Substance use and substance use disorders represent ongoing major public health crises. Specifically, the use of substances such as cocaine and heroin are responsible for over 50,000 drug-related deaths combined annually. We used a comparative meta-analysis procedure to contrast activation patterns associated with cocaine and heroin cue reactivity, which may reflect substance use risk for these substances. PubMed and Google Scholar were searched for studies with within-subject whole brain analyses comparing drug to neutral cues for users of cocaine and heroin published between 1995 and 2022. A total of 18 studies were included, 9 in each subgroup. Voxel-based meta-analyses were performed using seed-based d mapping with permuted subject images (SDM-PSI) for subgroup mean analyses and a contrast meta-regression comparing the two substances. Results from our mean analysis indicated that users of heroin showed more widespread activation in the nucleus accumbens, right inferior and left middle temporal gyrus, right thalamus, and right cerebellum. Cocaine use was associated with recruitment of dorsolateral prefrontal cortex during cue reactivity. Direct comparison of cue reactivity studies in heroin relative to cocaine users revealed greater activation in dopaminergic targets for users of heroin compared to users of cocaine. Differential activation patterns between substances may underlie differences in the clinical characteristics observed in users of cocaine and heroin, including seeking emotional blunting in users of heroin. More consistent research methodology is needed to provide adequate studies for stringent meta-analyses examining common and distinct neural activation patterns across substances and moderation by clinically relevant factors.

The growing prevalence of substance use disorders among the US population, particularly related to illicit substances lacking regulation (i.e., opiates, psychostimulants), thus represents a significant threat to public health. Though the variety of illicit drugs available for use has increased over time, stimulants (e.g., cocaine) and opioids (e.g., heroin, fentanyl, prescription medications) remain two of the most lethal classes of illicit drugs (Degenhardt et al., 2013). In fact, opioids and stimulants, account for almost 81,000 and 15,000 deaths annually, together representing approximately 80% of all substance-related deaths (Center for Disease Control and Prevention, 2023; National Institute on Drug Abuse, 2022). Given the high rates of mortality associated with cocaine and heroin use relative to other substances (Center for Disease Control and Prevention, 2023; Darke et al., 2023; Lewer et al., 2022; National Institute on Drug Abuse, 2022; Peacock et al., 2021; Schneider et al., 2021), we selected these substances as targets for the present study. These substances are also representative of the broader opioid and stimulant drug classes, whose use (and co-use) has surged nationally in what has been labeled as the ‘twin opioid-stimulant epidemics’. Lastly, although within the broader class of stimulant and opioid drugs, others have received considerable recent attention (e.g., fentanyl), cocaine and heroin remain the most well studied in fMRI studies of cue reactivity.

Nevertheless, differences in usage patterns between these classes of drugs have also been well documented. For example, heroin use is often associated with a more consistent, chronic, and high rate of use whereas cocaine tends to involve more episodic, intermittent, and moderate use (Hser, Evans, et al., 2008; Hser, Huang, et al., 2008; Sartor et al., 2014). Users of heroin also tend to begin use at a younger age and as a result have a longer duration of lifetime use than people who use cocaine (Hser, Evans, et al., 2008; Hser, Huang, et al., 2008). Onset of heroin, and opioid use often occurs in conjunction with traumatic events, and with a desire to numb both physical and emotional pain (Purdue et al., 2024; Shand et al., 2011), while users of cocaine may demonstrate higher levels of impulsivity and impaired inhibition, resulting in higher rates of risk-taking and reward-seeking behaviors such as sexual promiscuity compared to people who use heroin (Ahn & Vassileva, 2016; Bjork et al., 2022; Bornovalova et al., 2005; Lejuez et al., 2005; Mitchell & Potenza, 2014; Vassileva et al., 2014). As these behavioral differences do not appear to depend on acute drug effects, it is possible that they may reflect differences in susceptibility between those who use cocaine versus heroin.

Interestingly, whether one is in a social or non-social environment appears to impact drug choice (Caprioli et al., 2009; De Pirro et al., 2018). In both animal and human models, heroin use tends to be more common within non-social contexts, whereas cocaine use is often preferred in social environments (Badiani et al., 2011, 2019; Caprioli et al., 2009). Such contextual preferences also amplify the respective effects of each substance (De Pirro et al., 2018). In sum, these differences suggest that in addition to shared mechanisms, there may also be divergent mechanisms that are important to identify to better understand patterns of usage of cocaine and heroin. Identifying divergent mechanisms (i.e., pinpointing behavioral and neurobiological risk-factors for the development of a particular SUD, developing drug-specific treatment interventions) is imperative given the rapidly increasing rates of use and associated harm caused by these illicit substances.

A common component of substance use disorders that is closely tied to reuse and relapse is drug craving (i.e., the strong wanting or desire for a substance; Vafale & Kober, 2022). Characterizing the ways in which the brain differentially responds to craving in people who use cocaine and heroin may provide an important way to assess shared and distinct vulnerability mechanisms and may have implications for developing more targeted interventions to prevent their use. Cue reactivity paradigms are reliably related to subjective experiences of drug cravings and are thus widely used, well–supported laboratory assessments in samples of individuals using substances (Carter & Tiffany, 1999; Drummond, 2000), and have been used previously to reveal similarities and differences between individuals who use stimulants, opioids, or both (e.g., Hochheimer et al., 2023). In fact, a recent study found that when comparing cue induced and tonic craving between opioids and stimulant users in treatment, opiate cues consistently resulted in greater reports of craving (Hochheimer et al., 2023). Typically, such paradigms involve exposure to a drug cue or set of drug cues (e.g., images of a drug or associated paraphernalia, auditory drug cues via stories or sounds) and the subsequent measurement of self-reported craving or physiological responses to these cues (e.g., psychophysiological signals, patterns of neural activation; Carter & Tiffany, 1999; Courtney et al., 2016; Drummond, 2000).

Task-based fMRI studies implementing cue reactivity paradigms have been extensively used to characterize the neural circuitry of substance use disorders, including for identifying common and specific patterns associated with use of licit and illicit substances (e.g., Chase et al., 2011; Kuhn & Gallinat, 2011; Zilberman et al., 2019). However, to our knowledge, there have been no systematic comparisons that focus specifically on cocaine versus heroin drug cue reactivity, although such comparisons would be highly relevant given ongoing public health concerns associated with these substances and clinically relevant differences in their usage. Studies have revealed that targets of dopaminergic input such as the striatum, prefrontal cortex, and thalamus are recruited during cocaine use (Bartra et al., 2013; Haber & Knuston, 2010; Huang et al., 2018; Wang et al., 2016) and play a key role in craving and sustaining the cycle of addiction, consistent with rodent models (Everitt & Robbins, 2005; Goldstein & Volkow, 2002; Huang et al., 2018; Jentsch & Taylor, 1999). Similarly, heroin recruits dopaminergic targets such as the dlPFC, insula, and orbitofrontal cortex (OFC; Liu et al., 2021); it also seems to target other systems including the glutamatergic system (Gheraldini et al., 2015) which is conceptualized as the basis for heroin’s analgesic effect.

Further, prior meta-analyses studying users of these substances have largely utilized broad inclusion criteria, or examined outcomes other than functional activation (e.g., gray matter, structural differences; Hill-Bowen et al., 2022; Klugah-Brown et al., 2021; Long et al., 2022; Pando-Nuande et al., 2021). Although studies have begun to identify common neural circuits implicated in cocaine and heroin use, there is no existing meta-analytic synthesis of the extant literature to quantitatively evaluate similarities and differences in the functional neural underpinnings of cocaine versus heroin use. These methodological differences impact the ability to characterize similarities and differences in cue reactivity between users of each substance. Additionally, an important factor to consider is whether similarities/differences in neural recruitment during cue reactivity across users of cocaine and heroin vary as a function of individual differences in length of drug use/chronicity. Thus, a more stringent synthesis regarding the neural circuits implicated in cocaine and heroin usage is needed.

To address these gaps, we conducted a meta-analysis of fMRI studies aimed at comparing activation patterns during cue reactivity tasks for drug relative to neutral cues in people who use cocaine and people who use heroin. We applied strict inclusion criteria for our meta-analysis such as utilizing studies implementing within subjects, whole-brain analyses of drug versus neutral cues. We expected that studies of both people who use cocaine and people who use heroin will show significant engagement of reward-related and dopaminergic circuits during cue reactivity tasks, given the well-documented recruitment of these circuits by both substances. We also expected that people who use cocaine will exhibit greater activation in corticostriatal circuits than people who use heroin, given possible differences in levels of impulsivity, risk taking, and impaired inhibition observed in cocaine relative to heroin. We also explored the impact of length of use on neural correlates of cue reactivity, with the expectation that length of use would be negatively associated with activation in dopaminergic regions, including the prefrontal cortex and regions associated with reward processing such as the striatum.

2.1 Literature search and study selection

A literature search was conducted via PubMed and Google Scholar (up to April 2022) to identify applicable studies. Keywords used were (‘cocaine’, ‘heroin’, ‘substance use’, ‘cue-reactivity’, ‘fMRI’). Search strings used were (‘cocaine fMRI cue-reactivity', ‘heroin fMRI cue-reactivity') along with other combinations of the keywords. All reference lists from identified studies were checked manually for additional applicable studies. In the current meta-analysis, studies were included if (1) analyses were conducted via fMRI; (2) participants were medically healthy aside from their substance use adults over the age of 18; (3) participants were either abstaining from or currently engaging in drug use; (4) the study included a substance versus neutral condition whole-brain contrast within the substance using group; (5) the study utilized a cue reactivity paradigm; and (6) reported coordinates in either Montreal Neurological Institute (MNI) or Talairach and Tournoux space. Studies were excluded if (1) they lacked a substance versus neutral whole-brain contrast analysis; (2) included PET data; (3) there was insufficient reporting of fMRI data or findings; (4) they did not report coordinates from whole-brain analyses; (5) neither people who use cocaine nor heroin were included in the study; and (6) analyses were not separated according to drug type.

Studies were selected based on the above inclusion and exclusion criteria and were reviewed by multiple authors (J.D., N.S., D.F.). The corresponding author made final decisions regarding inclusion/exclusion conflicts. Study quality was assessed using a recently proposed checklist for cue reactivity studies (Ekhtiari et al., 2022). Given the recency of this publication compared to the dates of publication for our studies, we set our threshold for quality at 50% of criterion met within a category for at least 5 of the 10 proposed categories. All studies included in the current study exceeded this threshold meeting criterion for at least 7 of the 10 proposed categories. Figure 1 depicts the process of literature search and study selection. 1025 studies were first identified via the literature search; any duplicates from searches were removed. Next, 949 studies were excluded based on title/abstract review. The remaining 76 full text articles were assessed for eligibility. Exclusion of studies occurred for a variety of reasons, including: (1) participant groups were composed of people who use opiates, in general, as opposed to solely people who use heroin; (2) only between-group analyses were conducted comparing people who use substances to healthy controls—no within-subjects analyses examining responses to drug versus neutral cues; (3) lack of reporting of whole brain analyses, a focus on only connectivity or solely ROI analyses; and (4) lastly, an implementation of experimental tasks other than a drug versus neutral cue task (i.e., stroop word task, drug versus food cues, etc.). Of the 76 full-text studies, 58 studies were excluded based on the above criteria, leaving 18 studies remaining (total N = 447). Each subgroup—cocaine and heroin—was composed of 9 studies (Ncocaine = 273; Nheroin = 174) that were included in this meta-analysis (see Table 1). Group differences in sample characteristics, compared using independent samples t-tests, across included studies are presented in Table 3.

Table 1.

Characteristics of 18 fMRI studies included in the contrast meta-analysis, sub-group mean analyses, and the 15 included in exploratory meta-regression.

AuthorYearNTaskDrugSpaceIncluded in meta-regression
Duncan, E., Boshoven, W. Harenski, K., et al.  2007  10 Script guided mental imagery MNI 
Elton, A., Smitherman, S., Young, J., et al.  2015  20 Script guided mental imagery MNI 
Goldstein, R., Tomasi, D., Alla-Klein, N., et al.  2009  15 Word-cue task TAL 
Hanlon, C. A., Dowdle, L. T., Gibson, N. B., et al.  2018  55 Image-cue task MNI 
Hassani-Abharian, P., Ganjgahi, H., Tabatabaei-Jafari, H., et al  2015  20 CRET task MNI 
Kaag, A. M., Reneman, L., Homberg, J., et al.  2018  59 Image-cue task MNI 
Li, Q., Wang, Y., Zhang, Y., et al.  2012  24 Image-cue task MNI 
Li, Q., Wang, Y., Zhang, Y., et al.  2013  19 Image-cue task MNI 
Mei, W., Zhang, J. X., & Xiao, Z  2010  12 Image-cue task MNI 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clark, A. L., Santa Ana, E. J., et al.  2013  25 Image-cue task MNI 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clarck, A. L., & Brady, K. T.  2013  30 Image-cue task MNI 
Prisciandaro, J. J., McRae-Clark, A. L., Myrick, H., et al.  2014  38 Image-cue task MNI 
Walter, M., Denier, N., Gerber, H., et al.  2015  27 Image-cue task MNI 
Wang, W., Li, Q., Wang, Y., et al.  2011  14 Image-cue task MNI 
Wang, A.-L., Elman, I., Lowen, S. B., et al.  2015  32 Event related and image cue tasks TAL 
Xiao, Z., Lee, T., Zhang, J. X., et al.  2006  14 Image-cue task TAL 
Zhang, S., Zhornitsky, S., Wang, W., et al.  2020  52 Image-cue task MNI 
Zijlstra, F., Veltman, D. J., Booji, J., et al.  2009  12 Image-cue task MNI 
AuthorYearNTaskDrugSpaceIncluded in meta-regression
Duncan, E., Boshoven, W. Harenski, K., et al.  2007  10 Script guided mental imagery MNI 
Elton, A., Smitherman, S., Young, J., et al.  2015  20 Script guided mental imagery MNI 
Goldstein, R., Tomasi, D., Alla-Klein, N., et al.  2009  15 Word-cue task TAL 
Hanlon, C. A., Dowdle, L. T., Gibson, N. B., et al.  2018  55 Image-cue task MNI 
Hassani-Abharian, P., Ganjgahi, H., Tabatabaei-Jafari, H., et al  2015  20 CRET task MNI 
Kaag, A. M., Reneman, L., Homberg, J., et al.  2018  59 Image-cue task MNI 
Li, Q., Wang, Y., Zhang, Y., et al.  2012  24 Image-cue task MNI 
Li, Q., Wang, Y., Zhang, Y., et al.  2013  19 Image-cue task MNI 
Mei, W., Zhang, J. X., & Xiao, Z  2010  12 Image-cue task MNI 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clark, A. L., Santa Ana, E. J., et al.  2013  25 Image-cue task MNI 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clarck, A. L., & Brady, K. T.  2013  30 Image-cue task MNI 
Prisciandaro, J. J., McRae-Clark, A. L., Myrick, H., et al.  2014  38 Image-cue task MNI 
Walter, M., Denier, N., Gerber, H., et al.  2015  27 Image-cue task MNI 
Wang, W., Li, Q., Wang, Y., et al.  2011  14 Image-cue task MNI 
Wang, A.-L., Elman, I., Lowen, S. B., et al.  2015  32 Event related and image cue tasks TAL 
Xiao, Z., Lee, T., Zhang, J. X., et al.  2006  14 Image-cue task TAL 
Zhang, S., Zhornitsky, S., Wang, W., et al.  2020  52 Image-cue task MNI 
Zijlstra, F., Veltman, D. J., Booji, J., et al.  2009  12 Image-cue task MNI 
Fig. 1.

Flowchart PRISMA diagram outlining the process of literature search and study selection.

Fig. 1.

Flowchart PRISMA diagram outlining the process of literature search and study selection.

Close modal

2.2 Seed based d mapping

Seed-based d mapping (SDM) SDM-PSI v. 6.22 (formerly Signed Differential Mapping) is a specific statistical method used for meta-analyzing structural and/or functional differences in the brain across studies using a variety of different neuroimaging techniques, including fMRI, VBM, DTI, or PET (Albajes-Eizagirre, Solanes, Vieta, et al., 2019). The SDM method consists of a series of steps. To begin, coordinates of identified cluster peaks from individual studies are entered. Next, during pre-processing, the upper and lower bounds of possible effect sizes (Hedges g) and variances are calculated for each study. A mean-analysis is then performed where maximum-likelihood estimation (MLE) and MetaNSUE are implemented to estimate the most likely effect size for each included study and its standard error (Albajes-Eizagirre, Solanes, Vieta, et al., 2019). MetaNSUE is a meta-analytic method that decreases bias and permits inclusion of non-statistically significant unreported effects (NSUEs) (Albajes-Eizagirre, Solanes, Radua, 2019). MetaNSUE involves multiple imputations that are subsequently created by adding noise to the estimations within the previously calculated bounds (Albajes-Eizagirre, Solanes, Radua, 2019). Each of the imputed data sets is then meta-analyzed, and Rubin’s rules (a pooling of parameter estimates in an imputed data set to provide confidence intervals and p-values) are applied to combine the imputed meta-analyzed datasets. Lastly, a subject-based permutation test is conducted in which the maximum statistics, (i.e., the largest z-value reflecting the peak of activation across the whole brain in each study), from the combined meta-analysis images, are saved (Albajes-Eizagirre, Solanes, Radua et al., 2019). A family-wise error rate correction (FWER) for multiple comparisons is then performed using the distribution of the maximum statistic.

2.3 Meta-analysis procedure

Coordinates of peak activation and additional pertinent information (e.g., statistical values [p-value, t-value, z-value], statistical thresholds, and sample sizes) for each within-group drug cue > neutral contrast from the included studies were collected and converted into text files. Any p-values or z-values reported in the studies were converted to t-values using the SDM online converter. We then conducted a voxel-based meta-analysis using SDM-PSI v. 6.22 for all studies included, irrespective of drug type, and for each subgroup. We also conducted a subsequent linear model. The same parameters were utilized for all analyses and are outlined as follows. Pre-preprocessing for each group was conducted within a gray matter mask using the default settings within SDM-PSI: a 20-mm anisotropic full width half maximum (FWHM) kernel and 2-mm voxel size (Albajes-Eizagirre, Solanes, Vieta, et al., 2019). The mean analysis and permutation tests were then performed within each group of studies separately (cocaine, heroin), with the number of imputations set to 50 and the number of permutations set to 5000. The results of the meta-analyses of the subgroups were then corrected using the FWER correction with 5000 permutations and subsequently thresholded at p < 0.05.

2.4 Contrast analysis

We conducted a linear regression to compare the two subgroups, Heroin > Cocaine and Cocaine > Heroin. The inter-study heterogeneity of each cluster surviving correction was measured by the I2 index, which represents the proportion of total variation due to study heterogeneity (Higgins & Thompson, 2002). I2 > 50% commonly indicates considerable heterogeneity. The maps resulting from the contrast analyses were also corrected to a level of p<0.005 TFCE-FWER, with a cluster extent threshold of ≥ 10 voxels. Subsequent bias tests (e.g., funnel plots, I2 index, metabias test, and excess significance tests) were conducted for this analysis using the SDM software. The I2 value is generated during the extraction of the peaks and represents the percentage of heterogeneity across the data and studies included in the meta-analysis (Albajes-Eizagirre, Solanes, Fullana, et al., 2019). The funnel plot test identified if there was asymmetry in the plot (i.e., whether there were larger effect sizes in smaller studies). Last, the excess significance test indicates if there was publication bias (i.e., whether studies are only published if the results are significant; Albajes-Eizagirre, Solanes, Fullana, et al., 2019).

2.5 Exploratory meta-regression

We also tested whether length of use moderated activation by incorporating length of use and a substance type x length of use interaction into the model. Due to limited data in the included studies regarding demographic and clinical features, no other features were included as meta-regressors. The inter-study heterogeneity of each cluster surviving correction was measured by the I2 index, which represents the proportion of total variation due to study heterogeneity (Higgins & Thompson, 2002). I2 > 50% commonly indicates considerable heterogeneity. The maps resulting from the contrast analyses were also corrected with p<0.005 TFCE-FWER, with a cluster extent threshold of ≥ 10 voxels. Nilearn was used to visualize results from all analyses (Abraham et al., 2014).

3.1 Clinical and demographic differences

Assessment of group differences in age and length of use revealed a significant difference in age with cocaine participants being significantly older but no difference in length of use (see Table 3 for full description of study sample characteristics). Further, all studies excluded participants with a co-occurring SUD outside the drug of interest in the study; however, most studies permitted recreational tobacco and alcohol use. Four studies permitted inclusion of participants with mood disorders but did not specify whether any participants met criteria. Abstinence was reported variably across studies, making it difficult to directly compare and determine participants’ length of abstinence at the time of participation. When evaluating abstinence based on whether participants were abstinent at data collection (yes/no), participants across all studies were largely abstinent. Overall, participants were largely biological males. In fact, 10 of the 18 studies were an all-male sample. Of the remaining 8 mixed-sex studies, 7 of them were still composed of a majority of male participants (see Tables 2 & 3). In general, cocaine studies appeared to have more mixed-sex samples, though still largely male dominant. Treatment history was evaluated based on whether participants were undergoing medication-assisted treatment at the time of data collection. Across all studies, participants were not undergoing treatment except those in two of the heroin studies.

Table 2.

Clinical characteristics of heroin and cocaine subgroups.

StudyDrugAbstinent (Y/N)Co-useSexMean age (years)Length of use (mos)
Duncan, E., Boshoven, W., Harenski, K., et al. No Male 43.6 15.9 
Elton, A., Smitherman, S., Young, J., et al. No Male 40 176.4 
Goldstein, R., Tomasi, D., Alla-Klein, N., et al. Alcohol, tobacco, marijuana 12 M/ 3 F 43.6 188.4 
Hanlon, C. A., Dowdle, L. T., Gibson, N. B, et al. Nicotine 38 M/17 F 42.7 NA 
Hassani-Abharian, P., Ganjgahi, H., Tabatabaei-Jafari, H., et al No Male 31.9 36.24 
Kaag, A. M., Reneman, L., Homberg, J., et al. No other opiates Male 31.4 72 
Li, Q., Wang, Y., Zhang, Y., et al. No other opiates Male 32.8 78.6 
Li, Q., Wang, Y., Zhang, Y., et al. No Male 32.2 80.5 
Mei, W., Zhang, J. X., & Xiao, Z. No 12 M/1 F 33.5 54 
Prisciandaro, J. J., Myrick, H., Henderson, S. McRae-Clark, A. L., Santa Ana, E. J., et al. Tobacco, nicotine Male 48.8 215.6 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clarck, A. L., & Brady, K. T. Marijuana, alcohol, nicotine 24 M/6 F 48.4/31.2 NA 
Prisciandaro, J. J., McRae-Clark, A. L., Myrick, H., et al. No 33 M/5 F 47.7/43.96 213.5 
Walter, M., Denier, N., Gerber, H., et al. No 19 M/8 F 41.1 253 
Wang, W., Li, Q., Wang, Y., et al. No Male 36.1 94.8 
Wang, A.-L., Elman, I., Lowen, S. B., et al. No 17 M/15 F 29.19 NA 
Xiao, Z., Lee, T., Zhang, J. X., et al. No Male 33.2 85.2 
Zhang, S., Zhornitsky, S., Wang, W., et al. No 42 M/10 F 46.7/41.0 189.4 
Zijlstra, F., Veltman, D. J., Booji, J., et al. No Male 42.8 192 
StudyDrugAbstinent (Y/N)Co-useSexMean age (years)Length of use (mos)
Duncan, E., Boshoven, W., Harenski, K., et al. No Male 43.6 15.9 
Elton, A., Smitherman, S., Young, J., et al. No Male 40 176.4 
Goldstein, R., Tomasi, D., Alla-Klein, N., et al. Alcohol, tobacco, marijuana 12 M/ 3 F 43.6 188.4 
Hanlon, C. A., Dowdle, L. T., Gibson, N. B, et al. Nicotine 38 M/17 F 42.7 NA 
Hassani-Abharian, P., Ganjgahi, H., Tabatabaei-Jafari, H., et al No Male 31.9 36.24 
Kaag, A. M., Reneman, L., Homberg, J., et al. No other opiates Male 31.4 72 
Li, Q., Wang, Y., Zhang, Y., et al. No other opiates Male 32.8 78.6 
Li, Q., Wang, Y., Zhang, Y., et al. No Male 32.2 80.5 
Mei, W., Zhang, J. X., & Xiao, Z. No 12 M/1 F 33.5 54 
Prisciandaro, J. J., Myrick, H., Henderson, S. McRae-Clark, A. L., Santa Ana, E. J., et al. Tobacco, nicotine Male 48.8 215.6 
Prisciandaro, J. J., Myrick, H., Henderson, S., McRae-Clarck, A. L., & Brady, K. T. Marijuana, alcohol, nicotine 24 M/6 F 48.4/31.2 NA 
Prisciandaro, J. J., McRae-Clark, A. L., Myrick, H., et al. No 33 M/5 F 47.7/43.96 213.5 
Walter, M., Denier, N., Gerber, H., et al. No 19 M/8 F 41.1 253 
Wang, W., Li, Q., Wang, Y., et al. No Male 36.1 94.8 
Wang, A.-L., Elman, I., Lowen, S. B., et al. No 17 M/15 F 29.19 NA 
Xiao, Z., Lee, T., Zhang, J. X., et al. No Male 33.2 85.2 
Zhang, S., Zhornitsky, S., Wang, W., et al. No 42 M/10 F 46.7/41.0 189.4 
Zijlstra, F., Veltman, D. J., Booji, J., et al. No Male 42.8 192 

Abstinence reflects whether participants were free from substances at the time of data collection.

Table 3.

Group differences in age and length of use across subgroups.

HeroinCocainetp-value
Age (m, SD, range) 34.75, 4.48, 29.2-42.8 42.18, 4.88, 31-48.8 3.36 p < 0.01 
Length of use (m, SD, range) 109.29, 74.11, 36.2-253 154.31, 78.3, 15.9-215.6 1.14 p > 0.05 
Abstinent at time of study 3/9 1/9 — — 
Co-occurring SUDS 0/9 0/9 — — 
Treatment (medication assisted at time of study) 2/9 0/9 — — 
Biological sex (% male) 86% (174 total) 87% (304 total) — — 
HeroinCocainetp-value
Age (m, SD, range) 34.75, 4.48, 29.2-42.8 42.18, 4.88, 31-48.8 3.36 p < 0.01 
Length of use (m, SD, range) 109.29, 74.11, 36.2-253 154.31, 78.3, 15.9-215.6 1.14 p > 0.05 
Abstinent at time of study 3/9 1/9 — — 
Co-occurring SUDS 0/9 0/9 — — 
Treatment (medication assisted at time of study) 2/9 0/9 — — 
Biological sex (% male) 86% (174 total) 87% (304 total) — — 

3.2 Aggregate meta-analysis

Five clusters/peaks were identified as statistically significant across all studies for drug cues > neutral cues following the general meta-analysis in SDM (see Table 4 and Fig. 2). Regions within these clusters (peak x, y, z coordinates reported in MNI space) include the (1) right posterior cingulate (6, -44, 30), (2) right inferior frontal gyrus (48, 12, 32), (3) left inferior frontal gyrus (-50, 10, 28), (4) left superior frontal gyrus (-4, 54, 16), and (5) left inferior temporal gyrus (-46, -68, -8). One peak, the left inferior temporal gyrus, had an I2 value that exceeded 40% (I2 = 52%), suggesting that the between-study variance was low with most variance attributed to within-study sampling error, and therefore, the metabias test for this peak was not interpreted. The remaining peaks did not have I2 values that exceeded 25% and therefore, metabias tests were interpreted. Metabias tests on the remaining peaks revealed an additional cluster, the left inferior frontal gyrus that demonstrated bias/heterogeneity (p > 0.05). The T2 value was reviewed for this peak and revealed low dispersion (T2 = 0.01). All other findings were supported as Metabias test results were non-significant (all p’s > 0.2).

Table 4.

Location of neural activations for aggregate meta-analysis of Heroin and cocaine studies for drug > neutral cues.

IDRegionxyzVoxelspSDM - Z
Right posterior cingulate gyrus, BA 23 -44 30 1817 <0.001 5.915 
Right inferior frontal gyrus, opercular part, BA 44 48 12 32 620 <0.001 7.222 
Left inferior frontal gyrus, opercular part, BA 44 -50 10 28 185 0.002 6.058 
Left superior frontal gyrus, medial, BA 10 -4 54 16 223 0.01 4.970 
Left inferior temporal gyrus. BA 37 -46 -68 -8 75 0.02 4.932 
IDRegionxyzVoxelspSDM - Z
Right posterior cingulate gyrus, BA 23 -44 30 1817 <0.001 5.915 
Right inferior frontal gyrus, opercular part, BA 44 48 12 32 620 <0.001 7.222 
Left inferior frontal gyrus, opercular part, BA 44 -50 10 28 185 0.002 6.058 
Left superior frontal gyrus, medial, BA 10 -4 54 16 223 0.01 4.970 
Left inferior temporal gyrus. BA 37 -46 -68 -8 75 0.02 4.932 

Peaks organized by decreasing voxel size.

Fig. 2.

Activated peaks associated with the aggregate mean analysis of 18 cocaine and heroin cue reactivity studies for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. See Table 4 for a full list of activated peaks.

Fig. 2.

Activated peaks associated with the aggregate mean analysis of 18 cocaine and heroin cue reactivity studies for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. See Table 4 for a full list of activated peaks.

Close modal

3.3 Meta-analysis of subgroups

Fourteen clusters/peaks were identified as statistically significant in the Heroin subgroup for drug cues > neutral cues following the general meta-analysis in SDM (see Fig. 3). Regions within these clusters (peak x, y, z coordinates reported in MNI space) include the (1) left medial cingulate (-2, 30, 32), (2) right superior (6, 48, 42) and inferior (48, 14, 32) frontal gyri, (3) right anterior thalamus (6, 48, 42), (4) left amygdala (-28, -2, -22), (5) left inferior temporal gyrus (-48, -66, -6), and (6) right cerebellum (24, -74, -44). For a full list of significantly activated regions and their respective coordinates, see Table 5. Only one cluster/peak—right precentral gyrus/inferior frontal gyrus (46, 4, 30)—was identified as statistically significant in the cocaine subgroup for drug cues > neutral cues following the general meta-analysis (see Fig. 4). Considering the small number of studies (less than 10) that fell within each subgroup, bias test results were conducted but not interpreted (SDM-PSI v. 6.22). However, based on the results of the aggregate meta-analysis results, the left inferior temporal gyrus and left inferior frontal gyrus peaks revealed in the heroin-only meta-analysis may be biased.

Fig. 3.

Activated peaks associated with the mean analysis of 9 heroin cue reactivity studies where drug > neutral cue. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. See Table 4 for a full list of activated peaks.

Fig. 3.

Activated peaks associated with the mean analysis of 9 heroin cue reactivity studies where drug > neutral cue. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. See Table 4 for a full list of activated peaks.

Close modal
Table 5.

Location of neural activations for mean analysis heroin and cocaine subgroups for drug > neutral cues.

IDRegionxyzVoxelspSDM - Z
Heroin subgroup 
Left median cingulate/paracingulate gyri, BA 23 -2 -30 32 3194 <0.001 6.521 
Right superior frontal gyrus, medial, BA 9 48 42 2866 <0.001 4.885 
Right anterior thalamic projections -16 -2 1402 <0.001 5.766 
Right inferior frontal gyrus, opercular part, BA 44 48 14 32 1096 <0.001 5.554 
Left amygdala, BA 36 -28 -2 -22 815 <0.001 5.5 
Left inferior temporal gyrus, BA 37 -48 -66 -6 815 <0.001 5.830 
Right cerebellum, hemispheric lobule, VIIB 24 -74 -44 781 <0.001 5.082 
Right amygdala, BA 36 28 -2 -24 653 <0.001 5.186 
Left inferior frontal gyrus, opercular part, BA 44 -50 28 663 <0.001 5.65 
10 Right angular gyrus, BA 7 32 -60 52 664 <0.001 5.354 
11 Right inferior temporal gyrus, BA 37 48 -70 -4 670 <0.001 5.179 
12 Left inferior parietal (excluding supramarginal and angular) gyri, BA 2 -50 -32 46 590 0.005 4.263 
13 Left inferior parietal (excluding supramarginal and angular) gyri, BA 7 -36 -62 48 387 0.005 4.507 
14 Left caudate nucleus -8 10 10 249 0.003 4.629 
Cocaine subgroup 
Right precentral gyrus/inferior frontal gyrus, BA 44 46 30 144 0.008 5.394 
IDRegionxyzVoxelspSDM - Z
Heroin subgroup 
Left median cingulate/paracingulate gyri, BA 23 -2 -30 32 3194 <0.001 6.521 
Right superior frontal gyrus, medial, BA 9 48 42 2866 <0.001 4.885 
Right anterior thalamic projections -16 -2 1402 <0.001 5.766 
Right inferior frontal gyrus, opercular part, BA 44 48 14 32 1096 <0.001 5.554 
Left amygdala, BA 36 -28 -2 -22 815 <0.001 5.5 
Left inferior temporal gyrus, BA 37 -48 -66 -6 815 <0.001 5.830 
Right cerebellum, hemispheric lobule, VIIB 24 -74 -44 781 <0.001 5.082 
Right amygdala, BA 36 28 -2 -24 653 <0.001 5.186 
Left inferior frontal gyrus, opercular part, BA 44 -50 28 663 <0.001 5.65 
10 Right angular gyrus, BA 7 32 -60 52 664 <0.001 5.354 
11 Right inferior temporal gyrus, BA 37 48 -70 -4 670 <0.001 5.179 
12 Left inferior parietal (excluding supramarginal and angular) gyri, BA 2 -50 -32 46 590 0.005 4.263 
13 Left inferior parietal (excluding supramarginal and angular) gyri, BA 7 -36 -62 48 387 0.005 4.507 
14 Left caudate nucleus -8 10 10 249 0.003 4.629 
Cocaine subgroup 
Right precentral gyrus/inferior frontal gyrus, BA 44 46 30 144 0.008 5.394 

Peaks organized by decreasing voxel size.

Fig. 4.

Activated peaks associated with the mean analysis of 9 cocaine cue reactivity studies where drug cue < neutral cue. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. Image shows the precentral gyrus.

Fig. 4.

Activated peaks associated with the mean analysis of 9 cocaine cue reactivity studies where drug cue < neutral cue. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space. Image shows the precentral gyrus.

Close modal

3.4 Contrast analysis

Our primary interest was in examining differential recruitment of neural circuits at the whole-brain level between the cocaine and heroin subgroups in response to drug relative to neutral cues. We performed a contrast analysis to investigate this question. We observed increased activation in the Heroin > Cocaine contrast in several regions (see Fig. 5), including: (1) right inferior (48, -74, -4) and superior (30, -68, 50) parietal gyrus; (2) thalamus (6, -14, 2); (3) caudate nucleus (-10, 10, 10); (4) right superior frontal gyrus (6, 48, 40); and (5) right cerebellum (20, -80,- 40). For a full list of significantly activated regions and their respective coordinates, see Table 6. Low between-study heterogeneity was found for each significant peak (I2 = 0.74-32.2%). Funnel plots were symmetric, suggesting that none of the results were a function of a small subset of included studies or by studies with a small sample size. Metabias tests supported these findings as results were all non-significant (all p’s > 0.58). No significant clusters/peaks survived correction for the Cocaine > Heroin contrast. No study had an I2 value that exceeded 32%, suggesting that the between-study variance was low with most variance attributed to within-study sampling error.

Fig. 5.

Peak activations associated with the contrast analysis of heroin > cocaine subgroups for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space.

Fig. 5.

Peak activations associated with the contrast analysis of heroin > cocaine subgroups for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates were reported in MNI space.

Close modal
Table 6.

Location of neural activations for heroin > cocaine contrast of drug > neutral cue studies.

IDRegionxyzVoxelspSDM - ZBias test (p-value)I2
Right inferior temporal gyrus, BA 19 48 -74 -4 628 <0.001 4.313 0.603 32.199 
Right superior parietal gyrus, BA 7 30 -68 50 473 <0.001 4.387 0.598 11.118 
Right anterior thalamic projections -14 297 <0.001 4.345 0.641 7.040 
Left caudate nucleus -10 10 10 94 0.01 3.376 0.716 6.025 
Right superior frontal gyrus, BA 9 48 40 81 0.01 3.318 0.703 16.626 
Right cerebellum, crus II 20 -80 -40 67 0.03 2.655 0.771 0.739 
Left middle temporal gyrus, BA 37 -54 -62 54 0.02 3.169 0.712 4.254 
Left inferior parietal (excluding supramarginal and angular) gyri, BA 2 -50 -32 46 26 0.03 2.821 0.720 8.739 
IDRegionxyzVoxelspSDM - ZBias test (p-value)I2
Right inferior temporal gyrus, BA 19 48 -74 -4 628 <0.001 4.313 0.603 32.199 
Right superior parietal gyrus, BA 7 30 -68 50 473 <0.001 4.387 0.598 11.118 
Right anterior thalamic projections -14 297 <0.001 4.345 0.641 7.040 
Left caudate nucleus -10 10 10 94 0.01 3.376 0.716 6.025 
Right superior frontal gyrus, BA 9 48 40 81 0.01 3.318 0.703 16.626 
Right cerebellum, crus II 20 -80 -40 67 0.03 2.655 0.771 0.739 
Left middle temporal gyrus, BA 37 -54 -62 54 0.02 3.169 0.712 4.254 
Left inferior parietal (excluding supramarginal and angular) gyri, BA 2 -50 -32 46 26 0.03 2.821 0.720 8.739 

Peaks arranged by decreasing voxel size.

3.5 Exploratory meta-regression

An important consideration when examining activation differences across groups is whether they are moderated by the lifetime length of use of the substance, as resulting activation patterns to drug cues may vary in accordance with acute and chronic usage. To address this question, we conducted an exploratory meta-regression examining a whole-brain interaction between drug type and average length of use (in months). Fifteen of the 18 original studies were used for this analysis, as 3 studies did not provide average length of use data (see Table 1 for full list of included studies). This exploratory analysis revealed a negative interaction in clusters, including the left superior frontal gyrus (both the medial (-2, 62, 20) and dorsolateral areas (-18, 38, 46)), which suggests that activation here decreased as length of use increased in people who use heroin (see Table 7 and Fig. 6). Low between-study heterogeneity was found for each significant peak (I2 = 1.48-3.14%). Funnel plots were symmetric, suggesting that none of the results were a function of a small subset of included studies or by studies with a small sample size. The potential publication bias tests were all non-significant (p = 0.998). No significant clusters of activation survived whole-brain correction for people who use cocaine. No study had an I2 value that exceeded 32%.

Table 7.

Location of neural activations for exploratory meta-regression of interaction of drug (heroin and cocaine) for drug > neutral cues and average length of use.

IDRegionxyzVoxelspSDM - ZI2
Left superior frontal gyrus, medial, BA 10 -2 62 20 271 <0.001 -3.017 1.48 
Left superior frontal gyrus, dorsolateral, BA 9 -18 38 46 193 <0.001 -3.003 3.15 
IDRegionxyzVoxelspSDM - ZI2
Left superior frontal gyrus, medial, BA 10 -2 62 20 271 <0.001 -3.017 1.48 
Left superior frontal gyrus, dorsolateral, BA 9 -18 38 46 193 <0.001 -3.003 3.15 

Peaks arranged by decreasing voxel size.

Fig. 6.

Peak activations across 15 studies of the interaction between length of use and drug type for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates reported in MNI space.

Fig. 6.

Peak activations across 15 studies of the interaction between length of use and drug type for drug > neutral cues. Threshold was set at p < 0.005 TFCE-FWER and cluster size ≥ 10 voxels. Coordinates reported in MNI space.

Close modal

In the present study, we sought to investigate the differential neural underpinnings of these two major drug classes with respect to drug cue reactivity, a clinically significant process associated with recurrent use. We specifically focused on studies implementing cue reactivity paradigms, a gold-standard approach to the clinical neurobiology of craving, with a focus on responses to drug cues relative to neutral cues. Our results demonstrate that there are previously underappreciated differences in activation between people who use heroin and people who use cocaine. Specifically, increased (rather than decreased) activation in dopaminergic targets such as the striatum and mPFC, as well as the temporal and parietal gyri and cerebellum were found in people who use heroin that are not readily attributable to demographic or clinical differences across groups (see Tables 2 & 3).

4.1 Aggregate and substance specific effects

Aggregate mean analysis results revealed significant activation in the posterior cingulate gyrus, a region within the default mode network (DMN). The posterior cingulate, as well as the DMN at large, has been (e.g., right posterior cingulate gyrus) widely implicated in cue reactivity, as well as other paradigms, across substance use disorders (Devoto et al., 2020; Englemann et al., 2012; Janes et al., 2016; Wang et al., 2020). Further, our results also revealed significant activation in the superior and inferior frontal gyrus, which converge with recent meta-analyses, though not all specific to cue reactivity (Chase et al., 2011; Hill-Bowen et al., 2021; Klugah-Brown et al., 2020). In sum, our results are in line with previous findings as well as behavioral models of addiction, suggesting dysfunction in habit development, modulation of behavior, and cognitive control (Buhle et al., 2014; Everitt & Robbins, 2016; Klugah-Brown et al., 2020; Koob & Volkow, 2016; Wiers et al., 2015; Zilverstand et al., 2018).

Results from the mean analysis of heroin studies revealed significant activation in the left cingulate, the right superior and inferior frontal gyrus, the right anterior thalamic projections, the right and left amygdala, the left caudate nucleus, and the right cerebellum. These findings are consistent with rodent work indicating that acute heroin administration within the VTA enhanced calcium signaling in dopaminergic neurons as well as in serotonergic neurons within the dorsal raphe nucleus (DRN) (Wei et al., 2018), which, in turn, was associated with increased locomotor function. We note that similar patterns of results were reported in this study for administration of cocaine into the VTA and DRN (Wei et al., 2018), though we did not see consistent recruitment of dopaminergic regions in our mean analysis of cocaine studies. Instead, results for the cocaine mean analyses revealed significant peaks of activation in the right precentral gyrus. However, the influence of heterogeneity on cocaine results could not be rigorously tested due to a lack of data of clinical features. Our findings converge with some prior fMRI studies examining cue reactivity in people who use heroin and people who use cocaine. For example, Yang et al. (2009) conducted a study examining neural responses to cue reactivity in people who use heroin, reporting activation in the amygdala and cerebellum. Studies have also implicated increased recruitment of frontal regions and reward circuity, albeit with some variation in the specific peak locations when compared with our findings (Pollard et al., 2023; Yang et al., 2009). However, support for this result is variable, with other studies finding decreased activation in the left precentral gyrus (Moeller et al., 2010). It is difficult to compare these studies as methodology and task design varies greatly, however, indicating a need for replication of these findings using consistent task designs.

4.2 Between-substance effects

Interestingly, and contrary to our hypotheses, our direct comparison of cue reactivity between substances revealed increased recruitment of dopaminergic regions and other areas in people who use heroin relative to people who use cocaine. Given the robust literature on the involvement of reward processing regions such as the striatum and PFC across cocaine, heroin, and other substances (Cooper et al., 2019; Hobkirk et al., 2019; Klugah-Brown et al., 2020; Motzkin et al., 2014; Patel et al., 2013; Sell et al., 2001; Volkow et al., 2010; Yan et al., 2023), we did expect to find at least some regions of overlap, but our meta-analysis may have been underpowered to detect such effects. However, this could be explained by differential effectiveness of drugs eliciting cue reactivity; in other words, heroin cues were more effective at eliciting cue reactivity. A recent study found that when comparing cue induced and tonic craving between opioid and stimulant users in treatment, opioid cues consistently resulted in greater reports of craving (Hochheimer et al., 2023). Such results may, in turn, support the increased number of regions recruited in our sample of heroin cue reactivity studies. Alternatively, it is also possible that users of cocaine in the studies in our sample may be better able to engage regulatory mechanisms during their responses to cue-induced craving as suggested by dlPFC activation, a region consistently implicated in regulation of craving (Brevers et al., 2023; George & Koob, 2013; Hayashi et al., 2013; Kober et al., 2010; Li et al., 2017).

However, given the known neuropharmacological and clinical differences between these substances, the divergence in neural activation that emerged in our study is plausible. Support for this possibility comes from animal models that have demonstrated distinct recruitment of subpopulations of neurons within the striatum and PFC when comparing neural responses to cocaine and heroin (Badiani et al., 2011; Chang et al., 1998; Ettenberg et al., 1982; Pettit et al., 1984). Further, in spite of differing locations of activation associated with heroin and cocaine cues, there is evidence from other domains to suggest that differing locations of activation can be mapped back to the same network, ultimately implicating an overall common pathway (Jousta et al., 2022; Padmanabhan et al., 2019; Stubbs et al., 2023). Such a pattern suggests that viewing all substances as having the same neurobiological substrates because of common activation of regions such as the striatum and PFC, to some degree, may be an overgeneralization. It is, therefore, plausible that the activation differences we observed reflect preliminary neural underpinnings for the cognitive, behavioral, and clinical differences between heroin and cocaine use, including frequency of use and impulsive behavior, but replication and further research is needed (Bornovalova et al., 2005; Hser, Evans, et al., 2008; Hser, Huang, et al., 2008; Lejuez et al., 2005).

For example, cocaine and heroin have both been implicated in attentional dysfunction though the way in which each drug disrupts attention varies, such that individuals who use heroin display delayed attentional responses, whereas users of cocaine demonstrate increased interference with attention (Bjork et al., 2022). Opioid use has also been linked to impacts on decision making, including decreased response inhibition and delayed gratification (Psederska & Vassileva, 2023). Opiate use may affect other aspects of decision making, such as risky choices, differentially depending on length of abstinence (Psederska & Vassileva, 2023). Our results may parallel these findings, as both cocaine and heroin users show general activation in regions associated with attention and cognitive control (e.g., prefrontal cortex, parietal regions); however, users of heroin seem to show more significant activation in regions associated with cognitive control compared to users of cocaine, a finding which may be related to the differences in attentional disruptions found in behavioral tasks (Bjork et al., 2022; Psederska & Vassileva, 2023).

A recent study comparing whole-brain white matter abnormalities in users of both substances found that while both people who use heroin and people who use cocaine showed abnormalities in white matter, people who use heroin showed significantly decreased fractional anisotropy overall in the brain compared with people who use cocaine (Gaudreault et al., 2022), suggestive of decreased white matter integrity. Our results may dovetail with these in that we find studies of people who use heroin showing more diffuse activation than people who use cocaine in dopaminergic regions, as well as the right inferior and left middle temporal gyrus, the right thalamus, and the right cerebellum. Additional evidence from studies examining volumetric impacts of cocaine and heroin use have also converged on the involvement of the nucleus accumbens, which has shown significant gray matter reduction, in people who use both substances when compared to healthy controls; however, we note that this pattern was more robust in people who use heroin (Carlezon & Thomas, 2009; Ceceli et al., 2022; Seifert et al., 2015). Similar findings were reported in the vmPFC (Ceceli et al., 2022). These observed volumetric differences may suggest biomarkers of susceptibility to substance (ab)use or consequences of prolonged use, which may, in turn, be related to alterations in functional activation (Ceceli et al., 2022; Seifert et al., 2015). The incorporation of both functional and volumetric measurements, while controlling for contextual factors (e.g., duration of use, recency of use, age of onset) when examining the implications of substance use in future studies, may help address these questions (Oakes et al., 2007).

4.3 Length of use moderates heroin cue reactivity activation

Our exploratory meta-regression demonstrated that activation in the left superior frontal gyrus, both the medial and dorsolateral areas, decreased as length of use increased in people who use heroin, controlling for age. However, no significant relationship between cocaine and length of use was found. Literature examining the effect of duration of use on brain activation is limited; however, our findings fit within literature, suggesting that drug use is known to impact the prefrontal cortex (Goldstein & Volkow, 2012). A model put forth by Volkow et al. (2010) posits that one of the effects of chronic drug use is decreased reward sensitivity or hypoactivation in the prefrontal cortex. It is unclear whether this is a direct reflection of the effects of chronic drug use on the brain or perhaps better explained by the tolerance hypothesis—that is, that over time, one needs an increasing amount of a drug to produce the same effect (Peper, 2004). Additionally, Ferenzci et al. (2016) also found a similar pattern of deactivation in the PFC with length of substance use, though they also demonstrated a subsequent increase in BOLD activation in the striatum. While our findings did not show an increased BOLD response in the striatum, this relationship warrants more exploration as our lack of findings in this area may be a function of duration of use for study participants and our small sample size. This may require comparison across casual versus chronic use and examining developmental samples in future studies. Further, larger samples could evaluate both length of use and amount of use over time to explore this relationship further. Additionally, future studies should further examine the relationship between length of use and heroin as our study may not have incorporated enough variation in length of cocaine use to fully capture whether a relationship was meaningful.

4.4 Limitations

Limitations of this study include a small sample size due to limited availability of literature for within-group cue reactivity studies examining only people who use heroin or cocaine. The small sample size impacts the generalizability of these results to widespread populations of people who use cocaine and heroin. Another limitation of the current study is the lack of or inconsistencies in demographic and clinical data collected across studies, further limiting full consideration of these variables as moderators. Important co-variates such as additional/co-occurring diagnoses, amount of use, frequency of use, and so on need to be collected across all studies to enhance generalizability and provide opportunities for answering more complex and nuanced questions related to drug addiction. Further, there was variability across studies in the duration of use and time since recent use, preventing the ability to control for these factors in the analysis.

4.5 Future considerations

Given the clinical and behavioral findings that social context is related to drug choice, future studies should consider incorporating a social context into cue reactivity tasks. This would expand the clinical and behavioral findings to include the effects of social context on functional activation. Future studies should also consider implementing a standardized demographic and clinical data collection procedure that includes relevant contextual factors (e.g., time since last use, length of use, amount used) to enhance external validity and better understand the relationship between these factors and neural responses to drug cues (Ekhatiari et al., 2022). Further, future research should examine how the nuanced activation differences identified here may relate to differences identified in other areas of substance use, such as responsiveness to treatment (Lichenstein et al., 2021). Lastly, longitudinal studies may help answer whether and how such neural activation patterns during cue reactivity precede or succeed problematic substance use.

Taken together, our findings show that while there is some overlap in neural activation across heroin and cocaine use, there are distinct neural patterns uniquely associated with each substance. Additionally, these differential activation patterns in people who use heroin may be a function of length of use. Therefore, it is imperative that research continue to evaluate both the convergence and divergence of activation patterns across substances, as well as individual differences in length of use among other variables in order to relate such findings to behavioral differences as well as individualized treatments.

All text files used in this meta-analysis are accessible via the Open Science Framework at https://osf.io/gmu5v.

J.M.D. (Data Acquisition, Methodology, Writing – original draft, and Visualization), N.S. (Conceptualization, Data Curation), D.V.S. (Conceptualization, Methodology, Writing – review & editing, and Visualization), A.B.K. (Conceptualization, Writing – review & editing, and Visualization), and D.S.F. (Supervision, Formal Analysis, Writing – review & editing, and Visualization)

This manuscript is supported in part by funding from the National Institute on Mental Health (R15MH122927 to D.S.F.), the National Institute on Drug Abuse (R01DA053282, R01DA054201 to A.B.K.), and the National Institute on Aging (RF1-AG067011 to D.V.S).

The authors hereby declare that there are no competing financial or personal interests that could have appeared to influence the work reported in this paper.

Not applicable due to the meta-analytic study.

Abraham
,
A.
,
Pedregosa
,
F.
,
Eickenberg
,
M.
,
Gervais
,
P.
,
Mueller
,
A.
,
Kossaifi
,
J.
,
Gramfort
,
A.
,
Thirion
,
B.
, &
Varoquaux
,
G.
(
2014
).
Machine learning for neuroimaging with scikit-learn
.
Fronteirs in Neuroinformatics
,
8
,
14
. https://doi.org/10.3389/fninf.2014.00014
Ahn
,
W.-Y.
, &
Vassileva
,
J.
(
2016
).
Machine-learning identified substance-specific behavioral markers for opiate and stimulant dependence
.
Drug Alcohol Dependence
,
161
,
247
257
. https://doi.org/10.1016/j.drugalcdep.2016.02.008
Albajes-Eizagirre
,
A.
,
Solanes
,
A.
,
Fullana
,
M. A.
,
Ioannidis
,
J. P. A.
,
Fusar-Poli
,
P.
,
Torrent
,
C.
,
Solé
,
B.
,
Bonnín
,
C. M.
,
Vieta
,
E.
,
Mataix-Cols
,
D.
, &
Radua
,
J.
(
2019
).
Meta-analysis of voxel-based neuroimaging studies using seed-based d mapping with permutation of subject images (SDM-PSI)
.
Journal of Visualized Experiments
(
153
). https://doi.org/10.3791/59841
Albajes-Eizagirre
,
A.
,
Solanes
,
A.
, &
Radua
,
J.
(
2019
).
Meta-analysis of non-statistically significant unreported effects
.
Statistical Methods in Medical Research
,
28
(
12
),
3741
3754
. https://doi.org/10.1177/0962280218811349
Albajes-Eizagirre
,
A.
,
Solanes
,
A.
,
Vieta
,
E.
, &
Radua
,
J.
(
2019
).
Voxel-based meta-analysis via permutation of subject images (PSI): Theory and implementation for SDM
.
Neuroimage
,
186
,
174
184
. https://doi.org/10.1016/j.neuroimage.2018.10.077
Badiani
,
A.
,
Belin
,
D.
,
Epstein
,
D.
,
Calu
,
D.
, &
Shaham
,
Y.
(
2011
).
Opiate versus psychostimulant addiction: The differences do matter
.
Nature Reviews Neuroscience
,
12
(
11
),
685
700
. https://doi.org/10.1038/nrn3104
Badiani
,
A.
,
Caprioli
,
D.
, &
De Pirro
,
S.
(
2019
).
Opposite environmental gating of the experienced utility (‘liking’) and decision utility (‘wanting’) or heroin versus cocaine in animals and humans: Implications for computational neuroscience
.
Psychopharmacology
,
236
,
2451
2471
. https://doi.org/10.1007/s00213-019-05318-9
Bartra
,
O.
,
McGuire
,
J. T.
, &
Kable
,
J. W.
(
2013
).
The valuation system: A coordinate-based meta-analysis of BOLD fMRI experiments examining neural correlates of subjective value
.
Neuroimage
,
76
,
412
427
. https://doi.org/10.1016/j.neuroimage.2013.02.063
Bjork
,
J. M.
,
Keyser-Marcus
,
L.
,
Vassileva
,
J.
,
Ramey
,
T.
,
Houghton
,
D. C.
, &
Moeller
,
F. G.
(
2022
).
Attentional function and inhibitory control in different substance use disorders
.
Psychiatry Research
,
313
,
114591
. https://doi.org/10.1016/j.psychres.2022.114591
Bornovalova
,
M. A.
,
Daughters
,
S. B.
,
Hernandez
,
G. D.
,
Richards
,
J. B.
, &
Lejuez
,
C. W.
(
2005
).
Impulsivity and risk-taking propensity between primary users of crack cocaine and primary users of heroin in a residential substance-use program
.
Experimental and Clinical Psychopharmacology
,
13
(
4
),
311
318
. https://doi.org/10.1037/1064-1297.13.4.311
Brevers
,
D.
,
Baeken
,
C.
,
De Smet
,
S.
,
Catoira
,
B.
,
De Witte
,
S.
,
He
,
Q.
,
Maurage
,
P.
,
Schulze-Steinen
,
L.
,
Sescousse
,
G.
,
Vila Verde
,
C.
,
Vögele
,
C.
, &
Billieux
,
J.
(
2023
).
Stimulation of the dorsolateral prefrontal cortex modulates brain cue reactivity ot reward (un)availability
.
Cortex
,
164
,
51
62
. https://doi.org/10.1016/j.cortex.2023.03.008
Buhle
,
J. T.
,
Silvers
,
J. A.
,
Wagner
,
T. D.
,
Lopez
,
R.
,
Onyemekwu
,
C.
,
Kober
,
H.
,
Weber
,
J.
, &
Ochsner
,
K. N.
(
2014
).
Cognitive reappraisal of emotion: A meta-analysis of human neuroimaging studies
.
Cerebral Cortex
,
24
(
11
),
2981
2990
. https://doi.org/10.1093/cercor/bht154
Caprioli
,
D.
,
Celentano
,
M.
,
Dubla
,
A.
,
Luciantonio
,
F.
,
Nencini
,
P.
, &
Badiani
,
A
.
(
2009
).
Ambience and drug choice: Cocaine- and heroin-taking as a function of environmental context in humans
.
Biological Psychiatry
,
65
(
10
),
893
899
. https://doi.org/10.1016/j.biopsych.2008.12.009
Carlezon
,
W. A.
, &
Thomas
,
M. J.
(
2009
).
Biological substrates of reward and aversion: A nucleus accumbens activity hypothesis
.
Neuropharmacology
,
56
(
Suppl 1
),
122
132
. https://doi.org/10.1016/j.neuropharm.2008.06.075
Carter
,
B. L.
, &
Tiffany
,
S. T.
(
1999
).
Meta-analysis of cue-reactivity in addiction research
.
Addiction
,
93
(
3
),
327
340
. https://doi.org/10.1046/j.1360-0443.1999.9433273.x
Ceceli
,
A. O.
,
Huang
,
Y.
,
Krongberg
,
G.
,
Malaker
,
P.
,
Miller
,
P.
,
King
,
S.
,
Gaudreault
,
P.-O.
,
McClain
,
N.
,
Gabay
,
L.
,
Vasa
,
D.
,
Ekin
,
D.
,
Alia-Klein
,
N.
, &
Goldstein
,
R. Z.
(
2022
).
Common and distinct cortico-striatal volumetric changes in cocaine and heroin use disorder
.
Brain
,
146
(
4
),
1662
1671
. https://doi.org/10.1093/brain/awac366
Center for Disease Control and Prevention
. (
2023
).
Drug overdose
. https://www.cdc.gov/drugoverdose/deaths/other-drugs.html
Chang
,
J.-Y.
,
Janak
,
P. H.
, &
Woodward
,
D. J.
(
1998
).
Comparison of mesocorticolimbic neuronal responses during cocaine and heroin self-administration in freely moving rats
.
The Journal of Neuroscience
,
18
(
8
),
3098
3115
. https://doi.org/10.1523/JNEUROSCI.18-08-03098.1998
Chase
,
H. W.
,
Eickhoff
,
S. B.
,
Laird
,
A. R.
, &
Hogarth
,
L.
(
2011
).
The neural basis of drug stimulus processing and craving: An activation likelihood estimation meta-analysis
.
Biological Psychiatry
,
70
(
8
),
785
793
. https://doi.org/10.1016/j.biopsych.2011.05.025
Cooper
,
S.
,
Robison
,
A. J.
, &
Mazei-Robison
,
M. S.
(
2017
).
Reward circuitry in addiction
.
Neurotherapeutics
,
14
(
3
),
687
697
. https://doi.org/10.1007/s13311-017-0525-z
Courtney
,
K. E.
,
Schacht
,
J. P.
,
Hutchinson
,
K.
,
Roche
,
D. J. O.
, &
Ray
,
L. A.
(
2016
).
Neural substrates of cue-reactivity: Association with treatment outcomes and relapse
.
Addiction Biology
,
21
(
1
),
3
22
. https://doi.org/10.1111/adb.12314
Darke
,
S.
,
Duflou
,
J.
,
Peacock
,
A.
,
Chrzanowska
,
A.
,
Farrel
,
M.
, &
Lappin
,
J.
(
2023
).
Rates, characteristics and toxicology of cocaine-related deaths in Australia, 2000–2021
.
Addiction
,
111
(
2
),
297
306
. https://doi.org/10.1111/add.16055
De Pirro
,
S.
,
Galati
,
G.
,
Pizzamiglio
,
L.
, &
Badiani
,
A.
(
2018
).
The affective and neural correlates of heroin vs cocaine use in addiction are influenced by environmental setting but in opposite directions
.
The Journal of Neuroscience
,
38
(
22
),
5182
5195
. https://doi.org/10.1523/jneurosci.0019-18.2018
Degenhardt
,
L.
,
Whiteford
,
H. A.
,
Baxter
,
A. J.
,
Charlson
,
F. J.
,
Hall
,
W. D.
,
Freedman
,
G.
,
Burstein
,
R.
,
Johns
,
N.
,
Engell
,
R. E.
,
Flaxman
,
A.
,
Murray
,
C. L.
, &
Vos
,
T.
(
2013
).
Global burden of disease attributable to illicit drug use and dependence: Findings from the global burden of disease study 2010
.
Lancet
,
382
(
9904
),
1564
1574
. https://doi.org/10.1016/s0140-6736(13)61530-5
Devoto
,
F.
,
Zapparoli
,
L.
,
Spinelli
,
G.
,
Scotti
,
G.
, &
Paulesu
,
E.
(
2020
).
How the harm of drugs and their availability affect brain reactions to drug cues: A meta-analysis of 64 neuroimaging activation studies
.
Translational Psychiatry
,
10
(
1
),
429
. https://doi.org/10.1038/s41398-020-01115-7
Drummond
,
D. C.
(
2000
).
What does cue-reactivity have to offer clinical research
?
Addiction
,
95
(
Supplement 2
),
S129
S144
. https://doi.org/10.1080/09652140050111708
Duncan
,
E.
,
Boshoven
,
W.
,
Harenski
,
K.
,
Fiallos
,
A.
,
Tracy
,
H.
,
Jovanovic
,
T.
,
Hu
,
X.
,
Drexler
,
D.
, &
Kilts
,
C.
(
2007
).
An fMRI study of the interaction of stress and cocaine cues on cocaine craving on cocaine-dependent men
.
The American Journal on Addictions
,
16
,
174
182
. https://doi.org/10.1080/10550490701375285
Ekhtiari
,
H.
,
Zare-Bidoky
,
M.
,
Sangchooli
,
A.
,
Janes
,
A. C.
,
Kaufman
,
M. J.
,
Oliver
,
J. A.
,
Prisciandaro
,
J. J.
,
Wüstenberg
,
T.
,
Anton
,
R. F.
,
Bach
,
P.
,
Baldacchino
,
A.
,
Beck
,
A.
,
Bjork
,
J. M.
,
Brewer
,
J.
,
Childress
,
A. R.
,
Claus
,
E. D.
,
Courtney
,
K. E.
,
Ebrahimi
,
M.
,
Filbey
,
F. M.
,
Ghahremani
,
D. G.
, …
Zilverstand
,
A.
(
2022
).
A methodological checklist for fMRI drug cue reactivity studies: Development and expert consensus
.
Nature Protocols
,
17
(
3
),
567
595
. https://doi.org/10.1038/s41596-021-00649-4
Elton
,
A.
,
Smitherman
,
S.
,
Young
,
J.
, &
Kilts
,
C. D.
(
2015
).
Effects of childhood maltreatment on the neural correlates of stress- and drug cue-induced cocaine craving
.
Addiction Biology
,
20
(
4
),
820
831
. https://doi.org/10.1111/adb.12162
Englemann
,
J. M.
,
Versace
,
F.
,
Robinson
,
J. D.
,
Minnix
,
J. A.
,
Lam
C. Y.
,
Brown
,
V. L.
, &
Cincirpini
,
P. M.
(
2012
).
Neural substrates of smoking cue reactivity: A meta-analysis of fMRI Studies
.
Neuroimage
,
61
(
1
),
252
262
. https://doi.org/10.1016/j.neuroimage.2011.12.024
Ettenberg
,
A.
,
Pettit
,
H. O.
,
Bloom
,
F. E.
, &
Koob
,
G. F.
(
1982
).
Heroin and cocaine intravenous self-administration in rats: Mediation by separate neural systems
.
Psychopharmacology
,
78
,
204
209
. https://doi.org/10.1007/bf00428151
Everitt
,
B. J.
, &
Robbins
,
T. W.
(
2005
).
Neural systems of reinforcement for drug addiction: From actions to habits to compulsion
.
Nature Neuroscience
,
8
(
11
),
Article 11
. https://doi.org/10.1038/nn1579
Everitt
,
B. J.
, &
Robbins
,
T. W
.
(
2016
).
Drug addiction: Updating actions to habits to compulsion ten years on
.
Annual Review of Psychology
,
67
,
23
50
. https://doi.org/10.1146/annurev-psych-122414-033457
Ferenczi
,
E. A.
,
Zalocusky
,
K. A.
,
Liston
,
C.
,
Grosenick
,
L.
,
Warden
,
M. R.
,
Amatya
,
D.
,
Katovich
,
K.
,
Mehta
,
H.
,
Patenaude
,
B.
,
Ramakrishnan
,
C.
,
Kalanithi
,
P.
,
Etkin
,
A.
,
Knuston
,
B.
,
Glover
,
G. H.
, &
Deisseroth
,
K.
(
2016
).
Prefrontal cortical regulation of brainwide circuit dynamics and reward-related behavior
.
Science
,
351
, asc9698. https://doi.org/10.1126.science.aac9698
Gaudreault
,
P.-O.
,
King
,
S. G.
,
Malaker
,
P.
,
Alia-Klein
,
N.
, &
Goldstein
,
R. Z.
(
2022
).
Whole-brain white matter abnormalities in human cocaine and heroin use disorders: Association with craving, recency, and cumulative use
.
Molecular Psychiatry
,
28
(
2
),
780
791
. https://doi.org/10.1038/s41380-022-01833-y
George
,
O.
, &
Koob
,
G. F.
(
2013
).
Control of craving by the prefrontal cortex
.
Proceedings of the National Academy of Sciences of the United States of America
,
110
(
11
),
4165
4166
. https://doi.org/10.1073/pnas.1301245110
Gheraldini
,
C.
,
Di Cesare Mannelli
,
L.
, &
Bianchi
,
E.
(
2015
).
The pharmacological basis of opioids
.
Clinical Cases in Mineral and Bone Metabolism
,
12
(
3
),
219
221
. https://doi.org/10.11138/ccmbm/2015.12.3.219
Goldstein
,
R. Z.
,
Tomasi
,
D.
,
Alia-Klein
,
N.
,
Carrillo
,
J. H.
,
Maloney
,
T.
,
Woicik
,
P. A.
,
Wang
,
R.
,
Telang
,
F.
, &
Volcow
,
N. A.
(
2009
).
Dopaminergic response to drug words in cocaine addiction
.
The Journal of Neuroscience
,
29
(
18
),
6001
6001
. https://doi.org/10.1523/jneurosci.4247-08.2009
Goldstein
,
R. Z.
, &
Volkow
,
N. D.
(
2002
).
Drug addiction and its underlying neurobiological basis: Neuroimaging evidence for the involvement of the frontal cortex
.
American Journal of Psychiatry
,
159
(
10
),
1642
1652
. https://doi.org/10.1176/appi.ajp.159.10.1642
Goldstein
,
R. Z.
, &
Volkow
,
N. D.
(
2012
).
Dysfunction of the prefrontal cortex in addiction: Neuroimaging findings and clinical implications
.
Nature Reviews Neuroscience
,
12
(
11
),
652
669
. https://doi.org/10.1038/nrn3119
Haber
,
S. N.
, &
Knuston
,
B. N.
(
2010
).
The reward circuit: Linking primate anatomy and human imaging
.
Neuropsychopharmacology
,
35
,
4
26
. https://doi.org/10.1038/npp.2009.129
Hanlon
,
C. A.
,
Dowdle
,
L. T.
,
Gibson
,
N. B.
,
Li
,
X.
,
Hamilton
,
S.
,
Centerberry
,
M.
, &
Hoffman
,
M.
(
2018
).
Cortical substrates of cue-reactivity in multiple substance dependent populations: Transdiagnostic relevance of the medial prefrontal cortex
.
Translational Psychiatry
,
8
,
186
. https://doi.org/10.1038/s41398-018-0220-9
Hassani-Abharian
,
P.
,
Ganjgahi
,
H.
,
Tabatabaei-Jafari
,
H.
,
Oghabian
,
M. A.
,
Mokri
,
A.
, &
Ekhtiari
,
H.
(
2015
).
Exploring neural correlates of different dimensions in drug craving self-reports among heroin dependents
.
Basic and Clinical Neuroscience,
6
(
4
),
271
284
. PMID:
[PubMed]
; PMCID:
[PubMed]
.
Hayashi
,
T.
,
Ko
,
J. H.
,
Strafella
,
A. P.
, &
Dagher
,
A.
(
2013
).
Dorsolateral prefrontal and orbitofrontal cortex interactions during self-control of cigarette craving
.
Proceedings of the National Academy of Sciences of the United States of America
,
110
(
11
),
4422
4427
. https://doi.org/10.1073/pnas.1212185110
Hedegaard
,
H.
,
Minińo
,
A. M.
, &
Warner
,
M.
(
2020
).
National center for health statistics data (NCHS brief no. 394)
. Centers for Disease Control. https://www.cdc.gov/nchs/products/index.htm.
Higgins
,
J. P. T.
, &
Thompson
,
S. G
.
(
2002
).
Quantifying heterogeneity in a meta-analysis
.
Stat Med
,
21
(
11
),
1539
1538
. https://doi.org/10.1002/sim.1186
Hill-Bowen
,
D.
,
Riedel
,
M. C.
,
Salo
,
T.
,
Flannery
,
J. S.
,
Poudel
,
R.
,
Laird
,
A. R.
, &
Sutherland
,
M. T.
(
2022
).
Convergent gray matter alterations across drugs of abuse and network-level implications: A meta-analysis of structural MRI studies
.
Drug and Alcohol Dependence
,
240
,
109625
. https://doi.org/10.1016/j.drugalcdep.2022.109625
Hill-Bowen
,
L. D.
,
Riedel
,
M. C.
,
Poudel
,
R.
,
Salo
,
T.
,
Flannery
,
J. S.
,
Camillerio
,
J. A.
,
Eickhoff
,
S. B.
,
Laird
,
A. R.
, &
Sutherland
,
M. T.
(
2021
).
The cue-reactivity paradigm: An ensemble of networks driving attention and cognition when viewing drug and natural reward-related stimuli
.
Neuroscience & Biobehavioral Reviews
,
130
,
201
203
. https://doi.org/10.1016/j.neubiorev.2021.08.010
Hobkirk
,
A. L.
,
Bell
,
R. P.
,
Utevsky
,
A. V.
,
Huettel
,
S.
, &
Meade
,
C. S.
(
2019
).
Reward and executive control network resting-state functional connectivity is associated with impulsivity during reward-based decision making for cocaine users
.
Drug and Alcohol Dependence
,
194
(
1
),
32
39
. https://doi.org/10.1016/j.drugalcdep.2018.09.013
Hochheimer
,
M.
,
Strickland
,
J. C.
,
Rabinowitz
,
J. A.
,
Ellis
,
J. D.
,
Bergeria
,
C. L.
,
Hobelmann
,
J. G.
, &
Huhn
,
A. S.
(
2023
).
The impact of opioid-stimulant co-use on tonic and cue-induced craving
.
Journal of Psychiatric Research
,
164
,
15
22
. https://doi.org/10.1016/j.jpsychires.2023.05.018
Hser
,
Y.-I.
,
Evans
,
E.
,
Huang
,
D.
,
Brecht
,
M.-L.
, &
Li
,
L.
(
2008
).
Comparing the dynamic course of heroin, cocaine, and methamphetamine use over 10 years
.
Addictive Behaviors
,
33
,
1581
1589
. https://doi.org/10.1016/j.addbeh.2008.07.024
Hser
,
Y.-I.
,
Huang
,
D.
,
Brecht
,
M.-L.
,
Li
,
L.
, &
Evans
,
E.
(
2008
).
Contrasting trajectories of heroin, cocaine, and methamphetamine use
.
Journal of Addictive Diseases
,
27
(
3
),
13
. https://doi.org/10.1080/10550880802122554
Huang
,
A. S.
,
Mitchell
,
J. A.
,
Haber
,
S. N.
,
Alia-Klein
,
N.
, &
Goldstein
,
R. Z.
(
2018
).
The thalamus in drug addiction: From rodents to humans
.
Philosophical Transactions of the Royal Society B: Biological Sciences
,
373
(
1742
),
20170028
. https://doi.org/10.1098/rstb.2017.0028
Janes
,
A. C.
,
Betts
,
J.
,
Jenson
,
J. E.
, &
Lukas
,
S. E.
(
2016
).
Dorsal anterior cingulate glutamate is associated with engagement of the default mode network during exposure to smoking cues
.
Drug and Alcohol Dependence
,
167
(
1
),
75
81
. https://doi.org/10.1016/j.drugalcdep.2016.07.021
Jentsch
,
J.
, &
Taylor
,
J. R.
(
1999
).
Impulsivity resulting from frontostriatal dysfunction in drug abuse: Implications for the control of behavior by reward-related stimuli
.
Psychopharmacology
,
146
,
373
390
. https://doi.org/10.1007/PL00005483
Jousta
,
J.
,
Moussawi
,
K.
,
Siddiqi
,
S. H.
,
Abdolhai
,
A.
,
Drew
,
W.
,
Cohen
,
A. L.
,
Ross
,
T. J.
,
Desgpande
,
H. U.
,
Wang
,
H. Z.
,
Bruss
,
J.
,
Stein
,
E. A.
,
Volkow
,
N. D.
,
Grafman
,
J. H.
,
van Wijngaarden
,
E.
,
Boes
,
A. D.
, &
Fox
,
M. D.
(
2022
).
Brain lesions disrupting addiction map to a common human brain circuit
.
Nature Medicine
,
28
,
1249
1255
. https://doi.org/10.1038/s41591-022-01834-y
Kaag
,
A. M.
,
Reneman
,
L.
,
Homberg
,
J.
,
van den Brink
,
W.
, &
van Wingen
,
G. A.
(
2018
).
Enhanced amygdala-striatal functional connectivity during the processing of cocaine cues in male cocaine users with a history of childhood trauma
.
Frontiers in Psychiatry
,
9
,
70
. https://doi.org/10.3389/fpsyt.2018.00070
Liu
,
S.
,
Wang
,
S.
,
Zhang
,
M.
,
Xu
,
Y.
,
Liu
,
J.
, &
Yuan
,
K
.
(
2021
).
Brain responses to drug cues predict craving changes in abstinent heroin users: A preliminary study
.
Neuroimage
,
237
,
118169
. https://doi.org/10.1016/j.neuroimage.2021.118169
Klugah-Brown
,
B.
,
Di
,
X.
,
Zweerings
,
J.
,
Mathiak
,
K.
,
Becker
,
B.
, &
Biswal
,
B.
(
2020
).
Common and seperable neural alterations in substance use disorder: A coordinate-based meta-analysis of functional neuroimaging studies in humans
.
Human Brain Mapping
,
41
(
16
),
4459
4477
. https://doi.org/10.1002/hbm.25085
Klugah-Brown
,
B.
,
Jiang
,
C.
,
Agoalikum
,
E.
,
Zhou
,
X.
,
Zou
,
L.
,
Yu
,
Q.
,
Becker
,
B.
, &
Biswal
,
B.
(
2021
).
Common abnormality of gray matter integrity in substance use disorder and obsessive-compulsive disorder: A comparative voxel-based meta-analysis
.
Human Brain Mapping
,
42
,
3871
3886
. https://doi.org/10.1002/hbm.25471
Kober
,
H.
,
Mende-Siedlecki
,
P.
,
Kross
,
E. F.
, &
Oschner
,
K .N.
(
2010
).
Prefrontal-striatal pathway underlies cognitive regulation of craving
.
Proceedings of the National Academy of Sciences of the United States of America
,
107
(
33
),
14811
14816
. https://doi.org/10.1073/pnas.1007779107
Koob
,
G. F.
, &
Volkow
,
N. D.
(
2016
).
Neurobiology of addiction: A neurocircuitry analysis
.
Lancet Psychology
,
3
(
8
),
760
773
. https://doi.org/10.1016/S2215-0366(16)00104-8
Kuhn
,
S.
, &
Gallinat
,
J.
(
2011
).
Common biology of craving across legal and illegal drugs − a quantitative meta-analysis of cue-reactivity brain response
.
Behavioral Neuroscience
,
33
(
7
),
1318
1326
. https://doi.org/10.1111/j.1460-9568.2010.07590.x
Lejuez
,
C. W.
,
Bornovalova
,
M. A.
,
Daughters
,
S. B.
, &
Curtin
,
J. J.
(
2005
).
Differences in impulsivity and sexual risk behavior among inner-city crack/cocaine users and heroin users
.
Drug and Alcohol Dependence
,
77
,
169
175
. https://doi.org/10.1016/j.drugalcdep.2004.08.013
Lewer
,
D.
,
Brothers
,
T. D.
,
Van Hest
,
N.
, &
Hickman
,
M.
(
2022
).
Causes of death among people who used illicit opioids in england, 2001-18: A matched cohort study
.
The Lancet
,
7
(
2
),
126
135
. https://doi.org/10.1016/s2468-2667(21)00254-1
Li
,
Q.
,
Wang
,
Y.
,
Zhang
,
Y.
,
Li
,
W.
,
Yang
,
W.
,
Zhu
,
J.
,
Wu
,
N.
,
Chang
,
H.
,
Zheng
,
Y.
,
Qin
,
W.
,
Zhao
,
L.
,
Yuan
,
K.
,
Liu
,
J.
,
Wang
,
W.
, &
Tian
,
J.
(
2012
).
Craving correlates with mesolimbic responses to heroin-related cues in short-term abstinence from heroin: An event-related fMRI study
.
Brain Research
,
1469
,
63
72
. https://doi.org/10.1016/j.brainres.2012.06.024
Li
,
Q.
,
Wang
,
Y.
,
Zhang
,
Y.
,
Li
,
W.
,
Zhu
,
J.
,
Zheng
,
Y.
,
Chen
,
J.
,
Zhao
,
L.
,
Zhou
,
Z.
,
Liu
,
Y.
,
Wang
,
W.
, &
Tian
,
J.
(
2013
).
Assessing cue-induced brain response as a function of abstinence duration in heroin-dependent individuals: An event-related fMRI study
.
PLoS One
,
8
(
5
),
e62911
. https://doi.org/10.1371/journal.pone.0062911
Li
,
Y.
,
Yuan
,
K.
,
Bi
,
Y.
,
Guan
,
Y.
,
Cheng
,
J.
,
Zhang
,
Y.
,
Shi
,
S.
,
Lu
,
X.
,
Yu
,
D.
, &
Tian
,
J.
(
2017
).
Neural correlates of 12-h abstinence-induced craving in young adult smokers: A resting-state study
.
Brain Imaging and Behavior
,
11
,
677
684
. https://doi.org/10.1007/s11682-016-9544-3
Lichenstein
,
S. D.
,
Scheinost
,
D.
,
Potenza
,
M. N.
,
Carroll
,
K. M.
, &
Yip
,
S. W.
(
2021
).
Dissociable neural substrates of opioid and cocaine use identified via connectome-based modelling
.
Molecular Psychiatry
,
26
(
8
),
4383
4393
. https://doi.org/10.1038/S41380-019-0586-Y
Long
,
Y.
,
Pan
,
N.
,
Ji
,
S.
,
Chen
,
Y.
,
Zhang
,
X.
,
He
,
M.
,
Suo
,
X.
,
Yu
,
Y.
,
Wang
,
S.
, &
Gong
,
Q.
(
2022
).
Distinct brain structure abnormalities in attention-deficit/hyperactivity disorder and substance use disorder: A comparative meta-analysis
.
Translational Psychiatry
,
12
,
368
. https://doi.org/10.1038/s41398-022-02130-6
Mei
,
W.
,
Zhang
,
J. X.
, &
Xiao
,
Z.
(
2010
).
Acute effects of sublingual buprenorphine on brain responses to heroin-related cues in early-abstinent heroin addicts: An uncontrolled trial
.
Neuroscience
,
170
,
808
815
. https://doi.org/10.1016/j.neuroscience.2010.07.033
Mitchell
,
M. R.
, &
Potenza
,
M. N.
(
2014
).
Addictions and personality traits: Impulsivity and related constructs
.
Current Behavioral Neuroscience Reports
,
1
,
1
12
. https://doi.org/10.1007/s40473-013-0001-y
Moeller
,
F. G.
,
Steinberg
,
J. L.
,
Schmitz
,
J. M.
,
Ma
,
L.
,
Liu
,
S.
,
Kjome
,
K. L.
,
Rathnayaka
,
N.
,
Kramer
,
L. A.
, &
Narayana
,
P. A.
(
2010
).
Working memory fMRI activation in cocaine-dependent subjects: Association with treatment response
.
Psychiatry Research: Neuroimaging
,
18
,
174
182
. https://doi.org/10.1016/j.pscychresns.2009.11.003
Motzkin
,
J. C.
,
Baskin-Sommers
,
A.
,
Newman
,
J. P.
,
Kiehl
,
K. A.
, &
Koenigs
,
M.
(
2014
).
Neural correlates of substance abuse: Reduced functional connectivity between areas underlying reward and cognitive control
.
Human Brain Mapping
,
35
(
9
),
4282
4292
. https://doi.org/10.1002/hbm.22474
National Institute on Drug Abuse
. (
2022
).
Overdose Death Rates
.
NIDA
.https://nida.nih.gov/research-topics/trends-statistics/overdose-death-rates
Oakes
,
T. R.
,
Fox
,
A. S.
,
Johnstone
,
T.
,
Chung
,
M. K.
,
Kalin
,
N.
, &
Davidson
,
R. J.
(
2007
).
Integrating VBM into the general linear model with voxelwise anatomical covariates
.
Neuroimage
,
34
(
2
),
500
508
. https://doi.org/10.1016/j.neuroimage.2006.10.007
Padmanabhan
,
J. L.
,
Cooke
,
D.
,
Joutsa
,
J.
,
Siddiqi
,
S. H.
,
Ferguson
,
M.
,
Darby
,
R. R.
,
Soussand
,
L.
,
Horn
,
A.
,
Kim
,
N. Y.
,
Voss
,
J. L.
,
Naidech
,
A. M.
,
Brodtmann
,
A.
,
Egorova
,
N.
,
Gozzi
,
S.
,
Phan
,
T. G.
,
Corbetta
,
M.
,
Grafman
,
J.
, &
Fox
,
M. D.
(
2019
).
A human depression circuit derived from focal brain lesions
.
Biological Psychiatry
,
86
(
10
),
749
758
. https://doi.org/10.1016/j.biopsych.2019.07.023
Pando-Nuade
,
V.
,
Toxto
,
S.
,
Fernandez-Lozano
,
S.
,
Parsons
,
C. E.
,
Alcauter
,
S.
, &
Garza-Villarreal
,
E. A.
(
2021
).
Gray and white matter morphology in substance use disorders: A neuroimaging systematic review and meta-analysis
.
Translational Psychiatry
,
11
,
11
29
. https://doi.org/10.1038/s41398-020-01128-2
Patel
,
K. T.
,
Stevens
,
M. C.
,
Meda
,
S. A.
,
Muska
,
C.
,
Thomas
,
A. D.
,
Potenza
,
M. N.
, &
Pearlson
,
G. D.
(
2013
).
Robust changes in reward circuitry during reward loss in current and former cocaine users during performance of a monetary incentive delay task
.
Biological Psychiatry
,
74
(
1
),
529
537
. https://doi.org/10.1016/j.biopsych.2013.04.029
Peacock
,
A.
,
Tran
,
L. T.
,
Larney
,
S.
,
Stockings
,
E.
,
Santo
Jr.,
T.
,
Jones
,
H.
,
Santomauro
,
D.
, &
Degenhardt
,
L.
(
2021
).
All-cause and cause-specific mortality among people with regular or problematic cocaine use: A systematic review and meta-analysis
.
Addiction
,
116
,
725
742
. https://doi.org/10.1111/add.15239
Peper
,
A.
(
2004
).
A theory of drug tolerance and dependence I: A conceptual analysis
.
Journal of Theoretical Biology
,
229
(
4
),
477
490
. https://doi.org/10.1016/j.jtbi.2004.04.010
Perdue
,
T.
,
Carlson
,
R.
,
Daniulaityte
,
R.
,
Silverstein
,
S. M.
,
Bluthenthal
,
R. N.
,
Valdez
,
A.
, &
Cepeda
,
A.
(
2024
).
Characterizing prescription opiod, heroin, and fentanyl initiation trajectories: A qualitative study
.
Social Science & Medicine
,
340
,
116441
. https://doi.org/10.1016/j.socscimed.2023.116441
Pettit
,
H. O.
,
Ettenberg
,
A.
,
Bloom
,
F. E.
, &
Koob
,
G. F.
(
1984
).
Destruction of dopamine in the nucleus accumbens selectively attenuates cocaine but not self-administration in rats
.
Psychopharmacology
,
84
,
167
173
. https://doi.org/10.1007/bf00427441
Pollard
,
A. A.
,
Hauson
,
A. O.
,
Lackey
,
N. S.
,
Zhang
,
E.
,
Khayat
,
S.
,
Carson
,
B.
,
Fortea
,
L.
,
Radua
,
J.
, &
Grant
,
I.
(
2023
).
Functional neuroanatomy of craving in heroin use disorder: Voxel-based meta-analysis of functional magnetic resonance imaging (fMRI) drug cue reactivity studies
.
Am J Drug Alcohol Abuse
,
49
(
4
),
418
430
. https://doi.org/10.1080/00952990.2023.2172423
Prisciandaro
,
J. J.
,
McRae-Clark
,
A. L.
,
Myrick
,
H.
,
Henderson
,
S.
, &
Brady
,
K. T.
(
2014
).
Brain activation to cocaine cues and motivation/treatment status
.
Addiction Biology
,
19
(
2
),
240
249
. https://doi.org/10.1111/j.1369-1600.2012.00446.x
Prisciandaro
,
J. J.
,
Myrick
,
H.
,
Henderson
,
S.
,
McRae-Clark
,
A. L.
, &
Brady
,
K. T.
(
2013
).
Prospective associations between brain activation to cocaine and no-go cues and cocaine relapse
.
Drug and Alcohol Dependence
,
131
(
0
),
44
49
. https://doi.org/10.1016/j.drugalcdep.2013.04.008
Prisciandaro
,
J. J.
,
Myrick
,
H.
,
Henderson
,
S.
,
McRae-Clark
,
A. L.
,
Santa Ana
,
E. J.
,
Saladin
,
M. E.
, &
Brady
,
K. T.
(
2013
).
Impact of DCS-facilitated cue exposure therapy on brain activation to cocaine cues in cocaine dependence
.
Drug and Alcohol Dependence
,
132
(
0
),
195
201
. https://doi.org/10.1016/j.drugalcdep.2013.02.009
Psederska
,
E.
, &
Vassileva
,
J.
(
2023
).
Neurocognitive impulsivity in opiate users at difference lengths of abstinence
.
International Journal of Environmental Research and Public Health
,
20
,
1236
. https://doi.org/10.3390/ijerph20021236
Sartor
,
C. E.
,
Kranzler
,
H. R.
, &
Gelernter
,
J.
(
2014
).
Characteristics and course of dependence in cocaine-dependent individuals who never used alcohol or marijuana or used cocaine first
.
Journal of Studies on Alcohol and Drugs
,
75
(
3
),
423
427
. https://doi.org/10.15288/jsad.2014.75.423
Schneider
,
S.
,
Richeter
,
C.
,
Niethammer
,
R.
, &
Beisel
,
L.
(
2021
).
Fatal and non-fatal heroin-related overdose circumstances and patterns
.
Substance Use & Misuse
,
56
(
13
),
1997
2006
. https://doi.org/10.1080/10826084.2021.1963986
Seifert
,
C. L.
,
Magon
,
S.
,
Sprenger
,
T.
,
Lang
,
U. E.
,
Huger
,
C. G.
,
Denier
,
N.
,
Vogel
,
M.
,
Schmidt
,
A.
,
Raude
,
E.-W.
,
Borgwardt
,
S.
, &
Walter
,
M.
(
2015
).
Reduced volume of the nucleus accumbens in heroin addiction
.
European Archives of Psychiatry and Clinical Neuroscience
,
265
,
637
645
. https://doi.org/10.1007/s00406-014-0564-y
Sell
,
L. A.
,
Morris
,
J.
,
Bearn
,
J.
,
Frackowiak
,
R. S. J.
,
Friston
,
K. J.
, &
Dolan
,
R. J.
(
2001
).
Activation of reward circuitry in human opiate addicts
.
European Journal of Neuroscience
,
11
(
3
),
1042
1048
. https://doi.org/10.1046/j.1460-9568.1999.00522.x
Shand
,
F. L.
,
Degenhardt
,
L.
,
Slade
,
T.
, &
Nelson
,
E. C.
(
2011
).
Sex differences amongst dependent heroin users: Histories, clinical characteristics and predictors of other substance dependence
.
Addictive Behaviors
,
36
(
1–2
),
27
36
. https://doi.org/10.1016/j.addbeh.2010.08.008
Stubbs
,
J. L.
,
Taylor
,
J. J.
,
Siddiqi
,
S. H.
,
Schapner
,
L. W. V. J. F.
,
Cohen
,
A. L.
,
Drew
,
W.
,
Hanlon
,
C. A.
,
Abdolhai
,
A.
,
Wang
,
H. Z.
,
Honer
,
W. G.
,
Panenka
,
W. J.
, &
Fox
,
M. D.
(
2023
).
Hetergenous neuroimaging findings across substance use disorder localize to a common brain network
.
Nature Mental Health
,
1
,
772
781
. https://doi.org/10.1038/s44220-023-00128-7
Vafaie
,
N.
, &
Kober
,
H.
(
2022
).
Association of drug cues and craving with drug use relapse: A systemic review and meta-analysis
.
JAMA Psychiatry
,
79
(
7
),
641
650
. https://doi.org/10.1001/jamapsychiatry.2022.1240
Vassileva
,
J.
,
Paxton
,
J.
,
Moeller
,
F. G.
,
Wilson
,
M. J.
,
Bozgunov
,
K.
,
Martin
,
E. M.
,
Gonzalez
,
R.
, &
Vasilev
,
G.
(
2014
).
Heroin and amphetamine users display opposite relationships between trait and neurobehavioral dimensions of impulsivity
.
Addictive Behaviors
,
39
,
652
659
. https://doi.org/10.1016/j.addbeh.2013.11.020
Volkow
,
N. D.
,
Wang
,
G.-J.
,
Fowler
,
J. S.
,
Tomasi
,
D.
,
Telang
,
F.
, &
Baler
,
R.
(
2010
).
Addiction: Decreased reward sensitivity and increased expectation sensitivity conspire to overwhelm the brain’s control circuit
.
Bioessays
,
32
(
9
),
748
755
. https://doi.org/10.1002/bies.201000042
Walter
,
M.
,
Denier
,
N.
,
Gerber
,
H.
,
Schmid
,
O.
,
Lanz
,
C.
,
Brenneisen
,
R.
,
Riecher-Rössler
,
A.
,
Wiesbeck
,
G. A.
,
Scheffler
,
K.
,
Seifritz
,
E.
,
McGuire
,
P.
,
Fusar-Poli
,
P.
, &
Borgwardt
,
S.
(
2015
).
Orbitofrontal response to drug-related stimuli after heroin administration
.
Addiction Biology
,
20
(
3
),
570
579
. https://doi.org/10.1111/adb.12145
Wang
,
A.-L.
,
Elman
,
I.
,
Lowen
,
S. B.
,
Blady
,
S. J.
,
Lynch
,
K. G.
,
Hyatt
,
J. M.
,
O’Brien
,
C. P.
, &
Langleben
,
D. D.
(
2015
).
Neural correlates of adherence to extended-release naltrexone pharmacotherapy in heroin dependence
.
Translational Psychiatry
,
5
,
e531
. https://doi.org/10.1038/tp.2015.20
Wang
,
K. S.
,
Haiser
,
R. H.
,
Peechatka
,
A. L.
,
Frederick
,
B. B.
, &
Janes
,
A. C.
(
2020
).
Temporal dynamics of large-scale networks predict neural cue reactivity and cue-induced craving
.
Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
,
5
(
11
),
1011
1018
. https://doi.org/10.1016/j.bpsc.2020.07.006
Wang
,
K. S.
,
Smith
,
D. V.
, &
Delgado
,
M. R.
(
2016
).
Using fMRI to study reward processing in humans: Past, present, and future
.
Journal of Neurophysiology
,
115
,
1664
1678
. https://doi.org/10.1152/jn.00333.2015
Wang
,
W.
,
Li
,
Q.
,
Wang
,
Y.
,
Tian
,
J.
,
Yang
,
W.
,
Li
,
W.
,
Qin
,
W.
,
Yuan
,
K.
, &
Liu
,
J.
(
2011
).
Brain fMRI and craving response to heroin-related cues in patients on methadone maintenance
.
The American Journal of Drug and Alcohol Abuse
,
37
(
2
),
123
130
. https://doi.org/10.3109/00952990.2010.543997
Wei
,
C.
,
Han
,
X.
,
Weng
,
D.
,
Feng
,
Q.
,
Qi
,
X.
,
Li
,
J.
, &
Luo
,
M.
(
2018
).
Response dynamics of midbrain dopamine neurons and serotonin neurons to heroin, nicotine, cocaine, and MDMA
.
Cell Discovery
,
4
,
60
. https://doi.org/10.1038/s41421-018-0060-z
Wiers
,
C. E.
,
Gawron
,
C. K.
,
Gröpper
,
S.
,
Spengler
,
S.
,
Stuke
,
H.
,
Lindenmeyer
,
J.
,
Walter
,
H.
, &
Bermpohl
,
F.
(
2015
).
Decreased gray matter volume in inferior frontal gyrus is related to stop-signal task performance in alcohol-depdendent patients
.
Psychiatry Research: Neuroimaging
,
233
(
2
),
125
130
. https://doi.org/10.1016/j.pscychresns.2015.05.006
Xiao
,
Z.
,
Lee
,
T.
,
Zhang
,
J. X.
,
Wu
,
Q.
,
Wu
,
R.
,
Weng
,
X.
, &
Hu
,
X.
(
2006
).
Thirsty heroin addicts show different fMRI activations when exposed to water-related and drug-related cues
.
Drug and Alcohol Dependence
,
83
,
157
162
. https://doi.org/10.1016/j.drugalcdep.2005.11.012
Yan
,
H.
,
Xiao
,
S.
,
Fu
,
S.
,
Gong
,
J.
,
Qi
,
Z.
,
Chen
,
G.
,
Chen
,
P.
,
Tang
,
G.
,
Su
,
T.
,
Yang
,
Z.
, &
Wang
,
Y.
(
2023
).
Functional and structural brain abnormalities in substance use disorder: A multimodal meta-analysis of neuroimaging studies
.
Acta Psychiatrica Scandinavia
,
147
,
345
359
. https://doi.org/10.1111/acps.13539
Yang
,
Z.
,
Xie
,
J.
,
Shao
,
Y.-C.
,
Xie
,
C.-M.
,
Fu
,
L.-P.
,
Li
,
D.-J.
,
Fan
,
M.
,
Ma
,
L.
, &
Li
,
S.-J.
(
2009
).
Dynamic neural responses to cue-reactivity paradigms in heroin-dependent users: A fMRI study
.
Human Brain Mapping
,
30
,
766
775
. https://doi.org/10.1002/hbm.20542
Zhang
,
S.
,
Zhornitsky
,
S.
,
Wang
,
W.
,
Dhingra
,
I.
,
Le
,
T. M.
, &
Li
,
C.-S.R.
(
2020
).
Cue-elicited functional connectivity of the periaqueductal gray and tonic cocaine craving
.
Drug and Alcohol Dependence
,
216
,
108240
. https://doi.org/10.1016/j.drugalcdep.2020.108240
Zijlstra
,
F.
,
Veltman
,
D. J.
,
Booij
,
J.
,
van den Brink
,
W.
, &
Franken
,
I. H. A.
(
2009
).
Neurobiological substrates of cue-elicited craving and anhedonia in recently abstinent opioid-dependent males
.
Drug and Alcohol Dependence
,
99
,
183
192
. https://doi.org/10.1016/j.drugalcdep.2008.07.012
Zilberman
,
N.
,
Lavidor
,
M.
,
Yadid
,
G.
, &
Rassovsky
,
Y
.
(
2019
).
Qualitative review and quantitative effect size meta-analyses in brain regions identified by cue-reactivity addiction studies
.
Neuropsychology
,
33
(
3
),
319
334
. https://doi.org/10.1037/neu0000526
Zilverstand
,
A.
,
Huang
,
A. S.
,
Alia-Klein
,
N.
, &
Goldstein
,
R. Z.
(
2018
).
Neuroimaging impaired response inhibition and salience attribution in human drug addiction: A systematic review
.
Neuron
,
98
(
5
),
886
903
. https://doi.org/10.1016/j.neuron.2018.03.048
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. For a full description of the license, please visit https://creativecommons.org/licenses/by/4.0/legalcode.