**1. Introduction**

Posttraumatic stress disorder (PTSD) is an often debilitating mental disorder that may occur following trauma exposure [1]. PTSD is characterized by four diagnostic clusters—(1) the re-experiencing the traumatic event (e.g., recurrent memories, dreams, or flashbacks); (2) symptoms of avoidance (e.g., efforts to evade trauma reminders); (3) arousal (e.g., hypervigilance, sleep disruptions); and (4) negative

cognitions and mood (e.g., self-blame) [1]. Substance use disorder (SUD) is another psychiatric disorder characterized by 11 possible symptoms which involve negative consequences arising from one's substance use and inability to control one's substance use [1]. In the last version of the *Diagnostic and Statistical Manual of Mental Disorders* (the DSM-5; [1]), craving, the often intrusive desire to use the substance, was added as one of the 11 symptoms characterizing a SUD [1].

PTSD often co-occurs with SUD. Research has documented high rates of comorbidity between PTSD and alcohol use disorder (AUD) [2], cannabis use disorder (CUD) [3, 4], and other SUDs [5]. The prognosis of comorbid PTSD-SUD is worse than either disorder alone [6] with comorbid PTSD-SUD leading to greater functional impairment in comparison to those with only PTSD or a SUD [7].

It has been suggested that PTSD and SUD are likely functionally related to one another [8] in comorbid individuals. While the precise underlying mechanisms are not well understood, there are several learning theories that may help explain the high rates of substance misuse in people with trauma histories and help us understand the high comorbidity of PTSD with SUD. The first is the two-factor learning theory which was originally developed by Mowrer to explain the acquisition and maintenance of phobias [9] and which has more recently been applied by Stasiewicz [10] to the acquisition and maintenance of SUDs. Two-factor learning theory applies a combination of classical conditioning and operant conditioning mechanisms to the development and maintenance phases of these disorders, respectively. Applying this theory to the co-occurrence of PTSD and SUDs in traumatized individuals, trauma-relevant cues that were paired with the original traumatic experience (e.g., loud noises of gunfire paired with witnessing a comrade fatally injured in wartime) are thought to come to elicit negative affect through the process of classical conditioning [10]. Future exposures to the trauma cue alone (e.g., loud noises alone) motivate avoidance/escape behavior, including substance misuse, to reduce the associated negative affect and thereby experience relief [10]. Avoidance/escape behaviors like substance misuse are thus negatively reinforced in individuals with trauma histories/PTSD as they remove the aversive experience of negative affect. Therefore, substance misuse is maintained as a self-medication type of coping response through operant conditioning processes where behavior is repeated when it is followed by desirable consequences, in this case, relief from negative affect.

Another theory that is relevant to understanding the links of trauma/PTSD with SUD involves the role of classical conditioning in the development of conditioned craving—a strong urge to use the substance in response to exposure to the conditioned cues. It has long been known that drug-related stimuli that are frequently paired with drug-taking can come to elicit a conditioned craving response through the process of classical conditioning [11]. For example, a needle and other drug use paraphernalia that are frequently paired with heroin use can come to elicit craving when presented alone, for an injection drug user. Similarly, for a substance user with a trauma history/ PTSD, the frequent pairing of trauma cues (e.g., intrusive memories of the trauma, exposure to external trauma reminders) with substance use, as explained by the twofactor learning theory above [10], can come to create strong associations between trauma cues and substance use [12]. The result is that such trauma cues can become conditioned stimuli that elicit a conditioned craving response when presented on their own [13]. For example, if a young woman with sexual assault-related PTSD drinks alcohol each time she has an intrusive memory about her sexual assault, such trauma cues can come to elicit a strong craving for a drink, which may motivate her alcohol seeking and maintain her alcohol use.

The study of the above putative mechanisms under controlled, laboratory conditions is crucial for a better understanding of the intertwined relationships between trauma/ PTSD and substance misuse. Specifically, the use of cue-reactivity paradigms allows researchers to examine how substance-related and trauma cues may come to elicit craving and/or negative affective responses through the conditioning processes described above.

The cue-reactivity paradigm is broadly defined as a lab-based method in which participants are exposed to a set of stimuli meant to elicit a "reactivity" response—that is, a change from baseline in response to the stimulus [14]. In the context of addictions research, stimuli may be substance-related cues, such as a syringe or other drugrelated paraphernalia for an injection drug user [15]; these stimuli serve as analogs for real-life stimuli which may evoke a craving response outside of the lab. Indeed, research in this area has shown that relevant drug-related cues presented in the lab can elicit a heightened craving response among substance users [16, 17]. More recently, cue-reactivity paradigms have been used to study conditioned craving as a possible mechanism underlying the relationship between trauma/PTSD and SUD [18, 19]. Indeed, extant research has shown that in-lab exposure to cues representing trauma reminders (e.g., a video of a violent crime) activates both substance-related craving responses as well as increased negative affect [20].

Craving has been measured in a number of ways in substance- and trauma-related cue-reactivity research, including with subjective self-report measures, such as the Desire for Drug Questionnaire [21], and measures specific to the substances used, such as the Alcohol Urge Questionnaire [22] and the Marijuana Craving Questionnaire [23]. Craving has also been measured more objectively in cue-reactivity studies, albeit less commonly than via self-report. Specifically, physiological measures, such as salivary flow and heart rate monitoring, are often used as a more objective proxy measure of craving [24]. Craving has also been further differentiated into reward-related craving (i.e., a desire for reward or stimulation from a substance) and relief-related craving (i.e., a desire for a reduction in tension or negative affect from using a substance) using certain self-report measures [23].

While cue exposure paradigms are homogenous in their goal to elicit some form of reactivity (e.g., change from baseline in craving or emotional state in response to the stimulus), the types of cues and paradigms used in this area of research have varied widely. For example, cues may be standardized across participants in the study or may be personalized to the individual's own trauma history details; cues may be presented through the use of script-driven imagery (i.e., audio recordings, such as a retelling of a traumatic event) or *in vivo* (i.e., physical objects, such as drug paraphernalia); and cues maybe a photo or video stimuli (i.e., a video of an assault).

Indeed, it is evident that cue-reactivity paradigms vary widely in design, are used in an expansive variety of contexts and with a wide range of populations, with many different outcomes used to capture cue-reactivity effects. Thus, in this chapter, we intend to scope the extant cue-reactivity literature in the context of PTSD-SUD comorbidity research to identify patterns and variations in methodology, measures, and outcomes used in this growing field.

#### **1.1 Aims and objectives**

Our first aim was to examine how cue-reactivity paradigms have been used in samples of substance users with trauma histories. Specifically, we were interested in how these studies lead to a further understanding of the mechanisms underlying comorbid PTSD-SUD. Second, we intended to examine the different types of cues

used within the cue-reactivity paradigm as well as the specific effects, strengths, and weaknesses of variations in paradigm design. Specifically, we compared the merit of personalized vs. non-personalized cues, as well as other cue variations, in PTSD-SUD cue-reactivity research (e.g., *in vivo*, imagery-based). Third, we sought to assess the use of several measures of reactivity that have been examined using the cue-reactivity paradigm (i.e., craving [subjective, objective], negative affect) used in PTSD-SUD research. Lastly, we examined the types of participants who have been studied using a cue-reactivity methodology (e.g., trauma-exposed vs. suffering from PTSD).
