Home > Encyclopedia of Drugs, Alcohol, and Addictive Behavior > Expectancies

EXAMPLES OF RESEARCH STUDY

An example of research using balanced-placebo design is as follows: In a simulated bar setting, half the participants in a study are told they will receive a drink containing vodka and tonic, and half are told they will receive a drink containing only tonic. After this expectation is established, half of each group does receive vodka and tonic, while the other half receives only tonic, resulting in four groups: (1) those who expect vodka and tonic and receive vodka and tonic, (2) those who expect vodka and tonic and receive only tonic, (3) those who expect tonic and receive vodka and tonic, and (4) those who expect tonic and receive tonic. Thus, some of the people who expect alcohol receive only tonic, and some who expect only tonic receive a mix containing alcohol.

Behavioral observations following this manipulation reveal that the most powerful predictor of behavior after consuming the assigned drink is not whether the person actually receives alcohol, but whether that person believes he or she is drinking alcohol: People who expect alcohol in this experimental situation consume significantly more drink than those who are not expecting alcohol, regardless of whether or not they do receive alcohol in their drink. With the discovery of this phenomenon, even in people who are considered dependent on alcohol, this finding has been interpreted as providing contrasting evidence to the disease model's notion that "loss of control" is caused exclusively by the pharmacological effects of alcohol; the findings introduced the idea that cognitive factors are influential in a person's drug-related behavior. The presence of expectancy effects have also been identified in research on drugs other than alcohol, including TOBACCO and MARIJUANA (Marlatt & Gordon, 1985).

Most of the research on expectancies during the 1970s and 1980s was conducted on college students, with samples ranging from light to heavy social drinkers who were primarily Caucasian. This research has shown that the effect of a person's expectancies depends on whether the behavior involved is socially mediated: Stronger expectancy effects are found for social behaviors (e.g., aggression or sexual arousal) than for nonsocial behaviors (e.g., beliefs concerning motor coordination or memory skills); they are stronger for outcomes that are perceived as positive (e.g., sexual arousal) than as negative (e.g., poor motor coordination).

For socially mediated behaviors, expectancy research has revealed that college students of both sexes show less anxiety in social situations if they believe they have consumed alcohol. In addition, males show heightened sexual arousal when exposed to an erotic environment if they believe they have consumed alcohol (Marlatt & Gordon, 1985). Men and women of college age have also both been found to respond more aggressively when provoked after they believe they have consumed alcohol. Sex differences have been found on the effects of alcohol on anxiety with persons of the opposite sex: Women of college age have shown more anxiety in the company of an unfamiliar man when they believe they have consumed alcohol, while men of college age have shown reduced anxiety when in the company of an unfamiliar female. The results have been interpreted as reflecting gender differences regarding the acceptability of alcohol in social situations with a stranger of the opposite sex.

OTHER STUDIES

Other experimental work has revealed that specific outcomes can vary with the personal beliefs an individual holds regarding alcohol and with the phase of intoxication of an individual (Southwick et al., 1981). Overall, the results based on expectancy research point to the likelihood that people may have established cultural beliefs regarding the effects of alcohol in social situations and that these beliefs play some role in the behavioral effects of alcohol.

Research has also found that expectancies do predict drinking behavior over a one-year period for early adolescents (Christiansen et al., 1989); that expectancies tend to crystallize in people at a young age and that they tend to be resistant to change (Miller, Smith, & Goldman, 1990). Other studies on Caucasian adolescents and young adults have found that those who have mostly positive and only few negative outcome expectancies tend to experience more alcohol-related problems than those whose outcome expectancies are more evenly divided between positive and negative effects (Brown, Christiansen, & Goldman, 1987).

Since the late 1980s, researchers have begun to examine ethnic and racial differences in the expectancy variable. One study of college-age students (Daisy, 1989) revealed that Native-American students had significantly stronger expectancies for the positive social and physical effects of drinking than did Asian-American students. Caucasian students were found to have stronger positive expectancies for social and physical effects than did Asian-American students, but less than did Native-American students. These beliefs concerning the effects of alcohol were also found to be highly associated with the drinking patterns of the study participants: those people whose drinking pattern was considered heavy had stronger beliefs in the above expectancies than individuals who drank less. The study strongly suggests that ethnic differences exist in alcohol-related expectancies, and it confirms that expectancies are related to the amount of alcohol consumed.

The association between expectancies and drinking pattern has been consistent in the research and has therefore become targeted in substance-abuse treatment. Expectancies have been found to influence the way a person copes with high-risk situations after treatment aimed at abstinence (Marlatt & Gordon, 1985; Condiotte & Lichtenstein, 1981). In RELAPSE PREVENTION, positive-outcome expectancies are viewed as the source of urges or cravings for a substance. Treatment according to this perspective therefore includes changing a client's outcome expectancies: If a person believes that drinking will provide immediate relief from stress, then treatment focuses on helping that person consider the long-range implications of drinking—helping the person by adding the negative outcomes of drinking to the anticipated positive results of drinking—and thereby changing the composition of the person's outcome expectancies.

Self-efficacy expectancies, or how effectively one feels he or she can cope with a high-risk situation, are also examined in treatment. If a client lives a stressful lifestyle and believes that only alcohol provides relief from that stress, the therapist helps the client develop and utilize alternative methods for coping with stress. For example, clients can be taught to look forward to meditation or exercise or other positive-reward situations to help cope with stress and to reduce urges and the resulting temptation to drink. Treatment focuses on developing alternative coping strategies for a client's individual high-risk situations, and therefore includes an ongoing assessment of each client's high-risk situations.

Self-efficacy differs from overall motivation to quit or reduce substance use, since perceived control will vary across situations. In research on relapse prevention, self-efficacy has been found to be predictive of the first use of the substance after abstinence-based treatment: Those people who do not believe they can cope with either a specific situation or cope, in general, with the temptation to use a substance are more likely to relapse in the face of a high-risk situation than are people who believe that they are able to maintain their goal of abstinence in the same situation (Condiotte & Lichtenstein, 1981).

The Alcohol Expectancy Questionnaire (AEO), developed in the late 1980s, became the most commonly used alcohol expectancy instrument. Criticisms of the AEQ led to a conceptual model of drinking expectancy grounded in social learning theory. In this model, people acquire a set of alcohol expectancies regarding how alcohol will affect them during what is called the acquisition phase of the model. The behavioral outcomes of these beliefs were then hypothesized to be regulated by a process involving Drinking Refusal Self-Efficacy (DRSE).

In 1996, the Drinking Expectancy Profile (DEP) was developed, which had two interrelated subtests, the Drinking Expectancy Questionnaire (DEQ) and the Drinking Refusal Self-Efficacy Questionnaire (DRSEQ). When compared to the AEQ in a study, the DEP showed better predictive ability on the Alcohol Dependence Scale and for quantity of drinking and frequency of drinking in a student sample. Furthermore, the DEQ contained both negative and positive outcome expectancies, which yielded better information on alcohol-related outcomes.

Further research in the 1990s showed that alcohol expectancies can develop independently of the actual drinking experience, developing from vicarious learning before even tasting alcohol. Actual drinking behavior could later reinforce or modify the existing beliefs. Drinking refusal self-efficacy beliefs were also show to develop prior to one's drinking history.

A study in 2000 considered the 1992 Temptation Restraint Inventory (TRI) and DEP as indicators of problem drinking across a range of drinking parameters. It yielded a more comprehensive picture of the complex interrelationship between the variables that make up the individual drinker's motivation for risky and dependent drinking. The results showed that drinking restraint and related control and impaired-control issues were the strongest predictors of alcohol problems. Alcohol expectancies and drinking refusal self-efficacy, while reflecting some of these loss-of-control factors, tended to focus more on choices of whether to drink or not and thus predicted more frequent usage of alcohol. This study suggested that restraint, alcohol expectancy, and self-efficacy measured different cognitive domains (Connor, et al, 2000).

Another study in 2000 looked at psychosocial and behavioral factors as predictors of heavy drinking among adolescents and assessed students' expectancies about drinking. The study found that boys who reported positive drinking expectancies were over seven times more likely to become heavy drinkers than boys who had negative drinking expectancies. In fact, positive alcohol expectancy was the single strongest predictor of later heavy drinking among boys. However, the expectancy variables were not associated with later heavy drinking for the girls in the study (Griffin, et al, 2000).

(SEE ALSO: Coping and Drug Use; Disease Concept of Alcoholism and Drug Abuse; Ethnicity and Drugs; Prevention; Treatment; Women and Substance Abuse)

ALAN MARLATT

MOLLY CARNEY

REVISED BY MARY CARVLIN