ABSTRACT THE EFFECT ON WEIGHT REDUCTION OF COVERANT CONDITIONING THROUGH [X SELF-MANAGEMENT APPLICATION OF THE PREMACK PRINCIPLE By John Joseph Horan The purpose of this study was to determine the effect on weight reduction of coverant conditioning through a self-management appli- cation of the Premack principle. Coverants (a contraction of "covert operants" coined by Homme) are mental behaviors such as thoughts, images, feelings, reflections and so forth. An adaptation of the Premack principle--"For any pair of responses the more probable one will reinforce the less probable one"--provided a methodology through which certain coverants were reinforced. It was hypothesized that counseled clients who received training in a particular type of coverant conditioning would exhibit greater weight loss at the end of an eight-week period than would counseled and noncounseled clients who did not receive such training. Ninety-six female volunteers, mostly coeds between 20% and 30% overweight, were impartially assigned to one of four treatment groups. The first was delayed treatment control; the second was a placebo or "typical" treatment control. In Groups 3 and 4 an attempt was made to increase the frequency of coverant pairs which individual subjects identified as being incompatible with their over-eating habits. Nega- tive coverants involved the undesirable aspects of being overweight John Joseph Horan (e.g. "a shortened life span"). Positive coverants involved the desirable aspects of being properly proportioned (e.g. "clothes fitting better"). Group 3 was exposed to a scheduled coverant treatment designed to determine the necessity of invoking the Premack principle by omitting it as a reinforcement methodology. These subjects were simply told to think of the negative-positive coverant pairs at least seven times a day. Group 4 received training in coverant con- ditioning through a self-management application of the Premack prin- ciple. These experimental subjects were helped in identifying a specific highly probable behavior (i.e. a non-eating activity occur- ring at least seven times a day, such as "sitting down on a partic- ular chair"). They were then instructed to make the emission of this behavior contingent upon the thinking of a negative-positive coverant pair. All subjects were given the opportunity for a free physical exam prior to undergoing counseling. Those in Groups 2, 3, and 4 received a booklet containing information on obesity, a diet plan, and three individual counseling sessions lasting % hour each (the second and third session followed the first by one and eight weeks respectively). Those in Group 1 were told that due to a large number of applicants and a limited staff, their scheduling would have to be delayed. Pre- and posttreatment weights were taken at the same time of day in street clothes minus shoes, on a physician's scale to the nearest quarter-pound. After eight weeks the mean weight losses for Group 1 (Delayed Treatment), Group 2 (Placebo Treatment), Group 3 (Scheduled Coverant John Joseph Horan Treatment), and Group 4 (Reinforced Coverant Treatment) were +0.02, -3.l3, -2.72, and -5.66 pounds respectively. Using an analysis of covariance design with pretreatment weight as the covariate, Group 4 lost significantly more weight than did Group 1 (p <.03). All other pairwise comparisons were not significant. In Groups 3 and 4 the frequency of experiencing coverant pairs ‘was inversely related to weight differential (Pearson r s -.27). Both groups exhibited rapidly declining coverant pair frequency rates during the course of treatment. But on an overall basis, Group 4 experienced significantly more coverants than did Group 3 (p <.OS). Coverant conditioning appears to be a viable therapeutic adjunct. However, the applicability of the Premack principle in a self-manage— ment situation was not established. THE EFFECT ON WEIGHT REDUCTION OF COVERANT CONDITIONING THROUGH A SELF-MANAGEMENT APPLICATION OF THE PREMACK PRINCIPLE By John Joseph Horan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services and Educational Psychology 1970 I \.._ (3 Copyright by JOHN JOSEPH HORAN 1971 TO SUSAN ii ACKNOWLEDGEMENTS A number of people helped make this project possible. Credit should first of all be given to Dr. Lloyd E. Homme. Though I have not met him outside of his writings, Dr. Homme's theoretical inge- nuity gave rise to the experimental treatment here examined. I am deeply indebted to Dr. Richard G. Johnson and Dr. Mary H. Ryan. Dr. Johnson acted as my committee chairman and stimulated my thinking in a number of conceptual and methodological areas. Dr. Ryan served as medical consultant to this study and provided invalu- able technical assistance. Many other faculty members made perceptive suggestions which ultimately improved the quality of this experiment. I would espe- cially like to thank Drs.Robert C. Craig, Gregory A. Miller, Samuel A. Plyler, Norman R. Stewart, Howard S. Teitelbaum, and Dozier W. Thornton. Several graduate students generously donated their counseling time and skills. I am most grateful to Sue Brown, Sandy Ferguson, Lee Leininger, Ken LaFleur, John Malacos, and Alan Traines. My appreciation is also extended to Mary Anderson for her care in the typing of this manuscript, and to Linda Shawver for her proof- reading of the final draft. Finally, I would like to mention my wife Susan, but the effect of her love and quiet confidence cannot be operationally defined. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . vii Chapter I. RATIONALE . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . 1 Purpose . . . . . . . . . . . . 1 Theory and Supportive Research . . . . . . . . . 3 The Problem of Obesity . . . . . . . 3 Learning Theory, a Possible Solution . . . . 6 The Premack Principle and Contingency Management . . . . . . . . . ll Coverant Conditioning and the Treatment of Obesity . . . . . . . . . . 15 Other Behavioral Techniques in the Treatment of Obesity . . . . . . . . 22 II. METHODOLOGY . . . . . . . . . . . . . . . . . . . . 25 Overview . . . . . . . . . . . . . . . . . . . . 25 Sample . . . . . . . . . . . . . . . . . 26 Measures and Materials . . . . . . . . . . . . . 29 Procedures . . . . . . . . . . . . . . . . . . . 31 General Preliminaries . . . . . . . . . . . 31 Treatment Group Rationale . . . . . . . . . 32 Counselor Training . . . . . . . . . . . . . 34 General Counseling Format . . . . . . . . . 36 Specific Treatment Techniques . . . . . . . 38 Hypotheses . . . . . . . . . . . . . . . . . 43 Experimental Design . . . . . . . . . . . . . . 45 Statistical Analysis . . . . . . . . . . . . . . 45 III. RESULTS . . . . . . . . . . . . . . . . . . . . . . 48 Treatment Effects . . . . . . . . . . . . . . . 48 iv Page Chapter Counselor Effects . . . . . . . . . 53 Coverant Frequency and Weight Reduction . . . . 54 Effect of the Premack Principle on Coverant Frequency . . . . . . . . . . . 54 Subject Mortality . . . . . . . . . . . . 56 Status of Research Hypotheses . . . . . . . . . 57 IV. DISCUSSION . . . . . . . . . . . . . . . . . . . . 59 Conclusions . . . . . . . . . . . . 59 Limitations and Implications . . . . . . . . . 64 Summary . . . . . . . . . . . . . . . . . . . . 67 LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . . 70 APPENDICES Appendix A. Letter to Physicians Requesting Referrals . . . . 75 8. Newspaper Announcements . . . . . . . . . . . . . . 76 C. Questionnaire 1 (Application Blank) . . . . . . . . 77 D. Questionnaire 2 . . . . . . . . . . . . . . . . . . 78 E. Questionnaire 3 . . . . . . . . . . . . . . . . . . 82 F. Questionnaire 4 . . . . . . . . . . . . . . . . . . 84 G. Questionnaire 5 . . . . . . . . . . . . . . . . . . 85 H. Treatment Booklet . . . . . . . . . . . . . . . . . 86 I. Scrap Sheet . . . . . . . . . . . . . . . . . . . . 96 J. Display Packet . . . . . . . . . . . . . . . . . . 97 Table II. III. IV. VI. VII. VIII. IX. LIST OF TABLES Relationship Between Proper Weight And Selection Range . Personal And Demographic Characteristics Of The Sample . . . Counselor Case Loads Within Each Treatment Group . Statistics For Regression Analysis With 1 Covariate . . . . . Mean Pre And Post Weights And Weight Losses For Groups 1, 2, 3, and 4 Pre And Post Weights, And Weight Losses Of Subjects In Groups 1, 2, 3, and 4_ . Summary Of The Univariate Analysis Of Covariance . . . . . . All Pairwise Post Hoc Comparisons . Mean Weight Loss Achieved By Subjects Classified According To Counselor And Treatment Group . Subject Mortality Within Treatment Groups . vi Page 28 30 34 46 48 49 50 51 53 56 LIST OF FIGURES Figure Page 1. Pictorial representation of the experimental design . . . . . . . . . . . . . . 46 2. Daily coverant frequencies of Groups 3 and 4 . . . . . . . . . . . . . . . . 55 vii Theories exist in science because they are useful. There is no pretense that they are in any other sense ”correct" (Logan & Wagner, 1965, p. 104). viii CHAPTER I: RATIONALE Introduction Coverant conditioning is a generic term referring to the prin- ciples and procedures underlying the modification of mental behavior. Coverants (a contraction of "covert operants" coined by Homme in 1965) are mental behaviors such as thoughts, images, feelings, reflections, and the like. An adaptation of the Premack principle-- "For any pair of responses, the more probable one will reinforce the less probable one (Premack, 1965, p. l32)"--suggests a methodology through which certain coverants might be reinforced. Two studies (Stuart, 1967 and Harris, 1969) employed different forms of coverant conditioning as part of an overall treatment pro- gram designed to produce weight loss. Both reported considerable success insofar as reducing the weight of the participants was con- cerned. Neither, however, established the efficacy of coverant conditioning as a therapeutic adjunct. Such was not their intent. So in an experimental sense it is still not known whether coverant conditioning was an integral or superfluous component in either of the fairly complex treatment programs. Purpose In contrast to the Stuart and Harris projects, the major con- cern of this experiment was to determine the therapeutic effect of a particular type of coverant conditioning. Overeating was selected as a neurotic paradigm simply out of convenience; the outcome measure of weight loss is indisputable. Hence, the task of devel- oping a comprehensive and marketable method of weight reduction was not undertaken. Should coverant conditioning prove to be a viable therapeutic technique in this study, than similar procedures could easily dove- tail with any existing weight 1033 program. Furthermore, if cover- ant conditioning can be shown to alter maladaptive eating habits, then the implications for using this technique in the treatment of other behavior disorders would be greatly enhanced. Specifically, the primary purpose of this study was to inves— tigate Homme's (1965) contention that coverants of overweight coun- seling clients which are incompatible with overeating can be rein- forced through the use of a self-management application of the Premack principle, thereby effecting weight loss. The success of Homme's approach to the treatment of obesity actually rests upon two interlocking assumptions. 1. A high frequency of theoretically "incompatible" coverants will, in fact, interfere with overeating. 2. The Premack principle is applicable in a self-management situation. Therefore, the secondary purpose of this study was to examine the tenability of these assumptions. It was hypothesized that: l. Counseled clients who receive training in a particular type of coverant conditioning will exhibit greater weight loss at the end of an eight week period than will counseled and non-counseled clients who do not receive such training. 2. The frequency of experiencing incompatible coverants will be inversely related to weight differential. 3. Clients who attempt to reinforce their coverants through the use of a self-management application of the Premack principle will experience more coverants than will clients who attempt to follow a predetermined coverant frequency schedule. Theoryiand Supportive Research The Problem of Obesity Obesity occurs only when caloric intake consistently exceeds caloric expenditure. People differ from each other insofar as their minimum food requirements are concerned; for example, lumberjacks must consume more than fashion models in order to survive. Further- more, the daily caloric needs of a given individual may vary with the climate, his activity level, and a host of other factors. But obesity can occur only when one consistently exceeds his own par- ticular food requirements. Why some individuals consume much more than their energy needs demand, and consequently become obese, is a topic that has been sub- jected to a vast amount of speculation and research. Even a cursory bibliography of the area could comprise a volume in itself. No one doubts that the direct cause of obesity is overeating, but traditional theories about the cause of overeating range from the strictly somatic to the purely psychodynamic points of view. Advocates of the biochemical basis of obesity hypothesis have generally looked to studies of laboratory animals in support of their theories. Genetic determinants of obesity in certain strains of mice are well known (Danforth, 1927; Rytand, 1943; Mayer, 1953). Further- more, hypothalmic lesions surgically induced or produced through massive doses of gold thioglucose can precipitate a tendency to overeat (Brobeck, g£_§g,, 1942; Miller, g£_§l,, 1950; Anliker & Mayer, 1956; Mayer, 1953, 1968). Since the etiology of overeating in some experimental animals can be definitely traced to different and specific somatic sources, organic analogies have been cautiously drawn to cover humans as well (Stunkard, 1959a; Mayer, 1968). A concise but comprehensive account of the physiological fac- tors underlying obesity was published by Mayer in 1968. He used a four-way classification scheme (genetic, hypothalmic, other CNS, and endocrine) to present the numerous known organic origins of obesity in mice, rats, dogs, monkeys, farm animals, and man. In humans no less than thirty somatic sources were detailed. It should be kept in mind, however, that physiological factors cannot cause obesity. At most, they can only predispose some indi- viduals to overeat. And if overweight humans or lower animals (including those with identified organic debilities) can be made to adhere to a diet reasonable to their energy needs, they will in fact lose weight (Hamburger, 1951). Furthermore, the usual case of obesity is rarely complicated by any known biochemical disposition to overeat. It is conceivable that the failure to isolate possible somatic agents might be attri- butable to the relative youth and imprecision of the physiological sciences. Mayer (1968), for example, has suggested that slight, as yet undetectable, organic factors underlying a difference in caloric intake of only a few percent could result in substantial weight gain over the years. However, until these hypothetical entities are made measurable, "a priori" reasoning should not be substituted for EStablished law. The vast majority of overweight individuals are physiologically normal except for occasional complications caused by the obesity itself. Finally, since physiological theories concerning the etiology of overeating have been somewhat less than satisfactory, as might be expected, efforts to treat or control the problem on a strictly biochemical basis have proved to be conspicuously unsuccessful. An increasing number of amphetamine drugs are continually being devel- oped. However, the editors of Consumers Reports (1963) have indi- cated that: ...tolerance to these drugs is easily acquired, and after a period of several weeks or months, even increased doses lose their effectiveness altogether. In one careful five year study...resu1ts in patients treated by diet alone compared favor- ably with results in patients treated with a combination of diet and amphetamine drugs (p. 111). Similar conclusions were reached by Silverstone and Solomon (1965), Lynn (1969), and Ryan (1970). In well documented reviews of the medical literature, several authors (Richardson, 1946; Hamburger, 1951; Mendelson, 1966) have emphasized the irrelevance of organic factors in the etiology of the usual case of obesity. Instead, they have argued that the state of being overweight is a manifestation of neurosis, and have opted for formulations of a psychodynamic nature. But published reports authoritatively listing the dynamic causes of obesity are legion (see, for example, Fenichel, 1945; Shczek, 1955, 1957; Bruch, 1957; Kaplan, 1957; or Simon, 1963). Brosin's (1953) sym- bolism summary typifies this approach: ...the enlarged body may represent a fortress- like defense against a hostile world, a symbol of independence, importance and prowess, an intimidation to enemies, a symbol for a wished- for pregnancy, a means of discouraging suitors, a mask for emotions or a socially acceptable justification for underactivity that permits the person to take fewer risks and thus helps keep anxiety low (p. 975). The dynamic equation of oral frustration and affective starva- tion leading to caloric oversatiation is another recurring etiolog- ical theme. However, this parable and other "post hoc" explanations of the problem have proven to be grossly inadequate insofar as pre- dicting which individuals will become obese. As Stunkard (1959a) noted: Indeed it has not even been possible to define the psychological characteristics of obese persons which will consistently distinguish them from non-obese persons (p. 292). And as far as dynamically oriented approaches to treatment are con- cerned, be they individual or group, intensive or superficial, an oft quoted observation of 1958 rings true today: ...most obese persons will not stay in treatment for obesity. Of those who do stay in treatment most will not lose weight, and of those who do lose weight, most will regain it (Stunkard, p. 79). Since a scientific theory holds no value beyond its functional utility, the impotence of such armchair speculation (of a psycho- dynamic nature) in the generation of effective treatment programs renders it somewhat less than worthless, at least as far as the study of obesity is concerned. Learning Theory,,a Possible Solution The role of food (and hence eating) in the learning laboratory has traditionally been that of a reward or reinforcer (i.e. something which increases the frequency of a preceding response). Food pel- lets have been given to rats that have pulled certain levers, pigeons that have poked particular keys, chimps that have opened desired doors, and humans who have exhibited various and sundry socially appropriate behaviors. So while the reinforcing capacity of food is well known, little attention has been paid to the possi- bility that eating behavior itself may be capable of being reinforced. Through an ingenious series of experiments, Premack (1959, 1962, 1965) has shown that eating and drinking are both reinforcing and capable of being reinforced. In one study (1962) the drinking beha- vior of rats was reinforced by rewarding them with the opportunity to run, and conversely running behavior was reinforced by rewarding the rats with the opportunity to drink. Using children in an earlier study, Premack (1959) showed that the opportunity to operate a pinball machine could reinforce the eating of candy and vice versa. Premack's primary purpose in conducting these experiments was not to shed more light on the study of eating behavior "per se.” But rather he was attempting to reformulate some deeply entrenched, yet unproductive notions about the nature of reinforcement. (Premack's principle of reinforcement will be explored more fully in the following section of this chapter.) In any event, Premack has shown, even if only by implication, that consummatory responses are not solely the result of innate biochemical or psychological drives. Eating and drinking base rates are subject to the same modification by environmental consequences as any other operant pattern. A concise review of other research illustrating the relationship between obesity and environmental factors was presented by Albert Stunkard in a recent seminar sponsored jointly by the Department of Psychiatry and the Institute of Nutrition at Michigan State Univer- sity. Stunkard's talk was an elaboration of a paper he published in 1968 which summarized the contributions of sociology, social psychology, and experimental psychology to the understanding of obesity. In his review of a number of studies, particularly the work of Hollingshead and Redlich (1958), Nisbett (1968), Schachter (1967), Ferster, t l. (1962), Stuart (1967), and himself, Stunkard pointed out that: 1. Parental socioeconomic status and ethnic background are striking examples of social factors whose relationship to obesity is not only correlative, but also causative. (Stunkard's reasoning behind this latter claim is admit- tedly tenuous.) 2. The eating behavior of obese humans parallels that of experimentally obese animals. Environmental influences such as the availability of food, the palatability of food, and the time of day, dictate the eating behavior of overweight subjects to a far greater extent than sub- jects of normal weight. Conversely, with obese subjects as compared to non-obese subjects, there is a complimen- tary decrease in the control of eating by internal fac- tors (in the central nervous system). 3. Behavioral techniques in the treatment of obesity and anorexia nervosa have proved to be highly successful. The indisputable fact that eating behavior can be modified by its consequences provides the basis for a learning theory-oriented explanation and treatment of the obesity phenomenon. To a behavior- ist the usual case of obesity might be defined as nothing more than the end result of a series of maladaptive eating habits. No bio- chemical propensities need be found; no colorful psychodynamic determinants need be invoked. Similarly, the emaciated state of individuals commonly diag- nosed as having anorexia nervosa might also be defined as nothing more than the end result of a different series of maladaptive eating habits. The former syndrome would of course involve overeating, the latter, undereating. (Conceptual parsimony here is not meant to imply that extreme cases in either syndrome are free from serious physical or behavioral disturbance.) Overeating is really a generic term which can include any one of a number of individually identifiable maladaptive eating habits. The following undesirable habits were developed principally from the work of Ferster, _£.;L- (1962) and Stuart (1967). This list is cer- tainly not exhaustive. 1. Storing easily-eaten food 2. Eating rapidly 3. Eating everything on the plate 4. Combining eating with other activities 5. Eating when anxious 6. Eating when lonely or depressed An illustration and explanation of how these individual habits con- tribute to the gaining of weight can be found in the booklet Sgg§g_ and Nonsense About Obesity (see Appendix H). The use of the term "maladaptive" applied to such seemingly innocuous activities as "eating rapidly," might at first blush seem unwarranted. In some cultures, particularly those in which a paucity of food is to be found, the whole spectrum of habits implied by the concept of overeating could be very adaptive indeed. Even in this 10 country the person of normal weight who consistently "eats everything on his plate" is hardly guilty of engaging in a self-defeating acti- vity. However, in view of the indisputable health hazard brought on by excessive overweight (see Hundley, 1955, or Keys, 1955), the seriously obese individual who knowingly sustains his condition by not attempting to extinguish any one or more of these habits is cer- tainly exhibiting "maladaptive" behavior. Insofar as the behavioral treatment of obesity is concerned, identifying the complex interaction of events which precipitated the particular maladaptive habit system of a given individual is not at all essential. ”Post hoc” speculation about possible etio- logical agents (mother or father, positive or negative reinforce- ment, fixed or variable schedules, etc.) is of only academic utility. Premack's work (1959, 1962) implied that the laboratory production of overeating habits would be a relatively simple task which could be accomplished in a number of ways. The behavioral counselor is primarily concerned with the identification of the maladaptive response (problem clarification). From this vantage point he is free to employ a variety of techniques based on learning theory, in an effort to eliminate the undesirable activity. Changes in eating habits, however, are very difficult to effect. One reason the problem exists in the first place is that the aver- sive consequences of overeating (namely, becoming obese and all that is implied by such a condition) are delayed for months or years. The overweight individual is simply not aware of the fact that he has gained weight until a long time after he has overeaten. Even then he is probably not really sure of the particular maladaptive ll eating modes that preceded his obesity. Furthermore, the positive reinforcement for alterning one's undesirable eating behavior is also delayed for a long period of time. The overweight individual does not notice a loss of weight until a long time after he ceases to overeat. And mere habit alter- ation (e.g. eating more slowly) is not sufficient; to lose a sub- stantial amount of weight one must nearly refrain from eating. Dieting causes real and immediate physical discomfort which may not be offset by the miniscule reward of losing ideally a pound or two per week. Finally, unlike alcoholism and other addictions, the source of the obesity problem--food--cannot be entirely avoided. Survival demands repeated exposure to tempting stimuli. To recapitulate, theories are useful only insofar as they pro- vide understanding of a phenomenon and function in a predicting and controlling capacity. Traditional views of obesity as being the' result of biochemical or psychodynamic factors have proved unsatis- factory in every respect. Learning theory has provided a clear and consistent etiological explanation of the eating disorders. Its utility in the formulation of treatment programs will be discussed in the following three sections of this chapter. The Premack Principle and Contingency Management In the science of operant conditioning the construct "positive reinforcement" has traditionally been indistinguishable from the lay concept of a reward. Mice in a Skinner box will pull levers because this activity has been ”reinforced" with meal pellets; 12 house pets will sit on command because the performance of this trick has been "rewarded” with a bit of food. In this sense, a reinforcer has been viewed as someghipg_which can bring about a new behavior (through shaping), or somethi§g_which can modify the frequency of an already existing response. A Reinforcers have been said to obtain their power from biological or psychological needs. In other words, the subject is driven to act in a particular manner because he has observed that the emission of this response will bring about something (the reinforcer) which in turn will reduce a specific need. Since humans have been ascribed more sophisticated needs than lower animals (Maslow, 1943, 1954), intangible things such as a smile or a word of praise can also func- tion as reinforcers. But in the vast array of learning theory exper- iments, the most commonly employed reinforcers are probably food and water. When Premack (1959, 1962, 1965) showed that eating and drinking could not only reinforce other behaviors but were also in themselves capable of being reinforced, he was attempting to render obsolete traditional elements of learning theory. The drive reduction model was seen as inadequate because the discovery of each new reinforcer-- "lights, sounds, puzzles where a moment ago there had been only food (1965, p. l32)"--forced the positing of additional drives. In an exposition on instrumental learning, for example, Morgan (1961) defined reinforcement as: ...the presentation of an incentive satisfying a physiological motive immediately following the instrumental response (p. 684). His physiological definition then had to be expanded to include l3 "curiosity and exploratory drives" when such were found to reinforce other responses. In order to eliminate such conceptual redundancy, Premack (1965) has reformulated the construct of reinforcement: ...reinforcement involves a relation, typically between two responses, one that is being rein- forced and another that is responsible for the reinforcement. This leads to the following generalization: of any two responses, the more probable reSponse will reinforce the less probable one (p. 132). Premack has thus shifted the emphasis from the reinforcing stimulus (the tangible "thing") to the reinforcing event. If food functions as a reinforcer, it does so only because the experimenter has pre- viously denied the subject the opportunity to eat, that is, kept him in a state of partial starvation. Eating has therefore become a highly probable behavior (HPB). The response to be reinforced is called a lowly probable behavior (LPB). Should the opportunity to engage in an HPB be made to depend upon the emission of an LPB, the former will reinforce the latter. Such a process is called "contin- gency management." A hungry rat will press a bar, pull a string, or push a marble into a hole (all LPB's) in order to be fed (an HPB). If the contingencies are managed correctly, then the desired LPB will soon become highly probable. On the other hand, if the probability of eating is very low or zero, as might be the case with a totally satiated subject, then food will no longer function as a reinforcer. Should the contin- gency manager wish to increase the frequency of eating behavior in this case, he must first identify a functional HPB and then make the emission of this activity depend upon the occurrence of eating, 14 now the LPB. Hence, the familiar parental dictum, ”Finish your dinner, then you can go out and play" has found a home in modern reinforcement theory! Homme and his associates have reported a number of studies which successfully employed the Premack principle in the reinforcement of adaptive activities. With a group of nursery school children Homme & Tosti (1965) made the HPB of running down the hall contingent upon the emission of counting and consequently shaped counting behavior. And by using such unanticipated HPB's as pushing the experimenter around the classroom on a chair equipped with casters, a number of active three year olds were trained to sit quietly in a chair, focus on the blackboard, and learn to recognize various letters of the alphabet (Homme, t al., 1963; Homme, 1966a). In another study (Homme, 1966a), a group of high school dropouts completed a course in reading and math by making a coffee or cigarette break depend upon finishing a specified number of programmed learning frames (the LPB). The number of possible HPB's appears infinite, limited only by the contingency manager's ingenuity. One of Homme's associates, for example, noticed that at their first meeting, a young blind girl grabbed his hand and smelled it. The youth had been mute ever since her entry into a mental hospital. Instantly, the researcher recog- nized a functional HPB. He withheld his hand until the girl began to make some sounds. These were soon shaped into syllables, words, and sentences. Within a month the youngster had a 200 word vocabu- lary (Homme, 1966a). Contingency management has also been used in the treatment of 15 anorexia nervosa. Stunkard (1968, 1970) and his associates, for example, observed that such patients exhibit a striking amount of hyperactivity. Hence, they were restricted to their rooms and rewarded with passes only if their weight had increased by a half- pound from the preceding day. Eadlpatient subsequently responded with rapid and consistent gains in weight which continued after their discharge from the hospital. A Homme-Premack analysis would probably label eating as an LPB which soon became highly probable when the opportunity to leave the room (an HPB) was made contingent upon gaining weight. It should now be apparent that any successful example of posi- tive reinforcement can be neatly fitted into Premack's reformula- tion, even though the original researcher may have been operating out of a more traditional frame of reference. But the Premack principle accounts for a much broader range of events. According to Homme (1966b), it implies that "apy_behavior can reinforce apy_ lower probability behavior (p. 234)." In the studies reviewed up to this point, the experimenter has served as the contingency manager. It remains to be seen whether the Premack principle is really applicable in a self-management situation. Coverant Conditionipg and the Treatment of Obesity The human personality consists of cognitive, affective, motor, and somatic behaviors. In years gone by learning theorists have ignored or denied the existence of the first two types of behavior, concentrated on the third, and abandoned the fourth to the science 16 of physiology. Thus there exists a plethora of strategies for the 'production, maintenance, and elimination of maladaptive motor beha- vior, particularly the motor behavior of animals. From whence comes the cynical observation that every white mouse in captivity has, at least potentially, the opportunity to enjoy mental health. Human beings are not so fortunate; unlike the "albus rattus" their problems in living often involve a considerable amount of ideational material (cognitive and affective behaviors). And the behavioral therapies have been severely criticized for tending to ignore this fact. These mental behaviors may be called "coverants," a contraction of "covert operants” coined by Homme (1965). But only the name, not the concept is new; coverants have been previously identified as inner responses, inner states, or private events. Though coverants obey the same laws as operants (Dollard & Miller, 1950; Skinner, 1953, 1957; Homme, 1965), the great weight of learn- ing theory has pressed little for the prediction and control of such ‘mental activity, and focused instead on the enormous tasks of defi- ning these events physically, chemically, and electrically. Two partial exceptions to this generalization are, of course, desensitization and implosion, which deal with imagery at given points in the therapeutic hour. However, all of this covert acti- vity is directly elicited by the counselor. Though there may be some accidental "spill over,’ no provision is made for the client to systematically carry any of these mental behaviors beyond the confines of the counseling office. Learning theory has had little to offer the therapist in terms of modifying the covert activity of an individual engaged in the daily affairs of life. Such would 17 include techniques for altering the frequency of particular thoughts, images, fantasies, ruminations, and so forth. In an important series of papers, Homme (1965, 1966b; Homme & Tosti, 1965) has founded a technology for the control of coverants. The problem of covert response detection was dismissed with an extremely simple observation: Each S is a highly sophisticated computer when it comes to discriminating the occurrence or non- occurrence of a behavior in himself. Whether he is thinking about a chair or thinking about a table is a simple discrimination which he can make with great reliability (Homme, 1965, p. 503). Homme (1965) then suggested that coverant frequency can be increased through a self-management application of the Premack prin- ciple. Simply let the subject himself make the emission of an HPB contingent upon the occurrence of a desired coverant. In Homme's words: If there should be any reason for strengthening this coverant, it can easily be done. All that is required is that S, to whom it is private, demand that it occur immediately prior to the execution of some momentarily high probability behavior (pp. 504-505). In all of the research projects reviewed prior to this point, the experimenter acted in the role of contingency manager. He made the Opportunity to engage in an HPB depend upon the occurrence of the desired response (the LPB). Now Homme has argued that Nature dOesn't care who arranges the contingencies between lower and higher probability behaviors (Homme & Tosti, 1966). The client is Perfectly capable of serving as his own contingency manager. This would apply not only to the manipulation of motor behaviors but also t0 the reinforcement of mental activity as well. 18 The conditioning of coverants has strong theoretical implica- tions for the treatment of a number of personal behavior problems, particularly obesity (overeating). Upon request, most overweight individuals are able to think of a number of reasons for losing weight. These reasons may be highly idiosyncratic insofar as rank order of importance is concerned; they may also be of a negative or positive nature. For example, some overweight individuals may find the thought of "a shortened life span" more aversive (negative) than the idea of "diminished sexual attractiveness." Similarly, others may find the prOSpect of "clothes fitting better" more desirable (positive) than the fantasy of "running up a flight of stairs with- out getting tired." These reasons for losing weight can be viewed as a class of coverants. Because of their anxiety arousing potential they are undoubtedly lowly probable, If the frequency of such coverants which are "incompatible" with overeating (or the state of being obese) can be increased by making an HPB contingent upon their occurrence, then the incidence of maladaptive eating should conse- quently decrease, thereby producing a measurable decline in the Client's weight. Homme has never elaborated on the concept of "incompatibility." Nor has he provided a satisfactory explanation as to why he feels eating behavior itself cannot be used as an HPB. Several theoret- iCal models might be invoked to explain the impact of coverant con- tr01 on motor behavior; the latter point, however, is still research- able. Incompatibility as cognitive dissonance. According to Festinger 19 (1957), dissonance between two cognitions is psychologically dis- ”...the obverse of one ele- tressing. This relationship exists when ment would follow from the other (p. 13)." Individuals experiencing such a disparity are motivated to seek a state of consonance, which can be achieved by making one's behavior consistent with one's thoughts or vice versa. It therefore follows that overweight clients cannot entertain anti-obesity coverants along with ”obverse" cognitions which sanction continued overeating, without experiencing considerable dissonance. The former will either remain lowly prob- able, or if reinforced, will precipitate a change in eating behavior. Incompatibility as reciprocal inhibition. Wolpe & Lazarus (1968) have formally defined the principle of reciprocal inhibition: If a response inhibitory of anxiety can be made to occur in the presence of anxiety evoking sti- muli it will weaken the bond between these sti- muli and the anxiety (p. 12). They have also stated that this principle can be used to overcome responses other than anxiety: For example, when assertive behavior is insti- gated, while the expression of "positive” feelings produces conditioned inhibitions of anxiety, the motor actions involved in such expression inhibit and consequently displace the previous motor habit (p. 13). If some motor behaviors can inhibit other motor behaviors, it is conceivable that certain mental behaviors can do likewise. Thus, if the frequency of coverants antagonistic to overeating can be increased to the point of occurring repeatedly during the day (which implies in the presence of eating stimuli), then the bond between the stimulus to eat and the maladaptive eating response can be weak- ened if not broken. 20 Incompatibility as immediate punishment and reward. It is a well known axiom of learning theory that delayed reinforcement is not as effective as immediate reinforcement (Dollard & Miller, 1950; Reynolds, 1968). In fact, if what is thought to be a reinforcer is withheld for a sufficiently long enough time after the emission of a given response, no reinforcement of this particular activity will take place at all. Perhaps the primary reason why maladaptive eating habits do not extinguish themselves is that the aversive consequences of overeating and the positive reinforcement for refraining from eating are both typically delayed for months or years. Negative coverants pertain- ing to obesity may be viewed as the aversive consequences (i.e. punishment) of overeating. Similarly, positive coverants related to the state of being properly proportioned may be seen as the reward for refraining from eating. If the probability of such coverants can be dramatically increased (by making an HPB contin- gent upon their occurrence), then vicarious experiences of punish- ment for overeating and reward for dieting will occur immediately and frequently each day. Consequently the individual should exhibit much more adaptive eating behavior. Thus a number of psychological models might be invoked in order to explain the impact of coverant conditioning on the reduction of food intake. Each of those described, however, would not preclude the possibility of using eating behavior itself as an HPB. But Homme (1965) has argued that such a procedure would invite adaptation to the aversiveness of the coverants: S will typically report "this (thought) which seemed so horrifying to me at first, doesn't bother me anymore (p. 507)." 21 Therefore, Homme has maintained that a single neutral HPB should be selected from the literally hundreds that are available. Examples might be: combing one's hair, entering a particular room, or sit- ting down on a certain chair. The application of coverant conditioning to the treatment of obesity requires a preparatory step, which in itself may be viewed as a simple weight loss program. The client must first become acquainted with what constitutes a maladaptive eating habit. He should also have a rough idea of the nutritional value of the foods in his diet (i.e. that a pie and coffee work-break may be the caloric equivalent of a steak and a large potato). This information could be condensed and made available to the client in booklet form. Many treatment programs stop after the mere cognitive exchange of such knowledge. For example, the client is presented with a diet plan and simply advised to follow it. The self-control of coverants would enable the client to ”continue treatment," or better yet, would allow him to "manage his own treatment" between visits to the counselor. It might be argued that any successful weight reduction program nmkes at least informal use of a coverant conditioning mechanism. Consider the woman who finds the thought of her doctor's disapproval to be highly aversive. Should she enter treatment for obesity, the dreaded image of "weighing in each week and not losing” occurring on a random basis would undoubtedly influence her daily eating habits. Homme's paradigm forces the systematic, repeated occurrence 0f many such anti-obesity coverants. In addition to the problem of overeating, the coverant 22 conditioning technology can be used to increase the frequency of mental behaviors which are incompatible with cigarette smoking, stuttering, depression, possibly even alcoholism, drug addiction and a number of other self-defeating activities as well. Hark (1970), for example, concluded that in conjunction with the pres- ence of a counselor, Homme's technique was an effective tool for helping motivated groups of clients extinguish their smoking beha- vior. And Davison (1969) noted that a similar procedure enabled a bright youth with inconsistent parents to adaptively control his rebellious behavior. A variety of behavioral programs for the treatment of obesity have appeared in the literature. Operant and respondent techniques have been applied to both overt and covert behaviors with mixed success. But prior to this experiment, the effect on weight reduc- tion of coverant conditioning through a self-management application of the Premack principle has not been systematically examined. Other Behavioral Techniques in the Treatment of Obesity Ferster and his associates (1962) described an operant condi- tioning procedure which was specifically designed to break several of the maladaptive eating habits already mentioned. Yet he pro- Vided no account of his program's effectiveness. Stunkard (1968, 1970), after personal communication with Ferster, reported that his results were essentially no better than those achieved by traditional methods. Building on the work of Ferster, Stuart (1967) devised a number 0f step-by-step ”behavioral prescriptions" for the treatment of . .u. .5 .mo 23 obesity. Both operant and respondent techniques were employed in Stuart's program which, he reported, enabled the clients to obtain self-control over their maladaptive eating habits. Eating rates, for example, were slowed down by instructing the clients to put small amounts of food in their mouths, and to place their utensils on the table until after swallowing. Dramatic weight lossess occurred in all eight clients. However, since no control groups were employed, the efficacy of individual treatment components (inclu- ding that of extensive therapist contact) is not known. Stuart also made use of an aversion therapy technique involving coverants. This process, called "covert sensitization," was origi- nally described by Cautela (1966) as a treatment for compulsive behavior; it consisted of several specific procedures. After being told that the way to eliminate their eating compulsion was to asso- ciate the pleasurable object with an unpleasant stimulus, the sub- jects were trained in deep muscle relaxation. While relaxed they were asked to imagine themselves engaging in eating behavior. This scene was to be followed immediately by the imagination of an aver- sive event. Cautela had his client vicariously experience the sen- sation of vomiting. Stuart used an image of the client's husband seducing another woman. Both authors reported favorably for the technique in their case studies. However, in a controlled experi- ment Harris (1969) found that no additional weight loss could be attributed to covert sensitization. Another aversion therapy technique employing covert behaviors Was tested by Wolpe (1955). He reported the reduction of a food Obsession in a woman by pairing electro-shock and fattening food 24 associations. But in a carefully controlled study, Stollak (1967) closely examined this procedure and noted insignificant effects. Harris (1969) employed a multiplicity of behavioral techniques as part of an overall program designed to produce weight loss. Included was a form of coverant conditioning: ...among the suggestions were the reciting of the list of reasons for losing weight when tempted to overeat, the viewing of an unattrac- tive picture of oneself in a bathing suit when tempted...(p. 265). This procedure violates Homme's recommendation that eating beha- vior not serve as the HPB. Even so, Harris reported considerable success insofar as reducing the weight of her clients was concerned. But in view of the wide variety of factors involved, the relative efficacy (if any) of coverant conditioning was not established. In conclusion, then, it should be noted that behavioral pro- grams for the treatment of obesity have achieved sporadic success. No individual technique has, as yet, been validated. Coverant con- ditioning through a self-management application of the Premack prin- ciple appears to be a very promising procedure. But it too must be subjected to empirical test. CHAPTER II: METHODOLOGY Overview Ninety-six female volunteers, mostly coeds between 20% and 30% overweight, were impartially assigned to one of four treatment groups. The first was a delayed treatment control; the second was a placebo or ”typical" treatment control. In Groups 3 and 4 an attempt was made to increase the frequency of coverant pairs which individual subjects identified as being incompatible with their overeating habits. Negative coverants involved the undesirable aspects of being overweight (e.g. "a shortened life span"). Positive coverants involved the desirable aspects of being properly proportioned (e.g. "clothes fitting better"). Group 3 was exposed to a scheduled coverant treatment designed to determine the necessity of invoking the Premack principle by omitting it as a reinforcement methodology. These subjects were simply told to think of the negative-positive coverant pairs at least seven times a day. Group 4 received training in coverant conditioning through a self-management application of the Premack principle. These experimental subjects were helped in identifying a specific highly probable behavior (i.e. a non-eating activity occurring at least seven times a day, such as "sitting down on a particular chair"). They were then instructed to make the emis- sion of this behavior contingent upon the thinking of a negative- 25 26 positive coverant pair. All subjects were given the opportunity for a free physical exam prior to undergoing counseling. Those in Groups 2, 3, and 4 received a booklet containing information on obesity, a diet plan and three individual counseling sessions lasting % hour each (the second and third session followed the first by one and eight weeks respectively). Those in Group 1 were told that due to a large number of applicants and a limited staff, their scheduling would have to be delayed. Properly proportioned graduate students in counseling conducted the treatment sessions. To facilitate coverant identification, a questionnaire was mailed to all clients in Groups 3 and 4 prior to their first counseling interview. Additional questionnaires were used to obtain frequency ratings for experiencing coverant pairs. Pre- and posttreatment weights were taken at the same time of day in street clothes minus shoes, on a physician's scale to the nearest quarter pound. Sample The experiment was conducted at Michigan State University in tfhe Olin Memorial Health Center during the Spring of 1970. Partic- iPants in the study came from one of two sources: approximately two“thirds of the sample were respondents to announcements of a we”light control counseling program in the University newspaper; the remaining third were direct referrals from Health Center physicians. From an original subject pool of over 200 applicants, 96 females were selected who were roughly between 2070 and 30"'/o overweight. The 27 rationale for studying this weight range was developed, in part, by Stollak (1967, p. 61): ...all individuals more than 20% overweight are required to pay extra premiums on their life insurance policies (Keys, 1955), and over 30% overweight there is an increased probability of there being a physiological etiology and complication to the obesity (Olson, 1964). Furthermore, subjects who were only a few pounds overweight could conceivably reach their goal weight, and then stop losing, long before the scheduled end of the eight week programs. Such an event would confound any meaningful comparison of average weight losses between groups. On the other hand, with subjects considerably more than 30% overweight, there exists a strong likelihood of concomitant emotional disturbance requiring more intensive counseling than was offered. Maleswere excluded from the experiment because too few were available to allow for testing of the possible interactfon between treatment and sex. (Both Stunkard, 1959b, and Harris, 1969 have SUggested that males are more successful than females in losing Weight. ) Table I provides an illustration of the selection range corres- Pmlnding to the proper or goal weight of a given subject. It should be noted that the screening procedure was, in fact, "ulee crude than is immediately apparent. In the first place, the Selection range was subject to the error of misplaced precision. LTP‘iJle the quantities of 20% and 30% overweight can be exactly deter- mil"led from an algebraic equation which includes the variable "proper weight," the difficulty in arriving at a precise estimate of this 28 TABLE I Relationship Between Proper Weight And Selection Range Selection Range Proper Weight 20% Overweight 30% Overweight 100 120 130 105 126 136.5 110 132 143 115 138 148.5 120 144 156 125 150 162.5 130 156 169 135 162 175.5 140 168 182 145 174 188.5 150 180 195 latter concept is almost insurmountable. Mayer (1968) described ISOHe elaborate equipment and procedures which might have helped iJientify each subject's ideal weight, but such technology was finan- c=1ally beyond the scope of this experiment. Hence, the determina- tzilin of the proper weight for each subject was roughly based on an interpolation of her self-estimated goal weight and a "guestimate" (’15 this ideal weight made by physicians or trained personnel using Staudard charts and tables. In most instances, these two quantities were nearly identical. A second factor which occasionally caused a slight stretching 29 of the 20% to 30% qualifying range arose from the delay between the time of the screening weight and the time of the pretreatment weight which was used in the final analysis. The interval here was gener- ally less than a week; however, some borderline subjects gained or lost enough in this time period to place themselves slightly out- side of the selection range. The vast majority of participants in this study were unmarried, coed undergraduates, who had made many previous attempts at losing weight. Table II summarizes the personal and demographic data obtained from the application blanks. Measures and Materials The same physician's scale (with four ounce graduations) was used to measure the subjects' weights before and after treatment. In some instances another physician's scale and a portable scale were also employed. Both of these devices, however, were cali- brated with the standard measuring instrument. A number of questionnaires were distributed to the counseling groups at various times. These were: 1. An application blank requesting personal and demographic data was given to all groups. (See Appendix C.) 2. A pretreatment "fill in the blanks" handout designed to facilitate the identification of coverants was given to Groups 3 and 4. (See Appendix D.) 3. An extra page on Questionnaire 2, which attempted to ascer- tain the frequency of several personal, neutral, highly probable behaviors, was given to Group 4, (See Appendix E.) 4. A posttreatment questionnaire seeking coverant pair fre- quency ratings was given to Group 3. (See Appendix F.) 5. A posttreatment questionnaire requesting frequency ratings of joint coverant-HPB events was given to Group 4. (See Appendix G.) 3O m vouuoaum n vmuuoaom N vouuoaom ofi vouuoamm ma wouuoaom mm bouuoaom m vmuuoamm oq uoz usom A m uoz sumo» om A uoz m uoz umcuo uoz n uoz ooo.mmm A w uoz ucwvsumucoz N usom aw mums» om Ou OH «H ocoz 0 H950 a ooo.m~m ow ooo.-m o mumsvmuo 5 mouse N awash m~ cu AH a umownua nouumocwa N cmowuwfi< nmwcmam «H ooo.omm Ou ooo.oaw Ha acacmm «A 039 cu munch 0H cu 0 on vmuuwfiuoo uoz-uawcam o smaawh o bonowcoz ad ooo.m~w ou ooo.-w NN newcsh 5H 0:0 mm macs» m ou H ma nouuaeaoo -mfiwcam NN oasonumo N backwoz Hm ooo.on ou ooo.ow mm whosonaom 0602 ad was» H v 0H bmwuumz oa ucmumououm ma cowmmoswo a ooo.mm v NN smegmoum unwwmz mcamoa um munaouu< maca>mum mo wobssz suammno mo coHuwuso magnum Hmuwuwz oocmummmum msofiwwamm nacho oficzum mEoocH wawaum wcwbcmum mmwfio covz HouOH c3ovxmmum unmammEoo ufiuwwuquMWmao HH mqm113 The frequency of experiencing coverant pairs in Group 4 will exceed the frequency of experiencing coverant pairs in Group 3. Experimental Design This investigation used an elaboration of the "pretest-posttest control group design" described by Campbell and Stanley (1967). Figure 1 provides a pictorial representation. The pretest (initial weight) was essential because slight deviations from strictly random assignment precluded any assumption of pretreatment equivalence between groups. Statistical Analysis Hypotheses l, 2, 3, 4,,and 5. In order to investigate possible treatment effects an initial decision had to be made between the analysis of variance (ANOVA) and the analysis of covariance (ANCOVA) I = Impartial assignment 01 = Weight at first appointment (pretest) 02 = Weight at second appointment (posttest) The time interval between 01 and 0 was 8 weeks. (X) Control treatment X - Experimental treatment (X)1 02 Delayed Treatment Control Group (X)2 02 Placebo Treatment Control Group (X)3 0 Scheduled Coverant Treatment Control Group X 02 Experimental Treatment Group Figure l. Pictorial representation of the experimental design statistical models. By inspection of the coefficient of determina- tion, sometimes referred to as the square of the multiple r (see Table IV), it was observed that the pretest when used as a covariate accounted for 90% of the variance of the dependent variable. Hence, the ANCOVA was deemed proper. TABLE IV Statistics For Regression Analysis With 1 Covariate Variable Square of Multiple r Multiple r F p level Posttest 0.9037 .9506 703.5996 0.00001 df for Hypothesis 1 df for Error 75 Upon emergence of a significant omnibus F, the Scheffe post hoc technique was employed in order to ascertain which treatment groups differed from each other. The Scheffe method was chosen because it 47 allowed freedom in picking the contrasts while not sacrificing the overall alpha level of .05. Hypothesis 6. The relationship between coverant frequency and weight differential in Groups 3 and 4 was expressed in terms of a Pearson r. The significance of this correlation was determined from standard tables (Hays, 1963). Hypothesis 7. Differences in the number of coverants elicited by Groups 3 and 4 were plotted on a weekly basis for visual inspec- tion. The significance of the expected overall discrepancy was determined by a t test. CHAPTER III: RESULTS Treatment Effects The effect of the four counseling procedures on weight reduc- tion can be seen in Tables V and VI. Average weights (before and after treatment) and average weight losses of each group are pre- sented in Table V. All individual subject scores are presented in Table VI should a closer inspection of the data be sought. TABLE V Mean Pre And Post Weights, And Weight Losses For Groups 1, 2, 3, And 4 Mean Mean Mean Counselinngroup Pre Weight Post Weigpp~ Wepgpt Loss G1: Delayed Treatment 153.80 153.82 +0.02 G2: Placebo Treatment 148.98 145.85 -3.13 G3: Scheduled Treatment 143.26 140.54 -2.72 G4: Experimental Treatment 156.07 150.41 -5.66 No consensus exists in the literature as to the most appropriate method of program evaluation. Success, for example, might be defined in terms of the number of participants losing at least one pound per week, In Group 4, 53% of the subjects met such a criterion. This figure compares quite favorably to the 21% in Group 3, the 20% in GrOup 2, and the 5% in Group 1, who could be considered successful weight losers. 48 49 m~.m + m~.mw~ om.on~ > an.~ : co.ea~ om.mo~ D om.m . om.oo~ oo.ec~ wh.~ n n~.ao~ oo.ho~ a oo.m u om.~a~ om. mgm113 The frequency of experiencing coverant pairs in Group 4 will exceed the frequency of experiencing coverant pairs in Group 3. HA7 confirmed. Subject Mortalipy Of the 96 participants in this study, 16.6% terminated prema- turely. Seven dropped out prior to beginning treatment (i.e. were "no show" at their first and subsequent appointments); nine did not continue after coming to the first or second counseling session. A complete breakdown of subject mortality is presented in Table X. TABLE X Subject Mortality Within Treatment Groups Premature Terminators Counseling Group Original n Prior to During Total Final n Treatment Treatment G1: Delayed 24 2 ‘ 2 22 CZ: Placebo 24 l 3 4 20 G3: Scheduled 24 l 4 5 19 G4: Experimental 24 3 2 5 19 In Group 1 two subjects did not report for the final weighing. One had quit school and had moved out of the area; the other stated 57 that she was no longer interested in losing weight. In Group 2 one subject did not report for the initial interview. After several attempts to reschedule her it became apparent that she did not wish to begin treatment. Three other subjects did not return for the final weighing. One of these had left school early; the remaining two could not be reached for comment. After receiving Questionnaire 2, one subject in Group 3 decided not to begin treatment. Four other subjects did not report for the final weighing. One of these had been hospitalized because of per- sonal problems. Another refused to continue treatment because she felt that the counseling procedure was "humiliating." (The coun- selor recalled that this particular client had insisted that her's was a height, not a weight problem.) The remaining two made them- selves unavailable for comment; "evasion” ploys suggested they did not lose weight. After receiving Questionnaires 2 and 3, three subjects in Group 4 decided not to begin treatment. Two other subjects did not return for the final weighing. One of these maintained that coun- seling was of no benefit to her as her weight problem was not related to food intake, but rather to physiological malfunctioning. The other Subject had left the area prior to the time of her final appointment. Status of Research Hypptheses HA1: After adjustment for initial differences, the posttreatment weight of Group 4 will be lower than the posttreatment weight of Group 1. Confirmed. HA2: After adjustment for initial differences, the posttreatment weight of Group 4 will be lower than the posttreatment weight of Group 2. Not Confirmed. H : A5 A3: A4: A6: A7: 58 After adjustment for initial differences, the posttreatment weight of Group 4 will be lower than the posttreatment weight of Group 3. Not Confirmed. After adjustment for initial differences, the posttreatment weight of Group 3 will be lower than the posttreatment weight of Group 1. Not Confirmed. After adjustment for initial differences, the posttreatment weight of Group 2 will be lower than the posttreatment weight of Group 1. Not Confirmed. The frequency of experiencing coverant pairs will be inversely related to weight differential. Not Confirmed. The frequency of experiencing coverant pairs in Group 4 will exceed the frequency of experiencing coverant pairs in Group 3. Confirmed. CHAPTER IV: DISCUSSION Conclusions Apparently, motivation alone (defined as willingness to volun- teer for treatment), was not responsible for systematic weight loss. In fact, the subjects in Group 1 who volunteered but received no counseling, showed a slight gain of +0.02 pounds at the end of an eight-week period. On the other hand, comparable subjects in Group 4 who received instruction in coverant conditioning through a self-management application of the Premack principle exhibited a substantial weight loss of -5.66 pounds. The difference between these groups was statistically significant (p <.03). But perhaps even more important, over half of the experimental subjects exceeded the practical ideal of losing at least a pound per week. Group 2, the placebo or typical treatment control, showed a moderate weight reduction of -3.13 pounds. However, in comparison with Group 1, this loss was not statistically significant. Hence, the difference between these groups cannot properly be attributed to anything other than random error. It may very well be, though, that the Hawthorne effect, along with possible effects of the "typical treatment" content, contributed to the lower scores of Group 2 subjects. Each of the counselors in this group displayed a warm and supportive interest in their clients' weight problems, and the counseling technique itself was quite credible. 59 60 A similar interpretation can be made for the outcome of Group 3. The average weight loss of the scheduled coverant treatment control was only about 6.6 ounces lower than that of Group 2. Thus, although the posttreatment means of Groups 1, 2, and 3 varied in the expected direction, differences between these groups were not statistically significant. In this experiment, then, the control counseling pro- cedures were not proven to be better than no counseling at all. This finding mirrors the "real world," wherein commonly employed proced- ures (represented by Group 2) and occasional innovations (such as Group 3) rarely achieve success. The experimental subjects in Group 4 lost more weight than any of the control groups; however, differences between Groups 4 and 3, and Groups 4 and 2 were not statistically significant. Such gave rise to the following paradox: u = u ' u u e Mean posttreatment 1 2 3 weight of a given ”2 ” U3 = U4 group after adjust- ment for initial U4 f U1 differences Resolution of this seeming contradiction can be found in the logic of hypothesis testing. Groups 2 and 3 were not proved equal to Group 1, nor were they proved equal to Group 4. Differences between Groups 2 and 3 vs. Group 1, and Groups 2 and 3 vs. Group 4 were in the expected direction; however, they were not of sufficient magni- tude to permit rejection of equality. This combination of insignif- icant differences ("just misses") enabled a statement of inequality between Groups 4 and l to be made with a high degree of confidence. In retrospect, extensive similarity between the experimental counseling treatment (Group 4) and the control counseling treatments 61 (Groups 2 and 3) probably accounts for the failure to obtain signif- icant differences between the effects of these counseling techniques. Certain control procedures (planned similarities) were, of course, necessary in order to evaluate the effectiveness of coverant condi- tioning as a therapeutic adjunct. Not anticipated, however, was the extent to which the control subjects in Group 2 adopted the prin- ciples of coverant conditioning on an informal basis. Several suc- cessful weight losers in this group spontaneously reported using such motivational aids as ”I just kept thinking about how nice I could look on my wedding day" or ”My roommate and I posted a picture ' Also unex- of a fabulously skinny girl on the refrigerator door.‘ pected was the finding that the coverant frequency pattern experi- enced by Group 3 was comparable to that experienced by Group 4 (see Figure 2 on page 55). In spite of this high degree of similarity, the experimental group was the only one of four which lost a significant amount of weight. The probable conclusion, then, is that this loss can be attributed to the fact that Group 4 experienced a high frequency of coverants which were incompatible with overeating. On the other hand, it might be argued that coverant condition- ing was only indirectly effective: The experimental subjects worked harder at losing weight because they found their treatment to be much more intellectually appealing than did the subjects in either of the control procedures. Any- thing new works for a while, but as the novelty wears off, so does the therapeutic effect.. If so, then future research will bear this out. Presently, as an alternate explanation such a position is, at best, tenuous. 62 In Groups 3 and 4 the measured relationship between coverant frequency and weight differential was in the expected direction. Generally speaking, the clients who experienced the greatest number of coverant pairs lost the most weight. However, the Pearson r of -.27 between these two variables was much too slight (not signifi- cant at the .05 level) to draw strong conclusions about the theoret- ical incompatibility of certain coverants with motor behavior. Perhaps the primary reason for a strong relationship not emer- ging is the probable lack of validity in the questionnaires used to obtain coverant pair frequency ratings. In the first place these tabulations were based entirely upon verbal report, as no method exists for counting the private mental experiences of a particular subject without his consent. Failure to lose weight is accompanied by strong disappointment. For a variety of personal reasons, obvious instances of inaccurate reporting did take place. One subject in Group 4, for example, who had gained eight pounds, indicated that she "stopped trying to lose weight" about two weeks after her initial interview. A month long illness, as the story went, caused her to be put on medication with water-retention side effects. Yet on Questionnaire 5 she claimed a very high number of deliberate coverant-HPB experiences throughout the eight-week period. (This inconsistency was not pointed out to her; doing so would probably have increased her already high anxiety level and precipitated another prevarication.) Subject dishonesty, however, can work for the hypothesis as well as against it. A few successful reducers may have attributed their losses to inflated estimates of coverant experiences. 63 But apart from client honesty, the validity of Questionnaires 4 and 5 was also reduced by the difficult nature of the requested task. At the beginning of the second counseling session, the sub- jects were asked how often (on a per day average) they thought of the coverant pairs (Group 3) or made an HPB contingent upon the thinking of a coverant pair (Group 4). Responding to this inquiry was relatively simple. However, the same request was made at the end of the treatment program to cover the intervening seven weeks. Unfortunately, most clients were capable of giving only approximate tallies. This weakness was anticipated, but tolerated, because "in between monitoring" would probably have altered the nature of the self-management concept. Therefore, it appears that inaccurate reporting of coverant experiences, (both intentional and uncontrol- lable) may have seriously affected questionnaire validity, and con- sequently obscured the real relationship between frequency coverant elicitation and weight differential. Although Group 4 experienced significantly more coverant pairs than Group 3 (p‘<.05), both groups exhibited rapidly declining rates. The major differences occurred within the first few weeks of treat- ment. Hence, the sustained self-management applicability of the Premack principle was not demonstrated. In conclusion, then, coverant conditioning appears to be a viable therapeutic adjunct for the treatment of obesity, probably because the experience of "incompatible" coverants is inversely related to weight differential. However, the applicability of the Premack principle in a self-management situation has not been estab- lished. 64 Limitations and Implications In view of the recidivism of the obesity problem (those who are successful at losing weight often regain it), a major criticism of this study might be that no long range follow-up was conducted. Such a possibility, however, was precluded by the transient nature of the student population from which the bulk of the sample was drawn. Furthermore, as stated in Chapter I, the major concern of this experiment was to determine the therapeutic effect of a particular type of coverant conditioning which seemed to have strong implica- tions for the treatment of a number of personal problems in addition to obesity. This was to be an exploratory study, hence the task of developing a comprehensive and marketable method of weight reduction was simply not undertaken. Finally, very few members of the helping professions have had any success whatever in their attempts to effect a weight loss, much less prevent their clients from regaining. Perhaps two distinct research problems are involved here. Another limitation of this study was anticipated by Kiesler in his succinct description of the ”patient uniformity assumption." This "myth" implies that clients "at the start of treatment are more alike than they are different, and will consequently respond to a given type of therapy in a similar manner. On the contrary, overweight people are probably more different than they are alike. To assume that all will react identically to a counseling procedure involving coverant conditioning would be erroneous. The relatively low mortality rate of this study obscures the fact that the experimental treatment was not equally effective with 65 all clients. In fact, three Group 4 subjects showed conspicuous weight gain. Several other subjects who did not lose weight were reluctant to return; it was therefore necessary to conduct the final interview and weighing in their residence. These "therapeutic failures" naturally lowered the average weight loss of the experi- mental group. But the personal and social variables which might have helped identify potentially successful clients were, and remain, unknown. Since the coverant conditioning "ritual" resembles an obsessive neurosis (however artificial), one might speculate that compulsive personality types would be quite amenable to this form of treatment. Future research might also consider the demographic variable of "forced diet" and its relationship to treatment effectiveness. Many of the subjects in this experiment lived in dormitories and could not plan their own meals. It may well be that a substantial frac- tion of the clients who did not lose weight (in all groups) were those who had few, if any, food choice options. Keeping Kiesler's concepts in mind, Hark (1970) divided his sample of cigarette smokers on the basis of the Meyers-Briggs per- sonality types. But none of these categories were differentially more responsive to the coverant conditioning treatment. Theoretical poverty in this area is still the "status quo." Further research is also needed to determine whether the Pre- mack principle is really applicable in a self-management situation. Its role in this experiment was not clear. On one hand, only those subjects who received such training lost a significant amount of weight. But on the other hand, practically speaking, the average 66 number of additional coverant pair experiences attributable to this instruction was relatively small. Perhaps the answer to this puzzle lies in an unanticipated inconsistency between Questionnaires 4 and 5. The subjects in Group 3 were asked how often they thought of the coverant pairs; those in Group 4 were asked how frequently they made an HPB contin- gent upon these mental behaviors. It is quite probable that Group 4 clients experienced many additional coverants which were not connected to a highly probable activity. If so, then these uncounted, random, incompatible thoughts might be responsible for a substantial frac- tion of the total weight loss. Researchers in this area might seriously consider increasing the number of therapist-client contact hours. Many subjects in this study expressed the desire for more counseling sessions (i.e. between the second and final interviews). But to allow such at first seemed to rub against the grain of the self-management idea. Then again, maybe clients would elicit more coverant-HPB events, if they knew beforehand that they would be quizzed about the frequency of these experiences at the end of each week. There is nothing sacred about retaining the concept of self-management; finding ways to improve the treatment is a much more noteworthy endeavor. Homme's (1965) opinion that only non-eating HPB's should be employed is another researchable question. One client noted that the highly probable entry into her bedroom became less probable when it was made contingent upon the thinking of lowly probable coverants. If eating were used as an HPB in this case, than simple mechanics would have solved part of her obesity problem. 67 The nature of the incompatible coverant effect also needs clari- fication. Perhaps if some reliable scale could be devised which would measure "intensity of experience" it might be shown that the notion of incompatibility is really a function of both coverant fre- quency and intensity. If such is true then the weak correlation of -.27 between coverant frequency alone and weight differential (obtained in this study) is certainly understandable. Finally, this experiment employed combinations of negative and positive coverants. It is not known whether these pair components are differentially effective. Essentially, then, coverant conditioning appears to be a promi- sing therapeutic technique. Further research may clarify its impact and increase its effectiveness in the treatment of obesity and other personal problems. Summary The purpose of this study was to determine the effect on weight reduction of coverant conditioning through a self-management applica- tion of the Premack principle. Coverants (a contraction of "covert operants" coined by Homme) are mental behaviors such as thoughts, images, feelings, reflections and so forth. An adaptation of the Premack principle--"For any pair of responses the more probable one will reinforce the less probable one"--provided a methodology through which certain coverants were reinforced. It was hypothesized that counseled clients who received training in a particular type of coverant conditioning would exhibit greater weight loss at the end of an eight-week period than would counseled 68 and noncounseled clients who did not receive such training. Ninety-six female volunteers, mostly coeds between 20% and 30% overweight, were impartially assigned to one of four treatment groups. The first was a delayed treatment control; the second was a placebo or "typical" treatment control. In Groups 3 and 4 an attempt was made to increase the frequency of coverant pairs which individual subjects identified as being incompatible with their overeating habits. Negative coverants involved the undesirable aspects of being overweight (e.g. "a shortened life span"). Positive coverants involved the desirable aspects of being properly proportioned (e.g. "clothes fitting better"). Group 3 was exposed to a scheduled coverant treatment designed to determine the necessity of invoking the Premack principle by omit- ting it as a reinforcement methodology. These subjects were simply told to think of the negative-positive coverant pairs at least seven times a day. Group 4 received training in coverant conditioning through a self-management application of the Premack principle. These experimental subjects were helped in identifying a specific highly probable behavior (i.e. a non-eating activity occurring at least seven times a day, such as "sitting down on a particular chair"). They were then instructed to make the emission of this behavior contingent upon the thinking of a negative-positive cover- ant pair. All subjects were given the opportunity for a free physical exam prior to undergoing counseling. Those in Groups 2, 3, and 4 received a booklet containing information on obesity, a diet plan, and three individual counseling sessions lasting % hour each (the 69 second and third session followed the first by one and eight weeks respectively). Those in Group 1 were told that due to a large number of applicants and a limited staff, their scheduling would have to be delayed. Properly proportioned graduate students in counseling conducted the treatment sessions. To facilitate coverant identification, a questionnaire was mailed to all clients in Groups 3 and 4 prior to their first counseling interview. Additional questionnaires were used to obtain frequency ratings for experiencing coverant pairs. Pre- and posttreatment weights were taken at the same time of day in street clothes minus shoes, on a physician's scale to the nearest quarter-pound. After eight weeks the mean weight losses for Group 1 (Delayed Treatment), Group 2 (Placebo Treatment), Group 3 (Scheduled Coverant Treatment), and Group 4 (Reinforced Coverant Treatment) were +0.02, -3.l3, -2.72, and -5.66 pounds respectively. Using an analysis of covariance design with pretreatment weight as the covariate, Group 4 lost significantly more weight than did Group 1 (p<:.03). All other pairwise comparisons were not significant. In Groups 3 and 4 the frequency of experiencing coverant pairs was inversely related to weight differential (Pearson r a -.27). Both groups exhibited rapidly declining coverant frequency rates during the course of treatment. But on an overall basis, Group 4 experienced significantly more coverants than did Group 3 (p‘<.05). Coverant conditioning appears to be a viable therapeutic adjunct. However, the applicability of the Premack principle in a self-manage- ment situation was not established. LIST OF REF ERENC ES LIST OF REFERENCES Anliker, J. & Mayer, J. An operant conditioning technique for studying feeding-fasting patterns in normal and obese mice. Journal of Applied Physiology, 1956, 8, 667-670. Brosin, H. W. The psychology of overeating. New England Journal of Medicine, 1953, 248(23), 974-975. Brobeck, J. R., Tepperman, J., & Long, C. N. H. The effect of experimental obesity on carbohydrate metabolism. Yale Journal of Biology and Medicine, 1942-43, l5, 893-903. Bruch, H. The importance of overweight. New York: Norton, 1957. Campbell, D. T. & Stanley, J. C. Experimental and quasi-experimental designs for research. Chicago: Rand McNally & Co., 1967. Cautela, J. R. Treatment of compulsive behavior by covert sensiti- zation. Psychological Record, 1966, 1p” 33-41. Consumer Reports (Eds.) The medicine show. Mount Vernon, New York: Consumer Union, 1963. Danforth, C. H. Hereditary adiposity in mice. Journal of Heredity, 1927, 18, 153-162. Davison, G. C. Self-control through "imaginal aversive contingency" and "one-downsmanship": Enabling the powerless to accommodate unreasonableness. In J. D. Krumboltz & C. E. Thoresen (Eds.) Behavioral counseling. New York: Holt, Rinehart & Winston, 1969. Pp. 319-327. Dollard, J. & Miller, N. E. Personality and psychotherapy. New York: McGraw-Hill, 1950. Fenichel, O. The psychoanalytic theory of neurosis. New York: W. W. Norton Co., 1945. Ferster, G. B., Nurnburger, J. I., & Levitt, E. B. The control of eating. Journal of Mathetics, 1962, 1, 87-109. Festinger, L. A theory of cognitive dissonance. Evanston, Illinois: Row, Peterson, 1957. 70 71 Hamburger, W. W. Emotional aspects of obesity. Medical Clinics of North America, 1951, 35, 483-499. Hark, R. D. An examination of the effectiveness of coverant condi- tioning in the reduction of cigarette smoking. (Doctoral dissertation, Michigan State University), 1970. Harris, M. B. A self-directed program for weight control: a pilot study. Journal of Abnormal Psychology, 1969, 263-270. Hays, W. L. Statistics for psychologists. New York: Holt, Rinehart & Winston, 1968. Hollingshead, A. B. & Redlich, F. C Social class and mental ill- ness. New York: Wiley, 1958. Homme, L. E. Perspectives in psychology: XXIV Control of cover- ants, the operants of the mind. Psychological Record, 1965, 15, 501-511. Homme, L. E. Contingency management. Unpublished manuscript, Westinghouse Research Laboratories, Albuquerque, New Mexico, 1966. (a) Reviewed by Hark (1970). Homme, L. E. Contiguity theory and contingency management. Psycho- logical Record, 1966, $6, 233-241. (b) Homme, L. E., C. deBaca, P., Devine, J. V., Steinhorst, R., & Rickert, E. J. Use of the Premack principle in controlling the behavior of nursery school children. Journal of the Experimental Analysis of Behavior, 1963, 6(4), 544. Homme, L. E. & Tosti, D. T. Contingency management and motivation. Journal of the National Society for Programmed Instruction. 1965, 3, 14-16. Hundley, J. M. Need for weight control programs. In E. S. Eppright, P. Swanson, & C. A. Iverson (Eds.) Weight control. Ames, Iowa: Iowa State College Press, 1955. Pp. 1-17. Kaplan, H. I. & Kaplan, H. S. The psychosomatic concept of obesity. Journal of Nervous and Mental Disease, 1957, 125, 181-201. Keys, A. Weight changes and health of men. In E. S. Eppright, P. Swanson, & C. A. Iverson (Eds.) Weight control. Ames, Iowa: Iowa State College Press, 1955. Pp. 108-119. Kiesler, D. J. Some myths of psychotherapy research and the search for a paradigm. Psycholggical Bulletin, 1966, 65(2), 110-136. Logan, F. A. & Wagner, A. R. Reward and punishment. Boston: Allyn & Bacon, 1965. 72 Lynn, E. Drug abuse. Seminar sponsored by the Michigan Rehabili- tation Association, October, 1969. Maslow, A. H. A theory of human motivation. Psychological Review, 1943, 599 370-396. Maslow, A. H. Mbtivation and personality. New York: Harper, 1954. Mayer, J. Genetic, traumatic, and environmental factors in the etiology of obesity. Physiological Review, 1953, 33, 472-508. Mayer, J. Overweight, causesyycost, and control. Englewood Cliffs, New Jersey: Prentice-Hall, 1968. Mendelson, M. Psychological aspects of obesity. International Journal of Psychiatry, 1966, 2, 599-610. Miller, N. E., Bailey, C. J., & Stevenson, J. A. F. Decreased hunger and increased food intake in hypothalmic obese rats. Science, 1950, 112, 256-259. Morgan, C. T. Introduction to psychology. (2nd ed.) New York: McGraw-Hill, 1961. Nisbett, R. E. Determinants of food intake in obesity. Science, 1968, 159, 1254-1255. Olson, R. E. Obesity. In H. G. Conn (Ed.) Current therapy. Phil- adelphia, Pennsylvania: W. B. Saunders, 1964. Pp. 307-312. Premack, D. Toward empirical behavioral laws: 1. Positive rein- forcement. Psychological Review, 1959, 99(4), 219-233. Premack, D. Reversibility of the reinforcement relation. Science, 1962, 136, 255-257. Premack, D. Reinforcement theory. In D. Levine (Ed.) Nebraska symposium on motivation. Lincoln, Nebraska: University of Nebraska Press, 1965. Pp. 123-180. Reynolds, G. S. A primer of operant conditioning. Glenview, Illinois: Scott, Foresman & Co., 1968. RiChardson, H. B. Obesity as a manifestation of neurosis. Medical Clinics of North America, 1946, 39, 1187-1202. Ryarh PL Personal communication, 1970. 'thand, D. A. .Hereditary obesity of yellow mice. Proceedings of the Society for Experimental Biology and Medicine, 1943, 54, 340-341. 73 Schachter, 8. Cognitive effects on bodily functioning: Studies of obesity and eating. In D. C. Glass (Ed.) Neurophysiology and emotion. New York: Rockefeller University Press, 1967. Pp. 117-144. Silverstone, J. T. & Solomon, T. The long-term management of obesity in general practice. British Journal of Clinical Practice, 1965, 12(7), 395-398. Simon, R. I. Obesity as a depressive equivalent. American Medical Association Journal, 1963, 183(3), 208-210. Skinner, B. F. Science and human behavior. New York: Macmillan, 1953. Skinner, B. F. Verbal behavior. New York: Appleton-Century- Crofts, 1957. Stollak, G. E. Weight loss obtained under different experimental procedures. Psyghotherapy: Theorynyesearch and Practice, 1967, 4(2), 61-64. Stuart, R. B. Behavioral control of overeating. Behavior Research and Therapy, 1967, 5; 357-365. Stunkard, A. J. The management of obesity. New York State Journal of Medicine, 1958, 58, 79-87. Stunkard, A. J. Eating patterns and obesity. Psychiatric Quarterl , 1959, 33, 284-295. (a) Stunkard, A. J. The results of a treatment for obesity. Archives of Internal Medicine, 1959, 103, 79-85. (b) Stunkard, A. J. Environment and obesity: Recent advances in our understanding of regulation of food intake in man. Federation Proceedings, 1968, 21, 1367-1373. Stunkard, A. J. Obesityyand the social sciences. Seminar sponsored by the Department of Psychiatry and the Institute of Nutrition at Michigan State University, February 25, 1970. Suczek, R. F. Psychological aspects of weight reduction. In E. S. Eppright, P. Swanson, & C. A. Iverson (Eds.) Weight control. Ames, Iowa: Iowa State College Press, 1955. Pp. 108-119. Suczek, R. F. The personality of obese women. American Journal of Clinical Nutrition, 1957, 5(2), 197-202. wOlpe, J. Reciprocal inhibition as the main basis of psychothera- peutic effects. Archives of Neurology and Psychiatry, 1955, 12, 205-226. 74 Wolpe, J. & Lazarus, A. A. Behavior therapy_techniques. New York: Pergamon Press, 1968. APPENDI CES APPENDIX A LETTER TO PHYSICIANS REQUESTING REFERRALS To: From: LETTER TO PHYSICIANS REQUESTING REFERRALS MICHIGAN STATE UNIVERSITY Health Center INTER-DEPARTMENTAL CORRESPONDENCE Health Center Staff Date: February 24, 1970 Mary H. Ryan, M. D. Subject: see below For his doctoral dissertation in counseling, J. J. Horan is testing the application of a learning theory derivative in the treatment of obesity. This study will dovetail nicely with our existing weight loss programs. Please send prospective patienusto Office 1 for the application-questionnaire. Though all will be treated, we are particularly interested in obtaining 100 females who are between 20% and 30% overweight. It is essential that the applicants receive no other treatment besides that offered by the program, and that they remain as naive as possible (i.e. that they do not know they are part of a "study.") Thank you very much for your cooperation. 75 APPENDIX B NEWSPAPER ANNOUNCEMENTS Russmn and East European Studios Group will present a film, tonight, 7'00 p.m., 107 S. Kedzie, entitled I\ , 1: 808 Will hold a meeting tonight, 8:30 p.m., Old College Hall, Union. Orches‘is the Modern Dance Club will hold a meeting tonight, 7:00 p.m., Womens |.M., Dance Studio. "Save up to 20% . . .Sell your books rlimctly to students. All-U Book r-rlmnqo in Show Hall Lower . . r .» [mums Monday, March 30th. to ..i. collect books noon to 6 p.m., Tuustfuy through Friday of finals week . . . and during spring term registration (March 30, 31). Winter term we helped students exchange $7 000 vvorth of books: Maybe we - .:~ i.. ‘ \ VOU spring term." Afiflpfl mm, Last chance to reserve a spot in one of the weight loss c0unssling programs being oonductsd st Olin Health Center during the Spring quartsr. Apply tonight in Offios No. 1 betwssn 7 and 8 pm. The Michigan State Management Club will have a meeting, tonight, 7:00 p.m., Teak Room, Epp y Center. Mr. Barry Brown, Dlrsc r State of Michigan Department of Labor will speak on "Labqr Standards and the Working - Mons ights." Snyder Hall duplicate Uidgs club will A Vill‘ltllll control counseling program Will be conducted at Olin Health Center during the spring quarter. Those wishing to participate Imould apply in Office No. 1, tonight between 7:00 and 8:30 pm. present a duplicate bridge tournament tonight, 7:30 . ., Snyder Hall Cafeteria. Master points ‘ will,be awarded. Public invited. "Save up to 20 per csnt — Sell your Thu- Winged Spartans will have a presentation by Roger Odell, Eastern Flight Engineer, about traveling spots, air safety, airlines, and pilot careers. Several films will be shown, tonight, 7:30 p.m., Rm 38 - 39 Union. (,‘nllmi Lilo » the weekly meeting of C.llnl)l15 Crusade for Christ, will meet tonight, 9:13 p.m. in S. Wonders Lounge- Larry Tregoning - captain of the 1965 top-ranked U. of M. basketball team will speak. All wi-iuirrw to attend For rides call i.')l midi: A special stage performance of "The Apple Pie and White Picket Fences Carnival Show" will be presented tonight, 8:00 p.m., Wonders Kiva. The production is presented by MSU Tllnfaltrr .iritl TV majors. Free .itlllll:..~'.ltitl. New Ulllvally Conference will hold gt nit-cling, 7.30 p.m., Room 32, Union, this discussion will concern the Detroit geographical expedition .iiul inslllulu and the program for qnuu. lrlrm Faculty, graduate Situl”l|l> and other interested lllt'llllll‘rs oi the university currurlu nity are welcome. books directly to students. All - U Book'Exchanga in Show Hall Lower Loungs begins Monday, March 30th. We will collect books Noon to 6 p.m., Tuesday through Friday of finals week — And during Spring term registration (March 30, 31). Winter term helped studsms lsxchmge $7,000. worth of books; Maybe we can help YOU Spring term." The Anti ‘ ROTC Committee will meet 8:00 p.m., tonight, Room 34, Union. Learn to Soar. MSU Soaring Club will be flying March 30 and 31. Stop at our boot at registration for rides to slrmfl. Special introductory meeting April 1 — movie — 7:30 p.m., Room 30, Union. For information call Jim, 353-6931 or Bob, 882-3250. TONIGHT: HUBBARD HALL presents a dance - concert featuring BACKSTREET and light: by KALEIDOSCOPE, 8: .m. Hubblrd Hall classrooms, $.50 mission. The graduate History Wives Club will morn tonight, 7:30 p.m., at the home of Mrs. Madison Kuhn, 404 Cowley. East Lansing, Joan Oelslighe will speak on "Tho Peace Corps in Peru.“ Free air, Free newspapers, free people, free woman's liberations literature and free TRASH at the MAN AND NATURE BOOKSTORE. Open 9 am. to 5 pm. on Monday throwh Friday and Ill now open 12 to 6 pm. on Saturday. MAN AND NATURE BOOKSTORE is located on the third floor of the Student Service: Building. On Saturday the elevators don't run so use the stairs at the wsst side of the building. Dr. J. Stlober, School of Labor and Industrial Relations will speak on "Industrial Relations and Manpower ln lsrssl, tonight, 7:30 to 9:30 p.m., this is the final lecture in the Free University series lSRAEL: A Model for Social Change. Thursday, March 12, 1970 F—Tuesday, March 10, 76 1,970 APPENDIX C QUESTIONNAIRE 1 (APPLICATION BLANK) QUESTIONNAIRE 1 Application Blank Name Sex Age _____ Class Local Address Local Phone Home Address Home Phone Height Weight What weight do you think you would like to be? How long have you been overweight? Have you ever seriously tried to lose weight before? How often? Are you married? If so, is your spouse overweight? 25e you dating anyone in particular? Is this person overweight? Are (were) your parents overweight? Mother Father How many brothers and sisters do you have? How many are overweight? What is your approximate family income? per year. Religious preference Ethnic group Applicant's comments (if any): Physician's comments: 77 APPENDIX D QUESTIONNAIRE 2 QUESTIONNAIRE 2 The main purpose of this questionnaire is to get you thinking about why you might want to lose weight. All of your responses are con- fidential and will not be duscussed with anyone other than you and the staff member assigned to your case at the clinic. Please fill it out at your leisure. It is not essential that you answer every item. Try to express your real feelings whenever possible. Bring the completed questionnaire with you to your interview on: 78 79 Please indicate below as many of the negative aspects about being overweight that you can think of. These may range from completely conventional stereotypes such as "shortness of breath" to purely personal fears such as "losing the interest of one's spouse." Several examples are listed. 1. I could die long before my time because of being over- weight. 2. I have a hard time finding clothes to fit. 3. (some person) would be repulsed by the sight of me in a bathing suit. 10. ll. 12. 13. 14. 15. 80 16. Describe the most horrible fantasy or daydream related to being overweight that you can think of: Can you think of any other very uncomfortable experiences (real or imaginary) having to do with being overweight? 17. 18. Now, please indicate as many positive aspects about being properly proportioned that you can think of. Some of these may be just the opposite of the negative aspects you have already listed. Others may be completely different. Sev- eral examples are given. 1. Members of the opposite sex will find me attractive. 2. Perspiration won't be much of a problem anymore. 3. I'll feel better in every way. 10. 81 ll. 12. 13. 14. 15. 16. Describe the most desirable fantasy or daydream related to being properly proportioned that you can think of: Can you think of any other very pleasant experiences (real or imaginary) having to do with being properly proportioned? 17. 18. ON THE LINE IN THE LEFT MARGIN WRITE DOWN EXACTLY HOW OFTEN EACH OF THESE NEGATIVE AND POSITIVE THOUGHTS OCCURRED TO YOU DURING THE PAST 24 HOURS, PRIOR TO YOUR READING OF THIS QUESTIONNAIRE. If these thoughts did not occur during this 24 hour period, use the following code to describe their average frequency: never . once a month once a week once a day twice a day three or more times a day mmaoo‘m APPENDIX E QUESTIONNAIRE 3 QUESTIONNAIRE 3 The following list contains a variety of activities which you may engage in frequently each day, either out of habit or because they are somehow rewarding to you. combing your hair reading textbooks washing your hands putting on a coat using the lavatory leaving a classroom watching television programs entering your bedroom sitting down on a particular chair looking at your reflection in a mirror OQWVO‘ML‘WNI—i H Notice that each activity listed is fairly precise. "Getting dressed" is not listed, but "putting on a coat" is. "Sitting down” in general is not listed, but "sitting on a particular chair (e.g. an office chair or a favorite green recliner)" is. Such a list could be endless! Think of some more activities related to your life as an individual and write them down below. (Do not include eating or behaviors related to eating such as drinking Coke, chewing gum, or smoking cigarettes.) 1. 2. 3. 4. 5. Now, from the combined lists, select 3 activities which you know to occur at least 7 times a day. Activity Column A Col. X Col. Y Col. Z 1. 2. 82 83 In Column A write down the number of times these activities occurred during the past 24 hours. YOU ARE NOW ASKED TO KEEP TRACK OF HOW OFTEN THESE ACTIVITIES OCCUR DURING THE NEXT 3 DAYS. In Column X write down the number of times you engage in each activity during the next 24 hours. Column Y should contain the number of times each activity occurs in the following 24 hour period. Column 2 should indicate the frequency of each behavior in the final 24 hour period. APPENDIX F QUESTIONNAIRE 4 QUESTIONNAIRE 4 Exact answers to the following questions would, of course, be impos- sible. But please try to respond as accurately as you can. During the first week of the program you thought of the "negative and positive" aspect pairs about times per day. Did you continue thinking of these pairs beyond the first week? If no, please answer why on the back of this sheet. If yes, how often (per day average) did you think of these pairs during the second week? sixth week? third week? seventh week? fourth week? eighth week? fifth week? How intensely did you experience the emotions surrounding these thoughts? In other words, how emotionally involved did you get with the idea behind the words? 1 2 3 4 5 very low low medium high very high during the first week? during the fifth week? during the second week? during the sixth week? during the third week? during the seventh week? during the f0urth week? during the eighth week? Did you add new thought pairs that you felt would motivate you? How many? Did you ignore any thought pairs which you felt were no longer moti- vating? How many? 84 APPENDIX G QUESTIONNAIRE 5 QUESTIONNAIRE 5 Exact answers to the following questions would, of course, be impos- sible. But please try to respond as accurately as you can. You selected as your highly probable neutral behavior. During the first week of the program you made this behavior depend upon thinking the "negative and positive” thought pairs about times per day. Did you continue this system beyond the first week? If no, please answer why on the back of this sheet. If yes, how often (per day average) did you think of these pairs during the second week? sixth week? third week? seventh week? fourth week? eighth week? fifth week? How intensely did you experience the emotions surrounding these thoughts? In other words, how emotionally involved did you get with the idea behind the words? 1 2 3 4 5 very low low medium high very high during the first week? during the fifth week? during the second week? during the sixth week? during the third week? during the seventh week? during the fourth week? during the eighth week? Did you add new thought pairs that you felt would motivate you? How many? Did you ignore any thought pairs which you felt were no longer moti- vating? How many? 85 APPENDIX H TREATMENT BOOKLET So You Want To Lose Weight S E N S E A B O U T A N D O B E S I T Y N O N S E N S E 86 J. J. Horan Olin Health Center Michigan State University 1970 87 SOME MYTHS: You may have heard people say: "She eats all day and doesn't gain an ounce." or "Everything he eats turns to fat." Such talk is utter nonsense! What goes into the body is either used up and excreted, or stored "for a rainy day" as fat. You may have heard other people say: "He is fat because he has gland troubles." or "Psychological problems caused her to gain weight." This kind of talk is 99% nonsense! True, some overweight individuals have gland troubles and psychological problems as well. But so do a lot of skinny people! These disturbances do not cause obesity; they only complicate it. Obesity is a condition arising only from eating more than is necessary over a prolonged period of time. SOME FACTS: OBESITY OCCURS ONLY WHEN CALORIC INTAKE CONSISTENTLY EXCEEDS CALORIC EXPENDITURE! People do differ from each other insofar as their minimum daily food requirements are concerned. It's obvious that a lumberjack must consume more than a fashion model in order to survive. Furthermore, the daily caloric needs of a given individual may vary with the climate, his activity level, and a host of other factors. But obesity occurs only when one consistently exceeds his own Particular food requirements. Now you may wonder why some people eat more than they have to, 88 Lazyness, gluttony, or lack of will power have nothing to do with it! Some people eat more than is necessary simply because they have developed any one or more of what can be called "MALADAPTIVE EATING HABITS." We call these habits maladaptive because their end result (obesity) is highly undesirable. We do not always know how or why an overweight individual acquired his own particular system of maladaptive eating habits. All that can be said is that he learned them some time in his life. A hundred different people could have picked up the same bad habit from any one of a hundred different causes. Curiously enough, most of these habits may not even be known to the person who has them. But becoming aware of them is the first step toward breaking them! See how many of these maladaptive eating habits sound familiar to you. Notice that many of them involve eating, not as a result of hunger pangs, but as a response to something else. THE MALADAPTIVE HABITS: l. Non-Stop Nibbling and Easily-Estable Food Storing Non-stop nibbling is a maladaptive eating habit which can't possibly occur without the existence of another equally bad habit, namely, keeping food around the house which doesn't require pre- paration. Many overweight people tend to surround themselves with easily-eatable food. Candy dishes abound; cashew trays overflow. Cupboards are packed with popcorn, pretzels, potato chips, and sundry other munchie-crunchies. Rationalizations are equally as legion. "It's left over from a party" or "I keep it around for visitors" are frequently heard excuses. Yet a simple fact remains: Alcoholics living in wine cellars can't be cured. Neither can overweight people living in food storage bins! Are you a non-stop nibbler? Yes No Do You keep around a lot of easily-estable food? Yes No 89 2. Eating Rapidly Have you ever left a Thanksgiving dinner table feeling comfort- ably full, only to experience later on in the evening the sensation of being more stuffed than the turkey you ate? Many overweight people have a vague awareness of this pheno- menon, which may occur to a lesser extent on a daily basis. MOst, however, haven't the faintest idea as to what causes it. If pressed for an explanation they might even reply "I didn't mean to eat so much" or "I thought I stopped eating when I was full." Both excuses may be true! Overweight people, particularly those on diets, have been observed to eat much more rapidly than those of normal weight. Since it takes about 15 minutes after beginning to eat before one starts to feel the effects of his food, we can say that it takes at least 15 minutes after one stops before he feels the full effects of his meal. Because the amount you eat varies directly with how fast you eat, it's quite possible to unknowingly gorge yourself in a very short period of time. Do you eat rapidly? Yes No 3. Eating:Everything,On Your Plate Using such verbal sledgehammers as "It's a sin to waste food" or "The poor people in Korea would love that potato," many mothers succeed in cultivating the "clean plate" habit among their offspring. Hence many children grow into adulthood still letting the quantity of food placed before them rather than the quality of the hunger within them determine how much they eat. In fact a few unfortunate individuals are so indoctrinated that they suffer severe pangs of guilt if even the smallest morsel of food has to be thrown out. In this age of affluence an astronomical number of automobiles are junked every year. Expensive, still-wearable wardrobes are given away or stored unused simply because of slight changes in style. The United States produces more than four times the amount of food required to sustain its population. It makes no sense to save a few cents worth of over-nourishment and thereby incur the incredible health expense of obesity. (Of course, if cost is a factor, it's always possible to keep the left-overs for another meal, rather than finish them off in a hasty gulp.) Do you have the habit of cleaning the plate, regardless of how hungry you are? Yes No 90 4. Eating As A "Non-Pure" Experience Psychological learning theory tells us that if two events repeatedly occur at the same time, one is apt to become a stimulus for the other. If you frequently eat while you read, thumbing through a text- book may, in itself, become a stimulus for eating. If you snack while watching T.V., the "boob-tube" alone (not your stomach), may send you scurrying for a sandwhich. Do you combine eating with other activities? Yes No 5. Eating When Anxious Psychologists have classified several kinds of anxiety. But for our purposes we can briefly describe the anxious individual as a person who is "up tight" about something. We've all been anxious many times in our lives; it's an uncom- fortable, but inescapable experience. And we all have our own pet ways of reducing anxiety. For example, let's say you're anxious about an exam.' You've been studying for hours and you just can't look at the book any longer. You need to take a break, at least for a while. What do you do? 1. Turn on the T.V. 2. Go to sleep 3. Take a walk 4. Visit a friend 5. Grab a bite to eat All five choices probably sound familiar. You might even combine numbers 3, 4, and 5 by heading down to the Grille for a snack. But did you know that a pie and coffee study break may be the caloric equivalent of a steak and a large potato? Many overweight people frequently choose eating (and not some other activity) as a means of reducing anxiety. Do you eat when you get anxious? Yes No 6. Eating When Lonely Or Depressed The popular idea that fat people are friendly, happy-go-lucky, and without a care in the world is a delusion! True, some overweight people are personally content and 91 socially well adjusted, but many more obese individuals frequently experience the painful feelings of loneliness and depression. Though overweight men and women do not have a corner on the sadness market (those of normal weight can be just as unhappy), many obese individuals do have the maladaptive habit of eating as soon as they begin to feel lonely or depressed. Since eating can be a powerful anesthetic for this kind of mental anguish, a vicious circle can easily develop. Take the case of Kathy: A. Kathy was rejected by a boy. B. She then felt lonely and depressed. C Kathy found (subconsciously) that eating could soothe these painful feelings. D. But this kind of eating caused her to gain weight. E. The additional weight increased Kathy's likelihood of being rejected in the future. ./"‘\ I. ./ The vicious circle could easily be broken at point "C" if only Kathy would find some other way of handling her mental pains. Do you eat when you feel lonely or depressed? Yes No A WEIGHT LOSS ACTION PROGRAM: Losing weight (and keeping it off) requires concentrated effort in two areas: 1. Limiting your caloric intake (dieting) 2. Breaking your maladaptive eating habits After you're down to your goal weight you can begin eating comfort- ably (but sensibly!) again. And as long as you don't resume the bad habits you won't regain the weight. A successful weight control program should employ many of the following procedures. Adopt whatever suggestions you feel will help you lose weight and keep from regaining it. 1. Familiarize yourself with the dietary information contained in the 1000 Calorie Daily Menu Guide. Plan your meals accordingly. 10. 92 Get a good scale. Weigh yourself three times daily. Post the weight range sheet on the wall directly above the scale. Write down the time of day, your condition (e.g. clothed?, after eating?, etc.), and your weight, after each weighing. Remove food from all places in your house except the kitchen. With the possible exception of the foods on List 1 of the 1000 Calorie Daily Menu Guide, toss out, give away, or otherwise dispose of any food which doesn't require preparation. (And don't buy any more!) Reserve one spot in your residence for eating. If possible, eat nowhere else. Slow down your rate of eating. You might try interrupting your meal for two-minute intervals. And don't gulp! Focus on the flavor of the food. Prepare (or take) smaller portions. It's better to get a second helping that you want than it is to clean your plate when you're not really hungry. (Also, if you have to think about getting seconds, then you're not as likely to do so!) Make eating a "pure" experience. Do not combine eating with any other activity. Feeling anxious? Turn on the T.V. Take a walk. Work on a hobby. Try some exercises. Anything but eating! Feeling lonely or depressed? Visit a friend and talk it over. Immerse yourself in some group activity. But don't eat! % WEIGHT “MESH!!! N m w w M M M 6 w M M M m fl fl fl % N N n n fl M B N N m N M M fl 3 M fi % Dag & 13!! Eggfiggion Weighs Dex é Tina Condition Weigh; Do: a Tip; Qggdggiog Whi ht MNJMNJMNE m n MNEMNEMNLMNjMNLMNLMNL 2 1 3 1 4 1 s 1 6 1 7 1 8 1 94 «2...... a a...” .. 5...... .a . .3. so 3. .......... .3. ............ .55.... g “n ......... ‘8 '\— ....... i.‘::. ...o 8. ........ .3 1.....355 ...i .EO g .......... “a — ........... ’O‘E 0g: 3 ‘1! i 1.302.. .8 .1... .EOo.8..aIB..edo..s§.3.8 be .80 25.81.109.81 lags-IRE! 3 cone-sou. .5...“ an... .E0 a . . .A..-.11 .3 2 £11....) Ilauul... 13.8.... 53038159153558... 000.30... «I. m «0..— 111.050... .50 on .............. _ ................... 3m .80 n1 ........ 5.0 v: ......... 85.8 8020 .5090 .A..: u h: _ a h: n. 08.. . ............ ..0..0...< £1.10... .8095 .50 On .......... 10... n ............... .0... .13 .50 «v ......... :1... «1.510 ...E. ...A. £0.30 .50 en ........ 5.0 v: ..... .10 .45.. .50....-m ....0 an ......... A.— u N u .0 0"1... 1.9048013... .3100 50 an A... ..K o 3 _ .............. .511...“— E0 no .30.... B: ..a. h: :8... _ .............. :3 :5 .50 on ....... A. 2.. N u C 00.1 _ ......... .A... .625 3 1| Ado _ .38 .5003 .8130 2.041.... .E0n 15:39.: .0 .50 .. 30.55.98... .Ilns. no.5... 3 coca-sou. «E m «.3 .E0 2 ...... .3 S: ..A :0. :2 0.12.50. E35 0— ....o a :33 .N: . ........ 5.... .3 8...... .ED 3 ...... .8 v: ............. E... x. ...00) £093.... 18.303.050.98 .0 351-321.... .80 n. 321:... Ethan. Staggfi 08:010.; 1095 Q «a: .50 8— .......... a: N. ........ ggmn< .508. ..A..E.1 .~:_IE. .....:u.n._< 5:8 580.88.... 50 {ID 1104881 .320 .0 582.3000 .3 E0 0. 28.5105..- .u... a. seen... 5.1 A. 2.65.) 5:35.325... 009.21.... 1.5 m an... 815.... E g u...» 4.3.... 33.6 88...... 52.6 5...... :8. .208. .8 ..3 a: 158......»- EOIn 5.20 .81....» 4.5.8385... ESE-8533.3 .9 ...6 A ass-.3...- 8...... 8.5. «In .09....on 031000» N «a... 51.... .1018 .25.... .11 4.0.11.1. 0.9.3.3 .1818 .80 .8.— .880 .0818... 19.01 .115. 0.5....» 1.13 58.. .5.“ 8150.0 1.38.83 0935 8411.1 5.10 10. 8 8.51.1.0 El :06: 0110 :30 8.6.8. 0'. 5021!! :88. .83 I588! 60.31188. .11. 6381111881208: 6318.» i... EEIEIEAEB>§8 si§.§l5.0.8%0 63.68.819.03. A§»5.0.38¢.5 $300-$015!— Gig-50.0.8..." 383.80.318.0— Ana‘s-10.1.0: 5g. .nflggtggas Eggsiunxlevgigiu. .3550 18.5 8.0 5.. .10 :50... .1...» :0...» .03qu0 .0...— ..1 v..- .....89& .3813 3.0 .o 355.... «Ed. 05 385.38... 3.30.... no... 60500... .05.—i=0 on. E .8391 61.89.19.850: .1— 058 0.... E 68.. is... 5.... 8.... :8... u... .o a... u... ..o 10.2.... on 2.6:. 5:02 .09... 1... co 5.: 09.5.86 :98 05 E9... 10.8.0. 0n 11.2. .01. 16> 5 1030:. .108 15. 0115 3:0... n=8 3:28 80. 3832.33.58.333385 30.0 33!.— Agassi-.885 6368:1983. O. I-OI-u-O 09.06- HOC—g .Q 0 3 U. J‘.C.II o C“8fin 8 o 083 .51.... 6:32.31... 38:30 1... 2.... ..w .0. 62.8.6... .512. .o .25.... 8.»... 5.0—.— 33.‘ .m.: 05 .0 80.3.5 8.0510 0.1 51.. 3188—060 5 511032 3.0.0 AD 10.2 2F v... :33; 33 -. §< 0.1.0 coca—.5608. o... .. 133.3%.3550538185. III-838.88.... Alli-8.x. .iszuoggi‘fl 3085333838.... 8’8»... 8.0108. .050 ..o .0...) .88 88.3.0882? . 95 .56 389.26 is... :8 ...z .8380 8...... 3.80 2.6 2.. 3188033010903 .81.... I a: 4.8... .8. s .8...» asgoaxlaafluugufiia figufizggagnifilg .22... :5... ...8 . 0 01 «a. . 0 00 m aaaaa ..... 222.3... . 33.3353833319336 .282. .2088... a. 38... :38. £8. $3.A.fia§$§8§ . 5.3.5.3....88683633 ...al...§.a...3.8£:e§ .5. inexcusaapcfieaauso. g.ih39§8un= . 88.38... .88.:583535338! 3 2.6... Stags»... «.8... as... 2.. 5.3 3...... a. .2. 833...» as... .83....28.......8.§88.E§8 .395833.8.3328fiasz 81.80.510.— Ba:8»ae§8385.8.. 8.8.8.... 0.3— .0230 was... coo. APPENDIX I SCRAP SHEET NEGATIVE ASPECTS POSITIVE ASPECTS 1. Obesity 1. Properly Proportioned 2. Overweight 2. Normal Weight 3. Being Fat 3. Being Trim 4. Chubbiness 4. Slimness 5. Extra Weight 5. Well Distributed Weight 6 6. 7 7. 8 8. 9 9. 10. 10. 96 APPENDIX J DISPLAY PACKET DISPLAY PACKET I! 5. 97 "IIII’EII'I'LIEWIIMI