Affect and craving: Positive and negative affect are differentially associated with approach and avoidance inclinations

Abstract

Background

Research on reactivity to alcohol and drug cues has either ignored affective state altogether or has focused rather narrowly on the role of negative affect in craving. Moreover, until recently, the relevant analyses of affect and craving have rarely addressed the ambivalence often associated with craving itself. The current study investigated how both negative and positive affect moderate approach and avoidance inclinations associated with cue-elicited craving in a clinical sample diagnosed with substance use disorders.

Methods

One hundred forty-four patients (age range of 18–65, mean 42.0; n = 92 males) were recruited from an inpatient detoxification unit for substance abuse. Participants completed a baseline assessment of both positive and negative affect prior to completing a cue-reactivity paradigm for which they provided self-report ratings of inclinations to approach (use) and avoid (not use) alcohol, cigarettes, and non-psychoactive control substances (food and beverages).

Results

Participants with elevated negative affect reported significantly higher approach ratings for cigarette and alcohol cues, whereas those high in positive affect showed significantly higher levels of avoidance inclinations for both alcohol and cigarette cues and also significantly lower approach ratings for alcohol cues, all relative to control cues.

Conclusions

Results for negative affect are consistent with previous cue reactivity research, whereas results for positive affect are unique and call attention to its clinical potential for attenuating approach inclinations to substance use cues. Further, positive affect was related to both approach and avoidance inclinations, underscoring the utility of a multidimensional conceptualization of craving in the analysis.

Introduction

With nearly 9% of the U.S. population age 12 or older diagnosable with Substance Use Disorders (SUDs)—most notably alcohol and tobacco use problems (SAMHSA, 2010)—SUDs are undeniably a major public health concern. The high rates of comorbidity of SUDs with anxiety and mood disorders (e.g., Grant et al., 2004) serves to complicate this problem, but also directs attention to the need for a better understanding of links between affect and compulsive use of popular psychoactive substances (cf. Leshner, 1997). Despite strong links between affect and addictive behaviors, research on reactivity to alcohol and drug cues has typically presented substance use cues only, without regard to the person's affective state or has focused rather narrowly on the role of negative affect in craving. Moreover, until recently, the relevant analyses of affect and craving have rarely addressed the ambivalence often associated with craving itself (see Stritzke, McEvoy, Wheat, Dyer, & French, 2007, for a review). An ambivalence model of craving acknowledges both the desire to consume a substance (approach inclination) and the desire to not consume it (avoidance inclination). By examining the interplay of positive and negative affect with the desire to approach and the desire to avoid substance use, the present study sought to further elucidate the links between affect and subjective craving experiences for alcohol and tobacco.

Theories accounting for substance user's reactivity to drug related stimuli often focus on craving, which has been defined as cue-elicited motivation to consume the substance (e.g., Sayette et al., 2000, Tiffany, 1990, Tiffany and Conklin, 2000). Cue-elicited craving is thought to develop through a process of conditioning, in which drug-related cues are repeatedly paired with positively and /or negatively reinforcing drug effects (e.g., Baker et al., 2004, Carter and Tiffany, 1999); however, such conceptualizations fail to account for the ambivalence that substance abusers commonly display toward the drugs they abuse (e.g., Breiner et al., 1999, Stritzke et al., 2007, Tiffany, 1990). For example, ambivalence about use, defined as the simultaneous desire to use and to not use psychoactive substances, has been identified as a hallmark feature of addiction, and is central to many clinical formulations of substance use disorders (e.g., Heather, 1998, Orford, 2001). Further, Tiffany's (1990) cognitive processing model of craving posits that the subjective experience of craving may occur when the immediate gratification of urges to use are impeded by internal motivations to abstain (i.e., abstinence promotion). Despite acknowledgments of the importance of competing desires (Anton, 1999, Kavanagh et al., 2012, Tiffany, 1990), ambivalence is often overlooked in the study of craving.

To address this concern, Breiner et al. (1999) introduced an ambivalence conceptualization of craving (AMC). Consistent with qualitative (e.g., Smith-Hoerter, Stasiewicz, & Bradizza, 2004), quantitative (e.g., Curtin et al., 2005, Schlauch, Breiner, Stasiewicz, Christensen and Lang, 2012, Stritzke et al., 2004), and neurobiological findings (e.g., Barkby et al., 2012, Cacioppo et al., 1999), the AMC views craving as a complex experience highlighting the importance of not only craving or desires to use (approach inclinations), but also competing desires to not use (avoidance inclinations). These two dimensions of reactivity are thought to develop through different psychobiological systems following repeated, systematic exposure to reinforcing and punishing events associated with such substance use (Lang, 1995, LeDoux, 2000, Stasiewicz and Maisto, 1993), and are proposed to be orthogonal to one another resulting in four hypothetical quadrants (see Fig. 1 top panel; Breiner et al., 1999, McEvoy et al., 2004, Stritzke et al., 2007). Further, while these dimensions may be activated reciprocally (e.g., high on one dimension, low on the other), approach and avoidance inclinations are independent and can be activated simultaneously to elicit different motivational states that vary as a function of drinking and/or recovery status (see Fig. 1 bottom panel).

The addition of a distinct avoidance dimension has numerous advantages clinically and methodologically (see Stritzke et al., 2007 for a review). Further, it has been argued that measuring “craving” or “urge to use” exclusively in terms of approach inclinations without consideration of a separate, yet concurrent, avoidance inclination may misrepresent a motivational disposition that is actually a combination of both, thus significantly diminishing the utility of the information obtained (Breiner et al., 1999). Indeed, studies examining avoidance inclinations using both cue reactivity paradigms and self-report measures have provided support for its incremental validity in predicting substance-related variables in both non-clinical and clinical samples. For example, approach and avoidance can be independently measured as separate dimensions of cue reactivity (e.g. Curtin et al., 2005, Schlauch, Breiner, Stasiewicz, Christensen and Lang, 2012, Stritzke et al., 2004). Further, avoidance has been shown to moderate the effect of approach inclinations on drinking (Schlauch, Levitt, et al., 2012), is incrementally related to taking steps to make a change (Klein et al., 2007, Schlauch, Breiner, Stasiewicz, Christensen and Lang, 2012, Schlauch, Stasiewicz, et al., 2012), and distinguishes between clinically significant subgroups of smokers trying to quit (high approach, high avoidance) and not quit (high approach, low avoidance; Stritzke et al., 2004). Finally, avoidance inclinations may be more predictive of relapse among alcoholics, rather than increases in approach inclinations (Stritzke et al., 2007). This is important as it suggests that once internal or external cues trigger urges to use, people can resist them but it takes effort (Tiffany, 1990). The AMC provides a framework for which alcohol/drug use is not inevitable when approach is activated, but rather dependent on competing desires and effortful control such that the stronger one's avoidance inclinations are, the less likely the decisional balance tips in favor of use. However, despite these promising findings, to date no studies have examined the factors influencing both approach and avoidance inclinations.

The idea that substance use is intimately related to affect and the desire to regulate it is far from novel and enjoys considerable empirical support (cf., Kassel and Veilleux, 2010, Sher and Grekin, 2007). Thus, it is not surprising that there are numerous theoretical accounts of the relationship between approach inclinations and affect in the substance abuse literature (e.g., Baker et al., 1987, Baker et al., 2004; Kavanagh et al., 2005, Robinson and Berridge, 1993, Stasiewicz and Maisto, 1993, Tiffany, 1990). Collectively, these accounts provide varied predictions on the relationship between affect and craving, and point to both positive and negative affect as a precipitant of craving, as a consequence of craving, and as a defining feature of craving (i.e., craving as an affective state).

As a precipitant of craving, negative affect has received considerable empirical support. For example, laboratory procedures inducing negative affect consistently trigger cue-elicited craving (i.e., approach inclinations) in alcoholics (e.g., Cooney et al., 1997, Fox et al., 2007) and cigarette smokers (e.g., Conklin and Perkins, 2005, Maude-Griffin and Tiffany, 1996, Perkins and Grobe, 1992, Tiffany and Drobes, 1990). As a consequence of craving, negative affect is predicted (e.g.,) when access to the target substance is delayed or blocked (i.e., frustrative non-reward; Kavanagh et al., 2005, Stasiewicz and Maisto, 1993; Tiffany, 1990). In addition, negative affect may also arise in the form of guilt and/or anxiety when craving (i.e., approach inclination) is elicited in a person who is actively attempting to control his or her use (Kavanagh et al., 2005). Finally, negative affect has been described as the predominant emotional reaction in most or all craving experiences (Baker et al., 2004, Kavanagh et al., 2005, Nosen et al., 2012). In all cases, negative affect is predicted to be positively associated with approach inclinations (desire to use) and to varying degrees negatively associated with avoidance inclinations (desire to not use). For example, the predicted negative relationship between negative affect and avoidance inclinations might be stronger in those who have lower abstinence self-efficacy and weaker among those attempting to regulate or control their use of an addictive substance.

With regard to positive affect, it is also viewed as an antecedent (e.g., Baker et al., 1987) and consequence of craving (Kavanagh et al., 2005), and also as an appetitive-motivational response akin to the experience of craving itself (Robinson and Berridge, 1993, Stewart et al., 1984). Further, it has been argued that the potential for positive affect to elicit craving may be more active in the early stages of addiction (e.g., Tiffany, 2010). In contrast to negative affect, positive affect as a predictor of approach inclinations has received much less empirical support (c.f., Tiffany, 2010) and the conditions in which it is predicted to be a consequence of craving are limited to situations in which consumption is imminent and negative consequences for use are low (Kavanagh et al., 2005). Of the few reports finding significant relationships, positive affect may augment the impact of substance-related imagery on craving in non-clinical populations, with the combined effect as strong as negative affect (Maude-Griffin and Tiffany, 1996, Tiffany and Drobes, 1990). In an alcohol dependent sample, Mason, Light, Escher, and Drobes (2008) found a significant relationship between positive affect (experimentally induced) and higher ratings to beverage cues on a measure of craving strength (i.e., “How strong is your craving to drink alcohol”). Interestingly, no significant relationships were found between positive affect and craving measures of intentions to use, expectations of positive effects, or lack of control; however, several limitations were noted. Despite increased craving strength following the positive affect induction, manipulation checks revealed that participants' affective valence remained unchanged. Further, the images used to induce positive affect may have very well triggered craving responses themselves due to their associations with drinking behaviors (e.g., sporting events), a point acknowledged by the authors. Finally, although negative affect was successfully induced using images, no effect on cue-elicited craving was noted, a finding inconsistent with previous literature.

As an appetitive-motivational stimulus, positive affect is thought to share common features of the pleasurable or rewarding elements of substance use. Whether by activation or sensitization of neural systems of reward (e.g., Baker et al., 1987, Robinson and Berridge, 1993, Stewart et al., 1984), or enhancement of an appetitive-motivational process (Kavanagh et al., 2005), positive affect is predicted to maintain substance use approach behaviors. However, positive mood may also inhibit craving, or the accentuation of craving, by exerting influence on self-regulatory processes (e.g., Tice, Baumeister, Shmueli, & Muraven, 2007) such as self-efficacy and negative expectancies for use, with the latter posited to influence avoidance inclinations. Thus, in contrast to negative affect, the role of positive affect on craving is less straightforward and the nature of its influence appears to depend on such factors as substance availability, perceived negative consequences of use, and abstinence self-efficacy. Further, it is possible that those with greater positive affect in a clinical setting may have an increased inclination to avoid indulging because their positive affective state could very well reflect a sense of purpose and optimism in overcoming their addiction problems. Indeed, positive affect has been associated with greater restraint on smoking following a self-control depletion task (Shmueli & Prochaska, 2012). Therefore, studies examining both approach and avoidance inclinations and their relationships with positive and negative affective states may yield clinically useful information that would otherwise be obscured by traditional craving assessments (i.e. approach inclinations only).

Using data from a larger cue-reactivity study designed to examine the relationship between personality variables and cue-elicited craving towards various psychoactive substances, we examined how positive and negative affect might differentially influence inclinations to both approach (use) and avoid (not use) substances to which they were exposed during a visual cue-reactivity task. Participants sampled from an inpatient SUD detoxification unit reported on their affect states, making separate ratings for positive and negative affect prior to completing a standardized cue-reactivity paradigm (Curtin et al., 2005, Schlauch, Breiner, Stasiewicz, Christensen and Lang, 2012, Stritzke et al., 2004). Specifically, participants provided both approach and avoidance ratings in response to pictorial cues that included alcohol and cigarettes. The following predictions were derived from the existing literature and the ideas outlined above:

• Baseline negative affect would be positively associated with approach inclinations for alcohol and cigarettes because negative affect has been shown to predict increased inclinations to indulge in addictive behavior, perhaps as a maladaptive effort to cope.

• Baseline negative affect to would be negatively associated with avoidance inclinations for alcohol and cigarettes, as negative affect appears to decrease access to more adaptive coping responses in those with addiction.

• Baseline positive affect would be negatively associated with approach inclinations because positive affect might diminish the magnitude of desire or perceived need for a psychoactive substance. This may be most evident in a clinical sample of substance abusers who may have stronger negative reinforcement motives for use.

• Baseline positive affect would be positively associated with avoidance inclinations for alcohol and cigarettes, because of the recent negative consequences of their use and because positive affect appears to increase access to more adaptive coping options and/or alternative sources of reinforcement (e.g., greater self-regulation).