AN EXPLORATORY STUDY OF INTROVERSION - EXTROVERSION, LOCUS OF CONTROL AND THE EFFECTS OF ANTI-SMOKING TREATMENTS Thesis for the Degree of M. A. MICHTGAN STATE UNIVERSITY RGBERT JOHN SHAFFER 1975 lllfllfllflllllflljflflHUI”!!!1|!!!IIHIHIHIHHIHW 10372 7834 WW "- 32. ABSTRACT AN EXPLORATORY STUDY OF INTROVERSION-EXTROVERSION, LOCUS OF CONTROL AND THE EFFECTS OF ANTI-SMOKING TREATMENTS By Robert John Shaffer A survey of the literature dealing with anti-smoking treat- ments and relevant personality measures was conducted. A study was designed to investigate the relationships between extroversion as measured by the Eysenck Personality Inventory (EPI), the Firth Smoking Situation questionnaire (a measure of desire to smoke under a variety of arousing situations), Locus of Control (LOC) as measured by the Rotter I-E Scale and success of five treatment techniques, i.e., operant conditioning (0P), Covert sensitization (CS), will power (NP), group support (GS), and attention placebo (AP). It was predicted that the sample of lll smokers contacted through newspaper and television announcements would be more extroverted than the EPI norms, and more external than a normative group on the I—E scale. It was hypothesized that the Firth questionnaire would positively corre- late with extroversion, that the 0P condition would prove most success- ful for those scoring below the median on the EPI and I-E, that those scoring above the median on these measures would do best in the CS condition, but would relapse more quickly than other groups, and that Robert John Shaffer low scorers would be most successful among those in the WP condition. It was further hypothesized that extroversion would not change as a result of treatment, but that a positive correlation between post-test I-E scores and percent of estimated baseline at the end of treatment should exist. Comparisons were made using Pearson product-moment corre- lations, t:tests, and single-factor analysis of variances. It was found that the sample was indeed more extroverted than the EPI industrial norms. No significant change in extroversion as a result of treatment outcome was found. No other hypotheses were confirmed. Limitations of this study were discussed, and suggestions for further investigations were made. Thesis Committee Dr. Richard K. Russell, Chairman Dr. Dozier w. Thornton Approvedzc ' . Dr. J. Gordon Williams Date : u NWQ‘USI HUS“ AN EXPLORATORY STUDY OF INTROVERSION-EXTROVERSION. LOCUS OF CONTROL AND THE EFFECTS OF ANTI—SMOKING TREATMENTS By Robert John Shaffer A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1975 To the memory of my father ii ACKNOWLEDGMENTS I extend my greatest appreciation to my chairman and advisor, Dr. Richard K. Russell. His emotional support was as invaluable as was his scholarly advice. I also wish to thank the other members of my committee, Drs. Dozier N. Thornton and J. Gordon Williams. Their constructive criticism and helpful suggestions were always thought-provoking and broadening. The mutual affection and respect between my co-therapists, Ricki E. Kantrowitz and Meredith F. Taylor, and myself, helped to make this study a far from onerous task. Thank you. Finally, I wish to acknowledge my gratitude to those who made this study possible: The Michigan Lung Association, who pro- vided our funding, and all of those people serving as our subjects, who provided our inspiration. TABLE OF CONTENTS LIST OF TABLES ......................... Chapter I. INTRODUCTION . . ....... . ............ Behavioral Techniques ................ Personality Variables ................ Internal-external locus of control ........ Introversion and extroversion . . . . ....... Empirical studies of personality variables Changes in internality ...... . ....... Statement of the Problem ............... Hypotheses .................... II. METHODOLOGY . . . .................... Subjects ...................... Instruments and Measurements ............. Firth smoking situation questionnaire ....... Procedures . . . . . . . ............ Operant conditioning (DP). .......... Covert sensitization (CS) ............ . Group support (GS) ................ Will-power (WP) . . .- ............... Attention-placebo (AP) .............. No-contact (NC) .................. Therapists .................... Post-test measures ................ Data analysis ........... . ....... III. RESULTS ......................... IV. DISCUSSION .............. . ........ V. SUMMARY ......................... APPENDIX ............................ REFERENCES ........................... iv Page V I3 25 38 4o 42 Table TO. IT. 12. I3. 14. LIST OF TABLES Attendance and Smoking Data by Treatment Groups . . . Surrmary of Treatment Procedures ........ . . . . . Two Factor Repeated Measures Analysis of Variance on Smoking Rates During Treatment ........ One Factor Analysis of Variance on Smoking Rates at End of Treatment ........ . . ..... EPI (Form A) Pre-Test Data ............. I-E Scale Pre-Test Data ............... EPI Scores and Treatment Outcome Within the CS Condition ................... I-E Scores and Treatment Outcome Within the CS Condition ............... . . . . EPI Scores and Treatment Outcome Within the OP Condition . . . ..... . .......... I-E Scores and Treatment Outcome Within the OP Condition . . . . . . . . . . ......... EPI Scores and Treatment Outcome Within the GS Condition . . . . . . . . . . . . . ...... I-E Scores and Treatment Outcome Within the GS Condition . . . ................ One Factor Analysis of Variance on Smoking Rates at Three Months ................. I-E Scores and Treatment Outcome Within the WP Condition ....... . . ......... Page 18 23 27 28 Chapter I INTRODUCTION Cigarette smoking behavior leads to numerous aversive conse- quences. Unpleasant personal and social problems such as halitosis, stains and clothing burns are relatively benign. More serious are the increased risks of lung cancer, emphysema, and coronary artery disease (U.S. Public Health Service, 1962, 1964). Despite these immediate unpleasantries and long-term dangers, many smokers are unable to quit. Bernstein (l969, 1970) has reviewed some of the traditional techniques used to eliminate smoking behavior. Bans (with sanctions including the death penalty in 17th century Germany, Jarvik, 1970), and taxes have been legislated, but Brecher and Brecher (l964) note the lack of success of such prohibitions, and their repugnance in a free society. Public campaigns of various sorts including advertis- ing, lectures, and other media presentations have shown a similar lack of success (Cartwright, Martin, & Thomson, l960; Watne, Montgomery, & Petie, l964). Keutzer, Lichenstein, and Mees (l968) report the use of numerous psychologically-based group therapies (Lawton, l962, I967; Thompson & Wilson, 1966). Encouraging initial results rarely, however, last through follow-up. According to Bernstein (1970), this rapid relapse rate is also characteristic of the various large scale clinics (Ejup, 1964). Techniques of "re-education," lectures, personal 1 consultations, and in some cases, drugs, result in cure rates ranging from 30 to 85%, but relapse frequently follows. The first reported successful treatment by psychoanalysis is by Green (1923). Psycho- analysis is disadvantaged by being a long and expensive treatment process. Mann (1966) found active, "emotional" role playing to be more effective than "cognitive" role playing (i.e., taking a lung cancer patient's role as opposed to debating the issue). Clawson (1964) recommends hypnosis, but provides no group data. Edwards (1964), however, compared hypnosis with a nicotine substitute (Lobeline). No differences between groups were noted. Indeed, after initial improvements, subjects were smoking more upon follow-up. Drug treatments have used two main strategies: nicotine substitutes (e.g., Lobeline preparations), and mild psychotropics (e.g., Vallium, etc.). In reviewing treatments and double-blind drug studies, Bernstein states: "In general, it seems reasonable to conclude that suggestion is the primary and common “active ingredient“ in all anti-smoking preparations tested so far." (Bernstein, 1970, pp. 25-26.) Behavioral Techniques Partly as a result of the discouraging outcomes reported above, and partly due to different theoretical concepts of the nature of smoking behavior, behavioral or "learning-theory" based treatments have been attempted. Smoking behavior may be placed within an instrumental or operant conditioning paradigm. The behavior may be initiated and maintained by immediate positive reinforcement from the effects of nicotine and/or social factors (Jarvik, 1970). Raymond (1964) classed smoking as an addiction, but more precisely, it may be con- ceptualized as a habit, as defined by Hunt and Matarazzo (1970): A fixed behavior pattern overlearned to the point of becoming automatic and marked by decreasing awareness and increasing dependence on secondary, rather than primary, reinforcement (p. 67). Keutzer, Lichenstein, and Mees (1968) note that smoking behavior is maladaptive, occurs in discrete units and is thus easily recordable and observable, and is relatively frequent in occurrence. These factors combine to make smoking behavior an ideal area in which to investigate habit mechanisms in general, not to mention the im- portant salutory benefits of successful treatments. ‘ Bernstein (1969, 1970) and Keutzer et a1. (1968) have pro- vided comprehensive reviews of behavioral studies dealing with cigarette smoking reduction treatments. The earliest behavioral treatments utilized somatic aversion techniques to a great extent. Raymond (1964) paired on emetic drug with smoking in a respondent paradigm. The subject had remained free from smoking up to a one year follow-up. Intermittent electrical shocks were administered to 10 subjects by McGuire and Vallance (1964) as they inhaled cigarette smoke. Results at a one month follow-up revealed six abstainers, one improved, and three not improved. Hot smokey air was blown into the faces of subjects while smoking, until the cigarette was extinguished, at which point mentholated air was presented as a reinforcement and relief stimulus (Wilde, 1964). Ultimately, all five of seven subjects who had improved relapsed (Wilde, 1965). Slightly better results were obtained by Franks, Fried, and Ashem (1966) with a revision of this treatment. Resnick (1968, a,b) required normal, double, or triple their usual cigarette intake from three groups of smokers in order to reach a point of satiation. He reported smoking reduction among the latter two groups. Clairborn, Lewis, and Humbel (1972) however, found the treatment to be no more effective than a placebo. Mees (1966) utilized a method involving breath holding concurrent with the imagination of smoking. A six month follow-up found subjects smok- ing up to 87% of their pretreatment rates. Somatic aversion therapies have been criticized from various perspectives. Stoller (1968) condemns inflicting pain on subjects, as does Mayer (1970). Mayer also questions the long-term conse- quences of such therapies, in terms of effectiveness and possible psychological damage. Church (1963) notes that the effects of punishment in general are difficult to control. High relapse rates have been cited by Bernstein (1969, 1970) and Keutzer et a1. (1968). As a partial consequence of these points, non-somatic aversions and other behavior modification methods have been devised. Covert sensitization (Cautela, 1966) is a method in which a subject is taught to visualize vivid scenes with aversive conse- quences following imaginary indulgence in the maladaptive behavior. Wolpe's (1958) systematic desensitization, in which hierarchies of anxiety situations are successively imagined under conditions of deep muscle relaxation until anxiety is no longer elicited, has been used by Morganstern and Ratliff (1969) on the assumption that smoking is a response to anxiety. _ In an attempt to achieve greater generalization, various self-modification, self-control, or internalization approaches have been proposed. Homme (1965) has suggested the term "coverants" to refer to covert operants: to be used in a process in which a low probability thought such as "smoking is dangerous" is reinforced by a pleasant, high probability thought. Pumroy and March (1966) re- quested their subjects to simply cease smoking in a hierarchy of favored locations for smoking, from least popular to most popular. Other similar techniques also stressing control of operants and internalization of reinforcement are becoming more prevalent (Ferraro, 1973; Logan, 1973; Premack & Anglion, 1973). Combinations of treat- ments have been attempted. Tooley and Pratt (1967) successfully employed coverants, covert sensitization, and "contractual management" (i.e., non-smoking demands are met in exchange for social approval or reinforcement from significant others) sequentially with a young couple. Rather than confound treatment variables by using different methods sequentially, numerous outcome studies have been attempted to compare these techniques with others, or with a variety of different control groups. In an early study, Koenig and Masters (1965) com- pared electrical aversion, systematic desensitization, and supportive counseling in an attempt to reduce smoking behavior. No differences between groups emerged, although therapist effects were present. Interestingly, those therapists rated least empathic produced the best results. Also of interest was the finding that subjects could predict their own likelihood of success. Decreases in smoking be- havior eroded during follow-up periods. Mees (1966) investigated self administered electric shock, two breath-holding conditions (in lab and at home), and a placebo-shock group. No significant between-group effects were noted, nor were improvements lasting. Operant self-control, self administered electrical aversion and Transactional Analysis were studied by Ober (1967). All groups were improved at one month follow-up, smoking reduction ranging from 49% to 58%. There were no significant differences between groups. Rutner (1967) studied the effects of covert sensitization, coverants, breath-holding, contractual management, and self-monitoring. Groups were seen once, instructed in the method assigned, and told to apply it independently of the therapist. After three weeks, all five groups significantly reduced smoking. Contractual management was slightly more successful than the other methods. Keutzer (1968) employed group treatments of breath-holding, coverant instruction, negative practice (Yates, 1968, in which intense smoking supposedly builds a state of reactive inhibition, thereby making smoking aversive and abstinance reinforcing), a drug placebo. and a no-contact control group. All groups but no-contact control improved. There were no differences of significance between groups. As cited earlier, Clairborn et al. (1972) found no differences between Resnick's satiation technique (1968, a,b) and a placebo control. In a weight reduction context, Foreyt and Hagen (1973) found no differential weight loss between covert sensitization, placebo, or controls. The former two groups, however, expressed subjective aversion to formerly favorite foods. Sipich, Russell, and Tobias (1974) compared covert sensitization, self-control/will-power, self-monitoring, attention- placebo, and no-contact control. All but self-monitoring and no-contact controls showed a significant improvement through a six month follow-up. There were, however, no differences in effect be- tween the three groups. Bernstein (1969, 1970), Keutzer et al. (1968), McFall and Hammen (1971) and Ober (1968) note the tendency for non-specific and non-behavioral methods to be as successful as behavioral techniques. Keutzer et al. state: A consistent pattern to the reported results emerges: while virtually all behavior-modification treatments of smoking compare significantly favorable to non-treated control condi- tions, they do not yield significantly better results from those obtained with placebo-attention control treatment (where employed) or with non-behavioristic modification treatments (such as supportive counseling or Transactional Analysis) (1968, p. 526). Personality Variables In the light of the inconsistencies between theoretical pre- dictions and empirical results revealed by the comparative studies reviewed, it is apparent that other variables must be taken into account. Eysenck (1973) observes that large error-variances and consequent non-significance in some experiments are frequently the result of uncontrolled organismic variables. He proposes that personality is a systematic organismic variable of concern to smoking research. Among the aspects of personality that have received attention in regard to smoking behavior, two are especially salient. They are generalized expectancies of internal and external control of rein- forcement, and traits of introversion and extroversion. Internal-external locus of control. Rotter (1966) has de- fined internal-external (I-E) locus of control (LOC) as follows: When a reinforcer is perceived by the subject as following some action of his own but not being contingent upon his action, then, in our culture, it is typically perceived as the result of luck, chance, fate, as under the control of powerful others, or is unpredictable because of the great complexity of the forces surrounding him. When an event is interpreted this way ...we have labeled this a belief in external control. If the person perceives that the event is contingent upon his own be- havior or his own relatively permanent characteristics, we have termed this a belief in internal control (p. 1). It follows from the above definitions that externals would be less likely than internals to raise expectancies following success, or lower them after failure (Rotter, 1966). Rotter also proposes a roughly curvilinear relationship between I-E and "emotional health," with some cultural biases favoring internality. Internals are re- ported to be of higher socioeconomic status (SES), more alert to sources of reinforcement in the environment, and more active in improving the environment in regard to potential reinforcements than externals (Rotter, 1966). Abramowitz (1969) found internals report- ing less depression than externals, while Butterfield's external subjects revealed more debilitating anxiety than internals (Butterfield, 1964). Both of these studies, however, relied heavily on self reports, rather than objective measures. Introversion and extroversion. Eysenck and Eysenck (1968) describe introverts and extroverts (I/E): The typical introvert is a quiet, retiring, sort of person, introspective, fond of books rather than people...distant except to intimate friends. He tends to plan well ahead... does not like excitement,...likes a well ordered mode of life. He keeps his feelings under close control, seldom behaves in an aggressive manner. He is reliable, somewhat pessimistic... The typical extrovert is sociable,...needs to have people to talk to...does not like reading or studying by himself,...is generally an impulsive individual...generally likes change... carefree...optimistic, tends to be aggressive...and is not always a reliable person (p. 6). In an extensive series of books and studies, Eysenck and his associates (Eysenck, 1947, 1960, 1967, 1973; Eysenck & Eysenck, 1968, 1969; Eysenck & Rachman, 1965; Eysenck, Tarrant, Woolf & England, 1960) have developed and supported a type/trait theory of personality. This theory proposes that differences in level of cortical arousal lead to broad personality differences. Extroverts are believed to have low cortical arousal, introverts, higher levels. The reticular activating system (RAS) has been suggested as the source for these differences (Eysenck, 1967). This theory has two implications for learning and smoking behaviors: due to faster build up of reactive inhibition, extroverts will condition more slowly and extinguish faster than introverts (Eysenck & Rachman, 1965); extro- verts require more external sensory stimulation to maintain a positive affective tone, introverts require less (Eysenck, 1973). From the second implication, Eysenck (1973) argues that if nicotine provides 10 the basis for the maintenance of the smoking habit (Jarvik, 1970), and smoking may be initially arousing in its effects on cortical activity, with possible tranquilizing effects as it continues (Armitage, Hall, & Morrison, 1968), then more extroverts would be likely to initiate the smoking habit. Further, extroverts would smoke when bored to raise arousal; introverts, when tense to lower arousal. It follows for Eysenck that anti-smoking treatments should differ in accord with personality. Introverts would be less likely to be helped by arousing, aversion treatments, than by a method like systematic desensitization. The converse would be the case for extroverts (Eysenck, 1973). Empirical studies ofgpersonality variables. Empirical studies relating personality to smoking behavior have not been lack- ing. An early study of Matarazzo and Saslow (1960) found smokers to be more extroverted than non-smokers. This was also the finding of Straits and Sechrest (1960). In a comprehensive review of the empirical literature, Smith (1970) cites 22 of 25 studies in agree- ment on this point. The samples included: U.S. male and female adults, high school and junior high school students; British and Australian adult males; and Puerto Rican male and female high school students. Measurement devices included: The Maudsley Personality Inventory, Cattall's Sixteen Personality Factor Questionnaire, The Minnesota Counseling Inventory, peer ratings and special questionnaires. 0n the other hand, Keutzer (1968) found no significant relationship between extroversion and treatment outcome. It should be recalled, 11 however, that her treatments did not include those specified by Eysenck (1973) as having the greatest differential effects. I-E has also received much attention. Lilienfield (1959) found smokers to be more external than non-smokers. Smith's (1970) review also cites four of five investigations reporting smokers to be more external. Samples included: male and female adults, and high school and junior high school students. Instruments included: The James Test of Internal and External Control, The Neuropsychiatric Adjunct, and special questionnaires. Hjelle and Clouser (1970) re- port external females smoke significantly more cigarettes than internal females, while men show a strong tendency in this direction. Ryan (1973) emphasizes the differences between smokers, rather then between smokers and non-smokers, stating: "Among smokers, smoking behavior is highly dependent upon personality characteristics, so that one type of person may be expected to have one characteristic rate or pattern, and another person may have another rate or pattern." (Ryan, 1973, p. 240.) James, Woodruff, and Werner (1965) observed that smokers who claimed to believe the Surgeon General's report and quit were more internal than those who believed the report and did not quit. Moss (1973) found internals to be more likely to quit than externals. But Lefcourt (1965) found no relation between I-E and smoking. Rutner (1967), Keutzer (1968), and Best and Steffy (1971) also found no relation between treatment outcome and I-E. 12 Changes in internality. Although Rotter (1966) claims satisfactory test-retest reliability for his measure of I-E, studies have noted changes in scores. Some studies have deliberately attempted to raise internality. Smith (1970) found that the resolution of life crises preceded rises in internality. Brief psychotherapy had a similar effect in a study of Gillis and Jessor (1970). In a com- parison between client centered and a verbal reinforcement type of counseling for disabled subjects, Coven (1970) found the verbal reinforcement counseling to be superior in raising internal expect- ancies. Pierce and Schauble (1970) were able to directly teach internality to clients. Dua (1970) however, discovered that an active, behavioral technique was more effective in raising internal- ity than a re-educative treatment. Diamond and Shapero (1973) successfully raised internality through the use of encounter group techniques. Both professional group leaders and trained graduate students served as therapists. Inadvertant changes have also been observed. Berger and Koocher (1972) note that a group of narcotics addicts became significantly more internal in expectation when their institution was closing and they were preparing to cope with the outside environment. The importance of relative increases in internality lies in their cultural value, and the positive associations with the various aspects of mental health previously cited. It is possible that generalization of internal expectancies may increase the over-all effectiveness of an individual, especially in regard to obtaining reinforcement from the environment. 13 In summary, these studies find extroversion positively correlated to smoking. There is rather substantial evidence that externality is also linked to smoking behavior. However, the re- sults of treatment outcome investigations are hardly unanimous. These varying findings are possibly due to the different types of treatments used. Various sorts of intentional and unintentional manipulations and treatments have been shown to alter I-E expectan- cies. It is conceivable that effects from altered expectancies ‘ may generalize to situations other than those causing the altera- tion. It is also possible that increased effectiveness in obtain- ing reinforcements may result from increased internal expectancies. Statement of the Problem This study has been undertaken to further aid in resolving the following questions: Do smokers who seek outside assistance in quitting in fact differ in extroversion and LOC from normal popu- lations? Do introverts feel a greater need to smoke in arousing situations? Of greater practical importance, do different treatment techniques have differential effects as a function of the subject's LOC or extroversion? Will result of treatment correlate with changes in personality scores? Hypotheses. The above questions are stated formally as hypotheses below: 14 l. The sample of smokers will be more extroverted on the Eysenck Personality Inventory than a normative control group. Prior research on smokers (Smith, 1970), has overwhelmingly found smokers in general to be extroverted. It is believed that smokers who wish to quit would also be extroverted. 2. The sample of smokers will be more external on the Rotter I-E Scale than a normative control group. The majority of studies reviewed by Smith (1970) have found smokers to be external. It is likely that smokers who seek the aid of a clinic in quitting would be so as well. 3. There will be a significant linear negative correlation be- tween scores on the EPI and the Firth Smoking Situation Questionnaire. The question of the differential functions of smoking between intro- verts and extroverts, as deduced by Eysenck (1973), has implications for treatment methodology. A smoking situation questionnaire devel- oped by Firth (1971) should differentiate between introverts, who would smoke to lower their arousal level when tense, and extroverts, who would smoke to raise their arousal level when bored. 4. Of those subjects assigned to the covert sensitization condition (CS): a) Those scoring above the median on the EPI (extroverts) will show greater reduction in smoking than those below the median (introverts). b) Those scoring above the median on the I-E scale (exter- nals) will show a greater reduction than those scoring below the median (internals). 15 Different treatments should have different effects upon smokers with different personalities. An aversive treatment would be more successful in reducing the number of cigarettes smoked by extroverts than by introverts, due to its arousing features, and more effective with externals than with internals, because the in- volvement of externals in a more passive treatment will be easier to maintain than the involvement of internals, who might feel subtly manipulated (Rotter, 1966). 5. Of those subjects assigned to the operant conditioning treatment (0P): a) Those scoring below the median on the EPI will show a greater reduction than those scoring above the median. b) Those scoring below the median on the I-E Scale will show a greater reduction than those scoring above the median. An operant technique requiring much involvement but little effective ”arousal, should have the opposite pattern of success, i.e., best re- sults for introverts and internals. 6. Of those subjects assigned to group support condition (GS): a) Those scoring above the median on the EPI will show a greater reduction than those scoring below the median. b) Those scoring above the median on the I-E Scale will show a greater reduction than those scoring below the median. c) There will be a quicker relapse rate among those subjects within this treatment condition. 16 It is probable that a group support treatment would have significant initial success with extroverts and externals, due to the social participation and the diffusion of responsibility, but previous studies suggest that relapse will follow shortly after cessation of treatment. 7. Of those subjects assigned to the will-power condition (WP), those with scores below the median on the I-E Scale will show a greater reduction than those with scores above the median. A simple "will-power" treatment would be suited to reduce the cigarette con- sumption of internals, by permitting them to rely on their own inner resources. 8. As a result of treatment outcome: a) No significant change in EPI post-test scores is expected. b) There will be a significant positive correlation between change in I-E Scale scores and outcome. While various investigations of smoker's personality in relation to treatment outcome have been undertaken, little attention has been paid to changes in personality or generalized expectancies in relation to smoking treatment success. In Eysenck's theory, I/E is a constitu- tional temperment, thus any changes in this trait, as assessed by post-test scores, would reflect either error or physiological change. Thus no changes in I/E as a result of treatment outcome are expected. 'Internal-external LOC, on the other hand, is a psychological trait of expectancy, reflecting environmental and intra psychic or cognitive events. Therefore, successful treatment is likely to lead to increase in internality and possible generalized benefits. Chapter II METHODOLOGY Subjects Announcements of a free smoking clinic under the auspices of the Michigan Lung Association were made through local newspapers and a television station. One hundred and eleven smokers attended the organizational meetings where pre-test data were gathered. This included the pre-test personality measures discussed below, demographic data (see Kantrowitz, 1975, for further details on demography), an estimate of daily cigarette consumption, and slips of paper on which actual cigarette consumption was measured for a 10 day period. This initial sample consisted of 45 males and 66 females. The mean age of this group was 37.18 years, §D_of 9.67. Their estimates of daily smoking averaged 31.3, with an §D_of 7.34. Of the 111 subjects who provided pre-test data, 73 remained throughout the program to supply post-test data. Fourteen of the remainder did not attend the first treatment session. Twenty-four dropped out of the study or missed more than three of the eight treat- ments. Table 1, below, summarizes the distribution of subjects among groups and their treatment data. 17 18 Table 1. Attendance and Smoking Data by Treatment Groups. Groups OP CS WP GS AP ‘ NC Sex M F M F M F M F M F M F ATTENDANCE Session 1 6 7 8 7 3 ll 6 9 5 13 * Session 2 8 6 6 5 3 9 4 8 5 10 Session 3 8 4 6 5 3 9 5 7 5 10 Session 4 6 5 4 4 3 9 4 9 3 10 Session 5 5 4 5 5 3 9 5 7 2 11 Session 6 6 4 3 3 3 9 5 8 2 7 Session 7 4 3 2 2 3 9 5 6 3 8 Session 8 4 2 3 4 3 9 5 5 3 4 Final 3_ 6 4 4 6 3 9 5 8 4 8 7 9 Pre-treatment Estimate 35.8 27.8 32.0 31.8 29.1 31.6 Baseline Measure 26.7 21.9 19.92 27.8 20.9 25.0 Post-Treatment 7.7 6.9 13.9 7.6 11.23 24.9 *No contact made Instruments and Measurements I-E Scale. The Rotter I-E Scale is a 29 item forced choice questionnaire designed to assess belief in sources of reinforcement. There are six filler items, leaving a range of scores from 0 to 23. The higher the score, the greater the externality expressed. Text-retest reliability for one month is moderately good, ranging from .72 to .78 in two samples (Rotter, 1966). After two months, it dropped to .55 in one sample reported by Rotter (1966). The trend is toward a typical 19 drop of one point upon readministration. No significant correla- tions with the Marlow-Crown Social Desirability Scale were present. Evidence for satisfactory construct validity lies in correlations of .55 to .60 with the Likert type James-Phares I-E Scale (Rotter, 1966). Norms quoted by Rotter (1966) range from a low of 5.48 for female Peace Corps Volunteers, to a high of 10.00 for 18 year old males in the Boston area. EPI, The Eysenck Personality Inventory (Eysenck & Eysenck, 1968) is a 57 item yes-no questionnaire designed to assess degree of extroversion and neuroticism (i.e., "emotional ability"). It also contains a lie scale. There are two alternate forms available, with a mean difference of 2.10 points in favor of form B. Test-retest reliability for the whole instrument ranges from .84 to .94, and .80 to .97 for the separate forms. Split-half reliability ranges from .74 to .91. Factorial validity, construct validity, and con- current validity are all reported as satisfactory (Eysenck & Eysenck, 1968). Reference group norms for rather widely differing populations are provided by the manual. Correlations between I-E and the EPI are only .07 (Shriberg, 1972). Firth SmokinggSituation Questionnaire. As part of a factor analysis of demographic variables associated with smoking situations, Firth (1971), devised a 22 item questionnaire consisting of 12 high arousal, and 10 low arousal potential smoking situations. Subjects were asked to imagine themselves in these situations and rate their desire for a cigarette on a seven point scale. A copy of this instru- ment is included as the Appendix. 20 Procedures Potential subjects who had expressed an interest in attend- ing the clinic were contacted by telephone and invited to a group meeting, at which time the pre-test measures were administered, as well as instructions on baseline measurement of daily smoking. They were informed that the tests would help to assign them to the most effective group, and would aid in further research and treatment. The treatment program lasted six weeks, and consisted of eight sessions, two per week for the first two weeks, and one per week for the remaining four weeks. Six groups were formed with greatest priority given to a normal distribution of I-E Scale scores, smoking rate and, of course, scheduling within each condition. Four treatment conditions and two controls were formed: Operant conditioningg(0P). This program was based on Stuart and Davis (1974). It involved baseline determination and continued daily monitoring, immediate positive reinforcement of non-smoking behaviors by reference to a "reward deck" which consisted of indivi- dually determined benefits of abstinence written on cards designed to be carried and consulted when an urge to smoke is resisted, individu- ally determined tangible reinforcements given for continuing to meet the percentage criterion of gradual reduction, praise and social support from significant others for non-smoking behavior, and a gradual decrease in smoking through a) reduction in number of cigarettes smoked, and b) restriction of smoking to six locations, which are themselves gradually eliminated. Key elements of this condition were 21 reliance on individually tailored reinforcements, lack of punishing stimuli, and a general sense of subject participation. Covert sensitization (CS), This program followed the pro- gram outlined by Cautela (1970). The first session was devoted to teaching muscle relaxation. As treatment progressed, and facility in relaxation was acquired, less time was spent inducing relaxation. Ten situations determined by group consensus as desirable smoking situations were vividly imagined each sesSion. Imagined aversive stimulation (e.g., nausea) followed imagined smoking. Relief or escape trials, in which smoking was resisted, alternated in a ran- dom manner. Subjects were instructed to practice at home. Group Support (GS). A discussion group and therapist met for sessions that attempted to diminish smoking through peer support and peer pressure. The advantages of quitting, recounting of smok- ing experiences, and general support were emphasized. Willspower (WP), Subjects in this condition were informed that they could stop smoking by their own volition most easily. They were contacted exclusively by telephone after the first meeting. The telephone calls were meant to obtain information on smoking rates. No suggestions as to anti-smoking techniques were made, other than a reminder that their own capacities and resources could be relied upon. Attention:placebo (AP). An attention-placebo group was formed and told that smoking is the result of their unconscious atti- tudes and habits. These habits could be reconditioned through subliminal techniques. In a process used by Sipich et al. (1974), the supposed 22 "reconditioning" phrases were tachiscopically presented upon a screen. These were, in fact, nonsense syllables. The subjects were told that viewing these stimuli would be relevant to their treatment and would result in reduction in their smoking. It was felt that any mislead- ing information supplied was vital to the efficacy of the treatment, and justified by the scientific contribution of the investigation. No-contact (NC). No-contact controls were requested to record baseline smoking for one week, and take pre- and post-test smoking measures. It was determined that any condition resulting in markedly inferior treatment outcome would be given the option of further treatment. The subjects in the NC group were offered this opportunity. Table 2 summarizes treatment procedures over the eight sessions for each condition. Therapists. The therapists included one male and two female graduate students trained as a group in the treatments to be employed. One therapist was beyond the Master's level, the remain- ing two were first year clinical psychology graduate students. Each of the three therapists conducted one group from each of the six conditions. Post-test measures. The post-test measures consisted of the alternate form (B) of the EPI, readministration of the I-E Scale, and solicitation of subjective perceptions, complaints and suggestions from the participants. They were administered immediately following the final session. Follow-up calls recording daily cigarette con- sumption were made on a weekly basis for 10 weeks, once at three 23 mcosam_mu an .mmzumwo cm>wm pgoaaam a cowpmacmmmga acmspmwcp :owmgm>m aaocw umugouwg mcwxosm upqoumpsump new cowumxmpmm mocmcm»:_mz m mcocnmpwp x3 .mmzomwo cm>wm pgoaazm a cowpmucmmmca “cospmwcu commem>m mcwonm vcm Quota umucoomg mcwxosm ownoumwcump use cowumxmpmm 3mw>mm m mcogamep >5 cowpmooF mco x—co cw mcwonm .mmaomwo cm>wm ucoqqam a cowpmpcwmmga pewsgmwgp cowmgm>m mcwpmmmn yo xop op cowpoznmm nzogo umcgouwc mcwxosm owaoomwsumh new cowpmxmpmm mucmcmpcwms new zow>mm o mcocampmp an chmema to Rmm on compoaumm .mmzumwo cm>wm ycoaaam a cowgmgcwmmcg “cospmocp cowmgm>m umcwm_qu mcwmsu asogo wougoomc mcwxoem owaoomwzumh use cowpmxopmm pagow>mcmn use Pogpcoo msu m mcocam_mp an umpmcwstm mcowpmoop mcwonm 02H .mmzumwo cm>wm peoaazm a cowpmucmmmca ucmspmmgu cowmcm>m mcw_wmmn mo xom op cowaozvmm aaoco umugoumg mcwxosm owaoomwcomh new cowamxmpmm nosgoe xumu cowmcm>< v mcocampm» ma umpmcwEPFw mcopamuop mcwxosm oz» .mmaumwo cm>wm “Loaaam w coppmpcmmmga ucmspmmcu cowmgm>m mcw_mmmn eo am“ ob cowpusnmm azogw umugouog mcwonm owaoumwcomh use :owpmxmpmm nmcwmpaxw mm: :mxou .mmzo mcwxosm m mconnm_mp an emELoe um:_m_axm .mmaumwa =w>wm peonasm a cowpmucmmmca cowpmauwm m>mmgm>< Anaememwg cowumoo_ new xumu ucmzmm nsoso uwucoowg mcwxosm owaoomwsum» cowpmxmpmm mgwogoecwme meoom new mpnwm:MH m cowpmpcmmwca owaoumwsumh umsgoe xgucmcmw; cowpooog .mmaomwo upmcowpmc mpmcowpmg mcwcwmgu coppmxmpmm cowpmcmamga xomu ugmzmm aaogu pcmEpmmLH pcmEpmmgh mpmcowumg paprmmL» mwcanwuoca pcmsmcammmz _ cowmmmm mm a: m< mo 80 nzogu .mmczcmoocm pcmspmmgh mo accessm .N open» 24 months, six months, and again at one year. Due to the later treat- ment of the NC condition, no further follow-up was performed on this group. Data analysis. The resulting data, based on number of cigarettes smoked per day, was analyzed by means of two factor (time X condition) repeated measures analyses of variance (Keppel, 1973) for the six weeks of treatment, and the 10 weeks of follow-up. One factor analyses of variance were performed on end of treatment data and three month data. The data derived from EPI and I-E Scale scores, and the Firth questionnaire were correlated by the Pearson product moment correlation (Bruning & Kintz, 1968). Comparisons between high and low scorers on the EPI and the I-E Scale within the treatment conditions were performed via t_tests. The dependent variable for these comparisons was percentage of baseline estimate smoked at end of treatment. This is the measurement suggested by Keutzer et al. (1968) and Koenig & Masters (1965). Chapter III RESULTS A two factor repeated measures analysis of variance was perfonmed on the smoking data from the six weeks of treatment of all groups but NC. As can be seen from Table 3, significant effects were obtained for the time factor 5 (5,260) = 50.625, p_<.001, indi- cating that improvement occurred over the duration of treatments. The treatments, themselves, did not differ significantly from each other, nor did any significant interactions occur. A one factor analysis of variance on end of treatment data was performed in order to include the NC condition. Significance was obtained for conditions f (5,67) = 6.62, p_<.Ol. This data is summarized in Table 4. Post- hoc comparisons using the Scheffe’ test determined that CS and GS conditions differed from NC at the .05 level. No other comparisons were significant. Table 3. Two Factor Repeated Measures Analysis of Variance on Smoking Rates During Treatment Source Sums of Squares gf_ MS_ .5 Between (8) 15,585.25 56 -- -- Condition (C) 393.37 4 98.34 .34 Error B 15,191.88 52 292.15 -- Within (W) 18,538.17 285 -- -- Time (T) 8,695.59 5 1,739.12 50.63* T X C 911.05 20 45.55 .25 Error W 8,931.526 260 34.35 1.33 *p <.OOl 26 Table 4. One Factor Analysis of Variance on Smoking Rates at End of Treatment Source Sums of Squares gj_ MS, .5 Between (8) 3,189.77 5 637.95 6.62* Within (W) 6,460.34 67 96.42 -- Total (T) 9,650.11 72 -- -- *p_<.01 The initial hypothesis predicted that the smoking sample would be more extroverted on the EPI than a normal control group. To test for this, two t:tests were conducted, the first comparing the smoking sample to an American industrial sample, and the second to the American college norms. The American industrial sample consisted of 296 adult male industrial employees ranging in age from 19 to 64 years, with a mean of 43 (Eysenck & Eysenck, 1968). The college norms were derived from the scores of 1,003 male and female students. The findings indicate that the smoking sample was significantly more extroverted than the industrial norms, t_= 3.842, p_<.01, yet signi- ficantly more introverted than the college sample, t_= 3.273, p_<.01. These data are displayed in Table 5. Table 5. EPI (Form A) Pre-Test Data. n _)_(_ SQ _i_:_ value Sample 111 11.76 3.99 -- Industrial Norms 296 10.3 -- 3.84* College Norms 1,003 13.1 4.1 3.27* --data unavailable E <.Ol 27 The second hypothesis predicted that the sample of smokers would be more external on the I-E Scale than a normal control group. Previous research (Staats et al., 1974) has found differences in LOC related to age and sex variables. Thus the normative group chosen to test this hypothesis consisted of 200 rural and small town Ohioans, divided according to gender and four age groups, from 16 to over 60 years (Staats et al., 1974, Staats, personal communication). Eight t_tests were performed. No comparisons were significant. Thus it was not established that the smoking sample was more external. These findings are presented in Table 6. Table 6. I-E Scale Pre-Test Data Sample (N = 111) Age 16-25 26-45 46-60 60- Sex M F M F M F M F n 10 15 21 35 13 14 2 0 x 9.2 10.9 6.7 8.9 6.9 7.6 9.5 -- _s_1_)_ 4.9 4.3 10.0 3.4 3.5 5.2 2.5 -- Normative Groups (N = 200) n_ 25 25 25 25 25 25 25 25 x. 10.4 10.2 7.4 8.7 6.8 8.8 7.4 10.2 §D_ 4.2 3.9 4.2 3.9 3.4 3.2 3.4 3.2 t_va1ue .71 .59 .63 .16 .12 1.12 .82 -- The third hypothesis predicted a negative correlation between initial EPI scores and those on the Firth questionnaire. The correla- tion equaled +.036, which is clearly nonsignificant. There was no evidence relating smoking in different levels of arousal to degree of extroversion. 28 The fourth hypothesis stated that those subjects within the CS condition who were extroverted would be more successful than those who were introverted, and those who were more external would be more successful than those who were internal. As tables 7 and 8 indicate, neither t_ratio was significant, thus neither extroversion nor LOC appears relevant to outcome in the CS condition. Table 7. EPI Scores and Treatment Outcome Within the CS Condition. Low High (Introverts) (Extroverts) n_ 5 5 K_% of Estimated Baseline .226 .281 SD .128 .186 3 value .49 Table 8. I—E Scores and Treatment Outcome Within the CS Condition. Low High (Internals) (Externals) fl. 4 6 l_% of Estimated Baseline .270 .245 SD. .169 .151 3 value .063 The fifth hypothesis concerned the 0P condition. It pre- dicted that introverted and internal subjects would show a more pro- nounced reduction in smoking than those relatively more extroverted and external, respectively. As can be seen in tables 9 and 10, neither 3 ratio yielded significant differences. The evidence does not support any outcome differences in the OP condition based on extroversion or LOC. 29 Table 9. EPI Scores and Treatment Outcome Within the OP Condition. Low High (Introverts) (Extroverts) fl_ 6 4 l_% of Estimated Baseline .278 .275 gg; .450 .177 t_value .164 Table 10. I-E Scores and Treatment Outcome Within the OP Condition. Low High (Internals) (Externals) g_ 5 5 K_% of Estimated Baseline .235 .287 §2_ .287 .154 t_value .319 The sixth hypothesis predicted that greatest smoking reduction would occur among those subjects in the GS condition who were relatively more extroverted and more external. It was also hypothesized that this group would suffer a more rapid relapse. Neither of the predictions concerning personality were supported by the t_ratios, as shown in tables 11 and 12. A one-factor analysis of variance performed on the treatment conditions at three months follow-up disclosed no differences in rate of relapse among any of the groups, as seen in Table 13. The seventh hypothesis predicted that internal subjects assigned to the WP condition would be more successful than external subjects. As shown in Table 14, this t_ratio was not significant, and the hypothesis was not supported. I.‘ ,‘"3 “m- A 30 Table 11. EPI Scores and Treatment Outcome Within the GS Condition. Low High (Introverts) (Extroverts) n_ 3 9 l_% of Estimated Baseline .025 .252 §9_ .035 .226 t_va1ue 1.239 Table 12. I-E Scores and Treatment Outcome Within the GS Condition. Low High (Internals) (Externals) n. 6 6 K_% of Estimated Baseline .236 .155 §__ .248 .178 t_va1ue .589 Table 13. One Factor Analysis of Variance on Smoking Rates at Three Months. Source Sums of Squares gj. M§_ .5 Between (8) 387.20 4 96.8 .72 Within (W) 6,976.69 52 134.17 T0ta1 (T) 7,363.89 56 31 Table 14. I-E Scores and Treatment Outcome Within the WP Condition. Low High (Internals) (Externals) g_ 7 5 K_% of Estimated Baseline .514 .516 SD_ .357 .524 t_value .006 The eighth hypothesis predicted no change in EPI post-tests as a result of treatment outcome, but a positive correlation between outcome and post-test I-E Scale scores. The correlation between EPI (form A) - EPI (form B) + the published difference between the two forms (Eysenck & Eysenck, 1968), and the percent reduction in estimated baseline smoking was found to be -.056. This is nonsignificant, as expected. EPI scores did not differ as a result of treatment outcome. The correlation between I-E (pre) - I-E (post) and outcome was +.O8. This was also nonsignificant, contrary to expectations. Successful treatment did not significantly increase internality, nor did lack of success significantly lower it. Chapter IV DISCUSSION Overall comparisons between conditions revealed that sub- jects in all treatment conditions improved over the duration of treat- ment, but not at differential rates. At the end of treatment, GS and CS conditions were significantly different from the NC controls, but not from each other. There were no differences between any treatment conditions at 10 weeks follow-up. These findings add further support to those of Keutzer et a1. (1968), and Sipich et a1. (1974). Namely, treatment is superior to no treatment, but a behavioral treatment such as CS is not demonstrably superior to a non—behavioral treatment such as GS or the AP control. The finding that the AP group was not signi- ficantly inferior to DP and WP conditions suggests that placebo factors are involved in outcome. It was found that this sample of cigarette smokers attempt- ing to quit was more extroverted than an industrial sample, yet more introverted than a college sample. The first comparison confirms the Eysenckian theory, that is, smokers are relatively extroverted. The two populations were similar in regard to age, although there was apparently a somewhat greater variance in SE5 among the smoking sample. In addition, the smoking sample consisted of both males and females, while the industrial sample was confined to the former. It is not known what percentage of the industrial sample consisted of smokers, or more 32 33 specifically, smokers attempting to quit smoking. Yet when compared to a population considerably younger, and of likely high SES, the sample of smokers is found to be more introverted. Evidently, factors of age and social status play some role in extroversion. Several questions immediately arise: Do generations and/or social classes consistently differ in extroversion? If such is the case, are the relatively more extroverted members of such groups more likely to smoke or to smoke and wish to quit? These questions would be amenable to further research. A second characteristic of this sample of smokers was attitude toward LOC. Prior research has discovered significant differences in age and nearly significant differences in gender on the I-E Scale (Staats et al., 1974). Thus separate comparisons were made involving both sexes and four age groups. No differences between sample and normative groups were found. This result is not in con- formity with some of those reported by Smith (1970). Again, this find- ing is weakened by a lack of knowledge concerning the smoking habits of the normative sample. One important difference distinguishing this sample of smokers from other smoking samples is the fact that quitters or those attempting to quit smoking have I-E Scale scores closer to, or indistinguishable from, those of a normative population. A study con- cerning differences in I-E Scale scores among smokers, those wishing to quit, quitters and those who have never smoked, would be useful. On the basis of Eysenckian theory, a correlation between smoking in physically and psychologically arousing situations and high degree 34 of introversion was expected. None was found. Comments from subjects and patterns of scores on the Firth questionnaire suggested that indi- viduals do indeed tend to smoke in differential patterns based on arousability, yet a direct correlation with extroversion does not exist. A possible differential effect of physical as opposed to psychological stress may play some role in accounting for these re- sults. More likely, idiosyncratic reinforcement histories operating within individuals could be responsible. Other findings of this study reinforce those of Keutzer et a1. (1968), Ober (1968), and others: that is, personality variables such as extroversion as measured by the EPI, and attitudes toward LOC as measured by the I-E Scale, seem irrelevant to treatment outcome. This occurs even within several quite different treatment conditions, including behavioral methods such as an arousing sensitization treat- ment and an individualized operant treatment; and non-behavioral methods such as socially supportive groups and an independent will- power condition. Further, relapse seems to occur at uniform rates. The strength of these results has been weakened by several factors. Most noticeable is the relatively small n_within each of the treatment conditions. Additional data from more subjects would likely have resulted in a wider sampling of both personality scores and degrees of success. This, in turn, may have revealed trends that are not now apparent, or more firmly ruled them out. Perhaps more significantly, however, was the fact that in- sufficient cells could be filled to perform a three factor (EPI X I-E X conditions) analysis of variance (Keppel, 1973). Thus any 35 interactions between these variables and outcome could not be detected. It does not seem unrealistic to expect outcome differences between internal introverts and external extroverts, for example, given different treatment conditions. This area remains open for further investigation. Another factor that may have attenuated any trends is the success rate. Fifteen of 57 subjects in treatment conditions, or 26%, were completely abstaining at end of treatment, 8, or 14% were abstinent at 10 weeks, and 7, or 12.3% were still not smoking at the three month follow-up. Had there been more pronounced success, differences may have been more apparent. Therapists with greater clinical experience in the use of the techniques employed could have possibly increased this rate. Also, commitment may have been enhanced by the use of money deposits returned at the end of treatment. It should be noted, however, that this rate is not substantially different from that reported by previous researchers (Bernstein, 1970). It was also found that relative success or failure within treatment had no effect on EPI or I-E Scale post-tests. This was contrary to expectations in regard to the I-E Scale, but in accord with expectations concerning the EPI. Several possible explanations deserve consideration in regard to LOC. It is conceivable that rela- tive success within a treatment group that relies heavily on self control, (e.g., WP or OP) would be more likely to influence the I-E scale score than would success within a condition oriented towards external control (e.g., CS) or group processes (e.g., GS). Again, a 36 higher success rate might have induced more changes in LOC if such changes are actually a result of success. Follow-up post-tests correlated with follow-up smoking rates would also prove useful. The correlation between outcome and change in EPI scores was extremely small (r_= .08), as expected. However, determining a precise level of confidence is problematical when no significant differences are found. Thus, this finding must be accepted with some reserve. Nevertheless, due to the relatively large size of the sample (n_= 57, NC group excluded), it does support Eysenck's opinion that extroversion is a relatively stable trait within individuals. Subsequent treatment programs should be aware of the potency of placebo effects, and use them intentionally. It is important that commitment to a full course of treatment be main- tained. This may be accomplished by requiring monetary deposits or contracts. It is suggested that further studies in this area use larger samples, more mature or experienced therapists, and have a comprehensive knowledge of the demography (i.e., age, sex, SES and smoking habits) of any normative groups. If possible, local norms should be used. Specific questions that may merit further investi- gation include the following: What is the relationship between age, SES, and smoking? Is the extroversion and smoking relationship pre- dicated on absolute scores or scores relative to demographic groups? How do the states of contented smoking, quitting, and non—smoking relate to LOC? What are the interactions between traits, attitudes, 37 and treatments? Does success in different types of treatment have a greater or lesser influence on LOC change? And, finally, does LOC change as a function of relapse? As the relationships, if any, between personality, attitude, background and treatment technique are further explored, it may ulti- mately become possible to pair the individual smoker with the treat- ment providing the greatest likelihood of success. Chapter VI SUMMARY A survey of the literature dealing with anti-smoking treat- ments and relevant personality measures was conducted. A study was designed to investigate the relationships between extroversion as measured by the Eysenck Personality Inventory (EPI), the Firth Smoking Situation questionnaire (a measure of desire to smoke under a variety of arousing situations), Locus of Control (LOC) as measured by the Rotter I-E Scale and success of five treatment techniques, i.e., Operant conditioning (OP), covert sensitization (CS), will power (WP), group support (GS), and attention placebo (AP). It was predicted that the sample of 111 smokers contacted through newspaper and television announcements would be more extroverted than the EPI norms, and more external than a normative group on the I-E scale. It was hypothesized that the Firth questionnaire would positively corre- late with extroversion, that the OP condition would prove most success- ful for those scoring below the median on the EPI and I-E, that those scoring above the median on these measures would do best in the CS condition, but would relapse more quickly than other groups, and that low scorers would be most successful among those in the WP condition. It was further hypothesized that extroversion would not change as a result of treatment, but that a positive correlation between post-test I-E scores and percent of estimated baseline at the end of treatment should exist. 38 39 Comparisons were made using Pearson product-moment corre- lations, tftests, and single-factor analysis of variances. It was found that the sample was indeed more extroverted than the EPI industrial norms. No significant change in extroversion as a re- sult of treatment outcome was found. No other hypotheses were confirmed. Limitations of this study were discussed, and suggestions for further investigations were made. APPENDIX 4O .ouw— Agw> m? nuwcz .wEo; sweep L30» L04 uwmz on m>mc 30> .mv m w m N P o - uueeemap exam; :0 Same 6>0L3 as 6>63 36>. .03 m e m N _ o _- .E_._..._. DCwXVVLLO—L Ucm “cm—.Om> w DCmeQCugka wLm 30>. .mv m V m N _. O _.I Q’ m N F O F I .mvcmwge saw: mcwcm>0 00030 0 30030; 000 30> .vv m .mmco 0F003F0> 000 0000 >00 000 000;“ e_ 000 on mceou 000003; F000>0m 0330000 xoop 00 0>00 30> .mv m 3 m N F o _- .Egoe xmu 00000_P3500 0 00 ___w 00 0>0; 30> .Nv m 0 m N _ o F- .000 0 000 30_>Lmucw 00000005? 00 30030; 000 30> ._v m 3 m N P o - 3. _ _ _ .1- 3 3 000000 mcwmmu mc_mmu mx__mwc 300000 00000005 0: .mcowum3uwm 00000 0? 03050 00 30» 00% mpawmmoa 00 003000 00 0F303 0w 00:00:: 00: 0:0 0000003000 m3owgm> 0:0 0w wppmcmmwu 0 004 mewm00 030» 00 xpco 000mmcmu=w 000 02 0000 0002050; wmmmpa .00 0,303 000000300 0 00% 000000 030» 30000 30» 00:: 00 3000000000000 00053: 0;» 0:00 000;; 0032 30» 00_003p_m £000 00% 03:» _. 3 _ 4 3 3 _ m 3 m N F o F- mgwmmu mcwmmv mgwmmu mxwpmwv 300000 00000005 0: "000; 03000 0F00m 0:0 40 35000 00 umumuwucw 00 0P3ozm 03050 0» mgwmmu L30» 00 000300 00H .mco_003pwm 0:0 40 £000 cw 00 0P303 mupmgwmwo 0 00» mcw>0cu L30» xcwnu 30> 00:: 0000000? 000 uwawcommu mcowum3uwm 0;“ 0w mpmmg30> mcwmms_ 0» 000 30» 30—F00 00:0 mcowgmm3c 0;» 0H 0000000_pm03o acmxosm P000w003uwm xHozmmm< 1 4 .0000 000 >003 0 000 0030 0>00 30> .000000 00000 000> 0 00 000000003000 000000000 000000000 00 000>00 000 30> .00>00>00 0000 0>00 >00 000000 0 00000 300 000 00000 00000 0 00 000000030000 000 00000 0030 0>00 30> .0000>000300 00030 0 00 0003 >0000000 0 000 00 30> .00300 0000>00 000 00000 0 00 00000>000 000 30> .0000 000000000000 00 000 0000000000 000000 00000 0000 00 00 0>00 30> .000000 00000 0 00 00000 000 00 00000000 0000 0030 0>00 30> .00000 000000 0 000030 0000000 0003 00000000 000 30> .3000 300 000000 0000 000 30> 00 000 00 00000 00 0000 0000 0 0030000 0000000 000 30> .000000000 0000 0300 000 00000 000 000 00000000 0300 0003000 >0000000000 000 000 0030000 00 000>00 000 30> .0000 000 0 00 00 00 03000 00 00 0003 0000 0 00 00000 0 000 0000 0300 000 00 0>00 30> .00030 0000 00 0000 000 00000 >00> 000 30> .>0000 00000 000> 000 00000 0 00 00000000>000 0 0000 00 000>00 000 30> .0000000 003000000 000003000000 0 0>00 00 000 300 0000 0003000 0000 0000003 0.0000000 0 00 0000000 000 30> .000N0000 0 0000000 00000 00000>0 0300000 0 000>00 000 30> .08 .08 .03 .0: .0: .0: .0: .0: .0: .0: .0: .0: .0: REFERENCES REFERENCES Abramowitz, S. I. Locus of control and self-reported depression among college students. Psychological Reports, 1969, 25, 149-150. Armitage, A.K., Hall, G.H., & Morrison, C.F. Pharmacological basis for the tobacco smoking habit. Nature, 1958, 217, 331-334. Best, J.A., Steffy, R.A. Smoking modification procedures tailored to subject characteristics. Behavior Therapy, 1971, g, 177-1910 Berger, M.S., & Koocher, G.P. Locus of control in narcotic addicts: State or trait phenomenon? Journal of Clinical Psychology, 1972 9 _2_§_: 468-469 0 Bernstein, D.A. Modification of smoking behavior: An evaluative review. Psychological Bulletin, 1969, 21, 418-440. Bernstein, D.A. Modification of smoking behavior: An evaluative review. In W.A. Hunt (Ed.) Learning Mechanisms in Smoking. Chicago: Aldine, 1970. Brecher, R., & Brecher, E. Smoking; The Great Dilemma. Washington, D.C.: Public Affairs pamphlet No. 361,7The Public Affairs Committee, 1964. Bruning, J.L., & Kintz, B.L. Computational Handbook of Statistics. Glenview, 111.: Scott, Foresman and CE}, 1968. Butterfield, E. C. Locus of control, test anxiety, reactions to frustration and achievement attitudes. Journal of Personality. 1964, 32, 298-311. Cartwright, A., Martin, F.M., & Thomson, J.G. Efficacy of an anti- smoking campaign, Lancet, 1960, 1, 327-329. Cautela, J.R. Treatment of compulsive behavior by covert sensitization. Psychological Record, 1966, 16, 33-41. Cautela, J.R. Treatment of smoking by covert sensitization. Psychological Reports, 1970, gg, 415-420. Church, R. The varied effects of punishment. Psychological Review, 1963. 2.9.. 369-402. 42 43 Clairborn, W.L., Lewis, P., & Humble, S. Stimulus satiation and Clawson, Coven, A. Diamond, Dua, P.S. Edwards, Ejrup, B. Eysenck, Eysenck, Eysenck, Eysenck, Eysenck, Eysenck, Eysenck, Eysenck, smoking: A revisit. Journal of Clinical Psychology, 1972, 289 416-419. T.A. Hypnosis in medical practice. American Journal of Clinical Hypnosis, 1964, 6, 232-236. B. The effects of counseling and verbal reinforcement on the internal condition of the disabled. Dissertation Abstracts International, 1970, 31 1000. M.J., & Shapero, J.L. Changes in locus of control as a function of encounter group experiences: A study and replication. Journal of Abnormal Psychology, 1973, 82, 514-518. Comparison of the effects of behavioral oriented action and psychotherapy reeducation on introversion-extroversion, emotionality, and internal-external control. Journal of CounselingPsychology, 1970. 11, 567-572. G. Hypnosis and lobeline in an anti-smoking clinic. Medical Officer, 1964, 111, 234-243. Treatment of tobacco addiction: Experiences in tobacco withdrawal clinics. In Can We Hel Them Sto ? Chicago: American Cancer Society, Illinois Division, 1964. H.J. Dimensions of Personality, New York: MacMillan, 1947. H.J. The structure ofgpersonality. London: Methuen, 1960. H.J. The bjglogical bgsis of personality. Springfield, 111.: Charles C. Thomas, 1967. H.J. Personality and the maintenance of the smoking habit. In W. L. Dunn (Ed.) Smoking behaviors: Motives and incentives. Washington, D.C.: Winston and Sons, 1973. H.J., & Eysenck, S.B.G. Manual for the Eysenck personality inventory. San Diego, Califi: Educational and Industrial Testing Service, 1968. H.J., & Eysenck, S.B.G. Personality structure and measurement. London: Routledge and Kegan Paul, 1969. H.J., & Rachman, S. The causes and cures of neurosis. San Diego, Calif.: Knapp. 1965. H.J., Tarrant, M., Woolf, M, & England, L. Smoking and personality. British Medical Journal, 1960, 1, 1456-1460. 44 Firth, C.D. Smoking behavior and its relation to the smoker's immediate experience. British Journal of Social and Clinical Psychology, 1971,19, 73-78. Ferraro, D.P. Self control of smoking: The amotivational syndrome. Journal of Abnormal Psychology, 1973, 21 152-157. Foreyt, J.P., & Hagan, R.L. Covert sensitization: Conditioning or suggestion? Journal of Abnormal Psychology, 1973, 22, 17-23. Foss, R. Personality, social influence, and cigarette smoking Journal of Health and Social Behavior, 1973, 13, 279-286. Franks, C.I., Fried, R., & Ashem, B. An improved apparatus for the aversive conditioning of cigarette smokers. Behavior Re- search and Therapy, 1966, 5, 301-302. Gilles, J.S., & Jessor, R. Effects of brief psychotherapy on belief in internal control: An exploratory study. Egychotherapy: Theory, Researcn,_and Practice, 1970, 2, 135-137. Green, G.H. Some notes on smoking. International Journal ofgpsycho- analysis, 1923, A, 323-324. Hjelle, L.A., & Clouser, P. Internal-external control of reinforce- ment in smoking behavior, Psychological Reports, 1970, 2g, 562. Homme, L. Perspectives in psychology XXIV control or coverants, the operants of the mind. Psyghological Record, 1965, lg, 501-511. Hunt, W. A. & Matarazzo, J.D. Habit mechanisms in smoking. In W. A. Hunt (Ed.) Learning Mechanisms in Smoking, Chicago: Aldine, 1970. James, W.H., Woodruff, A.B., & Werner, W. Effect of internal and ex- ternal control upon changes in smoking behavior. Journal of Consulting Psychology, 1965, 22, 184-186. Jarvik, M.E. The role of nicotine in the smoking habit. In W. A. Hunt (Ed.) Learning Mechanisms in Smoking, Chicago: Aldine, 1970. Kantrowitz, R. E. Predictor variables and the determination of con- tinuing success in the modification of smoking behavior. Un- published Master's thesis, Michigan State University, 1975. Keppel, G. Design and analysis: A researcher's handbook, Englewood Cliffs, New Jersey: Prentice-Hall, 1973. 45 Keutzer, C.S. Behavior modification of smoking: The experimental investigation of diverse techniques. Behavior Research and Therapy, 1968, p, 137-157. Keutzer, C.S., Lichtenstein, E., & Mees, H.C. Modification of smoking behavior: A review. Psychological Bulletin, 1968, 19, 520-533. Koenig, K.P. & Masters, J. Experimental treatment of habitual smoking. Behavior Research and Therapy, 1965, 2, 235-243. Lawton, M.P. A group therapy approach to giving up smoking. Implied Therapeutics, 1962, fl, 1025-1028. Lawton, M.P. Group methods in smoking withdrawal, Archives of Environ- mental Health, 1967, 11, 258-265. Lefcourt, H.M. Personality correlates of cigarette smoking. Ottawa: Department of National Health andfiwelfare, 1965. Lilienfield, A. M. Emotional and other selected characteristics of cigarette smokers and non-smokers related to epidemiological studies of lung cancer and other diseases. Journal of the National Cancer Institute, 1959, 22, 259-282. Logan, F.A. Self-control as habit, drive, and incentive. Journal of Abnormal Psychology, 1973, 21, 127-136. Mann, L. The effects of emotional role-playing on smoking attitudes. Dissertation Abstracts, 1966, 2g, 4104-4105. Matarazzo, J.D., & Saslow, S. Psychological and related characteris- tics of smokers and non-smokers. Psychological Bulletin, 1960: _5_Z_o 404-413. Mayer, D.Y. A psychotherapist's note on behavior theory. British Journal of Psychiatry, 1970, 115, 429-433. McFall, R.M., & Hammen, C.L. Motivation, structure, and self-monitoring" Role of non-specific factors in smoking reduction. Journal of Consulting and Clinical Psychology, 1971, 21, 80-86. McGuire. J.R., & Vallance, M. Aversion therapy by electric shock: A simple technique. British Medical Journal, 1964, 1, 151-153. Mees, H.L. Placebo effects in aversive control: A preliminary report. Paper presented at the Joint Meeting of the Oregon-Washington State Psychological Associations, Ocean Shores, Washington, May, 1966. 46 Morganstern, K.P., & Ratliff, R.G. Systematic desensitization as a technique for treating smoking behavior: A preliminary re- port. Behavior Research and Therapy, 1969, 1, 397-398. Ober, D.C. The modifications of smokingibehavior. Unpublished doctoral dissertation, uniVersity of Illinois, 1967. Ober, D.C. Modification of smoking behavior. Journal of Consulting and Clinical Psychology, 1968, 22, 397-398. Pierce, R.M., Schauble, P.G., & Farkas, A. Teaching internalization behavior to clients. Psychotherapy: Theory, Research, and Practice, 1970, 1, 217-220. Premack, 0., & Anglion, B. On the possibilities of self control in man and animals. Journal of Abnormal Psychology, 1973, 81 137-151. Pumroy, D., & March, S.A. The evaluation of a cigarette smoking reduction method. Paper presented at the meeting of The Eastern Psychological Association, New York, April, 1966. Raymond, M.J. The treatment of addiction by aversive conditioning with apomorphine. Behavior Research and Therapy, 1964, 1, 287-291. Resnick, J.H. Effects of stimulus satiation on the over-learned maladaptive response of cigarette smoking. Journal of Consulting and Clinical Psychology, 1968, 22, 501-505, (a). Resnick, J.H. The control of smoking behavior by stimulus satiation. Behavior Research and Therapy, 1968, p, 113-114, (b) Rotter, J.B. Generalized expectancies for internal versus external control of reinforcement. Psychological monographs, 1966, §Q_(l, Whole No. 609). Rutner, I.T. The modification of smokinggbehavior through techniques of self-control. Unpublished MA thesis, Wichita State University, 1967. Ryan, F.J. Cold turkey in Greenfield, Iowa: A follow-up study. In W. A. Hunt (Ed.) Learning Mechanisms in Smoking, Chicago: Aldine, 1970. Shriberg, L.D. Intercorrelations among repression-sensitization, extroversion, neuroticism, social desirability and locus of control. Psychological Reports, 1972, 21, 925-926. Sipich, J.P., Russell, R.E., & Tobins, L.L. A comparison of covert sensitization and "nonspecific" treatment in the modification of smoking behavior. Journal of Behavior Therapy and Experi- mental Psychiatry, 1974, 2,-201-203. 47 Smith, G. H. Personality and smoking: A review of the empirical literature. In W. A. Hunt, (Ed.) Learning mechanisms in smoking. Chicago: Aldine, 1970. Smith, R.E. Changes in locus of control as a function of life crisis resolution. Journal of Abnormal Psychology, 1970, 2, 328-332. Staats, 5. Personal communication, June, 1975. Staats, S., & the 1972 Experimental Psychology Class. Internal versus external locus of control for three age groups. International Journal of Agingiand Human Development, 1974, 5. 7-10. Stoller, R.J. Sex andpgender. New York: Science House, 1968. Straits, B.C., & Sechrest, L. Further support of some findings about the characteristics of smokers and non-smokers. Journal of Consulting and Clinical Psychology, 1963, 21, 282. Stuart, R.B., & Davis, B. Slim chance in a fat world: Behavioral control of obesity. Champaign, 111.: Research Press, 1974. Thompson, O.S., & Wilson, T.R. Discontinuance of cigarette smoking: "Natural" and with "therapy". Journal of the American Medical Association, 1966, 12g, 1048-1052. Tooley, J.F., & Pratt, S. An experimental procedure for the extinction of smoking behavior. Psychological Record, 1967, 11, 209-218. U.S. Public Health Service. Smoking and health. Report to the Advisory Committee of the Surgeon General of the Public Health Service. Washington, D.C.: U.S. Department of H.E.W., 1962. U.S. Public Health Service. Smoking and Health. Report to the Advisory Committee of the Surgeon General of the Public Health Service. (Public Health Service publication No. 1103), Washington, D.C.: U.S. Department of H.E.W., 964. Watne, A.L., Montgomery, R.L., & Petit, W.W. A cigarette information program. American Medical Association Journal, 1964, 1§§_ 872-874. Wilde, G.J.S. Behavior therapy for addicted cigarette smokers. Bp;_ havior Research and Therapy, 1964, 2, 107-110. Wolpe, J. Psychotherapy by reciprocal inhibition, Stanford, Calif.: Stanford"University Press, 1958. Yates, A. The application of learning theory to the treatment of tics. Journal of Abnormal and Social Psychology, 1958, §§, 175-182. Typed and Printed in the U.S.A. Professional Thesis Preparation Cliff and Paula Haughey 144 Maplewood Drive : .- East Lansing, Michigan 48823 Telephone (51 7) 337-1 527 HICHIGRN STRTE UNIV. LIBRRRIES Ill 11 111 llllll Ill 111 ill 11 3 9 2 8 12 31037 7 34