Expectancy for social facilitation from drinking: the divergent paths of high-expectancy and low-expectancy adolescents

Personality characteristics ascribed to young women who drink alcohol

This study employs an implicit personality rating methodology from marketing research (the "shopping list" method) to investigate young adults' perceptions of young women who drink alcoholic beverages. Whereas there were some differences in perceptions of young women drinkers across the alcohol brands, the greatest differential in the personality ratings was caused by the mere presence or absence of alcohol. Young women who drink alcohol were described as being more "interesting" and "self-assured" than young women who do not drink. These positive traits were reliably ascribed to young women drinkers by both young adult men and young adult women and more strongly by those who themselves drink. Although it is possible that young women drinkers really are differentially interesting and self-assured, a more likely contributor to these perceptions is advertising aimed at the female drinker.

Expectancy influences the operation of personality on behavior

The authors investigated the moderating effect of expectancies on personality for 2 different addictive behavior processes: (a) drinking and (b) binge eating and purging characteristic of bulimia nervosa. Study 1 found that positive expectancies for social facilitation from drinking moderated the effect of extraversion on drinking behavior among undergraduate men and women. Study 2 found that the expectancy that eating will help manage negative affect moderated the effect of trait urgency on bulimic symptoms among undergraduate women. Thus, the relationships of the trait risk factors to these 2 addictive behaviors are stronger if one also holds certain expectancies for reinforcement from those behaviors.

Peer and partner drinking and the transition to marriage: a longitudinal examination of selection and influence processes

This study examined the longitudinal relationships among adult drinking, partner drinking, and peer drinking over the transition to marriage. Newlywed couples were assessed with respect to alcohol involvement, peer drinking, and risk factors and reassessed at their 1st anniversary. Husbands' premarital drinking was predictive of wives' drinking at the 1st anniversary, indicating partner influence. The results did not support a peer-influence hypothesis in that peer drinking at marriage was not predictive of husbands' or wives' drinking at the 1st anniversary. There was evidence, however, for a peer-selection effect with husbands' premarital drinking predicting peer drinking for both husbands and wives. Wives' premarital drinking was unrelated to the subsequent drinking of their peers or their husbands' peers.

Measuring alcohol expectancies with the implicit association test

Researchers have relied primarily on self-report questionnaires to measure alcohol expectancies. These questionnaires assess explicit expectancies about alcohol but donot provide any measure of the implicit processes that might also play an important role in determining drinking. The implicit association test (IAT; A. G. Greenwald, D. E. McGhee, & L. K. Schwartz, 1998), a reaction time task, measures differential associations of 2 target concepts with an attribute. In this study, the IAT provided a measure of the strength of associations of alcohol concepts to positive or negative outcomes in memory. This implicit measure of alcohol expectancies successfully predicted alcohol use in 103 undergraduates. The findings also supported the hypothesis that an implicit measure of expectancy can add to the predictive power of existing questionnaire-based measures.

Effects of negative and positive mood phrases on priming of alcohol words in young drinkers with high and low anxiety sensitivity

This study investigated whether potential emotional cues for drinking activate alcohol concepts in young drinkers. Participants were 84 university freshmen with high or low levels of anxiety sensitivity (AS). A verbal priming task measured activation (i.e., priming) of alcohol concepts (e.g., beer) by positive and negative mood phrases. Time to read alcohol target words was the dependent measure. Negative mood phrases consistently primed alcohol targets; positive mood phrases did not. Degree of negative mood priming did not differ as a function of gender or AS. Reported tendency to drink in bad moods predicted negative mood priming in women, whereas men showed negative mood priming irrespective of their reported drinking tendency. A general association between negative mood priming and severity of alcohol problems also emerged.

Changing expectancies: cognitive mechanisms and context effects

This article presents the proceedings of a symposium at the 2002 RSA Meeting in San Francisco, organized by Reinout W. Wiers and Mark D. Wood. The symposium combined two topics of recent interest in studies of alcohol expectancies: cognitive mechanisms in expectancy challenge studies, and context-related changes of expectancies. With increasing recognition of the substantial role played by alcohol expectancies in drinking, investigators have begun to develop and evaluate expectancy challenge procedures as a potentially promising new prevention strategy. The two major issues addressed in the symposium were whether expectancy challenges result in changes in expectancies that mediate intervention (outcome relations), and the influence of simulated bar environments ("bar labs," in which challenges are usually done) on expectancies. The presentations were (1) An introduction, by Jack Darkes; (2) Investigating the utility of alcohol expectancy challenge with heavy drinking college students, by Mark D. Wood; (3) Effects of an expectancy challenge on implicit and explicit expectancies and drinking, by Reinout W. Wiers; (4) Effects of graphic feedback and simulated bar assessments on alcohol expectancies and consumption, by William R. Corbin; (5) Implicit alcohol associations and context, by Barry T Jones; and (6) A discussion by Kenneth J. Sher, who pointed out that it is important not only to study changes of expectancies in the paradigm of an expectancy challenge but also to consider the role of changing expectancies in natural development and in treatments not explicitly aimed at changing expectancies.

Beyond invulnerability: the importance of benefits in adolescents' decision to drink alcohol

Some investigators propose that adolescents engage in risky behaviors mainly because they perceive themselves to be invulnerable to risk. However, studies have typically not included perceived benefits. In the current study, 5th, 7th, and 9th graders were surveyed about their perceptions of and experience with alcohol and tobacco. Results indicated that perceptions of the benefits were significantly related to drinking and smoking 6 months later, over and above perceptions of the risks, age of the respondent, and experience level. Further, the importance of benefits was replicated across 3 separate analyses. Experience with alcohol alone, especially positive experience, was also related to perception and behavior. These findings are discussed in terms of how to improve messages and influence adolescents' decisions regarding risk-taking behaviors.

Educational and occupational attainment and drinking behavior: an expectancy model in young adulthood

AIMS

The socio-economic status (SES) variables of education level and occupational functioning have been found to be correlated negatively with alcohol use. The present study examined prospectively the relationship between these functioning measures, alcohol expectancies and alcohol involvement. We propose that expectancies function as a mediator of the relationship between educational/occupational attainment and drinking behavior. We hypothesized that changes in young adult functioning are linked to changes in social context and/or the availability of non-alcohol reinforcers, which in turn affect the reinforcement expected from alcohol.

PARTICIPANTS AND DESIGN

Participants were 172 young adults from an ongoing longitudinal study of long-term clinical course of adolescent substance use treatment. Data from 6- and 8-year follow-ups were used in the present analyses. The treated sample (n=100) was recruited from in-patient substance abuse treatment programs for adolescents. A community sample (n=72) was matched on family history of substance abuse and SES at intake.

FINDINGS

A cross-lag panel analysis indicated that education had a unique longitudinal relationship with expectancy for both the treated and community sample, over and above previous alcohol use and expectancy. Occupational variables did not have a longitudinal relationship with alcohol use and expectancy for either sample. Expectancies mediated the education/drinking relationship for the treated sample only.

CONCLUSIONS

These results suggest one means through which changes in functioning may alter alcohol involvement over time: alteration of the reinforcement expected from alcohol.

Validation of eating and dieting expectancy measures in two adolescent samples

OBJECTIVE

To demonstrate the validity of two measures of eating and dieting expectancies (The Eating Expectancy Inventory [EEI] and the Thinness and Restricting Expectancy Inventory, [TREI]) for use with adolescents.

METHOD

Seventh (N = 392) and tenth graders (N = 300) completed the Bulimia Test-Revised (BULIT-R), the Eating Disorder Inventory (EDI-II), and two factors of the Three-Factor Eating Questionnaire (TFEQ).

RESULTS

Findings replicated for the two adolescent samples. The expectancy that thinness leads to overgeneralized life improvement correlated with measures of "successful" dieting, dieting plus disinhibition, and bulimic symptomatology. Expectancies for negative reinforcement from eating (e.g., eating helps manage negative affect and alleviate boredom) correlated with dieting plus disinhibition and bulimic symptoms, but not with successful dieting. Negative reinforcement from eating and reinforcement from thinness expectancies accounted for different bulimic symptom variance than that accounted for by the personality factors of perfectionism, interpersonal distrust, and ineffectiveness.

DISCUSSION

Results were consistent with prior work on college and clinical samples, thus supporting use of the expectancy measures with adolescents.

Brief prevention for adolescent risk-taking behavior

AIMS

Despite widespread prevention efforts to decrease adolescent risk-taking, substance use and driving after drinking (DD) are prevalent in the United States. The current study compared the efficacy of an abbreviated version of Drug Abuse and Resistance Education (DARE-A) to a new Risk Skills Training Program (RSTP).

DESIGN

Adolescent participation in drinking, drug use, DD and riding with a drunk driver was examined longitudinally. After baseline assessments, adolescents were randomly assigned to the RSTP. DARE-A or a no intervention control group and then completed 2-month post-test and 6-month follow-up assessments.

SETTING

Adolescents attended a mid-sized suburban high school.

PARTICIPANTS

The sample (N = 300) was comprised of 58% females and the age range was 14-19 years.

INTERVENTION

The RSTP was developed to target several risk behaviors and to examine the feasibility of conducting a brief personalized prevention program in a group setting. DARE-A focused on increasing knowledge and understanding the deleterious effects of substance use.

MEASUREMENTS

Risk-taking behavior, perception of peer risk-taking and positive and negative alcohol expectancies were assessed.

FINDINGS

RSTP participants decreased participation in several risk behaviors at post-test, but reductions were not maintained at 6-month follow-up. The control and DARE-A groups increased their positive and decreased their negative alcohol expectancies. The control group increased their alcohol consumption.

CONCLUSIONS

Results suggest that a brief, personalized, group prevention program is a feasible approach to reducing adolescent risk-taking. Strategies must be developed to solidify these positive changes so that they are longer-lasting.

College student heavy drinking in social contexts versus alone

Heavy drinking is common among college students and typically occurs in social contexts. Heavy drinking when alone, however, is less common. The present study hypothesized that students who drink heavily when alone (HD-Alone) would differ from college students who only drink heavily in social contexts (Social HD). Forty-nine HD-Alone students (at least one heavy-drinking episode when alone), 213 Social HDs, and 63 non-heavy drinkers (Non-HDs) were compared on alcohol-related consequences, drinking milestones, alcohol-outcome expectancies, and symptoms of depression. HD-Alone students reported more negative drinking consequences, earlier onset of regular drinking, more alcohol expectancies, less self-efficacy and motivation to reduce drinking, and higher depression scores than Social HDs and Non-HDs. Findings imply individual differences among heavy-drinking college students according to their drinking context.

Integrating disinhibition and learning risk for alcohol use

In this study the authors tested the acquired preparedness model of problem drinking, which holds that trait disinhibition, defined as neurotic extraversion by C. M. Patterson and J. P. Newman (1993), leads to the biased formation of positive over negative alcohol expectancies. Positive expectancies thus mediate disinhibition's influence on drinking. The authors also hypothesized that disinhibition moderates the expectancy-drinking relationship such that disinhibited individuals are more likely to act on their positive expectancies. In Study 1, positive expectancies both mediated and moderated the disinhibition-drinking relationship. In Study 2, learning task results indicated that disinhibited individuals sought reward, even when passive avoidance of punishment was indicated. Study 2 also replicated Study I hypotheses for men but generally not for women.

The role of family of origin food-related experiences in bulimic symptomatology

OBJECTIVE

With the goal of developing a model relating family of origin experiences to maladaptive cognitions to bulimic symptom formation, the authors developed a measure of family of origin food-related experiences called the Family History Inventory.

METHOD

A number (N = 662) of sixth to eighth-grade adolescents completed the inventory, eating and dieting expectancy measures, and the Bulimia Test-Revised (BULIT-R).

RESULTS

Fourteen scales were identified in the inventory. They emphasized family teasing about weight, negative maternal modeling regarding food, and family rules concerning eating. Eleven of the 14 scales correlated with the BULIT-R. Two superordinate factors called Family Teasing and Negative Maternal Modeling summarized 8 of the 14 subscales. Statistical tests were consistent with the hypothesis that eating and dieting expectancies mediate the influence of Family Teasing and Negative Maternal Modeling on bulimic symptomatology.

DISCUSSION

There was good evidence for the validity of the Family History Inventory. The theoretical implications of the mediation tests are discussed.

Social cognitive determinants of drinking in young adults: beyond the alcohol expectancies paradigm

In prior investigations of the psychology of drinking behavior, drinkers' positive expectancies regarding the effects of alcohol have been studied extensively. From a social cognitive point of view, however, several additional psychological factors also deserve attention, namely negative expectancies, social influence, and self-efficacy expectations. In a representative sample of 161 university students, this study examined to what extent inclusion of these additional social cognitive factors enhanced the predictive power of the predominant alcohol-expectancies model of drinking behavior, and to what extent all four social cognitive factors were related to the uptake and cessation of drinking behavior. The three additional social cognitive factors contributed 17% to the explained variance in drinking behavior, in addition to the 18% accounted for by positive expectancies. The constructs with the greatest predictive strength all pertained to the social effects and social context of drinking. The most important predictors of drinking behavior were found to differ for male versus female students, and for students living with their parents versus those living on their own. The data on drinking acquisition and cessation suggest that in this sample little change in drinking behavior could be expected. The social cognitive factors were strongly related to acquisition stages but only weakly to cessation stages. Recommendations for interventions aimed at lowering alcohol intake are given.

Alcohol expectancies and motives in a substance abusing male treatment sample

Although prior research has demonstrated the utility of both alcohol expectancies and drinking motives in the prediction of alcohol use and problems, the specific relationship between these domains has not been examined in a clinical sample. One-hundred, forty-seven veterans on an inpatient substance abuse unit completed questionnaires measuring alcohol expectancies and alcohol motives and provided information on their alcohol consumption and related problems. Covariance structure modeling was used to test four theoretically competing models. Findings indicated that: (1) motives mediate the effects of expectancies on use and problems and expectancies do not exert an independent influence on consumption and alcohol problems and (2) contrary to past findings, alcohol use only partially mediates the relationship between enhancement motives and alcohol problems.

Activation of alcohol expectancies in memory in relation to limb of the blood alcohol curve

Attempts to understand the mechanism by which alcohol expectancies might influence drinking have related activation of expectancies in memory to alcohol use. Limb of the blood alcohol curve, however, has not been considered. In the present study, 527 undergraduates completed the Anticipated Biphasic Alcohol Effects Scale and a drinking measure. Multidimensional scaling was used to map expectancies into memory network format, and likely activation of expectancies was empirically modeled. Heavier drinkers were most likely to activate positive and arousing expectancies associated with the ascending limb, whereas lighter drinkers were most likely to activate negative and sedating expectancies associated with the descending limb. These findings add to the literature suggesting that activation of expectancies in memory may be an important determinant of drinking behavior and a promising target for intervention strategies.

Psychometric evaluation of the marijuana and stimulant effect expectancy questionnaires for adolescents

Alcohol expectancies are important in the mediation and prediction of alcohol use. Expectancies for the effects of other drugs, although less well delineated, appear equally important. Therefore, development and validation of expectancy measures for drugs other than alcohol is necessary for evaluating the importance of these constructs. We examined the factor structure, reliability, and validity of the Marijuana Effect Expectancy Questionnaire (MEEQ) and the Stimulant Effect Expectancy Questionnaire (SEEQ) in clinical and community samples of adolescents as they moved into young adulthood (N=279). Confirmatory factor analyses (CFAs) supported the a priori factors, and we found good reliability for most individual scales. Temporal stability and convergent and discriminant validity of drug effect expectancies were supported in this sample of adolescents and young adults. Drug effect expectancies were associated with drug preference and drug use patterns over 2 years. Use of these measures may aid our understanding of the etiology and course of marijuana and stimulant involvement during adolescence and young adulthood.

Racial identity and its assessment in a sample of African-American men in treatment for cocaine dependence

Substance abuse treatment studies frequently include subjects from different ethnic and racial groups, but many investigations limit the examination of race and ethnicity to the use of nominal labels. This approach reveals little about the social or psychological significance of racial and ethnic group membership to the subjects of study or about the potential effects of these factors on substance-involved behaviors. In this study, a principal components analysis (PCA) with varimax rotation was conducted on the 50-item long form of the Racial Identity Attitude Scale (RIAS) (1) in a sample of 294 African-American men in treatment for cocaine dependence. The RIAS was developed to measure attitudes about race and racial status among blacks, but it has not been utilized widely in substance abuse research. Our findings provide evidence for the structural validity of this instrument in this sample of substance abusers. We discuss how recent advances in racial identity theory and its measurement may provide an important avenue for understanding the psychological consequences of racial group membership and for examining the potential effects of these factors on treatment response in studies of substance misuse.

Change in Alcohol Effect and Self-Efficacy Expectancies During Addiction Treatment

Alcohol effect expectancies and situational self-efficacy are important factors in addiction relapse. The extent to which these cognitive factors change during alcohol treatment and the relations between change in these two domains may facilitate our understanding of the transition out of addictive lifestyles. To evaluate such change, 101 male and female participants in an abstinence focused inpatient alcohol and drug treatment program completed measures of alcohol effect expectancies (AEQ) and self-efficacy (SCQ) during the first and fourth (final) week of inpatient treatment. Analyses confirmed our predictions that alcohol effect expectancies and self-efficacy were inversely correlated at the beginning of treatment, and that both alcohol expectancies and self-efficacy changed significantly during the four-week treatment program. Larger decreases in alcohol effect expectancies were evident for individuals who entered treatment with less confidence in their ability to resist drinking compared to those with more confidence in potential drinking situations. Results suggest that both expectancies and self-efficacy are useful factors to target in alcohol treatment.

Children's beliefs about drinking

This paper reports the results of a study in which age (grade level), racial/ethnic, and gender differences in beliefs and perceived norms about drinking were examined in a multi-ethnic urban sample of 4th through 7th grade children. Results showed that older children held beliefs and perceived norms that were more favorable toward drinking than younger children. The major difference between older and younger children lay in their differential estimates of the likelihood of certain consequences occurring and not in their evaluation of these consequences of drinking. Further, older children not only displayed less motivation to comply with their parents and greater motivation to comply with their peers, but they also perceived their parents, as well as their peers, as less disapproving of drinking than did younger children. There were few gender or race/ethnicity differences at these ages in children's beliefs and perceived norms about drinking.

Self-efficacy for refusal mediated by outcome expectancies in the prediction of alcohol-dependence amongst young adults

The present study examined the relative importance of outcome expectancies and self-efficacy [1] in the prediction of alcohol dependence [2] and alcohol consumption in a sample of young adult drinkers drawn from a milieu previously reported as supportive of risky drinking. In predicting alcohol dependence, outcome expectancies were found to mediate self-efficacy and the same pattern was found for both males and females. This suggests that male and female drinkers may become more similar as they progress along the drinking continuum from risky drinking to dependent drinking. However, in women, in comparison to men, a greater array of expectancies and self-efficacy scales were found to predict heavy drinking, as measured by quantity and frequency. These results suggest that heavy drinking women are particularly at risk of developing drinking related complications and that preventative education needs to take into account gender differences.

The utility of novelty seeking, harm avoidance, and expectancy in the prediction of drinking

To test the hypothesis that two temperament scales (Novelty Seeking and Harm Avoidance) are differentially related to alcohol expectancies and drinking patterns, 140 adolescents from an inpatient psychiatric facility completed several self-report questionnaires measuring temperament, alcohol expectancies, and alcohol consumption. Moderated multiple regression analyses indicated that Novelty Seeking was significantly related to frequency of drinking and problem drinking, but that Harm Avoidance was not related to these variables. Results of the MANOVA indicated that high novelty seeking and low harm avoidant (Type 2) individuals had a significantly higher frequency of drinking than did individuals who were high on Harm Avoidance and low on Novelty Seeking (Type 1). Results also showed that expectancy and Novelty Seeking contributed significant independent and overlapping variance in the prediction of amount of drinking. Although Novelty Seeking was related to expectations of social functioning, other hypothesized relationships between temperament and expectancy were not supported.

A PC-based software test for measuring alcohol and drug effects in human subjects

A new software-based visual search and divided-attention test of cognitive performance was developed and evaluated in an alcohol dose-response study with 24 human subjects aged 21-62 years. The test used language-free, color, graphic displays to represent the visuospatial demands of driving. Cognitive demands were increased over previous hardware-based tests, and the motor skills required for the test involved minimal eye movements and eye-hand coordination. Repeated performance on the test was evaluated with a latin-square design by using a placebo and two alcohol doses, low (0.48 g/kg/LBM) and moderate (0.72 g/kg/LBM). The data on 7 females and 17 males yielded significant falling and rising impairment effects coincident with moderate rising and falling breath alcohol levels (mean peak BrALs = 0.045 g/dl and 0.079 g/dl). None of the subjects reported eye-strain or psychomotor fatigue as compared with previous tests. The high sensitivity/variance relative to use in basic and applied research, and worksite fitness-for-duty testing, was discussed. The most distinct advantage of a software-based test that operates on readily available PCs is that it can be widely distributed to researchers with a common reference to compare a variety of alcohol and drug effects.

Alcohol expectancies in a Native American population

Native Americans, as a group, have a high prevalence of alcohol abuse and alcohol dependence, although specific risk factors for alcoholism among this population have yet to be clearly identified. One set of factors that may contribute to the development of alcoholism are expectations of alcohol's effects. Previous research has shown that heavy drinkers and alcoholics have higher alcohol-related expectancies. Some studies have also shown an association between alcohol expectancies and a positive familial history of alcoholism. To examine factors that are related to expectations of alcohol's effects in a Native American population, this study evaluated healthy, nonalcoholic Mission Indian men between the ages of 18 and 25 years using the short form of the Alcohol Expectancy Questionnaire (AEQ). The influence of recent drinking history, family history of alcoholism, and degree of Native American heritage on alcohol-related expectancies was determined using regression analyses for the total AEQ score and for each of the six AEQ subscales. Recent drinking history accounted for a significant proportion of the variance in the total score, as well as scale I (global positive changes) and scale VI (arousal and power) of the AEQ. Degree of Native American heritage and family history of alcoholism did not account for a significant amount of variability in alcohol expectancies. These results suggest that, consistent with findings in other populations, alcohol expectancies are related to drinking patterns in Mission Indians. However, no association with two other potential risk factors were found in this sample of Native Americans.

Alcohol expectancies among adolescents with conduct disorder: prediction and mediation of drinking

Among adolescents with conduct disorder, alcohol expectancies were examined for both predictive utility and mediation of other predictors of alcohol use (i.e., delinquency, family history, demographic and psychopathology variables). Data were collected from 260 adolescents with conduct disorder 11-18 years of age, who had been in either mental health residential facilities or community-based special education programs for adolescents with serious emotional disturbances. Zero-order correlations and structural path models assessed relationships between expectancies, alcohol use, and other predictors of alcohol use. Results indicated: (a) expectancies of enhanced social and cognitive behavior were significant (p < .05) univariate predictors of drinking, (b) among all of the selected predictors, expectancies of enhanced social behavior (i.e., Subscale 2 of the Alcohol Expectancy Questionnaire--Adolescent Form; AEQ-A) had the strongest association with alcohol use (r = .54) and mediated between 31% to 44% of the drinking variance associated with other significant predictors (p < .01). Results were discussed as supporting similar expectancy-drinking relationships among CD and nonclinical youth.