**Rationale** Addiction disorders are significant causes of mortality and morbidity globally. Both substance use and non-substance use addictive disorders (behavioural addictions) are increasingly prevalent conditions with high disease burdens (Merikangas & McClair, 2012; Robbins & Clark, 2015; Whiteford, Ferrari, Degenhardt, Feigin, & Vos, 2015). Addiction is characterized, in addition to debilitating symptoms, by an insidious course, social isolation, and high rates of relapse. So far, conceptualizations of addiction disorders have relied mostly on written inventories and verbal clinical interviews. Considering the many shortcomings of these approaches, there’s growing need for new, objective and theoretically sound means of integrating the etiological context of addiction, diagnosing addiction, and formulating patient-tailored treatment and monitoring (Ekhtiari, 2010; Volkow, Baler, & Goldstein, 2011). Long recognized as a central process in addiction (Wise, 1988), craving appears as a key symptom of substance use disorders (SUDs) in DSM-V (American Psychiatric Association, 2013) and interest in the neural underpinnings of craving has peaked more recently (Robinson & Berridge, 1993). While craving is a multifaceted phenomenon and can be conceptualized phenomenologically, cognitively or behaviourally, the utility of using conditioned drug-related cues to induce craving within controlled behavioural studies has long been established (Drummond, 2000, 2001). Cue-reactivity tasks present addicted individuals with series of conditioned drug cues, usually aiming to engage one sensory modality for the sake of simplicity and reliability (Rohsenow, Childress, Monti, Niaura, & Abrams, 1991). These tasks have been shown to be able to elicit craving in a laboratory setting (Carter & Tiffany, 1999) and they are widely incorporated in neuroscientific studies of addiction. Functional Magnetic Resonance Imaging (fMRI) is one of the most frequently utilized imaging modality in neurology and psychiatry (Matthews, Honey, & Bullmore, 2006) and neuroimaging studies of craving have mostly employed fMRI, using cue-reactivity tasks (Garrison & Potenza, 2014). fMRI literature on cue-reactivity has matured significantly in recent years and studies have painted a complex picture of brain activation in cue-induced craving involving disparate brain regions and networks (Chase, Eickhoff, Laird, & Hogarth, 2011; Kühn & Gallinat, 2011). Unfortunately, no activations pattern has been sufficiently consistent for wider clinical utility and translation (Tiffany & Wray, 2012). One major hindrance to application and replicability of fMRI cue-reactivity research is the considerable heterogeneity in study design. Relevant differences have been noted in participant characteristics, addiction type, craving regulation, urge intensity, task parameters, and data-analysis decisions (Jasinska, Stein, Kaiser, Naumer, & Yalachkov, 2014; Wilson & Sayette, 2015). To address the issue, multi-centric collaborations and agreed standards are needed (Ekhtiari, Faghiri, Oghabian, & Paulus, 2016) and cataloguing the methodological features of currently published research would facilitate the necessary discussions. While systematic reviews and meta-analyses of cue-reactivity fMRI studies have been published (Chase et al., 2011; Kühn & Gallinat, 2011; Zilverstand, Huang, Alia-Klein, & Goldstein, 2018), these studies have primarily focused on the results of the published cue-reactivity studies and have not attempted to systematically review and analyse the methodological parameters. Such a review would help provide a clear summary of the current status of cue-reactivity fMRI research in addiction medicine. It would help identify general patterns in study design, participant characteristics and task paradigms, and thus could highlight existing gaps and opportunities for further standardization. We plan for this review to be an extensive survey of every cue-reactivity fMRI study in populations of addicted individuals, including research focused on aetiology, diagnosis, treatment or monitoring and follow-up of disorders. We aim to classify studies into general categories. The categorization will probably evolve as the review process does, and will be based on such factors as the type of study (mapping, prediction, clinical trial or naturalistic longitudinal studies, neuro-feedback treatment) or neuroscientific scale (region, small network, large network, whole brain state) based on previous research (Ekhtiari et al., 2016). This review will also potentially help researchers interested in potential areas for conducting meta-analyses. Considering the scope of the project, no meta-analysis is currently planned for this review. Objective: The primary objective of this systematic review is to first discover existing original research literature on fMRI cue-reactivity studies using one or more fMRI-based protocols as a primary investigation method. Then, the methodological features of these studies will be surveyed and coded into a hierarchical database of study design parameters. To this end, our proposed review will seek to answer the following questions: 1. How many original studies have used fMRI as their main investigation technique in the field of cue-reactivity in populations of addicted individuals? 2. What is the most meaningful way by which to categorize existing original fMRI cue-reactivity studies in addiction medicine, and how many articles could fit in each category and sub-category? 3. What are, broadly, the most common task-designs utilized in cue-reactivity fMRI studies in addiction medicine? 4. What are the major characteristics of addicted populations included in fMRI studies in addiction medicine? **Methods** Eligibility criteria: Original studies will be selected according to the criteria outlined below. **Study design and methodology** We employed a broad perspective in the inclusion of studies, including all types of original research (i.e., basic research, observational studies and interventional studies) in this review. Only peer-reviewed studies were included. Of interest were studies which include one or more fMRI-based investigations as a major part of their methodology, whether as an observation or as a treatment tool. At least for some of the study population, the fMRI investigation had to include a cue-reactivity task, including the systematic presentation of craving-inducing cues and at least one other group of cues for comparison. Tasks involving direct consumption of the object of addiction (ie, administration of drugs or playing videogames during scanning) were not considered cue reactivity tasks, unless cues associated with the object of addiction were explicitly presented as well. Also, cues had to be “natural”, ie. they had to be associated with normal drug taking behaviour and not be novel conditioned cues associated with the object of addiction for the first time during the experiment. The fMRI protocol, setting and tasks, outcomes of interest used in analysis, and basic fMRI measurements had to be outlined. There were be no further limitations, and studies were classified into different groups based primarily on design. We expect selected original studies to belong to one (or more) of the following design/methodology categories, though modifications might be required considering the large number of studies expected to be included: 1. Mapping studies a. Group differences mapping i. fMRI-based comparison inside a group of addicted individuals, given different cognitive situations ii. Addicted individuals compared to normal groups based on some fMRI measure(s) iii. Groups in different stages of addiction compared based on some fMRI measure(s) iv. Addicted individuals with different comorbidities compared based on some fMRI measure(s) v. Groups treated in different ways compared based on some fMRI measure(s) vi. Addicted individuals compared, based on some fMRI measure(s), to those with other addictions, other psychiatric conditions or other medical problems b. Correlation studies i. Correlations between fMRI measures and one or more clinical or behavioural features ii. Correlations between fMRI measures and one or more psychological or cognitive functions iii. Correlations between fMRI measures and one or more biomarkers 2. Prediction studies (prospective cohorts) i. Predictions of treatment outcome based on some fMRI measure(s) ii. Predictions of natural course of addictive behaviour or substance use based on some fMRI measure(s) 3. Longitudinal use of fMRI a. Randomized clinical trials i. Case-control studies with post-intervention fMRI, both groups undergo fMRI after intervention or placebo and are compared based on some fMRI measure(s). ii. Crossover with post-intervention fMRI, differences in fMRI maps after an acute real or placebo intervention in a crossover setting, each subject scanned twice after active intervention and after control. iii. Case-control studies with pre- and post-intervention fMRI, differences in fMRI maps before and after a real or placebo intervention in a case–control setting. Each subject is scanned twice in a single session, once before and once after intervention. iv. Crossover with pre- and post-intervention fMRI, differences in fMRI maps before and after an acute real or placebo intervention in a crossover setting. Each subject is scanned four times, twice before and after sham and twice before and after active intervention in separate sessions. b. Naturalistic studies i. Treatment course. Differences in some fMRI measure(s) at different points during a treatment course ii. Natural course. Differences in some fMRI measure(s) during natural course of addiction 4. fMRI neuro-feedback treatment studies: real-time fMRI used for neuro-feedback training based on some fMRI measure(s) **Participants** In every study, there needed to be at least one human population or sub-population with more than one member, and at least one of the following with regards to participants had to be true: 1. At least one circumscribed group of participants had a diagnosis of at least one substance-use disorder or non-substance behavioural addiction disorder which was active, relapsed or in remission; with diagnosis made either before the study, as part of the study protocol during the investigation, or by the end of the study (i.e., with the diagnosis serving as an outcome measure). 2. At least one group of participants was included explicitly because they regularly consumed a potential object of addiction (substance or behaviour) and/or had a risky pattern of consumption that might lead to addiction, and the study was focused on their reactivity to cues of that substance or behaviour. 3. At least one group of participants had been assigned a) dependence- or b) urge/craving- or c) addiction-related scores or ordinal categories, based on a test of addictive substance-related or non-substance-related behavior patterns, and had been analysed based on their difference along this spectrum. No restriction will be placed on study participants based on demographic, biological, or clinical factors (such as any co-morbidities). **Language** Only publications in English will be included. Titles of seemingly relevant publications in other languages will be provided as an appendix, if they have at least an English translation of their title and/or abstract. **Information source** Existing research was identified and retrieved using the online portal of the National Library of Medicine (http://www.ncbi.nlm.nih.gov/pubmed/) including PubMed, PubMed Central and MEDLINE. Relevant articles were identified using a comprehensive search strategy for all terms related to addiction, fMRI and cue-induced craving, as detailed below. Search strategy: Considering the subject of the review was comprised of two main parts, a list of three sets of keywords was compiled (Table 1). These terms were adapted for use in PubMed (exact search strategy is outlined Table 2). The first set included synonyms of “functional magnetic resonance imaging”, the second included terms related to cue-induced craving, and the third included synonyms of “addiction” and various terms related to substance use disorders and behavioural addictions and addiction medicine. To help widen the search, no filter was used. Primary research and systematic reviews were manually separated **Data collection** Data will be filled into an electronic spreadsheet. There will be a calibration exercise so that authors can discuss and refine the form and consistency is maximized. Each eligible paper will be reviewed in duplicate by two authors in two separate spreadsheets independently, and results will be compared in the end. Inconsistencies will be resolved through discussion. For each study, demographic information, methodology and patient-important information will be obtained, either directly or by contacting the corresponding authors if needed. **Data items:** We will publication name, name of authors, year of publication, journal, country, study design, fMRI protocol (general categories will be defined), neuroimaging data analysis strategy used, neurocognitive aspects (neuroscientific scale, cognitive function studied, paradigms used), basic sample features (age, sex, number, comorbidity), cue-reactivity task features (number of cues, cue type, cue source, cue modality etc), focus (aetiology, diagnosis, prognosis, monitoring and treatment), addictive substance/behaviour, addiction severity (i.e., clinical diagnosis given, validated self-report measure of severity, or quantity/frequency of involvement) and type of treatment, if any exists.