The mood-induced activation of implicit alcohol cognition in enhancement and coping motivated drinkers

Abstract

In two experiments, we investigated whether different mood states activate specific types of implicit alcohol cognition among undergraduates classified as enhancement (EM) or coping (CM) motivated drinkers. Participants completed a Stroop task in Experiment 1 (n = 81), and an Extrinsic Affective Simon Task (EAST; [De Houwer, J. (2003). The Extrinsic Affective Simon Task. Experimental Psychology, 50, 77–85.]) in Experiment 2 (n = 79) following random assignment to listen to positive or negative musical mood induction procedures (MMIP). Consistent with hypotheses, only EM, and not CM, drinkers displayed an activation of implicit attention to alcohol cues (Experiment 1) and reward-alcohol implicit associations (Experiment 2) following exposure to positive MMIP. Contrary to hypotheses for CM drinkers, none of the groups, in either experiment, showed an activation of implicit alcohol processing following exposure to negative MMIP. Confidence that positive mood activates implicit alcohol cognition among EM drinkers is increased since this result emerged across two studies involving quite different methodologies. This research has implications for experimental cognitive research and it highlights the potential utility of treatment matching according to drinking motives (e.g., EM) to improve clinical outcomes.

Introduction

Drinking for internal or emotional reasons predicts heavy and problematic alcohol consumption, relative to drinking for external or social reasons. Within Cooper's (1994) motivational model of alcohol consumption there are two types of internal motives: (1) enhancement motives (EM; drinking to increase positive affect), and (2) coping motives (CM; drinking to decrease negative affect). While these two emotional motives for drinking have been reliably found to be associated with unique drinking consequences, there has been little research on the highly clinically relevant question of whether there are unique antecedents associated with the two “risky” EM vs. CM patterns of alcohol consumption. The purpose of the present research was to contrast the emotional antecedents for EM vs. CM drinking-related responding.

To first specify, we use the term “emotion” interchangeably with “mood” and “affect”. While some have argued that emotional, and affective, states have a shorter duration and stronger intensity than mood states, the intensity of all of these states can presumably be altered in the laboratory (see Martin, 1990), and they are all thought to be shorter in duration (or less stable) than traits or chronic emotional conditions (e.g., depressive disorder) (Kozma et al., 1990).

Given the nature of EM and CM, one would expect that alcohol-related responding (i.e., alcohol-related cognitions as well as actual drinking behaviour) is prompted for EM drinkers in response to positive emotions, and for CM drinkers in response to negative emotions. Previous research can be interpreted to suggest how this may occur. There may, for example, be drinking motive differences in personality (Stewart & Devine, 2000), or in the operation of neural networks/motivational systems (Gray, 1972), or in sensitivity to various effects of alcohol that could account for unique responding to emotional cues among EM and CM drinkers (for a detailed review of such theories see Birch, Stewart, & Zack, 2006). Overall, there has been scant experimental research conducted to investigate how individual differences, like drinking motives moderate relations between emotions and alcohol-related responding. There has been some prior research on whether EM and CM drinkers differ in the emotions that trigger cognitive alcohol-related responding specifically (e.g., explicit and implicit alcohol cognitions).

While there is often a lack of clarity as to the respective definitions for the terms “explicit” and “implicit” cognitions, some authors, like De Houwer (2006) and Wiers et al. (2002), have synthesized disparate conceptualizations of these terms. They suggest that explicit cognitions are often measured with questionnaires and other tasks that involve conscious, intentional selection of responses. In contrast, computerized reaction time (RT) tasks are often administered to measure cognitions with functional features characteristic of implicit or automatic cognitions. Participants are likely unaware or unconscious of the nature of the hypotheses under investigation with many RT tasks, and their responding is likely unintentional. While research on implicit cognition is still in its infancy, and the psychometric merits of measures of implicit cognition are not yet well-established, this research may be less prone to confounds associated with experiments involving explicit self-reports (e.g., demand characteristics, biased self-perception). Overall, it is important to examine how both explicit and implicit alcohol cognitions vary as a result of mood and drinking motives.

There has been some prior research on how explicit alcohol cognitions vary as a function of mood states and drinking motives. In a study by Birch et al. (2004), for example, participants with extreme scores on either the EM or CM subscale of Cooper's (1994) Drinking Motives Questionnaire (DMQ-R) were randomly assigned to listen to positive or negative musical mood induction procedures (MMIP). The self-reported strength of specific, explicit alcohol expectancy (AE) cognitions was measured with questionnaires before and after MMIP. Birch et al. (2004) found that the strength of reward AE (beliefs that alcohol will provide emotional rewards) increased following MMIP only for EM drinkers in the positive mood group, and conversely, the strength of relief AE (beliefs that alcohol will provide emotional relief) increased following MMIP only for CM drinkers in the negative mood group.

To our knowledge, only one study conducted to date (i.e., Stewart, Hall, Wilkie, & Birch, 2002) has examined relations between affective cues, EM and CM for drinking, and implicit alcohol cognition. Stewart et al. (2002) administered a RT task, called the primed Stroop, that is intended to measure implicit attentional bias that is automatically drawn to the semantic meaning of targets (e.g., alcohol words) following priming with another class of stimuli (e.g., neutral, positive, and negative mood cues). Stewart et al. (2002) considered implicit alcohol cognition activated in memory when participants showed longer colour-naming latencies (i.e., interference) for alcohol vs. non-alcohol targets. As expected, positive (but not negative) emotional primes activated implicit alcohol cognition for EM drinkers, and negative emotional (but not neutral) primes activated implicit alcohol cognition for CM drinkers. Unexpectedly, however, neutral primes also activated alcohol cognition for EM drinkers, and positive emotional primes also activated alcohol cognition for CM drinkers. Thus, results only partially supported hypotheses. Further research is needed, as this study did not specifically address whether drinking motives moderate relations between actual mood states and implicit alcohol processing. Indeed, exposure to lexical affective stimuli may not have any substantial influence on state affect (see Stein, Goldman, & Del Boca, 2000).

The purpose of the two experiments described in this paper was to further investigate the impact of emotions and drinking motives on cognitions that could be considered to be implicit in nature. This research also allowed for an integration of findings on implicit cognition with previous research on explicit cognition (e.g., Birch et al., 2004), and had the potential to yield unique information about the cognitive-motivational factors underlying problematic emotionally-motivated drinking. Clinically, findings from this research may highlight the potential utility of treatment matching according to drinking motives. Such treatment matching might improve cognitive-behavioural targeting of emotional cues that commonly trigger alcoholic relapse (Marlatt & Gordon, 1985).