xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002059706
007 cr bnu|||uuuuu
008 100208s2005 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001722
Bekman, Nicole M.
The role of sensation seeking in children's ability to learn alcohol expectancy associations
h [electronic resource] /
by Nicole M. Bekman.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 57 pages.
Thesis (M.A.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Sensation seeking is a personality characteristic associated with problematic alcohol use and positive alcohol expectancies, but little research has examined the relationship between sensation seeking and the acquisition of alcohol expectancy information. In a recent study (Steinberg, 2003), sensation seeking was associated with how quickly and accurately college-aged students were able to learn alcohol-expectancy word pairs in a paired associate learning task. In this age group, however, the individuals had fully developed alcohol expectancies that may have influenced their rates of learning. The current study sought to minimize the influence of previously held alcohol expectancies by exploring this relationship in children when the development of alcohol expectancies is just beginning. The participants in this study were fifth grade students. A series of regressions examined the relationship between sensation seeking, alcohol expectancies, current and predicted future drinking with the acquisition of alcohol and expectancy word pairs in a paired associate learning task. Although no statistically significant relationships were found, children with higher drinking frequency and males with higher Thrill and Adventure Seeking (TAS) demonstrated a minor advantage in their ability to match alcohol and expectancy words in cued-recall trials. Although the results of this study are inconclusive, they suggest that sensation seeking may play a role in the acquisition of alcohol expectancies. Future research with refined word pairs and a larger sample size is necessary to further clarify these trends.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Mark Goldman, PhD.
Paired associate learning, , ,
t USF Electronic Theses and Dissertations.
The Role of Sensation Seeking in Childrens Ability to Learn Alcohol Expectancy Associations by Nicole M. Bekman A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Mark Goldman, Ph.D. Douglas Nelson, Ph.D. Vicky Phares, Ph.D. Date of Approval: November 5, 2004 Keywords: Paired Associate L earning, School-Aged, Alcohol Use 5 th graders, Development Copyright 2005, Nicole M. Bekman
ii Table of Contents List of Tables iii Abstract iv Introduction 1 Expectancy theory 2 Development and modification of alcohol expectancies during childhood 3 Influences on alcohol expectancies in children 6 Sensation seeking 9 Educational significance 12 Specific aims 15 Method 17 Participants 17 Description of the sample 17 Measures 18 Procedure 24 Results 25 Descriptive statistics for the dependent measures 25 Descriptive statistics for the independent measures 28 Transformation of non-normally distributed variables 29 Correlational analyses 30 Gender differences on independent variables 32 Analyses of the relationship between the independent variables and task performance 33 Discussion 36 Reference List 40 Appendices 46 Appendix A: Paired Associate Learning Task Version A 47 Appendix B: Pair ed Associate Learning Task Version B 50 Appendix C: Alcohol Exp ectancy Questionnaire Adol escent Form: Scale 2 (Revised) 53 Appendix D: Sensation Seeking Scale for Children 54 Appendix E: Dem ographics and Drinki ng Questionnaire 56
iii List of Tables Table 1 Descriptive statistics of the dependent measures 26 Table 2 T-Test of differences in m eans between alcohol-expectancy and control-expectancy words 27 Table 3 T-Test of differences in mean s between control-control words in the two, randomly assigned group 27 Table 4 Descriptive statistics of the Alcohol-Expectancy word pairs scored using the category-correct scoring system. 28 Table 5 Descriptive statistics of the independent measures 29 Table 6 Descriptive statistics of the transformed variables 30 Table 7 Zero-order correlations between independent measures 30 Table 8 Zero-order correlations between independent variables and performance on the PALT for Alcohol-Expectancy word pairs 31 Table 9 Zero-order correlations between independent variables and performance on the PALT for Alcohol-Expectancy word pairs scored using the category-correct scoring system 35
iv The Role of Sensation Seeking in Childrens Ability to Learn Alcohol-Expectancy Associations Nicole M. Bekman ABSTRACT Sensation seeking is a personality char acteristic associated with problematic alcohol use and positive alc ohol expectancies, but little research has examined the relationship between sensation seeking a nd the acquisition of alcohol expectancy information. In a recent study (Steinberg, 2003), sensation se eking was associated with how quickly and accurately college-aged stude nts were able to learn alcohol-expectancy word pairs in a paired associate learning tas k. In this age group, however, the individuals had fully developed alcohol expectancies that may have influe nced their rates of learning. The current study sought to minimize the influence of previously held alcohol expectancies by exploring this relationship in children when the development of alcohol expectancies is just beginning. The participan ts in this study were fi fth grade students. A series of regressions examined the relationship between sensation seeking, alcohol expectancies, current and predicted future dr inking with the acquisition of alcohol and expectancy word pairs in a paired associate learning task. Although no statistically significant relationships were found, children with higher drinking frequency and males with higher Thrill and Adventure Seeking (TAS ) demonstrated a minor advantage in their ability to match alcohol and expect ancy words in cued-recall trials.
v Although the results of this study are inc onclusive, they suggest that sensation seeking may play a role in the acquisition of al cohol expectancies. Future research with refined word pairs and a larger sample size is necessary to further clarify these trends.
INTRODUCTION Research has shown that adolescents who are dependent on alcohol display memory impairment, distorted perception of spatial relationships, and weakened verbal skills (Brown, Tapert, Granholm & Delis, 2000) Teenagers who drink heavily are at greater risk for suicide (National Institute on Alcohol Abuse and Alcoholism; NIAAA, 1996, as cited in Leadership for a Drug Free America, 2002), injury (Hingston, Heeren, Jamanka & Howland, 2000), fatal crashes (Nat ional Highway and Safety Patrol; NHSP, 2001) and risky sexual behavior (Kaiser Fa mily Foundation, 2002). Additionally, Grant and Dawson (1997) found that more than forty percent of individuals who begin drinking before age 13 have an increased risk for developing alcohol abuse or alcohol dependency in the future. An extensive body of resear ch has formed attempting to understand and resolve this societal problem. Application of expectancy theory in this area has been valuable in efforts to understand peoples motivations to drink alcohol. Numerous studies have indicated that adolescents and adults expect ancies about alcohol influen ce the amount of alcohol that they consume (Brown, Goldman & Christia nsen, 1985, Christiansen, Smith, Roehling & Goldman, 1989, Darkes & Goldman 1993). There is also significant evidence that childrens expectancies about alcohol can influe nce their intentions to drink in the future (Austin & Meili, 1994) and are hypothesized to predict futu re drinking behavior (Dunn & Goldman, 1996; 1998; 2000).
2 Several researchers have e xplored the possibility that alcohol expectancies may serve as a mediator between identified ri sk factors for problem drinking and actual drinking (Finn, Sharkansky, Brandt, & Turc otte, 2000; Henderson, Goldman, Coovert, & Carnevalla, 1994). Expectancy research can he lp to better understand how risk factors for problems with alcohol transfer to actua l alcohol use over the course of a childs development and how childrens expectancies of the effects of alcohol may accelerate or inhibit the influence of other ri sk factors on drinking behavior. Expectancy Theory Formal expectancy theory was first a dvanced by Tolman (1932) and has been modified through years of research by M acCorquodale and Meehl (1954), Rotter (1954, 1981) and Bandura (1977). According to expectancy theory, human behavior can be understood in part by cognitions about the environment around us. These cognitions affect how people respond to different s timuli, and are influenced by their past experiences. People act in di fferent ways to achieve a ce rtain expected and desired outcome. These expectancies have been theori zed to be stable over time and experience, as specific behaviors continue to result in the anticipated outcome. Brown, Christiansen and Goldman (1987) described an expectancy as an if-then connection that occurs when a particular cue is observed an d a specific outcome is anticipated. Alcohol expectancies refer to an individuals knowledge and understanding of the effects and consequences of alcohol consump tion. These expectancies have been thought to be acquired early in life and are stored in a semantic memory network (Goldman, 1989; 1999). They influence peoples decisions about whether or not to drink alcohol. These expectancies are reinforced or modifi ed during the course of ones life based on
3 the individuals experiences. Those associations that have been strengthened through experience will have more influence on decision-making. Childrens expectancies about the effect s of alcohol develop well before the individual has any experience drinking alc ohol (eg. Noll, Zucker & Greenbaum, 1990; Dunn & Goldman 1996; 1998; 2000). Therefore, they must learn these expectancies through other means such as societal norms, pa rental behavior, various forms of media, and peer groups. How and when children acquire information about the effects of alcohol may vary due to the individual risk factors that increase the probability of developing problems with alcohol. Development and modification of al cohol expectancies during childhood Noll et. al. (1990) reported that preschool ag ed children were able to discriminate alcohol from other liquids, and also were awar e that adults usually drink alcohol rather than children. This indicates that at very young ages, childr en had developed a cognitive schema for alcohol and its use. Further e xploration of children s knowledge of alcohol (Miller, Smith & Goldman, 1990) revealed that children held expectancies about alcohol at all of the ages evaluated (ages 6-11). These expectancies were not static: older students had more positive alcohol expectancies than did st udents in the grades below them, and the largest increase in the endorse ment of positive expectancies was observed between grades three and four. This change is consistent with c ognitive developmental patterns typically occurring during early adolescence, including the beginning of development of abstract reasoning (Grabe r & Peterson, 1991). Abstract reasoning incorporates abilities such as understandi ng complex concepts and assimilating new information outside of ones personal experience, such as the variety of effects that one
4 may expect from alcohol use, even without any personal experience with alcohol. Additionally, between grades three and fi ve, childrens receptive and expressive communication abilities improve (Miller et. al., 1990), allo wing for increased reception of societal messages about al cohol and better communicati on of the childs understanding of alcohol to others, including researchers. Johnson and Johnson (1995) looked more clos ely at childrens expectancies and found that they have both positive (expectation of desirable outcomes) and negative (expectation of undesirable outcomes) expectan cies. These expectancies were examined across multiple school grades, in cluding first, fourth and seventh grade. At all ages children had significantly more negative expectancies of alcohol than positive and both positive and negative expectancies increased with age, indicating that older children were more aware of the effects of alc ohol than were younger children. Dunn and Goldman (1996) conducted an in-depth study of childrens alcohol expectancies. In an initial phase of this study, the author s elicited words from children that describe the effects of al cohol on adults. They then created a measure of childrens alcohol expectancies in which each item asks children how often alcohol causes a certain expected feeling in adults. These expectan cy words were either identical or closely matched to words generated by adults in previous work (Rather, Goldman, Roehrich & Brannick, 1992). They administered the ne w measure to students in classrooms from second to fifth grade. Individual Differences Scaling was used to map out these expectancies on two axes ( good-bad, and sedating-arousing) based on a score from a four-point Likert scale of how often these e ffects of drinking are e xperienced. Preference mapping analyses were then us ed to plot a vector thr ough the hypothetical expectancy
5 network to model the association pathways as a function of grade. This vector represented the judged frequency of occurren ce for each expectancy. This analysis showed that children in sec ond grade were more likely to have negative expectancies, such as dangerous or mean, and were more likely to make a value judgment (positive vs. negative) than judgment based on expected arous al or sedation. Fi fth graders, however, had more positive expectancies than did sec ond graders, such as cool or wild, and placed increased emphasis on arousal. This line of research was continue d (Dunn & Goldman, 1998; 2000) using the same techniques of Individual Differences Scaling and Preference Mapping to further explore changes in childre ns alcohol expectancies. In 1998, Dunn and Goldman explored differences between higher drinking and lower drinking children from third, sixth, ninth and twelfth grade. Within each grade, higher drinking children had more positive and arousing expectancies, and between grades, children in higher grades had more positive and arousing exp ectancies than those in the grade below them. In 2000, Dunn and Goldman replicated these findings wi th a different measure, Childrens First Associates. This method was used because it was thought to be a more direct means of retrieving uncontaminated memory contents. Again, similar results were found, where younger and lower drinking children were more likely to report negative outcomes, like bad, while older and higher drinking ch ildren would report more positive outcomes, such as happy. An additional study by Query, Rosenberg and Tisak (1998) examined differences between types of beverage to ensure that th e expectancies that ch ildren endorsed in those studies previously described were specific to alcohol and not applicable to all adult
6 beverages. Query and her colleagues compared second and third grade childrens expectancies of an alcoholic drink (beer) to their expectan cies of a non-alcoholic drink (iced tea). They found that these children ha d significantly more negative expectancies towards beer and positive expectancies towards i ced tea. This study se rved to clarify that alcohol is a salient construct to children and is one that is qualitatively different than other adult-related concepts. To study some of the motiva tions that children attri bute to adoles cent drinking, Johnson and Johnson (1996) explored childrens expectancies of the social consequences for drinking in first, fourth and seventh grade students. In all grades children agreed that parents would react negatively to adolescent drinking. However, more than twice as many fourth and seventh graders felt that their friends would approve of adolescent drinking than did first graders. Older children were also more likely to cite social motivation for drinking than were younger chil dren. Despite this pattern, at all grade levels children expected an adolescents friend to respond negativel y to someone refusing an alcoholic drink. These results, combined with other research concerning changes in alcohol expectancies around the same age, indica te that older children feel that social and peer approval strongly influence an adolescents decision to dr ink (i.e., that older children are more likely to describe alcohol as cool such as in Dunn & Goldman, 1996). Influences on alcohol ex pectancies in children The two most researched influences on childrens expectancies of alcohol are parental drinking and media, specifically alcohol advertis ing. Numerous studies have identified children of alcoholics (COAs) as a group at high risk for future alcohol abuse and dependence (Schuckit, 1994). However, how it is that some COAs experience these
7 problems while others do not is still unknown. Alcohol expectancies might play some role in this distinction. Studies compari ng young COAs to controls (Miller et. al., 1990; Kraus, Smith & Ratner, 1994) have found that young COAs have more negative expectancies of alcohol than their counterparts, indicating that at this age COAs expectancies may reflect their negative expe riences with an alcoholic parent. On the other hand, older adolescent COAs are more lik ely than their controls to have higher positive expectancies towards alcohol (Brown, Creamer & Stetson, 1987). The results of these studies indicate how complicated a nd difficult it can be to tease apart the relationship between risk factor s and mediational factors, such as alcohol expectancies. These studies also open up the possibilities of furt her research regarding the interaction of family history and alcohol expectancies to more fully explain the relationship. Besides exposure to alcohol information within the family environment, children also learn a significant amount of alcohol expectancy inform ation from media sources. Specifically, studies concerning the effects of media on childrens expectancies have shown significant effects of al cohol advertising. Austin a nd Meili (1994) examined the alcohol expectancies of a sample of children considered at-risk for alcohol abuse. The authors examined childrens perceptions of alcohol use by adults at home and people on television, and compared these to childrens ex pectancies of the effects of alcohol, as well as their intention to drink alcohol in the future. They also explored the extent to which the child felt that television was repres entative of real life, how often they saw alcohol in real life, what ki nds of television shows they were most likely to watch and how often. Results showed that both child rens identification with television, and
8 modeling at home were positive predictors of risky expectancies of alcohol use. These expectancies were in turn predictors of intention to drink. An additional study concerning alcohol advertising found that children had significantly higher positive expe ctancies of alcohol after watching and evaluating five beer commercials when compared to a control group that ev aluated five soda commercials (Dunn & Yniguez, 1999). Usi ng the Childrens Expectancy Measure and First Associate Expectancy Measure, the authors mapped childrens paths of association through a memory network. They found that ch ildren in the four th grade who were exposed to five beer commercials had more arousing and positive expectancies and were more similar to fifth-grade control students than fourth grade controls. In turn, fifth grade students who had seen beer commercials had more arousing and positive expectancies than fifth grade controls. Although the results of this study are striking, it is important to remember that the effects of these five beer commercials on students in a classroom could have temporarily changed childrens expectanci es, but it does not necessarily predict how these same children would react hours or days after seeing the same commercials. Additionally, the study did litt le to explain the long-ter m effects that hundreds of commercials can have on children over time as they are experiencing them in life. Both children in the experimental and control situation had probably seen beer commercials before this study. This study does not account for individual differences of exposure these students had to televisi on and alcohol commercials before their experience during their participation in research began. Both studies concerning the media did not discuss one more important variable: the likelihood that some children may be more susceptible to the influence of
9 commercials than other children. Austin and Meili (1994) explored some risk and protective factors that may infl uence alcohol expectancies. Some important possibilities include: influences from home, peer approval, level of suggestibility, level of intelligence, prior exposure to drinking and personality traits. Sensation seeking is a personality trait that may be particularly influential on peoples alcohol expectancies. Sensation Seeking Sensation seeking is defined by Zuckerman (1979) as the need for varied, novel and complex sensations a nd experiences and th e willingness to take physical and social risks for the sake of such experiences (p. 10). This personality trait is expressed behaviorally through different forms of risk -taking behavior, such as driving habits, health, gambling, financial activities, alcohol and drug use, sexual behavior, and sports. It also has been shown to play a signifi cant role in career choices and decision making, job satisfaction, social premarital and mar ital relationships, eating habits and food preferences, media and art preferences, hum or, fantasy, creativity and social attitudes (Zuckerman, 1994). The relationship between sensation seeking and alcohol use has led to research specifically examining the re lationship between sensation seeking and initiation of substance use (Martin et. al ., 2002), drinking habits (Stacy, Newcomb & Bentler, 1991), and alcohol expectancies (Finn et. al., 2000; Henderson et. al., 1994; Katz, Fromme & DAmico, 2000). Several studies have demonstrated that th ere may be developmental differences in the level of sensation seeking over the life sp an. Sensation-seeking has been shown to increase from adolescence to adulthood and then decrease with age throughout adulthood (Zuckerman & Neeb 1980). Specifically, incr eases in novelty seek ing, risk taking, and
10 sensation seeking during adolescence occur across species, including humans, rats and non-human primates (Spear, 2000). This may be evolutionarily adaptive because increases in sensation seeking encourage in dividuals to explore new things and new territories during a critical time period, whic h could help prevent inbreeding and promote gene variations (Spear, 2000). It has been argued that small amounts of risk taking may be considered developmentally appropria te experimentation, because adolescents engaging in some risk taking behaviors are more socially competent than both their abstaining and frequent risk taking counterparts (Shedler & Block, 1990). Sensation seeking has been associated with drinking a nd intentions to drink in children as young as ten years old (Brody, Flor, Hollen-Wri ght, McCoy & Donovan, 1999; Webb, Baer & McKelvey, 1995), indicating that sensation seeking may be considered a risk factor before actual problem behavior becomes apparent. Several researchers have hypothesized that people high in sensation seeking enjoy the effects of substance use while people low in sensation seeking avoid using drugs and alcohol because the experience is stressful to them (Klerbaur & Bardo, 1999; Zuckerman, Ballenger, & Post, 1984). Rothbart, Derr yberry and Posner (1994) theorize that personality variables may influence behavior through differences in the functioning of neural structures that control selective atte ntion to consequences and rewards. Since multiple studies have demonstrated that al cohol expectancies partially mediate the relationship between sensation seeking and alcohol use (Webb, Baer, Grancis & Caid, 1993; Henderson et. al, 1994; Fi nn et. al., 2000; Katz et. al ., 2000), Katz et. al. (2000) theorized that perhaps alcohol expectancies reflect these sens itivities to consequences and rewards.
11 Alcohol expectancy theory suggests that certain personality characteristics may place some individuals at risk for alcohol problems because they may be more likely to acquire more positive alcohol expectancy information or to gain positive alcohol expectancies at a faster rate than others do. Sensation seeking is one indicator of this type of risk. Past research has shown that alcohol expectancies and level of sensation seeking can both be used to predict drinki ng behavior and that alcohol expectancies partially mediate the relationship between sensation seeking and drinking. One possibility is that differen tial acquisition of expectancy information is driving this mediational relationship. It is possible that individuals w ho are high in sensation seeking are more likely to attend to and absorb information about positive or arousing alcohol expectancies and that these expectancies in turn encourage drinking behavior. In a previous study that examined this relationship (Steinberg & Goldman, 2003), the authors used a paired-associate learning task with cued-recall and free recall to determine if participants who scored higher on measures of sensation seeking were able to better learn alcohol-expectancy word pair s. This paradigm asks participants to remember a list of word pairs matching alc ohol words (eg. keg, beer), positive/arousing alcohol expectancy words (e g. happy, fun), and neutral words (eg. backpack, desk). Participants (university stude nts) who scored higher on measures of sensation seeking were able to freely-recall mo re alcohol and expectancy word pairs than lower sensation seekers. Additionally, participants who reported drinking more alcohol and more drinking-related problems learned alcohol and expectancy pairs at a faster rate during cued recall than those who reported less drin king and fewer problems. These results lend support to the theory that personality differe nces may be an important factor in the
12 acquisition of positive alcohol expectancies and that acquisition of expectancies may mediate the relationship between sensation seeking and problems w ith alcohol. This study does not, however, account for how an individuals level of experience with drinking may affect their expectancies and t hus, their performance on the paired associate learning task. Previous alcohol expectancies may have inte racted with the relationship between sensation seeking and acquisition of alcohol expectanci es, thus clouding the interpretation of the results of this study. Individuals with fully developed alcohol expectancies may be more likely to remember alcohol-e xpectancy pairs that are congruent with their already developed beliefs about alcohol. In other words, this task may have tapped into the streng th of pre-existing alcohol ex pectancies rather than the effect of sensation seeking on alcohol expectancy acquisition. Educational Significance Traditionally, schools have been involved in promoting efforts to reduce student involvement with drugs and alcohol. The pr evention programs available in schools have become increasingly guided by research, and have broadened their focus from the individual to include environm ental influences and social norms. Despite these efforts, 54 percent of fourth through sixth graders repor ted learning about the risks of drug use at school, but only 30 percent repo rted learning about how dange rous drinking and smoking can be (National Survey on Drugs and Alcohol 1995, as cited in Leadership for a Drug Free America, 2001). In order for school-bas ed prevention programs to be effective, more emphasis needs to be placed on alcohol, an d more research is necessary to establish the most effective means of preventing alcohol use.
13 Despite evidence that children primarily hold negative alcohol expectancies, previous prevention efforts, such as DARE, have attempted to teach these to children (teaching them what they alre ady know). As they grow older, however, they increasingly attend to the physiological responses to alcoho l, and begin to expect that alcohol will either have arousing or seda ting effects on their mood. Becau se of this pattern, Dunn and Goldman (1998) theorized that prevention e fforts will be more effective if they emphasize the sedating effects of alcohol, as most young people drink in order to experience the more desirable, arousing feelings that increase their ab ility to socialize. Kraus et al. (1994) attempted to use this information in thei r videotape prevention study. These researchers used two different tr eatment conditions, as well as two controls: an expectancy-related videotape with adult actors, expectancy-related videotape with puppets, alcohol-informational videotape and no videotape. Results showed that childrens expectancies as measured by the CO PE increased over time (four weeks) in all conditions except for the puppet tape. This out come was particularly striking because the puppet expectancy tape was not onl y able to resist increases in alcohol expectancies, but was the only condition to successfully decr ease these expectancies. The authors theorized that expectancy tape with adult actors might have been unsuccessful because children have goals of learni ng to be more like adults. Wooten (1995) used a more interactive ap proach when trying to challenge alcohol expectancies in adolescence. Modeling her study after Darkes and Goldman (1993), Wooten designed a preventative intervention that would directly involve middle school students in challenging the expectancies held by a college sample. Over multiple sessions, students were exposed to alcohol advertising and discu ssed the contradictory
14 messages about the effects that alcohol is pur ported to have. They then worked as a group to develop part of an expectancy ch allenge for college students in which the students would be receiving placebo or alcoholic drinks, and were asked to guess who had ingested alcohol. The de signed experiment was then im plemented and videotaped in a lab designed to look like a bar. The mi ddle-school aged subjects viewed the videotape of the college students participating in thei r study and discussed what they saw. In particular, the behavior of college students who had not ingested alcohol was pointed out and discussed. Pre and post-test measures of alcohol expectanci es showed that the expectancy-treatment group had significantly decreased expectancies even four weeks after the final session, while traditional alc ohol information and no treatment groups had increased expectancies over this time. Most recently, Cruz and Dunn (2003) have found another way to modify the expectancy challenge to make it salient to ch ildren. They lead children in a discussion that challenged commonly held beliefs about positive and arousing effects of alcohol by pointing out inconsistencies in common ideas about how alcohol will make someone feel (ex. happy or depressed, energetic or sick). Instead of enc ouraging students to say no to alcohol, the discussion attended to the negative and sedating effects associated with alcohol use. Finally, a quiz game was played to reinforce the main points of the earlier discussion. Analysis using Individual Differences Sca ling and Preference Mapping revealed that expectancy groups place far less emphasis on arousal vs. sedation effects, and much more importance on the positive vs. negative dimension. In particular, they reported more negative alcohol expectancies than other treatment groups. These results are extremely promising, given that this inte rvention was successful within a classroom
15 setting, and after only one session. Of course, further research is needed to replicate these results with a much longer follow-up period. The current study can serve to improve prevention efforts by focusing on specific risk factors, such as sensa tion seeking, in order to understa nd how this particular risk factor influences how children learn alcohol expectancy inform ation. If this process is better understood, than prevention efforts can be developed to target the children who are at risk, and the learning processes that lead to the acquisition of positive and arousing expectancy information. This study, as we ll as continued expectancy and prevention research, is crucial for efforts to reduce underage drinking. Specific Aims The current study aimed to further examine the acquisition of alcohol expectancies by measuring sensation seek ing and performance on a paired-associate learning task with children. Ideally, this study could be conducted at an age before children have acquired any alc ohol expectancies, so that we could examine this process without contamination of preexisting expectan cies. Research has shown, however, that children have already developed both positive and negative alcohol expectancies by the age of six (Miller et. al., 1990) and in fact, children as young as preschool can discriminate alcohol from othe r substances (Noll et. al., 1990). Rather than attempting to find children devoid of alcohol expectancies, if such childre n exist at all (Noll et. al., 1990), it may be beneficial to examine childrens alcohol expectancies during the time period when their expectancies are changing fr om negative to positive, but before alcohol use begins. As describe d earlier, previous literature (Dunn & Goldman, 1996, 1998, 2000; Johnson & Johnson, 1995, 1996; Miller et. al, 1990) has demonstrated that the
16 emphasis children place on different types of alcohol expectancies changes over time, specifically between third and sixth grade, a critical developmental period for the processing of alcohol expectancy information. By examining how childrens accuracy and speed of learning alcohol-expectancy word pairs in the paired-associate learning task may vary depending on the childs level of sensation seeking, this study was able to eliminate the contaminating effects of drinking experience on the results. Additi onally, this study was designed to tap the process of learning alcohol expectancy asso ciations during a developmental period when this process is occurring naturally as well. Th is allows us to get one step closer to the overall process of how individuals acquire alcohol expectancies. The following hypotheses were tested: 1. Children who scored higher on measures of sensation seeking w ould recall a greater proportion of word associations containing alcohol and expectancy content than those who scored lower on measures of sensation seeking. 2. Participants who scored higher on measures of sensation seeking would learn alcoholexpectancy word associations more quick ly across cued recall trials than would participants who scored lower on measures of sensation seeking. 3. Measures of sensation seeking would not diffe rentiate individuals in their ability to recall non-alcohol (control) a nd expectancy word pairs or control and control word pairs.
17 METHODS Participants One hundred and seventy-two fifth-grade st udents in after-sc hool programs were contacted regarding this st udy. These children attended af ter-school programs offered either by YMCA Latchkey or School Age Child Care (SACC), which is run by the School District of Hillsborough County. Thes e programs were chosen because they are the two largest after-school programs in th e area and are available to children in a representative sample of neighborhoods in Hillsborough County. An active informed consent procedure was used in which parents we re informed of the research and asked to provide permission for their child to participate in the study. Only students who returned the permission forms were allowed to particip ate. 67 percent of the children contacted returned their parental permission slips and of these, 87 percent agreed to participate in the study. The resulting sample included 97 fifth grade students, 81 percent of which were recruited from SACC programs. Due to this high response rate, this sample is likely to be representative of 5 th grade children in Hillsborough County after school programs. Description of the Sample All participants were fifth-grade students attending after-school programs in Hillsborough County. Their mean age was 10.69 years (SD = .585) with a range of 9 to 12 years. 54.6 percent of part icipants were male. This sample was diverse; 50.5 percent of participants identified themselves as White/Caucasian, 14.4 percent as Black/AfricanAmerican, 17.5 percent as Hisp anic/Latino(a), 5.2 percent as Asian, and 11.3 percent as
18 other. 82 percent of these students were enrolled in SACC (18 percent YMCA) and no site differences were found for any of the independent or dependent variables. Measures Paired-associate Learning Task (PALT). The paired-associate learning task was orig inally developed to test theories of memory and learning in cogniti ve psychology. Typically during this task, the participant is presented with a number of paired items, which may be objects, letters, words or nonsense words. After the participant has observe d all of the pairs, he or she is presented with the first item and asked to recall the second item in the pair. In most uses of this task, items are randomly paired to minimize the likelihood of a preex isting association that would affect the rate at which the participants learn th ese new associations. Pairedassociate learning tasks have also been used to examine individual differences in performance (e.g., Wang, 1983), including th e effects of alcohol on performance (Yohman and Parsons, 1985) and the effect of sensation seeking on performance (Pullis, 1980). Paired-associate learning tasks are used to assess memory and learning abilities for both research and clinical purposes. Pa rticularly, they are included in assessment scales to evaluate an individuals abilities compared to a normative sample (e.g. the Weschler Memory Scale III, WMS-III; We schler, 1997, the Childrens Memory Scale, CMS; Cohen, 1997 and the Test of Memory and Learning, TOMAL; Reynolds, 1994). The paired-associate learning task for this study followed the format and administration instructions given in the TO MAL, which is a memory measure used for children ages 519. The instructions for this task are as follows:
19 Listen carefully, Im going to say words tw o at a time. When I finish, I will tell you one of the words, and I want you to tell me which word goes with it. After giving this instruction, the administrator reads the first set of word pairs (Trial 1) at a rate of one wo rd per second with a two second pause between each pair. At the end of the list, th e administrator again pauses for two seconds before reading each word of the recall list and pausing for a res ponse. If the participant responds correctly then the administrator says, Right. If the participant answers incorrectly, than the administrator says, No, the word is _____. After the first list is completed, this pro cess is repeated two more times, for a total of three trials. Before trials 2 and 3, the ad ministrator says Listen carefully while I read the list again but in a different order. If the examinee correctly r ecalls all of the words within a trial, then the administrator di scontinues testing and gives credit for all remaining trials. The score, which is the number of pairs correctly re called at the end of all three trials, as well as th e slope of learning, were used as the dependent variables in this study. After the last trial of the cued-recall task, participants were asked to write down as many of the word pairs as they could fr om memory (free recall). Finally, after participants completed the remaining questionn aires they were asked to finish a delayed free recall task and a delayed reca ll task, with the same words from the three initial trials. These free recall tasks served as another pr obe of overall acquisition of information. Words included in this task fall into th ree categories: alcohol words, expectancy words and control words. The alcohol words, such as beer, wine and booze, were selected from childrens literature about alcohol in order to ensure th at children in fifth
20 grade would be able to read and understa nd the definition of each word. Expectancy words, such as happy, cool, and excited, were selected from previous research about childrens alcohol expectancies (Dunn & Goldman, 1996, 1998, 2000). These studies found that children in fifth grade were likely to hold these positive and arousing expectancies of alcohol. Fi nally, both the noun and adjectiv e control words were chosen using the MRC Psycholinguistics Database (Wilson, 1988). The words found using this database were matched to the alcohol and e xpectancy words based on word type (noun or adjective), number of syllables and the writte n word's frequency of occurrence as given in the norms of Kucera and Francis (1967) as cited by Wilson (1988). They were also chosen because they appear unrelated to the alcohol and expectancy words. All of these words were also within fifth grade readi ng level, as determined through the use of readability analysis software published by GAMCO education materials (Williams, 1994). These three types of words were arra nged into alcohol-expectancy, controlexpectancy and controlcontrol pairs, with a total of se ven pairs in each group. The final task that each participant completed consiste d of fourteen pairs. One group had fourteen pairs which included seven alcohol-expectan cy pairs and seven control-control pairs while the other group included seven control-expectancy pairs and seven control-control pairs. These two groups were compared to one another in the final analyses. This two-group design was chosen in order to control for several competing factors that could influence the possible conclu sions made about the results of the study. The mean scores of the participants in th e alcohol-expectancy gr oup were compared to the scores of participants in the control-expectancy group in order to demonstrate that
21 differences in learning were due to the pres ence of alcohol-expectancy pairs rather than simply due to the presence of expectancy wo rds on their own. The control-control pairs were used to determine whether the two groups are equivalent in the participants level of ability to complete a pair ed-associate learning task. Alcohol Expectancy Questi onnaire Adolescent Form : Scale 2 (AEQ-A2). Participants positive social expectancies for alcohol were assessed using 17 items from the original 90 item AEQ-A (Christi ansen, Goldman, & Brown, 1985). Both the adolescent and adult forms of the AEQ were developed to measure the degree to which individuals expect alcohol to cause different general and specific effects. The AEQ-A was derived both from age-appr opriate items included in the original AEQ, as well as interviews of adolescents between ages 12 to 19 years. Factor analysis of the AEQ-A revealed seven expectancy factors: global positive changes, changes in social behavior, improved cognitive and motor abilities, sexual enhancement, cognitive and motor impairment, increased arousal and relaxati on and tension reduc tion. Scale 2, which assessed expected changes in so cial behavior, had an internal consistency of .78 and testretest reliability of .56. This subscale has been found to have the strongest correlation with measures of alcohol consumption and wa s chosen for this study to limit the amount of time that is required of participants. Additionally, it was hypothesized that positive social expectancies would be more appeali ng to children high in sensation seeking and thus more easily learned than other alcohol expectancies (i.e. improved cognitive and motor abilities, relaxation and tension reduction, etc.). While this measure was originally create d for use with children from ages 12 to 19 years, it is currently the best available measure for examining the alcohol expectancies
22 of young people. Other child measures that ha ve been used in previous research were designed for specific types of da ta analysis and are not approp riate for the current study. There is no indication that the items included in this portion of the AEQ-A are inappropriate for the age group in the current sample. Sensation Seeking Scale for Children (SSSC). This scale was used to measure each part icipants level of sensation seeking. Developed through modification of the Sensat ion Seeking Scale (SSS; Zuckerman, Kolin, Price & Zoob, 1964), authors of the SSSC selected items from the SSS that were relevant to children between the ages of 7 and 12 y ears old (Russo, Lahey, Ch rist & Frick, 1991). These items were further refined (Russo et. al., 1994) when the authors added more child relevant items and deleted items that had poor internal consistency in a child sample. Also included in this revision were a set of appropriately modified items regarding substance use and sexual activity. The scale c onsists of 26 forced-choice items that form three factors: Thrill and Adventure Seeki ng (TAS), Drug and Alcohol Attitudes (DAA) and Social Disinhibition (SD). The corrected split-half reliability estimate for the SSSC was r (828) = .85 and the coefficient alpha was .83. Demographics and Drinking Questionnaire (DDQ). This measure was used to collect the pa rticipants age, gende r and ethnicity in order to describe the sample. In additi on, drinking was assessed using the following items (Dunn & Goldman, 1998): How often do you drink alcohol? and How much did you have the last few times you drank alcohol? This data was used to examine the effects that drinking experience may have on participants performance on the paired associate learning task. This measure also included items regarding locations where the
23 participant drank, whether or not he or she had parental perm ission to drink, and questions about how much and how often the participant expects to drink as an adult. This information helps to better understa nd the context in which childhood drinking occurs. Floridas Comprehensive Assessment Test (FCAT) The primary purpose of the FCAT in Fl orida schools is to assess student achievement of the high-order cogn itive skills represented by the Sunshine State Standards (SSS) in Reading, Writing, Mathema tics, and Science. Development of the SSS was begun by the Florida Department of Education staff in 1994 (Florida Department of Education, 2003). The FCAT is reported using two methods: the FCAT SSS Tests and the FCAT Norm-Referenced Tests. Both of these records were obtained from schools as measures of each childs le vel of achievement and used in order to account for differential learning ability on the paired-associate learning task. As the participants in this study were in the fifth grade, the FCAT scores obtained for each child were from their performance on the FCAT in fourth grade. The version of the FCAT given to students in fourth grade evaluate d his or her abilities in reading, writing and mathematics. Upon receipt of this information, the scores were matched to students data within the study and any identifying information associated with FCAT reports was destroyed. Procedure Students were informed of the research project at their after-school program and were given written information and informed consent forms to bring home for their parents to sign. Students whose parents comp leted the consent forms were then randomly
24 assigned to the alcohol-expectancy group or control-expectancy group. Each participant in the study was tested indivi dually at their after-school program location. At the beginning of the session, the administrator explained the informed consent to the participant, as well as a brie f outline of what the study enta ils. Then the administrator completed the PALT with the participant (both cued-recall and free-reca ll sections). The participant was then asked to complete a ll of the questionnaire s in order (AEQ-A2, SSSC, Demographics/Drinking Information). Af ter the measures were completed, each participant ended with a dela yed free recall task and a delayed recall task similar to earlier cued recall trials. Fi nally, all participants were th anked for their participation. This procedure took between 20-40 minutes for each child to complete. Participants were compensated for their time with entry into a drawing to win a gift certificate (for returning their permission slips) and a small toy after their assessment was completed.
25 RESULTS Descriptive Statistics for the Dependent Measures Three types of word pairs were used during the paired-associate learning task: alcohol-expectancy pairs, contro l-expectancy pairs and controlcontrol pairs. Given that these specific word pairs were designed for the purpose of this study, it was necessary to compare participants performance on the differe nt types of pairs in order to determine if they were functionally equivalent to one another. Table 1 provides descriptive information about the three types of pairs.
26 Table 1 Descriptive statistics of the dependent measures Trial N Min Max Mean SD Skewness Kurtosis Alcohol-Expectancy Cued Recall Trial 1 47 0 5 1.11 1.15 0.96 1.18 Cued Recall Trial 2 47 0 6 1.70 1.30 0.90 1.26 Cued Recall Trial 3 47 0 6 2.66 1.34 0.27 -0.37 Delayed Cued Recall 46 0 7 2.65 1.68 0.49 -0.05 Free Recall 46 0 6 1.67 1.43 0.66 0.22 Delayed Free Recall 46 0 5 1.50 1.43 0.77 -0.20 Control-Expectancy Cued Recall Trial 1 47 0 6 1.26 1.52 1.51 1.78 Cued Recall Trial 2 47 0 7 2.79 1.74 0.83 0.44 Cued Recall Trial 3 47 0 7 3.83 1.95 0.21 -0.89 Delayed Cued Recall 47 0 7 4.34 1.96 -0.32 -0.52 Free Recall 47 0 6 3.13 1.45 -0.10 -0.35 Delayed Free Recall 47 0 7 3.11 1.66 0.21 -0.40 Control-Control Cued Recall Trial 1 94 0 7 1.60 1.42 0.86 0.95 Cued Recall Trial 2 94 0 7 3.86 1.92 -0.15 -0.71 Cued Recall Trial 3 94 1 7 5.05 1.60 -0.63 -0.22 Free Recall 93 0 7 5.25 1.61 -0.83 0.26 Delayed Free Recall 93 0 7 3.53 1.46 -0.12 -0.67 Delayed Cued Recall 93 0 7 3.48 1.45 0.02 -0.07 An initial t-test was done to determine if subjects performed differently on the alcohol-expectancy pairs versus the control-expectancy pairs (see Table 2). In the first cued-recall trial, there were no significant di fferences between participants performance on the two pair types (t=-.469, n.s .). However, on all subseq uent trials, including cued recall, free recall and delayed trials, participants tested with alcohol-expectancy pairs performed significantly less well than did par ticipants tested with control-expectancy words (p < .002). A t-test was also done to ensure that the randomized groups did not
differ in their performance on control-control word pairs and no significant differences were found (see Table 3). Table 2 T-Test of differences in means between alcohol-expectancy and control-expectancy words Alcohol-Expectancy Control-Expectancy M SD M SD t p Cued Recall Trial 1 1.11 1.147 1.26 1.525 -.535 .594 Cued Recall Trial 2 1.70 1.301 2.79 1.744 -3.420 .001 Cued Recall Trial 3 2.66 1.340 3.83 1.948 -3.393 .001 Delayed Cued Recall 2.65 1.676 4.34 1.959 -4.461 .000 Free Recall 1.67 1.431 3.13 1.454 -5.004 .000 Delayed Free Recall 1.50 1.426 3.11 1.658 -5.013 .000 Table 3 T-Test of differences in means between control-control words in the two, randomly assigned group Group 1 Group 2 M SD M SD t p Cued Recall Trial 1 1.49 1.231 1.70 1.600 -.722 .472 Cued Recall Trial 2 3.96 1.922 3.77 1.925 .483 .631 Cued Recall Trial 3 5.00 1.757 5.11 1.448 -.320 -.106 Delayed Cued Recall 5.30 1.590 5.19 1.637 .337 .737 Free Recall 3.48 1.441 3.57 1.485 -.317 .752 Delayed Free Recall 3.63 1.597 3.34 1.290 .964 .337 Due to the limited range of scores on alcohol-expectancy word pairs, the data was also recoded to reflect whether or not the participant responded with a category-correct response. For example, if the correct pair was wine-happy and the participant gave the response cool, then when using the category-correct scoring system, the participant would score correctly because their response was one of the seven possible expectancy 27
28 word responses. Likewise, if the correct pair was barbeque-muddy and the participant gave the response careful, then that person also would have scored correctly because their response was one of the choices in the non-expectancy category. However, words that were not included on any of the lists would not earn a point, re gardless of whether they included alcohol or expectancy content. While this scoring system provided additional informa tion about the learning and retention of alcohol expectancy information, the control groups in this study did not provide a matched sample of words that form a category of their own. Due to this, any analyses using the category-correct scoring system were purely expl oratory in nature. The descriptive statistics fo r all trials using the alcohol -expectancy words and scored using the category-correct scoring system are listed in Table 4. Table 4 Descriptive statistics of the Alcohol-Expectancy word pai rs scored using the categorycorrect scoring system. N Min Max Mean SD Skewness Kurtosis Cued Recall Trial 1 47 0 7 3.06 2.12 .170 -.893 Cued Recall Trial 2 47 0 7 4.23 2.07 -.467 -.560 Cued Recall Trial 3 47 2 7 5.36 1.65 -.735 -.619 Delayed Cued Recall 46 0 7 4.93 1.88 -.553 -.654 Free Recall 46 0 6 2.52 1.62 .405 -.718 Delayed Free Recall 46 0 6 2.22 1.59 .562 -.595 Descriptive Statistics for the Independent Measures For this study, the independent variables we re ones that were potentially relevant to the development and learning of alcohol expectancies in children. These included: alcohol expectancies related to social behavior as meas ured by the AEQ-A2, all three
29 subscales from the SSSC, and drinking inform ation, including freque ncy and quantity of current drinking and intentions to drink in the future. Table 5 Descriptive statistics of the independent measures N Min Max Mean SD Skewness Kurtosis AEQ-A2 94 0 9 1.98 1.81 1.03 1.39 TAS 94 0 12 7.50 3.19 -0.62 -0.61 DAA 94 0 5 0.36 0.75 3.43 16.59 SD 94 0 7 2.27 1.79 0.74 -0.09 Total SSSC 94 1 21 10.13 4.55 0.02 -0.48 Drinking Freq 94 0 5 0.45 1.03 3.14 10.65 Drinking Quant. 94 0 2 0.29 0.50 1.47 1.24 Future Freq. 94 0 7 1.07 1.42 1.79 3.77 Future Quant. 94 0 5 1.09 1.16 0.72 -0.21 Transformation of Non-Norma lly Distributed Variables Several of the independent and dependen t variables demonstrated a non-normal distribution, including: AE QA-2, DAA, drinking frequenc y, drinking quantity, future drinking frequency, cued recall trial 1 for alc ohol-expectancy words and cued recall trials 1 and 2 for control-expectancy words. All of these variables were transformed by taking the logarithm (logt ), square root ( t 1/2 ) or reciprocal square root ( t -1/2 ) of the variable. These transformations served to improve the skewness and kurtosis for the majority of these variables.
30 Table 6 Descriptive statistics of the transformed variables N Min Max Mean SD Skewness Kurtosis t 1/2 AEQ-A2 94 0 3.00 1.17 0.79 -0.23 -0.92 t -1/2 DAA 94 -1 -0.41 -0.91 0.15 1.32 0.49 t -1/2 Drinking Freq 94 -1 -0.41 -0.91 0.17 1.61 1.31 t 1/2 Drinking Quant. 94 0 1.41 0.27 0.46 1.14 -0.56 t 1/2 Future Freq. 94 0 2.65 0.73 0.74 0.45 -0.94 log t Cued Recall Trial 1 94 0 0.85 0.27 0.25 0.33 -0.98 t 1/2 Cued Recall Trial 2 94 0 2.65 1.36 0.63 -0.52 0.37 Correlational Analyses As shown in Table 7, several of the in dependent measures were significantly correlated with one another. Specifically, the AEQ-A2 was moderately correlated with two of the SSSC subscales, as well as curre nt and estimated future drinking. All subscales of the SSSC were also correlated w ith current and estimated future drinking. Table 7 Zero-order correlations between independent measures 1 2 3 4 5 6 7 8 1. AEQ-A2 ( t 1/2 ) -2. TAS .03 -3. DAA ( t -1/2 ) .21* .25* -4. SD .32** .41** .42** -5. Total SSSC .18 .90** .49** .74** -6. Freq. ( t -1/2 ) .25 .29* .16 .30** .35** -7. Quant. ( t 1/2 ) .30** .34** .39** .42** .47** .89** -8. Future Freq. ( t 1/2 ) .31** .16 .06 .33** .24* .47** .49** -9. Future Quant. .32** .24* .16 .37** .34** .45** .51** .83** Note. p < .05; ** p < .01 Correlational analyses were also perfor med to explore the relationship between the independent variables and performance on the alcohol-expectancy word pairs. As
31 shown in Table 8, the majority of the indepe ndent variables did not demonstrate a clear relationship to scores on the alcohol-expectancy pairs of the PALT. Several of these, however, showed some relationship to task pe rformance when using a liberal alpha level of .05. These were selected for further investigation: DAA, Dr inking Frequency and Drinking Quantity. Table 8 Zero-Order correlations between independe nt variables and performance on the PALT for Alcohol-Expectancy word pairs CR 1 CR 2 CR 3 DCR FR DFR 1. AEQ-A2 ( t 1/2 ) -.10 -.16 -.06 -.20 -.16 -.17 2. TAS .15 .01 .03 .04 .12 .14 3. DAA ( t -1/2 ) -.10 -.29* -.02 -.02 -.07 .00 4. SD -.05 -.25 -.09 -.12 -.06 .03 5. Total SSSC .08 -.11 -.02 -.02 .06 .11 6. Freq. ( t -1/2 ) .14 -.09 -.01 -.35* -.09 -.24 7. Quant. ( t 1/2 ) .09 -.18 -.03 -.35* -.06 -.20 8. Future Freq. ( t 1/2 ) .03 -.14 -.04 -.16 .07 .00 9. Future Quant. .10 -.10 -.02 -.05 .10 .02 Note. p < .05 Further correlational analys es were conducted with the alcohol-expectancy pairs scored using the category-corr ect scoring system. Using this information, two more variables of interest were pu lled out for further analysis du e to a significant correlation with task performance. These variab les were AEQ-A2 and TAS (see Table 9).
32 Table 9 Zero-order correlations between independe nt variables and performance on the PALT for Alcohol-Expectancy word pairs scored us ing the category-correct scoring system CR 1 CR 2 CR 3 DCR FR DFR 1. AEQ-A2 ( t 1/2 ) .03 -.14 -.25 -.31* -.16 -.03 2. TAS .30* .19 .27 .14 .15 .16 3. DAA ( t -1/2 ) .01 -.26 -.05 -.10 -.01 .10 4. SD .00 -.10 -.08 -.15 -.10 -.03 5. Total SSSC .22 .07 .16 .04 .07 .11 6. Freq. ( t -1/2 ) .32* .07 .10 -.22 -.14 -.16 7. Quant. ( t 1/2 ) .23 -.08 .03 -.31* -.12 -.12 8. Future Freq. ( t 1/2 ) .18 -.14 -.07 -.23 -.07 .02 9. Future Quant. .27 -.06 .05 -.15 -.03 -.00 Note. p < .05 Finally, correlational analyses were r un in order to examine the relationship between the independent variables of interest and the rate of learning across trials. For these analyses, the dependent variables included the difference scores between each of the three trials, as well as the slope of the best fit line representing learning across all three trials. However, there were no si gnificant correlations between any of these variables of change and th e independent variables. Gender differences on independent variables Previous research has demonstrated that sensation seeking was more evident in males than in females (Russo, Lahey, Stuke s & Christ, 1993) and that the relationship between sensation seeking and alcohol expect ancies may be stronger in men than in women (McCarthy, Kroll & Smith, 2001). Thus it was important to look at possible differences due to gender in the current sample. A series of t-tests revealed no significant
33 gender differences as measured by the AEQ-A2, social disinhibition, or current or future drinking at this age. Males did, however, sc ore significantly higher th an females on thrill and adventure seeking and total sensa tion seeking (TAS: t=3.73, p<.001; SSSC: 2.70, p<.001). Further analyses usi ng these variables we re conducted separately for males and females. Analyses of the Relationship between the I ndependent Variables and Task Performance If the primary hypotheses were correct, participants who scored higher on measures of sensation seeking should have ha d a higher rate of ac quisition of alcoholexpectancy word pairs than participants w ho scored lower on measures of sensation seeking. There should have been no differen ces, however, in participants performance on control-control word pairs due to their level of sensation seeking. A series of linear regressions were performed in order to exam ine the relationship between the independent variables and performance on the alcohol -expectancy pairs of the PALT. The independent variables examined in these analyses included: AEQ-A2, TAS, DAA, Drinking Frequency and Drinking Quantity. De pendent variables incl uded all three cued recall trials, delayed cued reca ll, free recall and delayed fr ee recall. Each independent and dependent variable was entered separate ly in order to examine the relationship between all combinations. Additionally, an alyses examining the relationship between TAS and performance on the PALT were performed separately for male and female participants. Analyses were run using both general scoring and cat egory-correct scoring methods for these variables. Out of sevent y-two possible regressions, six were found to be significant at an alpha level of .05. When using a modified Bonferroni procedure in
34 order to control for Type I e rror, these six findings no longer reached levels of statistical significance. A separate set of analyses were run as hi erarchical regressions in order to control for the participants performance on the first cued recall trial. In these regressions, performance on the first cued recall trial was en tered in the first step and the independent variable of interest was entered in the sec ond step. Again, analyses were run with both standard and category-correct scoring, and male and female participants were run separately for analyses involving TAS. Of th ese sixty analyses, eight were significant at an alpha level of .05. Again, however, when an adjustment was made using the modified Bonferroni procedure these findings were no longer significant. In order to further examine the data for potential trends, each independent variable was subject to a median split. All ttests run using these median splits were also not significant, however these analyses provided further info rmation about trends in the data that were non-significant but consiste nt. For example, although non-significant, participants who scored lower on the AEQ-A2 performed consistently better on all trials of the PALT than participants who scored higher on the AEQ-A2. This was true across all three types of word pairs: alcohol-expectancy, control-expectancy and control-control word pairs. The same was true of the DAA subscale of the SSSC, which in many ways is similar to the AEQ-A2 in terms of content area. These results indica te that higher social alcohol expectancies may be associated with a dampening of overall learning. In order to verify this pattern in the future, a larger samp le would be necessary to detect so small of an effect size.
35 Similar to the trend found in median split analyses using the AEQ-A2 and the DAA, a consistent trend was also found in analyses based on the TAS. Among female participants, those who scored higher on the TAS actually performed better on PALT trials than did participants who scored lowe r on the TAS. This result was not consistent with the proposed hypotheses, however, because these participants scored higher on all three pair types rather than only on alc ohol-expectancy pairs. Additionally, these analyses were performed with a much sma ller sample size due to the gender split. Among male participants, this pattern was not found. No cons istent patterns were found when similar analyses were conducted using median splits of drinking frequency and quantity.
36 DISCUSSION The purpose of the current investigation was to explor e the relationship between the personality factor sens ation seeking and the acquis ition of alcohol expectancy information. Specifically, it was hypothesi zed that children who scored higher on measures of sensation seeking would learn alcohol-expectancy information presented in the form of word associations better and mo re quickly than children who scored lower on sensation seeking scales. It was also hypothesized that the same would not be true for control pairs. The afore-described resu lts, however, did not support these stated hypotheses as they were meas ured in the present study. Contrary to the hypotheses, there were no significant differences found on any PALT trials between people who scored higher on measures of sens ation seeking versus people who scored lower. Among female par ticipants, a trend was found in which high sensation seekers performed consistently be tter on all PALT trials than low sensation seekers. This difference, however, was not st atistically significant and crossed all word types, which is contradictory to the hypothesi zed results. In males, no consistent or significant patterns were found. In relation to other variables, children who scored lower on a measure of social alcohol expectancies and drug and alcohol attitudes were more likely to remember word pairs than childre n who scored higher on measures of both variables. This indicates that these variable s, which are quite sim ilar to one another in content, may be related to the way that childr en learned different t ypes of word pairs. However, these relationships are not significan t and may not provide specific information
37 concerning the acquisition of alcohol-exp ectancy information above and beyond the acquisition of information in general. Initial analyses of the pair ed associate learning task revealed that participants did not have equivalent scores across word pair lists. Specifically, participants performed worse onthe alcohol-expectancy word pairs th an on either the control-expectancy or control-control pairs. Although the current da ta does not contain information to explain these differences, one potential explanation is the effect that categories may have on the participants ability to learn distinct word pairs. Having two separate categories of words with similar content to one another could have made the alcohol-expectancy words more difficult to learn than control pairs because each word in the alcohol category or expectancy category could be easily confus ed with other words on those lists. Additionally, the task created for this study did not have comparable control groups in terms of comparing a category of nouns to a ca tegory of adjectives. Future research is needed to clarify the effects of categorization on th e results of participants performance in this study. Another concern with the PALT used in this study is th at due to the low scores on the alcohol-expectancy pairs, there was limited variability in the participants scores on these words. This floor effect may have potentially dampened any differences that could have been found between partic ipants who scored higher a nd lower on sensation seeking scales. Contrary to the hypotheses, it is possi ble that the expected results are not attainable in a sample of fifth grade child ren because they may rely on the development of positive alcohol expectancies. The range of alcohol expectancies in this sample was
38 quite low (9 out of a potential 17) and positively skewed, indicating that children in this study had relatively low alcohol expectancies. The results of this study can serve to inform future research in this area by pointing out necessary areas of improvement. Pilot testing several types of word pairs for the PALT will provide essential information to wards perfecting this task for the purposes of answering the questions th at this study began explori ng. An improved task would include pairs in each group that are better matched in terms of pair difficulty in order to provide a large enough range of scores on all pa ir types to compare groups. Additionally, more information may be provided by in cluding both positive and negative alcoholexpectancy pairs. This may help to dem onstrate how the learning of negative alcohol expectancy information is affected by sensat ion seeking and other variables of interest. Some limitations of this study also may be due to the sample that was used. The data was collected in after-school programs rather than from a more representative sample of children attending public school. Although it is unclear as to how this may have influenced the variables of interest, it is likely that participants from this population were more likely to have one or both pare nts employed outside the home. Additionally, although a power analysis confirmed that a sa mple of 94 participants is large enough to detect a medium effect size it is possible th at a larger sample would have provided the additional power to detect sma ller, more subtle differences. There is a growing interest in the literature regarding the ways in which alcohol expectancies form during childhood and con tinue to develop across adolescence and adulthood. This study sought to examine how se nsation seeking, a key personality factor associated with alcohol expectancies in adolescence and adulthood, may influence the
39 acquisition and retenti on of alcohol expectancy inform ation during the time at which childrens views and beliefs about alcohol begin to shift. Due to methodological limitations, no conclusions can be made about the potential pathways in which sensation seeking may affect the learning of alcohol e xpectancies based on the findings reported in this study. It does, however, offer the sc ientific community important information regarding ways in which this question can be thought about and examined in future research aimed at exploring this key proce ss. Ultimately, a better understanding of the development of alcohol expectancies in chil dren may provide society with the tools to intervene prior to the developm ent of problematic drinking pr oblems that are associated with high positive alcohol expectancies later in life.
40 REFERENCE LIST Austin, E.W., & Meili, H.K. (1994). Effect s of interpretations of televised alcohol portrayals on childrens alcohol beliefs. Journal of Broadcasting and Electronic Media, 38, 417-435. Bandura, A. (1977). Social Learning Theory New Jersey: Prentice. Brody, G.H., Flor, D. L., Hollett-Wright, N ., McCoy, J. K., & Donovan, J. (1999). Parent-child relationships, child temperam ent profiles and children's alcohol use norms. Journal of Studies on Alcohol, Supp. 13 45-51. Brown, S.A., Christiansen, B.A. & Goldman, M.S. (1987). The alcohol expectancy questionnaire: An instrument for the a ssessment of adolescent and adult alcohol attribution. Journal of Studies on Alcohol, 48 483-491. Brown, S.A., Creamer, V.A., & Stetson, B.A. (1987). Adolescent alcohol expectancies in relation to personal and pa rental drinking patterns. Journal of Abnormal Psychology, 96, 117-121. Brown, S.A., Goldman, M.S., & Christiansen, B. A. (1985). Do alcohol expectancies mediated drinking patterns of adults? Journal of Consulting and Clinical Psychology, 53, 512-519. Brown, S. A., Tapert, S. F., Granholm, E., & Delis, D. C. (2000). Neurocognitive functioning of adolescents: Effects of protracted alcohol use. Alcoholism: Clinical & Experimental Research, 24 164-171. Christiansen, B.A., Goldman, M.S. & Brow n, S.A. (1985). Alcohol expectancy questionnaire: Adolescent form. Asse ssing Alcohol Problems, A Guide For Clinicians and Researchers; NIAAA Tr eatment Handbook Series 4. Rockville, MD: National Institute of Alcohol Abuse and Alcoholism. Christiansen, B.A., Smith, G.T., Roehling, P.V., & Goldman, M.A. (1989). Using alcohol expectancies to predict adolescen t drinking behavior after one year. Journal of Consulting and Clinical Psychology, 50 336-344 Cohen, M.J. (1997). Childrens Memory Scale San Antonio, Texas: The Psychological Corporation. Cruz, I.Y. & Dunn, M.E. (2003). Lowering ri sk for early alcohol use by challenging alcohol expectancies in el ementary school children. Journal of Consulting & Clinical Psychology, 71, 493-503.
41 Darkes, J., & Goldman, M.S. (1993). Exp ectancy challenge and drinking reduction: Experimental evidence for a mediational process. Journal of Consulting and Clinical Psychology, 61, 344-353. Dunn, M.E., & Goldman, M.S. (1996). Empirical modeling of an alcohol expectancy memory network in elementary school ch ildren as a function of grade. Experimental and Clinical Psychopharmacology, 4, 209-217. Dunn, M.E., & Goldman, M.S. (1998). Age and drinking-related differences in the memory organization of al cohol expectancies in 3 rd 6 th 9 th and 12 th grade children. Journal of Consulting an d Clinical Psychology, 66 579-585. Dunn, M.E., & Goldman, M.S. (2000). Drinking -related differences in expectancies of children assessed as first associates. Alcoholism: Clinical and Experimental Research, 24 1639-1646. Dunn, M.E., & Yniguez, R.M. (1999). Experime ntal demonstration of the influence of alcohol advertising on the act ivation of alcohol expectancies in memory among fourth and fifth-grade children Experimental and Clinical Psychopharmacology, 7, 473-483. Finn, P. R., Sharkansky, E. J., Brandt, K. M ., & Turcotte, N. (2000). The effects of familial risk, personality, and expectancies on alcohol use and abuse. Journal of Abnormal Psychology, 109 122-133. Florida Department of Edu cation. (2003). FCAT Meas uring Student Learning. Retrieved on October 8, 2003, from: http://www.firn.edu/ doe/sas/fcat/fcatpub1.htm Goldman, M. S. (1989). Alcohol expectancies as cognitive-behavioral psychology: Theory and practice. Loberg, T. & Miller, W. R. et. al. (Eds) Addictive behaviors: Prevention and early intervention Swets & Zeitlinger, Lisse, Netherlands, 11-30. Goldman, M. S. (1999). Expectancy ope ration: Cognitive-neural models and architectures. In I. Kirsch (Ed.), How Expectancies Shape Experience Washington, DC: APA Books. Graber, J.A. & Peterson, A.C. (1991). C ognitive changes at adolescence: Biological perspectives. In K.R. Gibson & A.C. Petersen (Eds.), Brain maturation and cognitive development: Comparati ve and cross-cultural perspectives Hawthorne, NY: Aldine de Gruyter, 253279.
42 Grant, B.F. & Dawson, D.A. (1997). Age at ons et of alcohol use and its association with DSM-IV alcohol abuse and dependence: Results from the National longitudinal alcohol epidemiologic survey. Journal of Substance Abuse, 9 103-110. Henderson, M.J., Goldman, M.S., Coovert, M.D., & Carnevalla, N. (1994). Covariance structure models of expectancy. Journal of Studies on Alcohol, 55 315-326. Hingson, R.W., Heeren, T., Jamanka, A., & Howland, J. (2000). Age of Drinking Onset and Unintentional Injury Involvement After Drinking, Journal of the American Medical Association, 284 1527-1533. Johnson, H.L., & Johnson, P.B. (1995). Childre ns alcohol-related cognitions: Positive versus negative alcohol effects. Journal of Alcohol and Drug Education, 40 1-12 Johnson, P.B., & Johnson, H.L. (1996). Children s beliefs about social consequences of drinking and refusing to drink alcohol. Journal of Alcohol and Drug Education, 41, 34-43. Kaiser Family Foundation. (2002). Survey snapshot: Substance use and risky sexual behavior: Attitudes and practices among adolescents and young adults Menlo Park CA: The Henry J. Kaiser Foundation.. Katz, E. C., Fromme, K., D'Amico, E. J. ( 2000). Effects of outcome expectancies and personality on young adults' illicit drug use, heavy drinking, and risky sexual behavior. Cognitive Therapy & Research, 24 1-22. Klerbar, J.E. & Bardo, M.T. (1999). Individual differences in novelty seeking on the playground maze predict amphetamine conditioned place preference. Pharmacology, Biochemistry & Behavior, 63, 131-136. Kraus, D., Smith, G.T., & Ratner, H.H. (1994) Modifying alcohol-related expectancies in grade-school children. Journal of Studies on Alcohol, 55 535-542. Leadership for a Drug Free America. (2000). Making the link: Underage drinking and mental health. Washington D.C.: Leadership for a Drug Free America. Retrieved on October 7, 2003 from: http://www.alcoholfreechild ren.org/gs/audiences/youth.cfm Leadership to Keep Children Alcohol Free. (2001). Keep Kids Alcohol Free: Strategies for Action. Washington D.C.: NIH Publications. MacCorquodale, K.M., & Meehl, P.E. (1954) Preliminary suggestions as to a formulation of expectancy theory. Psychological Review, 60 53-60.
43 Martin, C. A., Kelly, T. H., Rayens, M. K., Brog li, B. R., Brenzel, A., Smith, W. J., et. al. (2002). Sensation seeking, puberty and nicotine alcohol and marijuana use in adolescence. Journal of the American Acade my of Child & Adolescent Psychiatry, 41 1495-1502. McCarthy, D. M., Kroll, L. S., Smith, G. T. (2001). Integrating disinhibition and learning risk for alcohol use. Experimental & Clinical Psychopharmacology, 9 389-398. Miller, P.A., Smith, G.T., & Goldman, M.S. (1990). Emergence of alcohol expectancies in childhood: A possible critical period. Journal of Studies on Alcohol, 51, 343349. National Highway Traffic Safety Administration. (2001). 2000 Youth Fatal Crash and Alcohol Facts. Washington, DC: US Department of Transportation. Retrieved on October 7, 2003 from the National Highway Traffic Safety Administration: http://www.nhtsa.dot.gov/people/in jury/alcohol/2002YFCAF/index.html Noll, R.B., Zucker, R.A., & Greenberg, G.S. (1990). Identification of alcohol by smell among preschoolers: Evidence for early soci alization about dri nking occurring in the home. Child Development, 61, 1520-1527. Pullis, C.L. (1979). Differential effects of sensation seeking trait on memorization under varying levels of task complexity Unpublished doctoral dissertation, Columbia University, New York. Query, L.R., Rosenberg, H. & Tisak, M.S. ( 1998). The assessment of young childrens expectancies of alcohol versus a control substance. Addiction, 93 1521-1529. Rather, B.C., Goldman, M.S., Roehrich, L., & Brannick, M. (1992). Empirical modeling of an alcohol expectancy memory network using multidimensional scaling. Journal of Abnormal Psychology, 101 174-183. Reynolds, C.R. & Bigler, E.D. (1994). Test of Memory and Learning Austin, TX: PRO-ED Inc. Rothbart, M.K., Derryberry, D., & Posner, M.I. (1994). A psychobiolog ical approach to the development of temperament. Found in Bates, J.E. & Wachs, Theodore D. (Eds.). Temperament: Individual differences at the interface of biology and behavior. APA science volumes. Washington, DC: American Psychological Association, 83-116. Rotter, J.B. (1954). Social Learning and Clinical Psychology. Englewood Cliffs, NJ: Prentice-Hall.
44 Rotter, J.B. (1981). The psychological situat ion in learning theory. In D. Magnussen (Ed.), Toward a psychology of situations: An interactional perspective Hillsdale, NJ: Lawerence Erlgaum Associates. Russo, M. F., Lahey, B. B., Christ, M. A., Frick, P. J. (1991). Preliminary development of a sensation seeking scale for children. Personality & Individual Differences, 12, 399-405. Russo, M. F., Lahey, B. B., Stokes, G. S., Ch rist, M. A. (1993). A Sensation Seeking Scale for Children: Further refinement and psychometric development. Journal of Psychopathology & Be havioral Assessment, 15 69-86. Shedler, J. & Block, J. (1990). Adoles cent drug use and psychological health: A longitudinal inquiry. American Psychologist, 45 612-630. Schuckit, M.A. (1994). A clinical model of genetic influences in alcohol dependence. Journal of Studies on Alcohol, 55 5-17. Spear, L.P. (2000). The adoles cent brain and age-related be havioral manifestations. Neuroscience & Biobehavioral Reviews, 24 417-463. Stacy, A.W., Newcomb, M.D. & Bentler, P.M. (1991). Social psychological influences on sensation seeking from adolescence to adulthood. Personality & Social Psychology Bulletin, 17 701-708. Steinberg, H. & Goldman, M.S. (2003). The role of individual differences in learning alcohol expectancy associations Unpublished doctoral dissertation, University of South Florida, Tampa, FL. The Weekly Reader. (1995). National Survey on Drugs and Alcohol. Middletown, CT: Field Publications. Tolman, E.G. (1932). Purposive behavior in animals and man New York: AppeltonCentury Crofts. Wang, A. Y. (1983). Individual differences in learning speed. Journal of Experimental Psychology: Learning, Memory, & Cognition, 9, 300-311. Webb, J.A., Baer, P.E., Francis, D.J. & Cai d, C.D. (1993). Relati onship among social and intrapersonal risk, alcohol expect ancies, and alcohol usage among early adolescents. Addictive Behaviors, 18 127-134. Webb, J.A., Baer, P.E. & McKelvey, R.S. (1995). Development of a risk profile for intentions to use alcohol among fifth a nd sixth graders. Journal of the American Academy of Child & Adolescent Psychiatry, 34 772-778.
45 Weschler, D. (1997). Wechsler Memory Scale III San Antonio, TX: The Psychological Corporation. Williams, G. (Ed.), Smith, B. (Programmer) & Tiffany, P. (Programmer). (1994). GAMCO Readability Analysis (Version 6.1.6) [Computer Software]. Big Springs, TX: Gamco Industries Inc. Wilson, M.D. (1988). The MRC Psycholi nguistic Database: M achine Readable Dictionary, Version 2. Behavioural Research Methods, Instruments and Computers, 20, 6-11. Retrieved October 7, 2003 from MRC Psycholinguistic Database: http://www.psy.uwa.edu.au/mrcdatabase/uwa_mrc.htm Wooten, B.T. (1995). Challenging alcohol expectancies: An application to adolescents. Unpublished Doctoral Dissertation, University of South Florida. Yohman, J.R. & Parsons, O.A. (1985). In tact verbal paired-associate learning in alcoholics. Journal of Clinical Psychology, 41 844-851. Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of arousal Hillsdale, NJ: Lawrence Erlbaum Associates. Zuckerman, M. (1994). Behavioral expressions and bios ocial bases of sensation seeking. New York: Cambridge University Press. Zuckerman, M., Ballenger, J.C., & Post, R. M. (1984). The neurobiology of some dimensions of personality. In Smythies, J.R. & Bradley, R.J. (eds), International review of neurobiology (pp. 391-436). Orlando, FL: Academic Press. Zuckerman, M., Kolin, E.A., Price, L., & Zoob, I. (1964). Development of a sensation seeking scale. Journal of Consulting and Clinical Psychology, 28 477-482. Zuckerman, M. & Neeb, M. (1980). Demogra phic influences in sensation seeking and expressions of sensation seeking in religion, smoking and driving habits. Personality and Individual Differences, 1 197-206.
47 Appendix A: Paired Associate Learning Task, Version A Paired Associate Learning Task Version A For all sections of this task, follow the PALT Instructions. Please follow the directions as closely as possible and write out the childs whole response. Cued Recall Directions : For Trials 1-3, record all responses verbatim in the response column. Score 1 point for each correct response and 0 points for each incorrect response. Trial 1 Cued Recall List A Trial 1 Response Alcohol Control Hill-Usual Barbeque (Muddy) Wine-Happy Liquor (Exciting) Mall-Soft Collar (Steady) Barbeque-Muddy Whiskey (Funny) Rum-Cool Bus (Careful) Collar-Steady Alcohol (Cheerful) Whiskey-Funny Rum (Cool) Beer-Friendly Mall (Soft) Liquor-Exciting Peg (Invisible) Bus-Careful Booze (Outgoing) Peg-Invisible Wine (Happy) Booze-Outgoing Pocket (Gentle) Pocket-Gentle Beer (Friendly) Alcohol-Cheerful Hill (Usual) Total Trial 1 Score Trial 2 Cued Recall List B Trial 2 Response Alcohol Control Pocket-Gentle Beer (Friendly) Booze-Outgoing Hill (Usual) Beer-Friendly Wine (Happy) Liquor-Exciting Rum (Cool) Collar-Steady Peg (Invisible) Whiskey-Funny Collar (Steady) Bus-Careful Alcohol (Cheerful) Wine-Happy Bus (Careful) Alcohol-Cheerful Whiskey (Funny) Hill-Usual Pocket (Gentle) Rum-Cool Mall (Soft) Peg-Invisible Barbeque (Muddy) Barbeque-Muddy Liquor (Exciting) Mall-Soft Booze (Outgoing) Total Trial 2 Score
48 Appendix A: Paired Associate Learning Task, Version A (Continued) Paired Associate Learning Task Version A Cued Recall Continued Trial 3 Cued Recall List C Trial 3 Response Alcohol Control Alcohol-Cheerful Peg (Invisible) Bus-Careful Booze (Outgoing) Hill-Usual Pocket (Gentle) Liquor-Exciting Whiskey (Funny) Pocket-Gentle Bus (Careful) Barbeque-Muddy Hill (Usual) Rum-Cool Wine (Happy) Booze-Outgoing Alcohol (Cheerful) Peg-Invisible Mall (Soft) Beer-Friendly Rum (Cool) Whiskey-Funny Barbeque (Muddy) Mall-Soft Beer (Friendly) Wine-Happy Collar (Steady) Collar-Steady Liquor (Exciting) Total Trial 3 Score Free Recall Directions : For Free Recall, record all responses verbatim in the response column. Score 1 point for each correct response and 0 points for each incorrect response. Free Recall Recall List (do not read) Response Alcohol Control Hill-Usual Wine-Happy Mall-Soft Barbeque-Muddy Rum-Cool Collar-Steady Whiskey-Funny Beer Friendly Liquor-Exciting Bus-Careful Peg-Invisible Booze-Outgoing Pocket-Gentle Alcohol-Cheerful Total Free Recall Score At this point, the participant mu st complete the AEQ-A, SSSC and DDQ.
49 Appendix A: Paired Associate Learning Task, Version A (Continued) Paired Associate Learning Task Version A Delayed Free Recall Directions : For Delayed Free Recall, record all responses verbatim. Score 1 point for each correct response and 0 points for each incorrect response. Recall List (do not read) Response Alcohol Control Hill-Usual Wine-Happy Mall-Soft Barbeque-Muddy Rum-Cool Collar-Steady Whiskey-Funny Beer Friendly Liquor-Exciting Bus-Careful Peg-Invisible Booze-Outgoing Pocket-Gentle Alcohol-Cheerful Total Delayed Free Recall Score Delayed Recall Directions : Read the first word of each pair and record all responses verbatim. Score 1 point for each correct response and 0 points for each incorrect response. Recall List (do not read) Response Alcohol Control Hill (Usual) Wine (Happy) Mall (Soft) Barbeque (Muddy) Rum (Cool) Collar (Steady) Whiskey (Funny) Beer (Friendly) Liquor (Exciting) Bus (Careful) Peg (Invisible) Booze (Outgoing) Pocket (Gentle) Alcohol (Cheerful) Total Delayed Recall Score At the end of this task, the par ticipant has comp leted the study.
50 Appendix B: Paired Associate Learning Task, Version B Paired Associate Learning Task Version B For all sections of this task, follow the PALT Instructions. Please follow the directions as closely as possible and write out the childs whole response. Cued Recall Directions : For Trials 1-3, record all responses verbatim in the response column. Score 1 point for each correct response and 0 points for each incorrect response. Trial 1 Cued Recall List A Trial 1 Response Alcohol Control Hill-Usual Barbeque (Muddy) Boat-Happy Iron (Exciting) Mall-Soft Collar (Steady) Barbeque-Muddy Basket (Funny) Grape-Cool Bus (Careful) Collar-Steady Antenna (Cheerful) Basket-Funny Grape (Cool) Keys Friendly Mall (Soft) Iron-Exciting Peg (Invisible) Bus-Careful Sock (Outgoing) Peg-Invisible Boat (Happy) Sock-Outgoing Pocket (Gentle) Pocket-Gentle Keys (Friendly) Antenna-Cheerful Hill (Usual) Total Trial 1 Score Trial 2 Cued Recall List B Trial 2 Response Alcohol Control Pocket-Gentle Keys (Friendly) Sock-Outgoing Hill (Usual) Keys-Friendly Boat (Happy) Iron-Exciting Grape (Cool) Collar-Steady Peg (Invisible) Basket-Funny Collar (Steady) Bus-Careful Antenna (Cheerful) Boat-Happy Bus (Careful) Antenna-Cheerful Basket (Funny) Hill-Usual Pocket (Gentle) Grape-Cool Mall (Soft) Peg-Invisible Barbeque (Muddy) Barbeque-Muddy Iron (Exciting) Mall-Soft Sock (Outgoing) Total Trial 2 Score
51 Appendix B: Paired Associate Learning Task, Version B (Continued) Paired Associate Learning Task Version B Cued Recall Continued Trial 3 Cued Recall List C Trial 3 Response Alcohol Control Antenna-Cheerful Peg (Invisible) Bus-Careful Sock (Outgoing) Hill-Usual Pocket (Gentle) Iron-Exciting Basket (Funny) Pocket-Gentle Bus (Careful) Barbeque-Muddy Hill (Usual) Grape-Cool Boat (Happy) Sock-Outgoing Antenna (Cheerful) Peg-Invisible Mall (Soft) Keys-Friendly Grape (Cool) Basket-Funny Barbeque (Muddy) Mall-Soft Keys (Friendly) Boat-Happy Collar (Steady) Collar-Steady Iron (Exciting) Total Trial 3 Score Free Recall Directions : For Free Recall, record all responses verbatim in the response column. Score 1 point for each correct response and 0 points for each incorrect response. Free Recall Recall List (do not read) Response Alcohol Control Hill-Usual Boat-Happy Mall-Soft Barbeque-Muddy Grape-Cool Collar-Steady Basket-Funny Keys Friendly Iron-Exciting Bus-Careful Peg-Invisible Sock-Outgoing Pocket-Gentle Antenna-Cheerful Total Free Recall Score At this point, the participant mu st complete the AEQ-A, SSSC and DDQ.
52 Appendix B: Paired Associate Learning Task, Version B (Continued) Paired Associate Learning Task Version B Delayed Free Recall Directions : For Delayed Free Recall, record all responses verbatim. Score 1 point for each correct response and 0 points for each incorrect response. Recall List (do not read) Response Alcohol Control Hill-Usual Boat-Happy Mall-Soft Barbeque-Muddy Grape-Cool Collar-Steady Basket-Funny Keys Friendly Iron-Exciting Bus-Careful Peg-Invisible Sock-Outgoing Pocket-Gentle Antenna-Cheerful Total Delayed Free Recall Score Delayed Recall Directions : Read the first word of each pair and record all responses verbatim. Score 1 point for each correct response and 0 points for each incorrect response. Recall List (do not read) Response Alcohol Control Hill (Usual) Boat (Happy) Mall (Soft) Barbeque (Muddy) Grape (Cool) Collar (Steady) Basket (Funny) Keys (Friendly) Iron (Exciting) Bus (Careful) Peg (Invisible) Sock (Outgoing) Pocket (Gentle) Antenna (Cheerful) Total Delayed Recall Score At the end of this task, the par ticipant has comp leted the study.
53 Appendix C: Alcohol Expectancy Questionnai re Adolescent Form : Scale 2 (Revised) AEQ-A2 Directions : These questions are about the effect s of alcohol. Read each sentence carefully and answer with your own feelings, thoughts, and beliefs about alcohol now We are interested in what you think about alcohol, and not wh at other people might think. If you think that the sentence is true, mostly true, or sometimes true than mark agree. If you think that the statement is false, or mostly false than mark disagree. Even if you have never tasted alcohol, you should answer each question in terms of how you think about alcohol It is important that you answer every question. Agree Disagree 1. People are harder to get along with after they have had a few drinks of alcohol. 2. Problems are caused by drinking alcohol. 3. People get a bad impression of those who drink alcohol. 4. Teenagers that want to ge t noticed drink alcohol. 5. Parties are not as much fun if people are drinking alcohol. 6. People feel more caring and giving after a few drinks of alcohol. 7. Drinking makes people more friendly. 8. Drinking alcohol is okay because it lets people join in with others who are having fun. 9. Sweet alcoholic drinks taste good. 10. Most alcoholic drinks taste good. 11. People act like better friends after a few drinks of alcohol. 12. Most alcohol tastes awful. 13. Having a few drinks of alcohol is a nice way to enjoy the holidays. 14. Its fun to watch others act silly when they are drinking alcohol. 15. Teenagers drink alcohol because th ey feel forced to do so by their peers. 16. Alcoholic drinks make parties more fun. 17. People get in better moods afte r a few drinks of alcohol.
54 Appendix D: Sensation Seeking Scale for Children Childs Interest and Preference Test (SSSC) Directions : Each of the items within this booklet contains two choices, A and B. Please circle the letter of your choi ce that best describes what you like or how you feel. In some cases you may find it difficult to decide between the two choices. Please circle the one that is most like you are. Do not circle both choices or leave any items blank. It is important that you answer all items with only one choice, A or B. We are interested only in what you like or how you feel, not in how others feel or how one is supposed to feel. There are no right or wrong answers, so please be honest in your answers. 1. A. Id like to try mountain climbing B. I think people who do dangerous thin gs like mountain climbing are foolish 2. A. Too many movies show peopl e falling in love and kissing B. I enjoy watching movies which show people kissing each other 3. A. I would like to try smoking marijuana B. I would never smoke marijuana 4. A. Its more exciting to be around kids older than myself B. I like to be with kids my own age or younger 5. A. Id never do anyt hing thats dangerous B. I sometimes like to do thi ngs that are a little scary 6. A. I think riding fast on a skateboard is fun B. Some of the daring acts of skate board riders seem very scary to me 7. A. I like to be with larg e groups of kids with so mething exciting happening B. I like quiet times with only 1 or 2 friends 8. A. I would not like to learn to fly an airplane B. I think it would be fun to learn to fly an airplane 9. A. I dont like to swim in wa ter that is over my head B. I like to swim in deep water 10. A. I would like to try jumping from a plane with a parachute B. I would never try jumping from a plane with a parachute 11. A. People probably feel good af ter drinking alcoholic drinks B. Something must be wrong with people who need a few drinks to feel good 12. A. I like kids who make jokes even if they sometimes hurt other kids feelings B. I dont like kids who think its fun to hurt other kids feelings
55 Appendix D: Sensation Seeking Scale for Children 13. A. I dont like it when people get drunk, talk loud and act silly B. When people get drunk, it se ems like they are having fun 14. A. Sailing on the ocean in a small boa t would be dangerous and foolish B. I think it would be fun to sail on the ocean in a small boat 15. A. I think skiing fast down a s nowy mountain would be dangerous B. I think skiing fast down a snowy mountain would be exciting and fun 16. A. Id never touch a bug or snake B. Bugs or snakes are fun to hold and play with 17. A. I think it would be exciting to go on a date B. Im not interested in dating yet 18. A. I enjoy the feeling of riding my bike fast down a big hill B. Riding a bike fast down a big hill is too scary for me 19. A. I think its too dangerous for people to take drugs B. I sometimes wonder what it would feel like to be high on drugs, even though I know it would be dangerous 20. A. I dont like being around ki ds who act wild and crazy B. I enjoy being around kids who sometimes act wild and crazy 21. A. I dont think Id like the feeling of getting drunk B. I think I might like to find out what it feels like to get drunk 22. A. I dont do anything I think I might get in trouble for B. I like to do new and exciting things, even if I think I might get in trouble for doing them 23. A. Riding dirt-bikes or motorcycles seems like a lot of fun B. It seems scary and dangerous to ride dirt-bikes or motorcycles 24. A. I like to do wheelies on my bike B. Kids who do wheelies on their bi kes will probably get hurt sometimes 25. A. The worst thing a kid can do is be rude to his/her friends B. The worst think a kid can do is be boring around his/her friends 26. A. If I could, Id see a movie with an R rating B. Im not interested in movies made for older people
56 Appendix E: Demographics and Drinking Questionnaire Demographics and Drinking Questionnaire Directions : Read the questions and check or circ le the option that best describes you. 1. You are a: Girl Boy 2. Circle your grade: 1 2 3 4 5 6 3. Circle your age: 6 7 8 9 10 11 12 13 4. Your school is: ____________________________________________ 5. Check the item that best describes your family: Black White Hispanic Asian-American Other__________ Directions : For the next six questions, drinking alcohol means drinking any drink with alcohol in it such as beer wine, wine coolers, whiske y, rum, vodka, gin, and alcoholic mixed drinks. A drink is one b eer, a glass of wine, a shot of alcohol, or one mixed drink. Remember, your answers will be kept confidential. 1) How often do you drink alcohol? A. Never had a drink of alcohol B. Less than 4 drinks in life C. Drink 1 or 2 times a year D. Drink 3 to 8 times a year E. Drink 1 or 2 times a month F. Drink once a week G. Drink twice a week H. Drink 3 times a week I. Drink 4 times a week J. Drink almost every day K. Drink 1 or 2 times a day 2) How much alcohol did you have the last few times you drank? A. Dont drink alcohol at all B. A few sips of a drink C. Usually 1 drink or less D. Usually 2 drinks E. Usually 3 drinks F. Usually 4 drinks G. Usually 5 drinks H. Usually 6 drinks I. Usually 7 drinks or more
57 Appendix E: Demographics and Dr inking Questionnaire (Continued) 3) The last few times that you drank alcohol, were you: A. Never had a drink of alcohol B. At a religious event C. Celebrating a holiday or special occasion D. At home E. At a friends house F. At a party G. Other ________________________________ 4) Usually when you drink alcohol, do you ha ve permission from your parents or guardians? A. I dont drink alcohol B. Yes C. No 5) When you are an adult (21 or older) how often do you think you will drink? A. Wont drink B. Drink 1 or 2 times a year C. Drink 3 to 8 times a year D. Drink 1 or 2 times a month E. Drink once a week F. Drink twice a week G. Drink 3 times a week H. Drink 4 times a week I. Drink almost every day J. Drink 1 or 2 times a day 6) When you are an adult (21 or older), how much alcohol do you think you will have when you drink? A. Wont drink alcohol at all B. A few sips of a drink C. Usually 1 drink or less D. Usually 2 drinks E. Usually 3 drinks F. Usually 4 drinks G. Usually 5 drinks H. Usually 6 drinks I. Usually 7 drinks or more