.doi.org/10.1016/ j.addbeh.2008.06.009. Schmidt, N. B., Richey, J. A., Buckner, J. D., & Timpano, K. R. (2009). Attention training for generalized social anxiety disorder. Journal of Abnormal Psychology, 118, 5–14, http://dx.doi.org/10.1037/a0013643. Schoenmakers, T. M., De Bruin, M., Lux, I. F., Goertz, A. G., Van Kerkhof, D. H., & Wiers, R. W. (2010). Clinical effectiveness of attentional bias modification training in abstinent alcoholic patients. Drug and Alcohol Dependence, 109, 30–36, http://dx.doi. org/10.1016/j.drugalcdep.2009.11.022. Sharpe, L., Ianiello, M., Dear, B. F., Perry, K. N., Refshauge, K., & Nicholas, M. K. (2012). Is there a potential role for attention bias modification in pain patients? Results of 2 randomised, controlled trials. Pain, 153, 722–731, http://dx.doi.org/10.1016/j.pain. 2011.12.014. Shiffman, S. (2009). Ecological momentary assessment (EMA) in studies of substance use. Psychological Assessment, 21, 486–497, http://dx.doi.org/10.1037/a0017074. Waters, A. J., Shiffman, S., Bradley, B., & Mogg, K. (2003). Attentional shifts to smoking cues in smokers. Addiction, 98, 1409–1417, http://dx.doi.org/10.1046/j.13600443.2003.00465.x. Zack, M., Belsito, L., Scher, R., Eissenberg, T., & Corrigal, W. A. (2001). Effects of abstinence and smoking on information processing in adolescent smokers. Psychopharmacology, 153, 249–257, http://dx.doi.org/10.1007/s002130000552.