Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Dec 15.
Published in final edited form as: Drug Alcohol Depend. 2019 Dec 16;207:107803. doi: 10.1016/j.drugalcdep.2019.107803

Inhibitory-Control Training for Cocaine Use Disorder and Contingency Management for Clinic Attendance: A Randomized Pilot Study of Feasibility, Acceptability and Initial Efficacy

Craig R Rush 1,2,3,✉, Justin C Strickland 2, Erika Pike 1, Christina R Studts 4, William W Stoops 1,2,3,5
PMCID: PMC6953399  NIHMSID: NIHMS1547964  PMID: 31864166

Abstract

Background.

Cocaine abusers have impaired inhibitory control. This study determined the feasibility, acceptability, and initial efficacy of inhibitory-control training to cocaine or neutral images in cocaine use disorder patients.

Methods.

Participants were randomly assigned to inhibitory-control training to cocaine (N=20) or neutral (N=20) images. Feasibility was assessed by percent of patients eligible for participation after a behavioral qualification session, time-to-target enrollment, percent of clinic visits attended, percent of participants who completed 80% or more training sessions, and percent of follow-up visits attended. Acceptability was determined using a Treatment Acceptability Questionnaire. Initial efficacy was determined during training and a follow-up phase with urine samples tested qualitatively and quantitatively for cocaine. Participants in both conditions received monetary incentives delivered on an escalating schedule for clinic attendance.

Results.

The groups were well matched and no differences on demographic or substance use variables were observed. Attendance was stable during the treatment period with high overall attendance in both groups (average sessions attended: cocaine images group = 97%; neutral image group = 90%). No group differences were observed in the percentage of follow-up sessions attended (95% for the cocaine-image group; 88% of neutral image group). Ratings on the Treatment Acceptability Questionnaire were high (i.e., mean scores ≥ 80 for all items rated on 101-unit visual analog scales). Participants in the cocaine- and neutral-image conditions did not differ significantly in terms of cocaine use during the training nor follow-up phase. Inhibitory-control training improved stop signal performance but not delay discounting.

Conclusion.

The procedures were feasible and acceptable. Inhibitory-control training to cocaine images did not reduce cocaine use relative to the neutral image training condition. The inability to detect significant differences in cocaine use across the groups is not surprising given the small sample size. More research is needed to determine the utility of inhibitory-control training for cocaine use disorder. Future trials should determine whether inhibitory-control training to cocaine images augments the efficacy of other behavioral interventions.

Keywords: Cocaine, Clinical Trial, Inhibitory Control, Impulsivity, Treatment

1. Introduction

Cocaine use and cocaine use disorder are unrelenting public health concerns. For example, over two million Americans were current (i.e., past month) cocaine users in 2017 and approximately one million met cocaine use disorder criteria (Center for Behavioral Health Statistics, 2018). Cocaine use produces multiple health problems and is associated with numerous negative health consequences including overdose death (Butler et al. 2017; Center for Disease Control and Prevention, 2018; Havakuk et al. 2017). Recent data suggest a nearly three-fold increase in cocaine-related overdose deaths between 2013 and 2017 (Centers for Disease Control and Prevention, 2018). Another study showed cocaine overdose rates in Black men and women are comparable to opioid overdose rates in non-Black men and women in the United States (Shiels et al., 2018).

Behavioral therapies are considered the “standard of care” for cocaine use disorder, but relapse rates are high (Covi et al., 2002; Higgins et al., 2004; McKee et al., 2007; Vocci and Montoya, 2009). Despite the conduct of more than 100 blinded, randomized, fully placebo-controlled studies of over 60 compounds, the Food and Drug Administration (FDA) has not approved a pharmacotherapy for cocaine use disorder (Czoty et al., 2016). These findings, along with the epidemiology data described above, underscore the need to identify novel interventions for cocaine use disorder.

Our laboratory has conducted a series of studies designed to systematically determine the role of impaired inhibitory control in cocaine use disorder (Alcorn et al., 2017; Fillmore and Rush, 2002, 2006; Pike et al., 2013, 2015, 2017; Vansickel et al., 2008). In the seminal study, we compared cocaine users (N=22) and controls (N=22) (Fillmore and Rush, 2002). Response inhibition and execution were measured by a Stop-signal paradigm using a choice reaction time task that engaged subjects in responding to Go signals when Stop signals occasionally informed them to inhibit the response. Cocaine users displayed significantly poorer ability to inhibit their behavioral responses than controls. Specifically, cocaine users required more time to inhibit responses to Stop-signals and displayed a lower probability of inhibiting their responses. Cocaine users did not differ from controls in their ability to execute responses as measured by their speed and accuracy of responses to Go signals. Deficits in inhibitory control in cocaine users has been systematically replicated by several other investigative teams as summarized in a literature review (Czermainski et al., 2017).

We extended our previous work on inhibitory control by showing that cocaine users fail to inhibit pre-potent responses following cocaine images significantly more often than following neutral images using the Attentional Bias-Behavioral Activation task (ABBA) (Pike et al., 2013, 2015, 2017). In the ABBA task, cocaine images serve as a Go cue in that they generally predict when a response will be required (i.e., on 80% of trials). The trials of interest are the 20% when a participant sees a cocaine image, is prepared to respond, but must inhibit responses at the last moment. In the initial study, separate groups of cocaine users completed the ABBA task in which they saw cocaine or neutral images as the Go cue (Pike et al., 2013). Participants in the cocaine image Go cue condition (N=15) had a significantly higher proportion of inhibitory failures to the No-Go target than their counterparts (N=15) in the neutral cue condition.

The findings described above are important because they identified a specific deficit in inhibitory control that could contribute to cocaine use disorder. Impaired inhibitory control in the presence of drug cues suggests that individuals would have an especially difficult time avoiding or discontinuing drug use in the presence of drugs or paraphernalia, thus contributing to continued use. Specifically, the presence of drug paraphernalia may signal the presence of a drug, thus making it more difficult for individuals to inhibit initiation of drug use. The ability to inhibit responding in the presence of drug cues could also affect the ability or inability to stop taking drugs once use has already been initiated, as the discontinuation of use would have to occur in the presence of both interoceptive and exteroceptive drug cues, as well as drug-induced impairment of inhibitory control (Fillmore et al., 2002). Research on the relationship between drug cues and substance use suggest that drug cues are associated with motivational states to obtain or use drugs (Ryan, 2002). Thus, individuals with impaired inhibitory control that is exacerbated by drug cues may have an especially difficult time resisting or discontinuing drug use in the presence of drug-related cues, which may contribute to the high incidence of relapse.

Consistent with the findings described above, response inhibition is associated with meaningful clinical outcomes (e.g., Goncalves et al., 2014; Verdejo-García et al. 2012; Vergara-Moragues et al., 2017; see Verdejo-García et al. 2016 for a review). The first trial determined the association between response inhibition as measure by a Stroop task in cocaine dependent patients enrolled in Therapeutic Communities (Verdejo-García et al. 2012). Stroop inhibition was correlated with treatment retention such that time in treatment decreased as inhibition worsened. The second trial also used cocaine dependent patients enrolled in Therapeutic Communities (Vergara-Moragues et al., 2017). Patients who failed to complete treatment had worse response inhibition as measured by a Stroop test relative to those who completed. In another study, one-month of abstinence resulted in improved inhibitory control in cocaine-dependent patients (Goncalves et al., 2014). The authors of a recent trial concluded improved response inhibition may contribute to the efficacy of cognitive behavioral therapy for cocaine used disorder (Devito et al., 2018). Specifically, methadone-maintained patients with cocaine-use disorder participated in a randomized trial of treatment as usual alone or combined computer-based cognitive behavioral therapy. Patients completed a Drug Stroop task before and after treatment. Patients able to achieve at least 3 weeks of abstinence showed the largest improvement on the Drug Stroop task (i.e., reduced slowing to drug-related stimuli relative to neutral stimuli).

Despite documented deficits in inhibitory control and their relationship with clinical outcomes in cocaine use disorder patients, we are unaware of published studies that used inhibitory-control training to reduce cocaine use or improves treatment outcomes. The goal of this study, therefore, was to determine the feasibility, acceptability, and initial efficacy of inhibitory-control training to cocaine or neutral images. Feasibility was assessed by percent of patients eligible for participation after completing the behavioral qualification session, time-to-target enrollment, percent of clinic visits attended, percent of participants who completed 80% or more training sessions, and percent of follow-up visits attended. Acceptability was determined using a Treatment Acceptability Questionnaire. Following inhibitory-control training to cocaine or neutral images, participants returned thrice weekly for qualitative and quantitative urinalysis. During each of these visits, the assigned inhibitory-control training was repeated. We hypothesized inhibitory-control training to cocaine and neutral images would be feasible and acceptable. We further hypothesized inhibitory-control training to cocaine images would reduce the number of cocaine (i.e., benzoylecgonine) positive urine samples relative to the neutral image training condition.

2. Materials and Methods

2.1. Sample and Behavioral Qualification Session

Participants were recruited using community posting, print advertisement, and word-of-mouth to complete an 11-week randomized study (i.e., 2 weeks of abstinence induction; 6 weeks of inhibitory-control training; and 3 weeks of follow up). Study inclusion criteria were: 1) 18 years of age or older; 2) self-reported cocaine use; 3) provision of a cocaine- or benzoylecgonine-positive urine sample during screening; 4) meeting DSM-IV criteria for cocaine abuse or dependence based on a structured clinical interview (SCID), 5) treatment seeking; and 6) ability and willingness to commit to the protocol. Individuals with current or past medical or psychiatric illness that would interfere with study participation (e.g., physical dependence on any drug requiring medically managed detoxification) were excluded. Those meeting the inclusion criteria also participated in a behavioral qualification session in which they completed a cocaine-cue ABBA task (Pike et al., 2013, 2015, 2017). The flow of study enrollment is depicted in the CONSORT diagram (Figure 1). The University of Kentucky Institutional Review Board (IRB) reviewed all study procedures and all participants provided informed consent prior to participation.

Figure 1.

Figure 1.

Open in a new tab

CONSORT Flow Diagram.

Individuals were required to have at least 8% response errors to cocaine Go cues on the ABBA to be randomized to a treatment condition. This stipulation was included to ensure all participants had pre-morbid inhibitory control impairment. Due to miscommunication, three patients were inadvertently included that did not meet this criterion. Two were included in the ICT-Cocaine group while one was part of the ICT-Neutral group. Each of these individuals made 4% response errors.

2.2. General Procedures

Patients were required to visit the research clinic three times per week (Monday, Wednesday, Friday) for an eight-week period. The first two weeks comprised an Abstinence Induction phase. During the Abstinence Induction phase, participants who provided cocaine-free urine samples (based on qualitative testing) earned bonus payments. The first cocaine-free sample resulted in a $10 bonus payment, and the bonus amount increased by $10 for each subsequent consecutive cocaine-free urine sample. Participants providing three consecutive cocaine-free urine samples received an additional $20 bonus. These bonuses were paid to participants on the day samples met each criterion. The Abstinence Induction phase was based on prior contingency management research demonstrating it to be an effective behavioral approach for reducing substance use in a variety of populations (see reviews in Davis et al., 2016; McPherson et al., 2018).

Participants also received modules from Yale’s Compliance Enhancement: A Manual for Psychopharmacotherapy of Drug Abuse and Dependence, delivered by a trained masters-level drug counselor during the Abstinence Induction phase (Carroll et al., 1999). This manual-guided supportive behavioral treatment was conducted weekly. The drug counselor was masked to the assigned condition for each participant. Compliance Enhancement was designed to facilitate compliance with study procedures and provide clinical recommendations for reducing or ceasing cocaine use. We selected this platform rather than a more robust behavioral intervention like Cognitive Behavioral Therapy so as not to overshadow the effects of inhibitory-control training in this initial pilot study.

Prior to the first training session in Week 3, participants were assigned using adaptive covariate randomization to one of two conditions: Inhibitory-control training to inhibit responding to cocaine (ICT-Cocaine) or neutral images (ICT-Neutral). Adaptive covariate randomization is used to minimize both treatment and covariate imbalance, especially in smaller clinical trials (Hedden et al., 2006). In this study, adaptive covariate randomization was intended to ensure balance in number of participants assigned to each condition, as well as on two important covariates: sex and urine sample test result on the last day of the Abstinence Induction phase (positive or negative, with missing samples coded as positive). Participants then completed the assigned inhibitory-control training sessions (see Inhibitory-Control Training for details). Training tasks were completed at each study visit from Weeks 3 to 8. Delay discounting and Stop-signal tasks were performed at baseline and at study follow up (see details below). Additional cognitive-behavioral measures were also collected at varied timepoints throughout the training session (data to be presented elsewhere). At each study visit, participants provided observed urine specimens tested for cocaine.

During the training phase, participants in both conditions also received monetary incentives delivered on an escalating schedule for continued attendance. In addition to a $40 session payment at all visits, participants received an additional $10 attendance incentive at the first clinic visit. This attendance incentive increased by $3 for each consecutive visit, with an additional $10 bonus delivered for each non-overlapping set of three consecutive visits. Missed visits reset the incentive back to the initial $10; however, three consecutive visits following a missed visit returned the attendance incentive to the amount prior to the miss.

Three weekly follow-ups were completed at the end of the initial eight-week period. During the weekly follow-ups, participants returned to the research clinic and provided an observed urine sample tested using qualitative and quantitative measures for the presence of cocaine.

2.3. Inhibitory-Control Training

The inhibitory-control training procedure was based on a modified version of the ABBA (Pike et al., 2013; Weafer and Fillmore, 2012). This computerized task takes approximately 15 minutes to complete and consists of five blocks of 50 trials. A trial involves the following sequence: 1) a fixation point (+) is presented for 800 ms, 2) a blank white screen is presented for 500 ms, 3) a cue image (cocaine or neutral; 15 cm × 11.5 cm) is presented in the center of the computer monitor against a white background for one of five stimulus onset asynchronies (SOA) (i.e., 100, 200, 300, 400, 500 ms), and 4) a Go (green rectangle) or No-Go target (blue rectangle) is displayed until 5) a response occurs (i.e., the participant presses the forward slash [/]) or 1000 ms elapse. A 700 ms interval separated all trials.

Traditionally in the ABBA, 20% of trials in which participants are prepared to respond (i.e., following presentation of a Go cue) require them to inhibit instead, because a No-Go target is presented (Pike et al., 2013). To be consistent with previous studies showing positive effects of similar inhibitory-control training in reducing alcohol use (Houben et al., 2011, 2012), the task was modified such that No-Go cues predicted No-Go targets 100% of the time and Go cues predicted Go targets 100% of the time. Therefore, for the ICT-Cocaine group, the No-Go cues were cocaine images, and participants were trained to inhibit responses preceded by cocaine images 100% of the time. Go cues in the inhibitory-control training task were matched-neutral images. For the ICT-Neutral group, Go and No-Go cues were vertical and horizontal rectangles, respectively. This control condition was selected to avoid ethical concerns associated with training a control group with cocaine images that were 100% predictive of a Go response (i.e., the reverse of the ICT-Cocaine group).

Each training task took approximately 15 minutes to complete and participants completed the task three times per session. By the end of the first training day, participants were required to respond with no more than 5% errors on the training task in order to continue with the study. To engage participants in the training and incentivize completion, they were able to earn a $15 bonus per training day (i.e., $5 each time they completed the task). Participants were informed that they would lose $0.50 of this bonus for each mistake on the training task (i.e., failing to respond when required to or responding when required to inhibit). As it was not possible to mask the research assistant who was conducting the training with participants to the training condition delivered, one research assistant was assigned to conduct training tasks and a different research assistant was masked to study condition and conducted the remaining outcome assessments.

2.4. Study Measures

2.4.1. Feasibility.

The primary feasibility measures were: 1) percent of patients eligible for participation after completing the behavioral qualification session, 2) time-to-target enrollment, 3) percent of clinic visits attended, 4) percent of participants who completed 80% or more training sessions, and 5) percent of follow-up visits attended.

2.4.2. Acceptability.

The primary acceptable measure was a modified version of the Treatment Acceptability Questionnaire (TAQ) completed at the end of the training phase (Raiff et al., 2013). Responses to the individual questions were assessed using a 100-mm visual analog scale (1 = lowest satisfaction, 100 = greatest satisfaction). Participants were asked: 1) overall satisfaction; 2) acceptability of receiving vouchers; 3) acceptability of thrice weekly clinic visits; 4) acceptability of providing observed urine samples; 5) acceptability of the computer program to deliver inhibitory-control training; 6) acceptability of additional inhibitory-control training delivered throughout the study; and 7) acceptability of receiving compliance enhancement. Additional acceptability measures evaluated whether participants believed that their program needs were met (response options: did not meet my needs at all, only met a few of my needs, met most of my needs, met almost all of my needs), satisfaction with the program (Quite Dissatisfied, Dissatisfied, Mildly Satisfied, Very Satisfied), and willingness to participate again in the future (Definitely Not, Probably Not, Probably So, Definitely So). One participant in the ICT-Neutral group had missing TAQ data on this measure because they did not attend the final training session or any follow-up visits.

2.4.3. Initial Efficacy.

Initial efficacy was assessed using observed urine samples collected at thrice weekly clinic visits during training and once weekly during follow up. Urine samples were analyzed using qualitative and quantitative test results for the cocaine-metabolite benzoylecgonine. Qualitative analyses dichotomized urine samples based on testing strips with a 300-ng/mL benzoylecgonine testing threshold (CLIAwaived, Inc.; San Diego California). Quantitative urine testing was conducted using ELISA analysis for benzoylecgonine (BQ Kits, Inc; San Diego California).

Initial efficacy was further assessed with a delay discounting and Stop-signal task completed at baseline and during study follow-up visits. Delay discounting was evaluated using an adjusting amount task presenting choices between $10 available after specified delays (i.e., 1, 2, 30, 180 and 365 days) and a smaller amount available immediately (e.g., Would you rather have $10 in 30 days or $2 now?) (Richards et al., 1999). Indifference values were determined reflecting the smallest amount of money an individual would select to receive immediately instead of the delayed standard amount ($10) at each specified delay. A modified version was used to evaluate delay discounting rates for cocaine, in which participants chose between $10 of cocaine available at delay and a smaller amount available immediately. All selections were hypothetical, and no money or cocaine was delivered. The primary outcome from delay discounting tasks was area under the curve (AUC) for the indifference points by delay (Myerson et al., 2001), in which lower values indicate steeper discounting.

Finally, a Stop-signal task was used to evaluate response inhibition (Fillmore and Rush, 2001; Fillmore and Vogel-Sprott, 1999). In this task, participants were instructed to identify and respond to letters presented on a computer screen (X and O). A Stop-signal (500 ms 900 Hz tone) was presented on approximately 30% of trials to indicate that a participant should withhold a response. Stop-signals were presented at one of five delays (10, 70, 150, 230, and 300 ms). The outcome was the mean proportion of inhibitory failures following the Stop-signal. Participants completed the Stop-signal task twice at baseline and twice at each follow-up session to help isolate within-session practice effects.

2.5. Data Analysis

The current study was designed as a proof-of-concept feasibility and acceptability study evaluating the implementation of an inhibitory-control training protocol for cocaine use disorder. Therefore, the sample size was not determined based on power to detect clinically meaningful effects. Instead, the target sample size was selected to allow a realistic examination of study design and implementation, as well as to facilitate identification of potential modifications needed for a larger, hypothesis-testing efficacy trial (see Leon et al., 2011 for considerations in feasibility of study design).

Demographic and substance use history were first compared between conditions using independent samples t-tests (continuous) or Fisher’s exact tests (categorical). Additional tests showed that there were no significant differences between randomized participants and those not randomized due to not meeting criteria assessed in the behavioral qualification session, save baseline responding on the ABBA task.

Feasibility of clinic visits was evaluated over the eight-week period. Overall participation rates were compared between groups using independent samples t-tests. Acceptability ratings on the TAQ were also compared between groups using independent samples t-tests. Delay discounting AUC values were evaluated with a 2 × 2 × 2 mixed ANOVA with Group (ICT-Cocaine versus ICT-Neutral) as a between-subject factor and Visit (Baseline versus Follow-Up 1) and Commodity (Money versus Cocaine) as within-subject factors. Stop-signal data were evaluated using a 2 × 2 × 2 mixed ANOVA with Group (ICT-Cocaine versus ICT-Neutral) as a between-subject factor and Visit (Baseline versus Follow-Up 1) and Time (Time 1 versus Time 2 completion) as within-subject factors. One participant in the ICT-Cocaine condition and three in the ICT-Neutral condition did not complete the first follow-up session that included these cognitive-behavioral measures; these four participants were not included in this analysis.

Initial efficacy was evaluated using generalized linear mixed effect models with random intercepts (random effect of participant). These models tested the fixed effect of Group and its interaction with a linear effect of time (Group × Week interaction). Separate models tested qualitative (dichotomous) and quantitative (continuous) results during the six-week training phase. An additional set of models tested urine results during the three follow-up visits. Missing urine samples were treated as positive in all analyses. Sensitivity analyses treating these data as missing revealed similar results (data not shown). Benzoylecgonine results were log-transformed prior to analysis to correct for substantive variable skew. All analyses were conducted using R statistical language and the nlme and lme4 packages (Bates et al., 2015; Pinheiro et al., 2017).

3. Results

3.1. Demographics and Substance Use History

Table 1 contains demographic and substance use history by treatment condition. Overall, randomized participants were an average of 49.4 years old with 12.5 years of education. A majority of participants were male (70%) and African American (77.5%) and reported current tobacco cigarette use (92.5%). Participants reported an average of 13.0 days of cocaine use in the past month and had been using for approximately 17.6 years. No significant differences on demographic or substance use variables were observed between the ICT-Cocaine and ICT-Neutral groups, p values > .06. Patients that were not randomized demonstrated significantly better performance on the baseline ABBA task, p < .001.

Table 1.

Demographic and Substance Use Variables by Group

ICT-Cocaine (n = 20) ICT-Neutral (n = 20) Non-Randomized (n = 33)
Demographics
Age 49.2 (9.6) 49.6 (9.9) 51.3 (6.3)
Male 75.0% 65.0% 66.7%
Race
 Caucasian 2 4 7
 African American 18 13 26
 Other 0 3 0
Education (Years) 12.6 (1.4) 12.4 (1.3) 12.1 (2.0)
Baseline Response Inhibition 18.4 (11.0) 16.2 (8.4) 4.4 (9.7)
Monthly Income ($) 534 (553) 686 (614) 736 (869)
BSI-Anxiety 46.2 (11.7) 48.3 (13.3) 50.7 (12.5)
ADHD 3.4 (5.2) 4.2 (5.0) 3.4 (3.5)
BDI 6.8 (9.4) 8.1 (7.4) 10.0 (9.5)
Alcohol and Cigarette Use
Drinks/Week 9.7 (8.1) 10.6 (13.7) 6.9 (10.7)
MAST 11.8 (11.7) 13.9 (12.0) 11.7 (11.0)
DAST 11.6 (4.7) 12.2 (4.7) 11.0 (5.5)
Tobacco Cigarette Use 95.0% 90.0% 84.9%
 Cigarettes/Day 7.3 (6.1) 9.2 (7.6) 11.4 (7.5)
Cocaine Use
Lifetime Cocaine 2831 (2805) 1532 (1166) 2723 (1904)
Past Month Cocaine Use 12.6 (11.0) 13.4 (7.4) 14.4 (8.3)
Years Cocaine Use 19.0 (10.0) 16.1 (8.1) 21.1 (10.5)
Smoked Cocaine Use 90.0% 95.0% 97.0%
Past Month Drug Use
Amphetamines 5.0% 5.0% 0.0%
Benzodiazepines 5.0% 5.0% 9.1%
Opioids 5.0% 20.0% 15.2%
Cannabis 65.0% 65.0% 54.6%
Open in a new tab

Note. All values presented as mean (standard deviation) or counts/percentages. ICT = Inhibitory-Control Training (Cocaine or Neutral Images); MAST = Michigan Alcohol Screening Test; DAST = Drug Abuse Screening Test; BDI = Beck Depression Inventory; ADHD = Attention Deficit/Hyperactivity Disorder.

3.2. Feasibility

Figure 2 (top panel) contains a time-to-enrollment plot based on expected enrollment over a planned 3-year period and actual enrollment of randomized participants. Actual enrollment approximately matched planned enrollment during the first 20 months of the study. Enrollment then accelerated, and the planned sample size was achieved by month 27.

Figure 2.

Figure 2.

Open in a new tab

Study enrollment and weekly clinic attendance. Plotted in the top panel is the planned study enrollment divided over a three-year period against the actual percentage of total planned participants randomized to a training condition. Plotted in the bottom panel are overall weekly clinic attendance rates for the thrice weekly clinic visits during the first eight-weeks of the trial. Attendance rates are separated by intervention group (Inhibitory-Control Training-Cocaine [ICT-Cocaine] = circles; Inhibitory-Control Training-Neutral [ICT-Neutral] = squares).

Weekly attendance rates by intervention group are also plotted in Figure 2 (bottom panel). Attendance was generally stable over the 8-week period with high overall attendance in both groups (average sessions attended: ICT-Cocaine = 96.9%; ICT-Neutral = 89.6%). Although the percentage of sessions attended by participants in the ICT-Cocaine group was higher, these differences were not statistically significant, t38 = 1.63, p = .11, d = 0.51. All participants in the ICT-Cocaine group and 90% of the participants in the ICT-Neutral group completed at least 80% of the training sessions. No group differences were observed in the percentage of follow-up sessions attended (95.0% of ICT-Cocaine and 88.3% of ICT-Neutral), t38 = 0.89, p = .38, d = 0.28.

Participants received average total abstinence incentive bonuses during the training phase of $615 (SD = $96) in the ICT-Cocaine group and $573 (SD = $201) in the ICT-Neutral group. These amounts did not significantly differ by group, t38 = 0.83, p = .41, d = 0.26.

3.3. Acceptability

Figure 3 plots mean ratings on the TAQ by treatment condition. Ratings were generally high with average scores above 80 for all items. No between-group differences were observed, t37 values < 1.15, p values > .26, d values < 0.37. Approximately 95% of participants reported they were very or mildly satisfied with the study procedures and 79% reported that the sessions met most or all of their needs. All participants reported they would definitely or probably participate again. No significant group differences were observed for these secondary acceptability measures, p values > .08.

Figure 3.

Figure 3.

Open in a new tab

Study acceptability measures. Mean values with 95% confidence intervals for acceptability measures on the Treatment Acceptability Questionnaire (TAQ) completed at the end of the follow-up phase. All items were completed on a 100-point visual analog scale (VAS) (1 = lowest satisfaction, 100 = greatest satisfaction). Questions were: 1) overall satisfaction (Overall); 2) acceptability of receiving vouchers (Payment); 3) acceptability of thrice weekly clinic visits (Session Timing); 4) acceptability of providing observed urine samples (Urine Observation); 5) acceptability of the computer program to deliver cocaine-based inhibitory-control training (Training); 6) acceptability of additional cocaine-based inhibitory-control training delivered throughout the study (Booster Sessions); and 7) acceptability of receiving compliance enhancement (Compliance Enhancement).

3.4. Initial Efficacy

Figure 4 plots qualitative (top panel) and quantitative (bottom panel) urinalysis results over the 8-week abstinence induction and training phases. General increases in the number of cocaine positive urines were observed over the 8-week period. For example, 75.8% (73.3% ICT-Cocaine, 78.3% ICT-Neutral) of urines samples during Week 1 (Abstinence Induction Phase) tested positive whereas 90% (85% ICT-Cocaine, 95% ICT-Neutral) collected during Week 8 (Training Phase) tested positive. No significant group effects were observed overall or when interacting study week for either quantitative or qualitative outcomes, p values > .19. During the follow-up phase, 70% of samples provided in the ICT-Cocaine group were positive whereas 85% of the samples in the ICT-Neutral group were positive. No significant differences were observed for these qualitative (p = 0.13), or the quantitative (p = 0.46) results when evaluated in mixed effect models.

Figure 4.

Figure 4.

Open in a new tab

Qualitative and quantitative cocaine results. Plotted is the percentage of positive cocaine urine samples (top panel) and mean (95% confidence interval) ng/mL benzoylecgonine (bottom panel) for each week during the primary study phases. Quantitative urine results are presented in the analyzed log-transformed value. The vertical dotted line separates the Abstinence Induction (left) and Training (right) phases of the trial. Results are separated by intervention group (ICT-Cocaine] = circles; ICT-Neurtal] = squares).

Figure 5 shows performance on the cognitive-behavioral tasks at baseline and follow-up visits. A 2 × 2 × 2 mixed ANOVA revealed no significant main effects or interactions on delay discounting outcomes, p values > .08, ηp2 < .09. A 2 × 2 × 2 mixed ANOVA for the Stop-signal task revealed a significant main effect of Visit, F1,34 = 18.11, p = .001, ηp2 = .35, reflecting an overall decrease in errors from baseline to follow up. A Visit × Time interaction, F1,34 = 4.37, p = .044, ηp2 = .11, was also observed reflecting greater improvements in performance from Time 1 to Time 2 at baseline compared to follow-up.

Figure 5.

Figure 5.

Open in a new tab

Cognitive-behavioral data at baseline and follow up. Presented are mean values for performance on a monetary delay discounting task (top panel), cocaine delay discounting task (middle panel), and Stop-signal task (bottom panel). Data are presented for the inhibitory-control training for the Cocaine (ICT-Cocaine; black bars) and Neutral (ICT-Neutral; white bars) groups. T1 and T2 for the Stop-signal task refer to the first and second task completion within each session. Error bars are standard error of the mean (SEM).

4. Discussion

The present randomized pilot study determined the feasibility, acceptability, and initial efficacy of inhibitory-control training for cocaine use disorder. Forty treatment-seeking patients who met diagnostic criteria for cocaine use disorder were randomly assigned to inhibitory-control training with cocaine-related or neutral images for 6 weeks. Participants attended the clinic thrice weekly to provide a urine sample for qualitative and quantitative analysis, and to complete additional inhibitory-control training. Participants in both conditions received monetary incentives delivered on an escalating schedule for clinic attendance. The methods were both feasible and acceptable to participants. ICT-Cocaine did not significantly decrease the percent of cocaine-positive urines samples during the training or follow up phase relative to ICT-Neutral. Below we discuss these findings in the context of the extant literature.

The methods used were both feasible and acceptable, which is consistent with the results from previous trials that determined the feasibility and acceptability of novel interventions for substance use disorders (e.g., Agarwal et al., 2015; Gilchrist et al., 2017; Lindsay et al., 2014; Strickland et al., 2019). The targeted sample size (N=40) was identified and enrolled over a 27-month period. The randomization procedures resulted in two demographically similar groups. In no instance did the groups differ significantly on a demographic or drug-use variable.

The feasibility and acceptability findings are concordant with those from a previous trial in which inhibitory-control training and working memory training were delivered via a crowdsourcing platform (Amazon Mechanical Turk [mTurk]) for alcohol use disorder (Strickland et al., 2019). A majority of participants in that study indicated they were satisfied with the study procedures (95%) and would participate again (97%). Importantly, the majority of participants (81%) reported they would consider incorporating the training task in their daily life, suggesting that these types of brief interventions may also be feasibly adopted in pragmatic clinical trials.

Clinic attendance was generally stable and high over the 8-week treatment period in both groups. Clinical trials, particularly in the area of substance use disorder, are adversely affected by high rates of patient dropout and poor clinic attendance. Rates of attrition frequently approach between 30–50% by 12 weeks in trials enrolling stimulant-dependent individuals (Grabowski et al., 2004). High rates of attrition render data analysis difficult and often leave end-point outcomes with insufficient power for analysis. To circumvent this problem, the present trial employed monetary incentives to reinforce clinic attendance. Similar procedures targeting reductions in drug use have been effectively used for a variety of substance use disorders (for a review, see Davis et al., 2016). However, monetary incentives are also effective for other targeted behaviors (e.g., Gaalema et al., 2016; Kurti et al., 2013; Petry et al., 2018). The results of a recent trial are particularly germane to the findings of the current study (Petry et al., 2018). In that trial, cocaine-use disorder patients (N=360) were randomly assigned to receive incentives for clinic attendance for six weeks or usual care. After six weeks, patients (N=308) were re-randomized to receive incentives or usual care for another six weeks, with assignment stratified on current functioning. Patients assigned to receive monetary incentives attended 82% of the clinic days which was significantly more than their counterparts assigned to usual care (68%). These previous findings, along with those of the present trial, attest to the positive effect of incentives on clinic attendance. Future trials that determine the efficacy of behavioral or pharmacological interventions for cocaine use disorder should consider using incentives to achieve high levels of clinic attendance or adherence to the trial procedures.

Participants in the cocaine ICT-Cocaine group did not provide fewer cocaine positive urine samples than participants who completed ICT-Neutral group during the training phase. Similarly, participants in the cocaine ICT-Cocaine group did not provide fewer cocaine positive urine samples during the follow-up visits than participants in the ICT-Neutral group (i.e., 70% versus 85%, respectively). The inability to detect statistically significant differences is not surprising given the small sample size. Consistent with the recommendations of Leon and colleagues (2011), this pilot study was designed to assess the feasibility and acceptability of our study procedures as well as the potential effects of inhibitory-control training to cocaine or neutral images on outcomes of interest. Pilot studies are not, however, efficacy trials. Thus, it is inappropriate to power them based on sample size requirements to detect statistically significant effects. We selected a sample size (N=20/group) to allow us to realistically examine each aspect of our study design and implementation in preparation for a larger, appropriately powered clinical trial, which will focus on efficacy.

As noted above, the authors of a recent report concluded improved response inhibition may contribute to the efficacy of cognitive behavioral therapy for cocaine use disorder (Devito et al., 2018). In this trial, methadone-maintained patients with cocaine-use disorder were randomly assigned to treatment as usual alone or combined with computer-based cognitive behavioral therapy. Patients completed a Drug Stroop task before and after treatment. A significant interaction of time (pre versus post-treatment), trial type (drug versus neutral) and abstinence duration (> 3 weeks versus < 3 weeks) in patients that completed the trial. The graphical display of these data suggested the interaction was attributable to patients able to achieve > 3 weeks abstinence showed larger improvement on the Drug Stroop task (i.e., reduced slowing to drug-related stimuli relative to neutral stimuli). Future research should determine whether inhibitory-control training augments the efficacy of Cognitive Behavioral Therapy (CBT). Because inhibitory-control training is inexpensive and need not be administered by a clinician, it could easily be incorporated into other behavioral treatment approaches to further reduce drug use.

To the best of our knowledge, this is the first report of inhibitory-control training for cocaine use disorder. Inhibitory-control training with cocaine did not significantly reduce cocaine use in the present trial relative the neutral images group. The present findings are discordant with previous trials that determined the efficacy of inhibitory-control training for alcohol use disorder (Houben et al., 2011, 2012; Strickland et al., 2019). In our previous trial, participants with alcohol use disorder were recruited from the crowdsourcing website Amazon mTurk and randomly assigned to an inhibitory control, working memory, or control training condition (Strickland et al., 2019). Participants completed the training tasks daily over a two-week period. Modest, but statistically significant, reductions in alcohol consumption were observed (e.g., 0.5 drinking days/week) in the inhibitory-control training group. The results of these previous studies, along with those from the current trial, suggest more research is needed to determine the efficacy of inhibitory-control training for substance use disorders.

Repeated inhibitory-control training improved performance on a Stop-signal task regardless of assigned condition (i.e., cocaine or neutral images), which is consistent with the results of a previous study (Alcorn et al., 2017). Alcorn and colleagues demonstrated improved performance on a Stop-signal task following acute inhibitory-control training to cocaine or neutral cues. By contrast, repeated inhibitory-control training did not change discounting rates on a delay discounting task in the present study. Previous research has suggested improving delay discounting (i.e., making discounting rates less steep) did not alter response inhibition in a sample of predominantly cocaine users (Bickel et al., 2011). These findings collectively support the notion that response inhibition and delay discounting measure independent facets of impulsivity (Wang et al., 2016).

Future research should determine whether inhibitory-control training augments the efficacy of Cognitive Behavioral Therapy (CBT). Such trials should, however, consider a few methodological modifications. First, more intense training should be provided. In the present trial, training sessions were about 45 minutes in duration. Increasing both the intensity and frequency of training might result in larger reductions in cocaine use. Second, the follow-up period should be longer with more frequent urine testing. In the present trial, the follow up period was only three weeks in duration and urine samples were tested only once weekly. More frequent urine testing during the follow up is needed because once weekly testing raises the possibility some cocaine use went undetected. Third, because inhibitory-control training to cocaine and neutral image did not produce differential effects, future trials might consider including a “sham” condition that does not include any type of inhibitory-control training. Fourth, future trials should consider combining training with interventions targeting other facets of impulse control. Difficulty suppressing pre-potent responses, sometimes referred to as motor impulsivity, is one facet of impulsivity (Rubenis et al., 2018; Smith et al., 2014; Stevens et al., 2014). Inhibitory-control training was chosen for study based on our previous worked that showed deficits in cocaine users in suppressing pre-potent responses (Fillmore and Rush, 2002, 2006; Pike et al., 2013, 2015, 2017). Another facet of impulsivity is difficulty delaying gratification, defined by behavioral researchers as a preference for smaller, immediate rewards over larger, delayed rewards, and typically measured using Hypothetical Discounting tasks (Green et al., 1997; Rachlin et al., 1991). Working memory training has been shown to remediate excessive delay discounting in a sample who were largely cocaine users (Bickel et al., 2011). Working-memory training also reduces cocaine use (Schulte et al., 2018). Cocaine-using participants (N=38) completed a battery of working memory training tasks (i.e., backward-digit span, complex span and visuospatial tasks) daily for 25 days. Cocaine use, craving and response inhibition (i.e., Stop Signal task) were assessed during two lab visits (i.e., baseline and follow-up). The primary aim of this study was to determine the efficacy of N-acetylcysteine (0 or 2400 mg/day) for cocaine use disorder. However, the efficacy of working-memory training could be determined by comparing baseline and follow-up data in participants assigned to placebo (N=21). These participants reported using cocaine 2.1 days/week prior to working-memory training versus 1.4 days/week following working-memory training. Perhaps combining inhibitory-control training and working-memory training would result in larger reductions in cocaine use than was observed with either training alone.

Future studies might also consider targeting impulse control in patients with other stimulant-use disorders. Chronic exposure to methamphetamine, like cocaine, results in more impulsive responding on Stop-signal and Go/No-Go tasks (for a review, see Smith et al., 2014). The seminal study compared response inhibition on a Stop-signal task in methamphetamine users (N=11) who had been abstinent for 5–7 days and controls (N=43) (Monterosso et al., 2005). Stop-signal reaction time (i.e., latency to inhibit an initiated motor response) was significantly longer for methamphetamine abusers, suggesting that methamphetamine use is associated with a specific deficit in inhibiting a pre-potent response. Other investigators systematically replicated these findings in human laboratory studies and preclinical experiments (Furlong et al., 2016; Leland et al., 2008; Tabibnia et al., 2011; Tolliver et al., 2012, cf. van der Plas et al., 2009). Whether inhibitory-control training alone or combined with an intervention that targets another facet of impulsivity would effectively reduce methamphetamine use remains to be determined.

In conclusion, the present randomized pilot study determined the feasibility, acceptability and initial efficacy of inhibitory-control training for cocaine use disorder. Repeated inhibitory-control training was feasible and acceptable in both study conditions. Patients assigned to ICT-Cocaine did not provide fewer cocaine-positive urines samples during the trial than their counterparts who completed ICT-Neutral. Future research is needed to determine whether inhibitory-control training might augment the efficacy of other behavioral interventions like Cognitive Behavioral Therapy.

Highlights:

1) Inhibitory-control training for cocaine use is feasible and acceptable; 2) Inhibitory-control training improved stop signal performance but not delay discounting; 3) Future trials should determine if inhibitory-control training augments the efficacy of Cognitive Behavioral Therapy (CBT)

Acknowledgements:

The authors gratefully acknowledge the staff of the University of Kentucky Laboratory of Human Behavioral Pharmacology for technical assistance. This study complied with all laws of the United States of America.

Funding: This work was funded by a grant from the National Institute on Drug Abuse (NIDA) (R34 DA038869; Rush, CR [PI]). This funding agency had no role in study design, data collection, data analyses preparation of presentations, or submission of publications. Content is solely the responsibility of the authors and does not necessarily represent the official views of NIH.

Footnotes

Publisher's Disclaimer: 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.

Conflict of Interest: The authors declare no relevant conflicts of interest.

ClinicalTrials.gov Identifier:

References

  1. Agarwal RP, Kumar A, Lewis JE, 2015. A pilot feasibility and acceptability study of yoga/meditation on the quality of life and markers of stress in persons living with HIV who also use crack cocaine. J. Altern. Complement. Med 21, 152–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alcorn JL III, Pike E, Lile JA, Stoops WW, Rush CR, 2017. A pilot investigation of acute inhibitory control training in cocaine users. Drug Alcohol Depend. 174, 145–149 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bates D, Maechler M, Bolker B, Walker S, 2015. Fitting linear mixed effects models using lme4. J. Stat. Soft 67, 1–48. [Google Scholar]
  4. Bickel WK, Yi R, Landes RD, Hill PF, Baxter C, 2011. Remember the future: working memory training decreases delay discounting among stimulant addicts. Biol. Psychiatry 69, 260–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Butler AJ, Rehm J, Fischer B, 2017. Health outcomes associated with crack-cocaine use: Systematic review and meta-analyses. Drug Alcohol Depend. 180, 401–416. [DOI] [PubMed] [Google Scholar]
  6. Carroll KM, O’Malley SS, Nuro KF, 1999. Compliance enhancement: A manual for the psychopharmacotherapy of drug abuse and dependence. Yale University Psychotherapy Development Center Training Series No. 4 [Google Scholar]
  7. Center for Behavioral Health Statistics, 2018. 2017. National Survey on Drug Use and Health: Detailed tables, Substance Abuse and Mental Health Services Administration, Rockville, MD. [Google Scholar]
  8. Centers for Disease Control and Prevention, 2018. National Center for Health Statistics. Multiple Cause of Death 1999–2017 on CDC WONDER Online Database. Released December 2018. Accessed March 1, 2019: https://www.drugabuse.gov/related-topics/trends-statistics/overdose-death-rates.
  9. Covi L, Hess JM, Schroeder JR, Preston KL, 2002. A dose response study of cognitive behavioral therapy in cocaine abusers. J. Subst. Abuse Treat 23, 191–197. [DOI] [PubMed] [Google Scholar]
  10. Czermainski FR, Willhelm AR, Santos AZ, Pachado MP, de Almeida RMM, 2017. Assessment of inhibitory control in crack and/or cocaine users: a systematic review. Trends Psychiatry Psychother. 39, 216–225. [DOI] [PubMed] [Google Scholar]
  11. Czoty PW, Stoops WW, Rush CR, 2016. Evaluation of the “pipeline” for development of medications for cocaine use disorder: A review of translational animal laboratory, human laboratory and clinical trial research. Pharmacol. Rev 68, 533–562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Davis DR, Kurti AN, Skelly JM, Redner R, White TJ, Higgins ST, 2016. A review of the literature on contingency management in the treatment of substance use disorders, 2009–2014. Prev. Med 92, 36–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. DeVito EE, Kiluk BD, Nich C, Mouratidis M, Carroll KM (2018). Drug Stroop: Mechanisms of response to computerized cognitive behavioral therapy for cocaine dependence in a randomized clinical trial. Drug Alcohol Depend. 183: 162–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fillmore MT, Rush CR, 2002. Impaired inhibitory control of behavior in chronic cocaine users. Drug Alcohol Depend. 66, 265–273. [DOI] [PubMed] [Google Scholar]
  15. Fillmore MT, Rush CR, 2006. Polydrug abusers display impaired discrimination-reversal learning in a model of behavioural control. J. Psychopharmacol 20, 24–32. [DOI] [PubMed] [Google Scholar]
  16. Furlong TM, Leavitt LS, Keefe KA, Son JH, 2016. Methamphetamine-, d-Amphetamine-, and p-Chloroamphetamine-Induced neurotoxicity differentially effect impulsive responding on the stop-signal task in rats. Neurotox. Res 29, 569–582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gaalema DE, Savage PD, Rengo JL, Cutler AY, Higgins ST, Ades PA, 2016. Financial incentives to promote cardiac rehabilitation participation and adherence among Medicaid patients. Prev. Med 92, 47–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gilchrist G, Swan D, Shaw A, Keding A, Towers S, Craine N, Munro A, Hughes E, Parrott S, Strang J, Taylor A, Watson J, 2017. The acceptability and feasibility of a brief psychosocial intervention to reduce blood-borne virus risk behaviours among people who inject drugs: a randomised control feasibility trial of a psychosocial intervention (the PROTECT study) versus treatment as usual. Harm Reduct. J 14, e14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gonçalves PD, Ometto M, Bechara A, Malbergier A, Amaral R, Nicastri S, Martins PA, Beraldo L, dos Santos B, Fuentes D, Andrade AG, Busatto GF, Cunha PJ, 2014. Motivational interviewing combined with chess accelerates improvement in executive functions in cocaine dependent patients: a one-month prospective study. Drug Alcohol Depend 141: 79–84. [DOI] [PubMed] [Google Scholar]
  20. Grabowski J, Rhoades H, Stotts A, Cowan K, Kopecky C, Dougherty A, Moeller FG, Hassan S, Schmitz J, 2004. Agonist-like or antagonist-like treatment for cocaine dependence with methadone for heroin dependence: Two double-blind randomized clinical trials. Neuropsychopharmacology 29, 969–81. [DOI] [PubMed] [Google Scholar]
  21. Green L, Myerson J, McFadden E, 1997. Rate of temporal discounting decreases with amount of reward. Mem. Cognit 25, 715–723. [DOI] [PubMed] [Google Scholar]
  22. Havakuk O, Rezkalla SH, Kloner RA, 2017. The cardiovascular effects of cocaine. J. Am. Coll. Cardiol 70, 101–113. [DOI] [PubMed] [Google Scholar]
  23. Hedden SL, Woolson RF, Malcolm RJ, 2006. Randomization in substance abuse clinical trials. Subst. Abuse Treat. Prev. Policy 1, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Higgins ST, Heil SH, Lussier JP, 2004. Clinical implications of reinforcement as a determinant of substance use disorders. Ann. Rev. Psychol 55, 431–61. [DOI] [PubMed] [Google Scholar]
  25. Houben K Havermans RC, Nederkoorn C, Jansen A, 2012. Beer a no-go: Learning to stop responding to alcohol cues reduces alcohol intake via reduced affective associations rather than increased response inhibition. Addiction 107, 1280–1287. [DOI] [PubMed] [Google Scholar]
  26. Houben K, Nederkoorn C, Wiers RW, Jansen A, 2011. Resisting temptation: Decreasing alcohol-related affect and drinking behavior by training response inhibition. Drug Alcohol Depend 116, 132–136. [DOI] [PubMed] [Google Scholar]
  27. Kurti AN, Dallery J, 2013. Internet-based contingency management increases walking in sedentary adults. J. Appl. Behav. Anal 46, 568–5. [DOI] [PubMed] [Google Scholar]
  28. Leland DS, Arce E, Miller DA, Paulus MP, 2008. Anterior cingulate cortex and benefit of predictive cueing on response inhibition in stimulant dependent individuals. Biol. Psychiatry 63, 184–190. [DOI] [PubMed] [Google Scholar]
  29. Leon AC, David LL, Kraemer HC, 2011. The role and interpretation of pilot studies in clinical research. J. Psychiatric. Res 45, 626–629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lindsay JA, Minard CG, Hudson S, Green CE, Schmitz JM, 2014. Using prize-based incentives to enhance daily interactive voice response (IVR) compliance: a feasibility study. J. Subst. Abuse Treat 46, 74–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. McKee SA, Carroll KM, Sinha R, Robinson JE, Nich C, Cavallo D, O’Malley S, 2007. Enhancing brief cognitive-behavioral therapy with motivational enhancement techniques in cocaine users. Drug Alcohol Depend. 91, 97–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McPherson SM, Burduli E, Smith CL, Herron J, Oluwoye O, Hirchak K, Orr MF, McDonell MG, Roll JM, 2018. A review of contingency management for the treatment of substance-use disorders: adaptation for underserved populations, use of experimental technologies, and personalized optimization strategies. Subst. Abuse Rehabil 9, 43–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Monterosso JR, Aron AR, Cordova X, Xu J, London ED, 2005. Deficits in response inhibition associated with chronic methamphetamine abuse. Drug Alcohol Depend. 79, 273–277. [DOI] [PubMed] [Google Scholar]
  34. Myerson J, Green L, Warusawitharana M, 2001. Area under the curve as a measure of discounting. J. Exp. Anal. Behav 76, 235–243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Petry NM, Alessi SM, Rash CJ, Barry D, Carroll KM, 2018. A randomized trial of contingency management reinforcing attendance at treatment: Do duration and timing of reinforcement matter? J. Consult. Clin. Psychol 86, 799–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pike E, Marks KR, Stoops WW, Rush CR, 2015. Cocaine-related stimuli impair inhibitory control in cocaine users following short stimulus onset asynchronies. Addiction 110, 1281–1286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Pike E, Marks KR, Stoops WW, Rush CR, 2017. Influence of cocaine-related images and alcohol administration on inhibitory control in cocaine users. Alcohol Clin. Exp, Res 41, 2140–2150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pike E, Stoops WW, Fillmore MT, Rush CR, 2013. Drug-related stimuli impair inhibitory control in cocaine abusers. Drug Alcohol Depend. 133, 768–771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team, 2017. nlme: Linear and nonlinear mixed effects models. Available at: https://cran.r-project.org/package=nlme, R package version 3.1–137.
  40. Rachlin H, Raineri A, Cross D, 1991. Subjective probability and delay. J. Exp. Anal. Behav 55, 233–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Raiff BR, Jarvis BP, Turturici M, Dallery J, 2013. Acceptability of an internet-based contingency management intervention for smoking cessation: Views of smokers, nonsmokers, and healthcare professionals. Exp. Clin. Psychopharmacol 21, 204–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rubenis AJ, Fitzpatrick RE, Lubman DI, Verdejo-Garcia A, 2018. Impulsivity predicts poorer improvement in quality of life during early treatment for people with methamphetamine dependence. Addiction 113, 668–676. [DOI] [PubMed] [Google Scholar]
  43. Ryan F, 2002. Detected, selected, and sometimes neglected: Cognitive processing of cues in addiction. Exp. Clin. Psychopharm 10, 67–76. [DOI] [PubMed] [Google Scholar]
  44. Schulte MHJ, Wiers RW, Boendermaker WJ, Goudriaan AE, van den Brink W, van Deursen DS, Friese M, Brede E, Waters AJ, 2018. The effect of N-acetylcysteine and working memory training on cocaine use, craving and inhibition in regular cocaine users: Correspondence of lab assessments and Ecological Momentary Assessment. Addict Behav. 79: 24–31. [DOI] [PubMed] [Google Scholar]
  45. Smith JL, Mattick RP, Jamadar SD, Iredale JM, 2014. Deficits in behavioural inhibition in substance abuse and addiction: a meta-analysis. Drug Alcohol Depend. 145, 1–33. [DOI] [PubMed] [Google Scholar]
  46. Stevens L, Verdejo-Garcia A, Goudriaan AE, Roeyers H, Dom G, Vanderplasschen W, 2014. Impulsivity as a vulnerability factor for poor addiction treatment outcomes: a review of neurocognitive findings among individuals with substance use disorders. J. Subst. Abuse Treat 47, 58–72. [DOI] [PubMed] [Google Scholar]
  47. Strickland JC, Hill JC, Stoops WW, Rush CR, 2019. Feasibility, acceptability, and initial efficacy of delivering alcohol use cognitive interventions via crowdsourcing. Alcohol Clin Exp Res, ePub ahead of print. [DOI] [PubMed] [Google Scholar]
  48. Tabibnia G, Monterosso JR, Baicy K, Aron AR, Poldrack RA, Chakrapani S, Lee B, London ED, 2011. Different forms of self-control share a neurocognitive substrate. J. Neurosci 31, 4805–4810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tolliver BK, Price KL, Baker NL, LaRowe SD, Simpson AN, McRae-Clark AL, Saladin ME, DeSantis SM, Chapman E, Garrett M, Brady KT, 2012. Impaired cognitive performance in subjects with methamphetamine dependence during exposure to neutral versus methamphetamine-related cues. Am. J. Drug Alcohol Abuse 38, 251–259. [DOI] [PubMed] [Google Scholar]
  50. van der Plas EA, Crone EA, van den Wildenberg WP, Tranel D, Bechara A, 2009. Executive control deficits in substance-dependent individuals: A comparison of alcohol, cocaine, and methamphetamine and of men and women. J. Clin. Exp. Neuropsychol 31, 706–719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Vansickel AR, Fillmore MT, Hays LR, Rush CR, 2008. Effects of potential agonist-replacement therapies for stimulant dependence on inhibitory control in cocaine abusers. Am. J. Drug Alcohol Abuse 34, 293–305. [DOI] [PubMed] [Google Scholar]
  52. Verdejo-García A, 2016. Cognitive training for substance use disorders: Neuroscientific mechanisms. Neurosci Biobehav Rev 68: 270–281. [DOI] [PubMed] [Google Scholar]
  53. Verdejo-García A, Betanzos-Espinosa P, Lozano OM, Vergara-Moragues E, Gonzalez-Saiz F, Fernandez-Calderon F, Bilbao-Acedos I, Perez-Garcia M, 2012. Self-regulation and treatment retention in cocaine dependent individuals: A longitudinal study. Drug Alcohol Depend, 122: 142–148. [DOI] [PubMed] [Google Scholar]
  54. Vergara-Moragues E, Verdejo-García A, Lozano OM, Santiago-Ramajo S, González-Saiz F, Betanzos Espinosa P, Pérez García M, 2017. Association between executive function and outcome measure of treatment in therapeutic community among cocaine dependent individuals. J Subst Abuse Treat 78: 48–55. [DOI] [PubMed] [Google Scholar]
  55. Vocci FJ, Montoya ID, 2009. Psychological treatments for stimulant misuse, comparing and contrasting those for amphetamine dependence and those for cocaine dependence. Curr. Opin. Psychiatry 22, 263–268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wang Q, Chen C, Cai Y, Li S, Zhao X, Zheng L, Zhang H, Liu J, Chen C, Xue G, 2016. Dissociated neural substrates underlying impulsive choice and impulsive action. Neuroimage 134, 540–549. [DOI] [PubMed] [Google Scholar]
  57. Weafer J, Fillmore MT, 2012. Alcohol-related stimuli reduce inhibitory control of behavior in drinkers. Psychopharmacology 222, 489–498. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES