v A STUDY OF THETREATMEM 0F BLINDISMS USING PUNISHMENT AND POSITIVE , ‘ REINFORCEMENT III LABORATORY AND NATURAL ‘ SEII'INGS. . Dissertation for the Degree of Ph. D.— MICHIGA‘N‘ STATE UNIVERSITY r BRUCE B. BLASCH ' 1975 V This is to certify that the thesis entitled A STUDY OF THE TREATMENT OF BLINDISMS USING PUNISHMENT AND POSITIVE REINFORCEMENT IN LABORATORY AND NATURAL SETTINGS presented by Bruce B. Blasch has been accepted towards fulfillment of the requirements for Ph. 0. degree In E] ementary and Special Education Major professor Date May 15, I975 0-7639 Pf ‘5fi = IIIIIIIIIIIIIIIIII 31 ”\a-<. .k—A O o :1: MN ‘\ ‘) ABSTRACT ‘.°i.1 A STUDY OF THE TREATMENT OF BLINDISMS USING PUNISHMENT AND \H/’ POSITIVE REINFORCEMENT IN LABORATORY AND NATURAL SETTINGS By Bruce B. Blasch The purpose of this thesis was to determine if a combination of punishment and positive reinforcement is effective in the reduction of blindisms, to determine the effects of such treatment on concomitant blindisms, and to observe the generalization of these effects to various stimulus situations. In addition, the literature reviewed indicated a need to test effective procedures for controlling blindisms and to introduce variations in the application of the treatment pro- cedures to generate suggestions for procedures in classroom settings. Six subjects (cases) were used each representing an independent experimental study. The subjects ranged in age from l6 to 20, with 1.0. scores from 85 to 134. Of the two males and four females, five of the subjects were blind due to retrolental fibroplasia and one student was blind due to optic atrophy. The visual acuity of the subjects varied from no light perception in either eye to light perception in both eyes. A multiple baseline technique was used with each of the six cases. In three of the cases, an A-B-A-B reversal technique was also Bruce B. Blasch applied. The study of six individual cases facilitated the use of a combination of replication designs (i.e., intrasubject direct replication and systematic replication). The hypotheses were tested by evaluating changes in the frequency and duration of the blindism. The blindisms (sterotypic behaviors) observed in this study were head-rolling, rocking and eye-poking. These behaviors were recorded in terms of frequency and time (duration) of the blindism. The punishment or aversive stimulus used was a screeching sound of chalk on a blackboard. The positive reinforcer was money. The experi- ment took place at the Western Pennsylvania School for the Blind in a combination of three stimulus situations for each subject,a laboratory setting and two regular classrooms. For every subject there was first a period of observation to establish a baseline, followed by periods of treatment or of observation without treatment. In every instance for all six cases, when punishment and positive reinforcement were introduced, there was a marked decrease in the frequency and duration of the blindisms. The trend of the data offered moderate support for the generalization of a reduction of a blindism in a laboratory setting to a reduction of the blindism in a classroom setting, however there was a variability of results. This variability of results was also found in the data testing the hypothesis dealing with the generalization of a reduction of a blindism in one classroom setting to another class- room setting. while some of the cases offered moderate support for these hypotheses, there were instances also of contradictory findings, so that these results must be regarded as inconclusive. Bruce B. Blasch The treatment and reduction of one blindism contributed to a simultaneous reduction in a second untreated blindism in four of the six cases. No data were available in one case and there was an in- crease in the second untreated blindism in the sixth case. The results of a neutral stimulus (CS) paired with an aversive stimulus (US) during initial treatment sessions demonstrated that the CS served as an effective aversive stimulus in repeated treatment sessions. Finally, the results demonstrated that the reduction of blindism in a laboratory setting contributed to a markedly reduced frequency and duration of this blindism over a period of time without further treatment. A STUDY OF THE TREATMENT OF BLINDISMS USING PUNISHMENT AND POSITIVE REINFORCEMENT IN LABORATORY AND NATURAL SETTINGS By Bruce B. Blasch A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Elementary and Special Education 1975 ACKNOWLEDGMENTS How and where to begin expressing appreciation to people who have contributed to my educational progress is difficult to decide. At the outset, I should like to extend my appreciation to the faculty of the Department of Elementary and Special Education, particularly Mrs. Lou Alonso and Dr. Charles Mange, who granted me permission to study with them. It would be virtually impossible to list the names of the many friends and professional associates who were involved in the development of the ideas set forth in this dissertation. In particu- lar, I found each member of my dissertation committee to be extremely understanding, helpful, and patient. I should first like to thank Dr. Edwin Keller, my committee chairman, for his support, direction, and his willingness to listen and to help resolve a dissertation problem and to Dr. Charles Mange who supplied effective guidance in specific problematic parts of the dissertation. I was appreciative of the kindness of Mrs. Lou Alonso who shared her expertise in the initial isolation of the problem in blind children. I would also like to thank Dr. Harvey Clarizio and Dr. Hapkiewicz for guidance and help with the research designs and techniques utilized in this study. I would like to thank the staff of the Western Pennsylvania School for Blind Children who allowed this study to be done in their school and to the students that participated. ii I would also like to extend my appreciation to Mr. Bledsoe, Ms. Holmes, Ms. Cohen, and Ms. King for their technical assistance in the preparation of this manuscript. I am particularly indebted to Mr. Donald Blasch who has been responsible for my professional interest in work for visually impaired individuals and has served as a non-directive force to complete this study. I am also appreciative to Dr. Richard Welsh, Dr. Robert Crouse, and Mr. Robert LaDuke for their candid encouragement. These acknowledgments would not be complete without including a special thank you to my wife, Barbara, and my three sons, Erik, Ian, and Kyle. Their presence was my motivation for completing the final stages of this study. TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES . LIST OF FIGURES Chapter I. INTRODUCTION Purpose Research Questions . Organization of the Remainder of the Thesis II. REVIEW OF LITERATURE Literature Related to Blindisms Literature Related to Similar Mannerisms The Use of Behavior Modification Techniques The Use of Punishment . . . Behavior Modification in Natural Settings . III. METHODOLOGY Hypotheses Subjects . Settings . . Phases of the Studies The Behavior Modification Procedures (The Independent . Variables) . . The Laboratory Setting The Classroom Settings . . Measurement of Emission of Blindisms (The Dependent Variable). . . Observer Training and Inter- Observer Agreement Experimental Procedures . . . . The Laboratory Sessions . The Classroom Sessions Treatment of the Data The Individual Experiments . iv Page ii vi ix Chapter Page IV. RESULTS . . . . . . . . . . . . . . . . . . 53 Introduction . . . . . . . . . . . . . . . 53 Case Study Form . . . . . . . . . . . . . . 54 Case 1 . . . . . . . . . . . . . . . . . 56 Case 2 . . . . . . . . . . . . . . . . . 67 Case 3 . . . . . . . . . . . . . . . . . 75 Case 4 . . . . . . . . . . . . . . . . . 85 Case 5 . . . . . . . . . . . . . . . . . 95 Case 6 . . . . . . . . . . . . . . . . . 104 Discussion . . . . . . . . . . . . . . . ll4 Summary of Findings . . . . . . . . . . . . . l20 V. SUMMARY, CONCLUSIONS, LIMITATIONS, PROBLEMS, AND IMPLICATIONS . . . . .- . . . lZl Summary . . . . . . . . . . . . . . . . . lZl Conclusions . . . . . . . . . . . . . l23 Limitations of the Study . . . . . . . . . . . 125 Problems . . . . . . . . . . l26 Implications for Future Research. . . . . . . . . l27 APPENDIX . . . . . . . . . . . . . . . . . . . . 129 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . l3] Frill}, .II’IIE III: II>IIEIIIIILIII15IIIIEI Iii-IE}! Table mNOSU'I-th 10. ll. 12. 13. I4. 15. LIST OF TABLES Definitions of Positive Reinforcement, Punishment and Negative Reinforcement Experimental Procedure for Case l Experimental Procedure for Case 2 Experimental Procedure for Case 3 Experimental Procedure for Case 4 Experimental Procedure for Case 5 Experimental Procedure for Case 6 Case l::“FrequeNCy and Time Scores fer the Primary Blindism Obtained in each of.the Experimental Phases for each of the Stimulus Situations for Case l . . . . Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case l . Case l: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase Case 1: Means and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase Frequency and Time Scores for the Primary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 2 . Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 2 . . . . . . Case 2: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase . . Case 2: Means and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase vi Page 39 45 47 48 49 50 52 57 58 59 64 68 69 70 74 Table Page l6. Frequency and Time Scores for the Primary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 3 . . . . . . . . . . 78 17. Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 3 . . . . . . . . . . 79 18. Case 3: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase . . 83 19. Case 3: Means and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase . . B4 20. Frequency and Time Scores for the Primary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 4 . . . . . . . . . . 87 2l. Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 4 . . . . . . . . . . 88 22. Case 4: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase . . 89 23. Case 4: Means and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase . . 94 24. Frequency and Time Scores for the Primary Blindism Obtained in each of the Experimental Phases for each of the ‘ Stimulus Situations for Case 5 . . . . . . . . . . 96 25. Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 5 . . . . . . . . . . 97 26. Case 5: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase . . . . . 98 27. Case 5: Means and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase . . lO3 28. Frequency and Time Scores for the Primary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 6 . . . . . . . . . . 106 29. Frequency and Time Scores for the Secondary Blindism Obtained in each of the Experimental Phases for each of the Stimulus Situations for Case 6 . . . . . . . . . . lO7 vii Table Page 30.--Case 6: Means and Percentage Scores for the Primary Blindism by Stimulus Situation and Experimental Phase . . lOB 31. Case 6: Mean and Percentage Scores for the Secondary Blindism by Stimulus Situation and Experimental Phase . . ll3 viii Figure \0 CD \I Ch 01 45 (A) N o o o o o o o o ——l o d d N —l . . LIST OF FIGURES Rocking: Cumulative Frequencies for Case l . Rocking: Cumulative Time in Seconds for Case l Rocking: Cumulative Frequencies for Case 2 . Rocking: Cumulative Time in Seconds for Case 2 Rocking: Cumulative Frequencies for Case 3 . Rocking: Cumulative Time in Seconds for Case 3 Head Rolling: Cumulative Frequencies for Case 4 Head Rolling: Cumulative Time in Seconds for Case 4 . Rocking-Head Nodding: Cumulative Frequencies for Case 5 Rocking-Head Nodding: Cumulative Time in Seconds for Case 5 . . . . . . . . . . . . Rocking: Cumulative Frequencies for Case 6 . Rocking: Cumulative Time in Seconds for Case 6 ix Page 60 6l 71 72 80 Bl 90 9l 99 100 109 110 CHAPTER I INTRODUCTION The most common treatment variables for behavior modification studies have been social reinforcement,--praise and other forms of attention from the teacher,--(e.g., Clark and Walberg, l968; Schutte and Hopkins, 1970; Sibley, Abbot and Cooper, l969: Thomas, Becker and Armstrong, l969; and Ward and Baker, 1968); token economies,--points or counters which can later be exchanged for reinforcers,--(e.g. Glynn, 1970: O'Leary and Drabman, l97l; Packard, 1970; and Tyler and Brown, l968) and contracts with the student that desirable reinforcers will be available at the completion of a specified task (e.g. MacDonald, Gallimore and MacDonald, l970; and Smith, Brethower and Cabot, l969). Punishment or the combination of punishment and positive reinforcement have been studied relatively little in educational settings, partly as a consequence of the stress on positive reinforcement and partly because educators and those working with them are apprehensive lest punishment be badly misused by untrained or insensitive teachers. However, the use of punishment often occurs, although its use has often been maligned and little understood as a method of behavior modification. Because the issue was unsettled with regard to the efficacy of the combined use of punishment and positive reinforcement, and because the use of punishment has not been widely investigated, there was a need to explore the effectiveness of punishment and positive reinforcement. The use of behavior modification in remedying behavioral deficits of children has been demonstrated in numerous instances (Harris, Wolf, and Baer, l964; Allen, Henke, Harris, Baer and Reynolds, l967; Baer and Wolf, 1968; Hart, Reynolds, Baer, Brawley and Harris, 1968; Buell, Stoddard, Harris and Baer, l968). These studies have singled out specific behavior problems for the child under study and have demon- strated that remediation can be achieved systematically. The resulting changes in behavior were rarely questioned and were clearly desirable. However, the quantitative evaluation of con- comitant behavior has been of little concern. The literature reviewed revealed one such study in which Buell, Stoddard, Harris and Baer (l968) made an attempt to deal with the problem of allied social behavior changes while studying a specific motor deficit. Therefore, there was a need to determine whether a reduction in one behavior would be accompanied by a reduction in a second allied behavior. Finally, blindisms (sterotypic movements in blind children: e.g., rocking, unusual movements with the hands, rotating movements of the head, forward tilt of the head and eye poking) are apparent in most, if not all, congenitally blind children. These blindisms are undesirable in that they draw attention to the individual's excep- tionability, but they also have other adverse consequences. They contribute to poor and inefficient work and/or study habits. They cause fatigue, pain, general discomfort and they disrupt communication (Scott, 1969). These behaviors may hinder the social acceptance of blind individuals by drawing attention to their differences. Sighted people may intrepret these behaviors as indications of emotional mal- adjustment or other types of mental disorders. Many authors have offered interpretations of these blindisms (Cutsforth, 1951; Holand. l97l; Lowenfeld, 1964: Smith, Chetnik and Adelson, 1969; Spencer, 1960; and Thurrell and Rice, l970). However, the literature reviewed for this study failed to reveal any systematic and effective procedures for controlling blindisms in the school setting. Therefore, a need existed to explore several variations of treatment procedures which could be used by teachers in classroom situations to effectively reduce blindisms. Purpose Since the combined use of punishment and positive reinforce- ment has been studied relatively little, one purpose of this study was to explore the use of punishment and positive reinforcement in the reduction of blindisms among blind adolescents. As few studies have made an attempt to deal with the effects of the treatment of one symptom on a presumably related symptom, a second purpose of this study was to determine if a reduction in one blindism is accompanied by a reduction in a second blindism. However, the above purposes deal only with principles of behavior modification and it has been pointed out that there is a need to test effective procedures for controlling blindisms. Therefore, a third purpose was to determine if the proposed treatment procedures are effective in controlling blindisms. Finally, a fourth purpose of this study was to introduce variations in the application of the treatment procedures (i.e., punishment and positive reinforcement) to generate suggestions for the application of these behavior modification procedures in a school setting. Research Questions l. Can blindisms be effectively reduced by a combination of punishment and positive reinforcement with short-term treatment procedures applied in a school setting? 2. Will the reduction of blindisms in a laboratory setting lead to a simultaneous reduction of blindisms in classroom settings? 3. Will reduction of blindisms in one classroom setting lead to a simultaneous reduction of the blindisms in other classroom settings? 4. Will a reduction in one blindism lead to a simultaneous reduction in second untreated blindism? - 5. Will a neutral stimulus (CS) paired with an aversive stimulus (US) during initial treatment serve alone as an effective aversive stimulus in a repeated treatment session? 6. Will successful reduction of a blindism in a laboratory setting be reflected in a reduced manifestation of the blindism in classrooms over a period of time without further treatment? Organization of the Remainder of the Thesis Chapter II contains a summary cf the literature, including literature related to: blindness and blindisms; similar mannerisms; behavior modification techniques; the use of punishment; and behavior modification in natural settings. A description of the methodology used in this study is presented in Chapter III. A summary of the results is presented in Chapter IV. A discussion and summary of the study is found in Chapter V. CHAPTER II REVIEW OF LITERATURE Literature Related to Blindisms Many professionals working with visually handicapped indi- viduals have interpreted the presence of blindisms in many ways. One of the prominent interpretations is that of Cutsforth (l95l). He believes that while the normal child's life develops to include an ever-widening range of stimulation, the blind child must find his only stimulation within bodily reach. From the onset of blindness, the blind child constitutes the greatest part of his own environment. Therefore, he finds in himself the stimulation and motivation to action that the seeing child finds in his visual environment. Cutsforth also discussed the nature of the stimulation that the tactual environment provides. He indicates that the greater part of the blind child's environment consists of the self, as soon as it is differentiated from a meaningless mass of clothing and blankets. The body becomes at once the source and the object of stimulation. Thus, the patterns for bodily stimulation and manipulation are estab- lished in the cradle; and it may be counted on that they will persist until other, and equally adequate forms of stimulation are substituted. These acts of automatic self-stimulation to be found among the blind are commonly known under the generic term of blindisms. Children who have a very slight degree of vision usually adopt the blindism that 6 consists of fluttering the hands and fingers before the eyes so that the movement of light and shadow may be observed. Tactual stimulation takes the form of thrusting the fingers into the eyes, nose, or mouth, or manipulating appendages such as ears, nose, lips, or locks of hair. Proprioceptive stimulation is produced through bodily swaying, rolling or tilting the head, arm motion and shoulder shrugs, and exaggerated genuflections. Some believe blindisms to be nervous habits of one sort or another. Supporting such a view, Spencer (l960) discussed the values in taking a blind child for a walk, stating that it helps to cut down on the amount of nervous tension marked by the eye rubbing and other mannerisms that may be seen in some blind children. In an anecdotal report describing two blind children, Spencer notes that "Johnny's tension mounts and he rubs his eye, a habit often exhibited in moments of frustration. However, it is normally outgrown, com- pletely, by the time a blind child reaches school age." Lowenfeld (l964) when discussing blindisms agrees with the lack of stimulation hypothesis and states that it can be easily under- stood that a child who does not get visual stimulation from the out- side world will turn to his own body for stimulation. The blind child will continue to do this if he finds it satisfactory or pleasurable. Can the blind child be kept sufficiently stimulated to avoid development of blindisms? Lowenfeld believes that it would not be wise to attempt such stimulation. The blind child cannot hope to match the many impressions which the seeing child receives from his environment, and the same holds true of his outlets for activity. Therefore, Lowenfeld maintains that it would appear almost normal for the blind child to resort to some self-stimulation. As the blind child grows older, he will be able to develop interest in a greater variety of aetivities, and in due time they will supplant his mannerisms because of their greater power to provide satisfaction. Because of this, although these mannerisms are quite commonly observed among young blind children, they may decrease in the lower grades and pos- sibly disappear almost completely in the high school child and in blind adults. Lowenfeld points out that when a blind person develops some peculiar habits, as seeing people do, the public has been unduly ready to ascribe to his blindness these mannerisms, which otherwise would receive little or no attention or interpretation. Lowenfeld does not believe that much can be done about the mannerisms while the child is quite young and that, in fact, the child who is forced to stop one mannerism may develop another one. But as the child grows older and becomes capable of being active in a variety of ways, more able to control himself by his own will power, and more open to reasoning, the time is ripe for a natural abandonment of these mannerisms. He goes on to say that if the child is provided with opportunities for experiences and is kept busy, if he is encouraged to experience his own will, and if he knows that these mannerisms are not acceptable to others, he will in all likelihood give them up as time goes on. Many parents are concerned about these mannerisms because they fear the mannerisms are a sign of their child's lack of mental capacity. Lowenfeld feels that this is not at all true unless these habits persist, and there are other indications of mental retardation. He believes that the mannerisms as such must be considered as quite "normal." While there have been many interpretations of blindisms, Lowenfeld's (1964) and Cutsforth's (1951) interpretation, indicating a lack of stimulation as the cause, is the most prominent. Even though there has been a great deal of speculation as to the origins of these blindisms, a review of the literature failed to reveal any studies that dealt with the systematic control of these stereotypic behaviors. Literature Related to Similar Mannerisms Many interesting and bold interpretations of habits, not "blindisms" specifically, were reported in a study by Dunlap (1945). He defines tics as "obsessive motor performances such as thumbsucking and fingernail biting, and recurrent movements of an annoying sort, such as jerking the head, twisting the shoulders, making facial gri- maces, etc." Dunlap states that tics are generally symptoms of basic maladjustments, the sources of which are various. He states that in eliminating the tic, there must be, on the part of the patient: an understanding of the habit and its detrimental effects; the ideal of abolishing it and the desire to be free from it: and faithful carrying out of the practice prescribed. He believes that tics are sometimes due to inadequate outlets for certain normal desires such as sex. Spasmodic twisting of the head is an example. Elimination of a tic of this type involves producing the tic voluntarily. Sometimes after 10 a few such sessions, the tic is permanently abolished, but a tic of a different sort takes its place. If other tics develop as substi- tutes for the original one, these should be treated concurrently. Dunlap states, "It may be that the basal maladjustment of the patient is little modified by the abolition of the tic." It has been hypothesized that some ticks may be drive-reducing conditioned avoidance responses, originally evoked in a highly trau- matic situation (Yates, 1958). In such a traumatic situation, intense fear has been aroused and a movement of withdrawal or aggression was made. If the movement produced or coincided with the cessation of the fear inducing stimulus, it acquired strength through reinforcement. On subsequent occasions, through stimulus generalization (including internal symbolization), conditioned fear (anxiety) may be aroUsed, which has then been reduced by the performance of the movement. In this way the tic, eventually elicited by a large variety of stimuli, achieves the status of a powerful habit. It has been demonstrated (Yates, 1958) that animals placed in a highly traumatic situation develop conditioned avoidance re- sponses which apparently reduce the anxiety associated with the original situation and which are highly resistant to extinction. It has been suggested that the tic is an avoidance response arising originally in a highly traumatic situation, especially in childhood. Next, the kind of response evoked may be determined partly at least by the mode of response characteristic of the subject in any stressful situation. In terms of Hullian learning theory (Yates 1958), the reaction potential of the tic at a given moment may be conceived as 11 a multiplicative function of the habit strength (er) of the tic (determined mainly by the number of times it has previously been evoked) and the momentary drive strength of anxiety (D), which fluc- tuates from time to time. Yates reported a successful experiment on the extinction of four tics in a female psychiatric patient of high average intelligence. He based his method of treatment on a theoretical model treating the tic as a simple learned response which has attained its maximum habit strength. His general hypothesis was that massed practice of the tic leads to a significant decrement in the ability of the subject to respond voluntarily, and eventually leads to extinction of the tic by the process of building up a negative behavior pattern of "not performing it." His results confirmed this hypothesis. He described several experiments in detail but the main outcome was that very prolonged periods of massed practice, followed by prolonged rest periods, produced the largest declines. Lovaas, fl (1967) worked with children whose behavioral repertories were restricted to three simple categories: (a) self stimulation (stereotyped, repetitive behavior such as rocking, twirling, spinning, etc.); (b) tantrums, including self-destructive behavior (e.g., head-banging); and (c) vocal output involving mostly vowels, tongue clicking, etc. The method used to establish nonverbal imitation involved a set of successive discriminations. Therefore, the children were positively reinforced (food) for closer and closer approximation to the attending adult's behavior. The procedure was on a one hour a day, five days a week basis, involving sixty behavior items or tasks. 12 One autistic girl, for example, was engrossed in self-stimulatory behaviors for as much as 99% of her day. After one year of training in nonverbal imitation, she demonstrated a preference for engaging in appropriate play (drawing and painting) and the pathological be- haviors were reduced greatly. Macpherson (1967) attempted to apply some of the procedures used in behavior therapy in symptomatic relief of a patient suffering from Huntington's chorea. Prior to training, the patient attempted to reduce involuntary movement in her legs by pressing her hand down on her knee. Generally, the result of this was that involuntary movement was temporarily avoided and the patient's attempt at control was immediately reinforced. However, when the involuntary movement did take place, the marked increase in muscle tone which accompanied it added to the explosive and gross character of the movement. Since the response of increased muscle tone was immediately reinforced it had become established as a habit. The training involved three stages: (1) training in relaxation; (2) training in attending to interoceptive afferent input associated with the onset of involuntary movement; and (3) training in reciprocal inhibition of involuntary movements by deep muscle relazation. The patient showed marked improvement and treat- ment was no longer needed after six weeks.. As pointed out, attention toward a particular motor behavior to be extinguished may, in fact, increase the frequency of the un- desired behavior. Madsen, gt_al._(l968) observed the reinforcing function of "sit-down" commands. After a baseline of standing-up behavior was established, the teachers were instructed to attend more 13 to standing up behavior and this behavior, in fact, tripled. Finally, teachers praised incompatible behavior (sitting in seat) and standing up behavior fell far below the baseline. Therefore, attention to the inappropriate behavior of standing served to increase the frequency of this behavior. Barrett (1962) describes an application of free operant methods to the control of multiple neuromuscular tics in a 38 year old subject. By use of a tape recorder, a positive stimulus (music) could be re- moved or an aversive stimulus (noise) presented when a tic occurred. The contingency arrangement was programmed so that each tic produced a 1.5 second interruption of music. If the patient did not tic for at least 1.5 seconds, he could hear the music until it was automati- cally interrupted by the next tic. This tic-contingent interruption of music by white noise proved to be very effective. The above studies indicate that there are a variety of interpre- tations as to the causes of various stereotypic behaviors (Dunlap, 1945: Yates, 1958). Some of these same behaviors are labeled blindisms when emitted by blind individuals. Yates viewed these behaviors as learned, and successfully controlled them by utilizing massed practice. Lovaas, SELEl;.(1957) positively reinforced appropriate modeling. Barrett (1962) effectively used positive reinforcement and punishment to control multiple neuromuscular tics. Based on the success of the studies cited, it would appear that control of blindisms could also be achieved. 14 The Use of Behavior Modification Techniques During the past few decades an experimental analysis of behavior has produced several powerful and reliable techniques for controlling behavior (Holland and Skinner, 1961). Although these procedures were originally established with lower organisms, research has shown that, if principles of behavior are determined clearly and then applied in a systematic manner to human beings, the final result is a much more effective training program (Watson, 1967). The extension of these procedures to human behavior was made when Lindsley (1956) successfully applied the methodology of operant conditioning to the study of psychotic behavior. Following Lindsley's example, numerous investigators have demonstrated that, in its essentials, the behavior of mentally defective individuals (Orlando and Bijou, 1960), stutterers (Flanagan, Goldiamond and Agrin, 1958) and economically deprived children (Hall, Lund and Jackson, 1968) is subject to the same controls. Data have provided evidence that maladaptive behavior can be modified by the teacher in a variety of situations and contingencies. Hart, Reynolds, Baer, Browley, and Harris (1968) demonstrated the posi- tive effect of adult social reinforcement contingent on the cooperative play of a nonsocial disruptive five year old. Hewett (1967) developed an engineered classroom design in four public school systems and a hospital setting. He stated that "one of the aspects that most impressed observers is the purposeful, controlled and productive atmosphere in the classroom." Allen, Turner and Everett (1970) found that the deciding factor in improving the behavior of children in a Head Start class was the teacher's behavior and the use of appropriate 15 reinforcement techniques. They stated that "successful behavior modification depends on correct teacher-child interaction." A study dealing with the handling of tantrums, irrelevant verbal behavior and baby talk (Zimmerman and Zimmerman, 1962) used attention as the rein- forcer. At the conclusion of this study the subject was working more efficiently in class and was reported to be making good progress. His speech was generally characterized by relevancy and maturity. In working with undesirable behavior, Girardeau and Spradlin (1964), Gordon and Hollis (1965) and Colwell (1966) recommended extinction to eliminate undesirable behavior patterns while simul- taneously using positive reinforcement to develop desirable behavior that would be incompatible with the unwanted behavior. Girardeau and Spradlin (1964) used token reinforcement to eliminate temper tantrums in a retarded girl. They also differentially reinforced verbal be- havior in a girl who had a reputation for continually complaining and had repeatedly alienated herself from institutional members because of this behavior. She was reinforced for making positive statements while extinction was used to eliminate her noxious verbal behavior. Gordon and Hollis (1965) noted there was minimal interaction between attendants and retardates, except when retardates were engaged in undesirable acts, e.g., those who were in imminent danger from self-destructive acts, those who became overtly aggressive, those who soiled themselves, or those who tore or otherwise were destroying their clothing." Such findings have been noted in other studies dealing with both children and adults and the thinking has been that 16 much of the maladaptive behavior was due to reinforcement of this behavior and ignoring of the appropriate behavior. Flanagan, Goldiamond and Azrin (1958) worked with three male stutterers. The S's read from printed pages and every time they stuttered, the experimenter pressed a microswitch and initiated a 30- minute period of response-contingent stimuli. Use of noise as an aversive stimulus was contingent upon stuttering or'could be escaped by not stuttering. When termination of a noxious stimulus was made contingent upon stuttering, stuttering increased. When onset of a noxious stimulus was made contingent upon stuttering, stuttering de- creased. For one 5, stuttering was completely suppressed, and this suppression continued beyond the termination of the aversive con— tingency. Wolfe, Risley, and Mees (1964) used operant conditioning procedures on the behavior problems of an autistic child. The be- havior problems included temper tantrums, not wearing glasses, bed- time problems, verbal problems and eating problems. A combination of mild punishment and extinction, and also successive approximation, were procedures employed on these problems. A follow-up report six months after the child's return home indicated he wore his glasses, had no tantrums, no sleeping problems, and was more verbal. Bandura (1961) who has reviewed much of the literature con- cerned with the direct manipulation of "symptomatic“ behavior, con- cludes: 0n the whole the evidence, while open to error, suggests that no matter what the origin of the maladaptive behavior might be, 17 a change in behavior brought about through learning procedures may be all that is necessary for the alleviation of most forms of emotional disorder. If the teachers' behaviors are 'correct,‘ then the behaviors of the student will become progressively better; that is, the student will be learning. At the same time, the 'correct' behaviors of the teachers will be reinforced by the progress of the student and, thereby, be maintained. Bandura further stated that "more and more cases of non-learning are, simply enough, reinforcement problems." In summarizing "correct" teaching behavior for behavior modification, Larsen (1970) provided further rationale for the use of the techniques. He stated: In any instructional setting, the presence of a reinforcer and its administration contingent upon appropriate responding must be assured . . . very little appropriate student behavior is generated in the absence of reinforcement. Reinforcement is a highly individual thing, and especially in dealing with the multiply handicapped, wide individual differences with respect to what is, and what is not, reinforcing can be expected. For efficiency in teaching, Larsen stressed the need for precise assessment of the behavior of the student prior to his beginning the instructional program and equally precisely, his behavior when he completes the program. The studies reviewed in this section indicate that behavior modification techniques have been successfully used to control a variety of undesirable behaviors. While none of the studies dealt specifically with stereotypic behaviors in blind children some of the blindism-like behaviors were controlled in other exceptional children. The Use of Punishment Several behavioral techniques can be used to reduce the rate of an undesirable behavior pattern. Generally, the best known of 18 these techniques are extinction, reinforcement of an incompatible response, and punishment. Extinction technique involves the discon- tinuation of reinforcement for a response. Conditioning one or more responses that are incompatible with the undesirable response repre- sents the incompatible response procedure. Punishment may involve either of two procedures: "the presentation of an aversive stimulus, such as electric shock, following the response. The second involves the removal of a positive reinforcer, such as food, following the response" (Vakelich and Hake, 1971). These three behavioral techniques can be used in combination to increase the likelihood of eliminating the undesirable response, i.e., undesirable behavior patterns are extinguished and punished while incompatible responses are reinforced. For many years theorists avoided the use of punishment re- garding it as unpredictable and likely to produce undesirable side effects (Kanfer and Phillips, 1970). The use of punishment has also been studied relatively little in educational settings, partly as a consequence of the stress on positive reinforcement, ethical consider- ations, and the realization that punishment could be badly misused by untrained or insensitive teachers. Recent research has clarified parameters governing the diverse effect of punishment, and resulted in a re-examination of its supposed harmful side effects (Kanfer and Phillips, 1970). Reviews of the punishment research such as those by Church (1963), Solomon (1964), Azrin and H012 (1966), Boe and Church (1968), Campbell and Church 19 (1969) and Brush (1971), have fostered renewed investigation and research with punishment. Punishment has often been the procedure of choice because the reinforcement history and current maintenance of the undesired behavior was unknown. Without such knowledge the use of extinction or the use of the same reinforcer with an incompatible alternative behavior is not possible. The punishment procedure using an aversive stimulus has been shown to be effective with responses that need to be reduced immediately (Bucher and Lovaas, 1967; Tate and Baroff, 1966; Lovaas and Simmons, 1969), and responses that cannot be eliminated with any other procedure (Risley, 1968). Punishment studies utilizing the procedure of removing posi- tive reinforcers have also eliminated undesirable behavior patterns (e.g., Wolf, £E_El;: 1964; Zeilberger, Sampan and Sloane, 1968; Bostow and Bailey, 1969; Sloane, gt_al;, 1967). However, punishment by means of a reduction in positive reinforcement was not as effective as the punishment procedure using an aversive stimulus in eliminating un- desirable climbing behavior (Risley, 1968). Lovaas and Simmons (1969) found punishment by a reduction in positive reinforcement to be effec- tive in eliminating self-destructive responses, but it took so long that its practical value in preventing physical injury was questionable. The incompatible response procedure has a similar drawback. Lovaas, Freitag, Gold and Kassorla (1965) and Peterson and Peterson (1968) have shown that this procedure can reduce self-destructive behavior, but 20 the time required to condition incompatible responses makes the procedure less practical than punishment by using an aversive stimulus. H012 and Azrin (1963) have listed four criteria on which to measure the effectiveness of procedures designed to eliminate be- havior: the immediate effect, the enduring effect; whether or not the effect is reversible; and whether or not suppression is complete. They compared the following procedures which reduce response ratio: stimulus change; extinction; satiation; physical restraint; and punish- ment. Overall, punishment was the most effective. Therefore, neither the incompatible response procedure nor punishment by the removal of a positive reinforcer are as fast or as effective as punishment by an aversive stimulus. Among the parameters that influence the effectiveness and durability of punishment discussed by Kanfer and Phillips (1970) are the stimulus intensity and contingency schedule, the immediacy and frequency of punishment, the possibility of undesired escape responses, the absence of positive reinforcement as a predictable sequel to punishment, and simultaneous development of alternative desirable behaviors which produce positive reinforcement. "Available evidence does not support earlier suspicions that punishment and other oper- ations with aversive stimuli need have harmful side effects, when the procedures are properly constructed and applied to suit the indi- vidual circumstances" (Kanfer and Phillips, 1970). Investigators who have used aversive control have reported the effects on other areas of behavior as well as on the target behavior. Risley (1968) monitored a number of other behaviors while 21 punishing climbing by a shOck, and autistic rocking by shouting and shaking, in a six-year-old girl. Punishment was applied in the laboratory and at home, both by the investigator and by the mother. In both cases, the target behaviors were rapidly eliminated. Among the side effects noted were increases in similar be- havior (e.g., climbing on a chair) when the target behavior patterns were suppressed (i.e. climbing a book case). However, when this response was also punished, no other substitute appeared. No general avoidance of the situations or of punishing agents was reported. Instead, avoidance responses were highly specific. No other be- havior patterns were suppressed, nor did any aggressive behavior occur. Despite the stimulus control exerted bythe investigator over response rate, the girl increased her eye contact with him after punishment began, thereby enhancing other training activities, and she otherwise behaved no differently with the experimenter. Risley concludes: The most significant side effect was the fact that eliminating climbing and autistic rocking with punishment facilitated the acquisition of new desirable behaviors. . . . Some deviant be- haviors, maintained by unknown variables, interferred with the establishment of new behaviors. This interference was not pri- marily due to a physical incompatibility between the behaviors. This interference, which might be termed 'functional incompati- bility,‘ suggests that the elimination of such deviant behaviors may be a necessary prerequisite to the establishment of new behaviors (pp. 25-26). Bucher and Lovaas (1968) report similar results for their clinical use of punishment. After self-destructive behaviors were suppressed by shock and generalization was promoted by use of several punishing agents in several environments, their autistic boy exhibited less avoiding of adults and less Crying. f.Eliminating.previously 22 necessary physical restraints, he also rapidly developed many desirable behaviors. Comparable positive side effects were noted in other children. The generalization of a response suppression produced by punish- ment has been dealt with in several studies (Corte, Wolf, and Locke, 1971; Azrin, 1956, Brethower, and Reynolds, 1962). When punishment is associated with one set of "safe" stimuli (SA) (Azrin, 1956), in some cases, a contrast effect may boost responding in the safe context (SA) above what it had been before treatment (Brethower, and Reynolds, 1962). Birnbauer (1968) provides a clinical illustration of the effects of punishment so highly discriminated as to thwart therapeutic goals. The subject, an adolescent retardate, had to be kept in re- straints constantly because of his biting and other destructive acts. In laboratory sessions intense shock was contingent on.destructive‘. acts, and the specific target behaviors were quickly eliminated. How- ever, verbal warnings, paired with shock in an effort to make them conditioned aversive stimuli, were effective only when spoken by the person who had actually administered shock (SD). Concurrent attempts to reduce another destructive act (i.e. napkin-tearing during meals) by reinforcing competing responses and time out contingent upon the target behavior were tried. These procedures had no effect. Only when shock was administered for this specific response was it sup- pressed. Birnbrauer concluded that application of punishment requires safeguards "against the formation of discriminations--between responses, between a response at one time from the same response at other times, between situations, and between people" (p. 209). 23 Punishment may use a variety of aversive stimuli. Punishment by electric shock has been used effectively with retarded and au- tistic children to increase social behavior (e.g., Lovaas, Schaeffer, and Simmons, 1965), to decrease self destructive and other deviant behavior (e.g., Lovaas and Simmons, 1969; Risley, 1968; Tate and Baroff, 1966) and as an aversive stimulus for inattentive behavior and in- correct responding in a picture naming task with retarded children (Kirsher, Pear and Martin, 1971). Noise has also proven to be an effective, non-detrimental stimulus. Wickes (1958) treated a number of Ss for persistent enuresis with the use of a loud buzzer. Azrin (1958) demonstrated that intense noise (95 db of white noise) can serve as an aversive stimulus to modify behavior. These studies demonstrate that response-contingent noise produces large and stable modifications of performance. In a study by Azrin, g§_gl;(l968) a sound was used as an aversive stimulus to correct rounding of the back or slouching. An apparatus was developed that provided a warning stimulus followed by an aversive tone for the duration of slouching. Slouching was thereby punished by the onset of the tone, and non-slouching was reinforced by tone termination and postponement. The experiment involved twenty- five adults who wore the apparatus during their normal working day during alternate periods in which the aversive tone was connected and disconnected experimentally. Also, a miniature time meter was used to record the duration of slouching. The results showed that slouching decreased for each subject. Two subjects were used as an experimental control in which slouching terminated the tone. The 24 result was an increase of slouching, demonstrating that the postural changes were controlled by the scheduled relation between the aversive tone and the slouching response. The condition of spasmodic torticollis is one in which there is a disorder of the cervical muscles, resulting in abnormal movements or positioning of the head. This condition may be a symptom of a neurological disorder, or it may be hysterical in origin. Brierley (1967) devised a technique in which the occurrence of the undesirable habit of inclining the head was associated with an aversive stimulus. The apparatus consisted of a headgear which could be positioned firmly over the top of the head. A clip slid along the head band and carried a small mercury switch. As the head tipped a circuit was completed causing a shock to the head. Two patients were treated on a weekly basis in three, three-minute reading sessions divided by five-minute rest periods. Symptoms disappeared after about the tenth session. The patients were treated for four and nine months respec- tively and in a one year follow-up study, symptoms had not recurred. Punishment suppresses old behaviors whereas positive reinforce- ment strengthens new. By breaking up an old behavior pattern, punish- ment can provide the occasion for the positive reinforcement of new or more advantageous behavior that was of greater value to the subject (Kanfer and Phillips, 1970). Kircher, Pear and Martin (1971) demon- strated that mild shock that was used in conjunction with positive reinforcement was more effective and faster in teaching retarded children to name pictures than other procedures. In addition, the combined procedure produced a greater decrease in the ratio of 25 incorrect responses to correct responses. Because of these effects, the absolute number of correct responses tended to be increased and the learning rate was much higher than that produced by the other procedures. MacMillan, Forness,and Trumbull (1973) emphasize the fact that punishment is particularly effective if at the same time the social- izing agents (e.g., teachers) provide information concerning alterna- tive desired behaviors. The teacher must therefore provide and rein- force alternatives to the punished behavior. It is often important that these be competing behaviors or responses that are incompatible with the undesirable behavior. In essence, the teacher clearly communicates that behavior A is not appropriate but behavior 8 is appropriate. Frequently, as is the case with blindisms, children are exposed to constant criticism or punishment by teachers and/or parents. Azrin (1959) reported that extended periods of punishment diminished its effectiveness. A child becomes adept at "tuning out" the berating mother or teacher: repeated spankings become old hat, criticisms tend to be accepted as a part of life and cease to arouse much anxiety. The studies reviewed above suggest that the judicious use of punishment and positive reinforcement can be a very fast, effective and enduring technique for controlling behavior. Such techniques have been effectively used on behaviors similar to blindisms. While blind children may have long histories of receiving criticism about their undesirable behavior, they may have tuned out this consequence of their behavior. Finally, electric shock and noise have proven to 26 be effective aversive stimuli; however, the use of shock is frequently not permitted. Behavior Modification in Natural Settings The previous studies have demonstrated the success of behavior modification as a technique of changing behavior. However, this evidence is limited to laboratory or highly controlled classroom settings. Thus the usefulness of current behavior modification techniques in non-reSearch oriented human service settings such as public schools, prisons, mental hospitals and institutions for handi- capped individuals is unclear. Reppucci and Saunders (1974) point out that The point to be made is not simply that there is a large gap between what is real and what is imagined about the social application of behavior modification; rather, it is that there are reasons for this gap, and that some of these reasons fall outside the domain of behavioral technology as it is presently elaborated. Although there is little question that behavior modification techniques have potent, predictable effects under carefully controlled conditions, psychologists have only a slight comprehension of their effects under less-than-optimal conditions usually encountered in natural settings. Most of the academic literature in the field leaves one with the impression that implementation of an effective modification program is a straightforward, trouble-free affair, and that all one really requires for success is an understanding of learning theory and the techniques of behavior analysis (e.g., Bijou, 1970; Ullmann, and Boren, 1968; Tharp, and Wetzel, 1969). In natural settings, the behavior modifier faces a variety of problems that do not relate directly to theoretical issues in behavior modification and that are either nonexistent or relatively inconspicuous in the laboratory or special research situations, where the investi- gator has almost complete control over the contingencies of rein- forcement. Failure to appreciate the importance of these problems may be the primary reason why demonstration projects so often fail when efforts are made to transfer them out of the isolated classroom, ward, or building and into the natural setting (pp. 649- 650 . 27 Reppucci and Saunders (1974) further discuss eight problems that confront the change agent in natural settings and tend to miti- gate the attainment of optimal results. The first problem, the insti- tutional constraint is often labeled as "red tape" or "administrative matters." These are constraints that occur by virtue of common institutional procedures and arise with great frequency, regardless of the particular individuals who occupy specific positions. The second problem, external pressure, often dictated by political, eco- nomic, or administrative considerations is always a potential force for change. However, the direction of change valued by the external pressure does not necessarily coincide with what is seen as desirable by the behavior modifier. Therefore the process of change can seesaw under the alternating influence of the behavior modifier and external pressure. It is important to develop a common vocabulary among staff in the natural setting. The third problem is therefore the problem of language (i.e., communication). The fourth problem of two popu- lations concerns the situation where the behavior modifier is not able to work directly with his subject. In natural settings the indigenous personnel must be utilized to perform the actual behavior modification operations. Therefore the behavior modifier can influence the be- havior of subjects only by modifying the behavior of the staff. The problem of limited resources can be quite severe in natural settings. Many very desirable behavior changes are possible "in principle" or in the research laboratory, but not in the "real world." In short, since all of the operations normally performed by 28 elaborate equipment or skilled research assistants in the controlled laboratory must be left to indigenous staff and the creativity of the behavior modifier in most uncontrolled natural settings, the precise measurement of behavior necessary for optimal behavior modification cannot be obtained. In the natural setting the problem of labeling occurs since behavior modification programs encompass a wide range of activities that often bear value-laden labels such as education, recreation, therapy and rehabilitation. In such cases the indigenous staff may respond to the label of the activity rather than the function. The seventh problem is one of perceived inflexibility. A variety of circumstances, some of which have been mentioned already, make it difficult to establish and maintain a behavior modification program in the natural environment. Because of this, behavior modi- fiers working in the natural environment must struggle constantly to ensure the basic integrity of the programs they develop while at the same time not becoming unduly and unrealistically rigid; that is, they must strive for flexibility but within a theoretical context. The eighth problem discussed by Reppucci and Saunders (1974) concerns the problem of compromise. Objectivity may be readily lost in natural settings, which are characterized by the values or prefer- ences that govern their operations. Therapeutic contingencies of reinforcement are by definition statements of values in the sense that certain of the subject's behaviors are rewarded over others, and a project is not likely to be supported by the setting unless the values of the behavior modification programs are at least somewhat 29 concordant with the values of powerful individuals in that setting. Therefore, the behavior modifier is permitted entry to the setting only as long as he embraces certain of the values of the institution and its staff. This situation can lead to compromises (trade offs) that may jeopardize the objectivity and integrity of the research. The fact that some applied research efforts have successfully and closely adhered to principles of good design and research method- ology suggests that compromise on the fundamentals of scientific analysis is not always necessary, although circumstances and problems peculiar to each research setting will effect the degree of sophisti- cation in design and execution that each study attains. MacMillan, Forness, and Trumball (1973) state that classroom teachers are often faced with the problem of eliminating or weakening certain behaviors that are either interfering with a child's learning or hindering his social adjustment. This appears to be particularly true of teachers of exceptional children. MacMillan, Forness, and Trumball further state that based on the literature the only viable alternatives available to the classroom teacher are extinction and counter-conditioning (reinforcing a behavior that is incompatible with the undesirable behavior). Punishment has all but been rejected theoretically, yet remains one of the most commonly used behavioral devices by parents and teachers. Johnston (1972) discusses several issues concerning the use of punishment in natural settings. He maintains that the behavior modifier must attempt to define and control, as carefully as feasible, factors in the natural setting such as the reinforcers for the response to be punished. If punishment 30 procedures are less successful than expected, the behavior modifier must be ready to admit that the uncontrolled factor in the natural setting may have hindered the effects of the punishing stimuli. Johnston states that a great deal of concern is frequently expressed in applied settings over whether or not the punishing effects will generalize to other settings in which the response has not been punished and if undesirable responses that have not been punished will similarly decrease in frequency. The punishment of a response in one situation and not in others is likely to increase the sharpness of the control, but it may well decrease the probability that the response will be reduced in nonpunishment situations. D In many cases the teacher serves as a S discriminative stimu- lus: that is, a stimulus during which if the child responds in a certain way he will receive punishment (or reinforcement). This is in contrast to the SA situation in which a different teacher may represent the absence of any contingency (i.e., punishment or rein- forcer). This was the case in a study reported by Ramey (1974): (l) the elimination of the self-abusive behavior remained person and place specific. That is, the behavior was eliminated only when the child was both (a) in the classroom and (b) in the physi— cal presence of the teacher. The child was quickly able to discriminate that the device would not be employed by others. It was amazing to see how quickly a severely retarded child could learn that discrimination. Face slapping behavior that would go on in the hallway would stop immediately upon entering the class- room where the teacher was present. This effect of aversive conditioning has been noted in a number of studies (Risley, 1968; Birnbauer, 1968; Lovaas, and Simmons, 1969; Corte, Wolf, and Lock, 1971; Azrin, and Holtz, 1966). Lovaas (1969) has noted if punishment to suppress self-destruction is to be maxi- mally therapeutic (i.e., durable and general), it has to be administered by more than one person, in more than one setting. Practical problems prevented that being done in this case, with the results being the lack of generalization effect. (2) While 31 the face slapping was eliminated, the crying that frequently accompanied the behavior was not eliminated, although it was significantly reduced. The studies cited above point out a variety of problems when applying principles of behavior modification in a natural setting as outlined in this study. The problems of institutional constraints, external pressures, language, two-populations, limited resources, labeling, perceived inflexibility, and compromise affect the degree of sophistication in design and execution that a study attains when being conducted in a natural setting. The use of punishment and the generalization of the effects derived from the aversive stimulus can be greatly influenced by the natural setting. The use of a small residential institution such as the one used in the present study, may in effect magnify the problems described above or introduce unique problems not mentioned. CHAPTER III METHODOLOGY This study consists of six individual case studies utilizing experimental designs employing research strategies that have become traditional in the study of operant conditioning. This chapter is in two parts: the first describes general aspects of the research methodology that apply to all of the individual case studies: the last part of the chapter describes the individual studies. In stressing a functional analysis of behavior within a be- havior modification framework, Sidman (1960) emphasizes that "a group function may have no counterpart in the behavior of the individual. . . " (p. 53). He maintains, for example, that replications across successive individual subjects (each dealt with singly) are more power- ful in testing the reliability of a central tendency than are group data which cannot indicate the nature of exceptional cases. Therefore, Sidman maintains that replication of observed functional relationships within the same individual is the most powerful investigative tool. The study of six individual subjects facilitated the use of a combination of replication designs. Sidman states, "The soundest empirical test of the reliability of data is provided by replication." Direct replication was achieved by making repeated observa- tions on the same subject under each of several conditions (intra- subject). Systematic replication was achieved by varying the method 32 33 of presentation of the independent variable as well as the situation in which it was presented. According to Sidman (1960) ". . . every successful systematic replication demonstrates that the finding in question can be observed under conditions different from those pre- vailing in the original experiment. Where direct replication helps to establish generality of a species,systematic replication can accomplish this and, at the same time, extend its generality over a wide range of different situations." The A-B-A-B reversal technique and the multiple-baseline technique were used. In the A-B-A-B reversal technique, a behavior is measured and examined over time to establish a baseline. Then, the treatment or independent variable is applied and the behavior continues to be measured to determine if in fact behavior has been affected. If a change has occurred, the independent variable is discontinued to see if the behavioral change just brought about depends on this variable. If so, the behavioral change should be lost or diminished. The independent variable is then applied again to see if the behavioral change can be recovered. The behavioral change may be reversed briefly again, and so forth. Hypotheses Following are the hypotheses of the study. Hypothesis 1: The application of punishment and positive reinforce- ment reduces a blindism. Hereafter, this hypothesis has been referred to as the "Treatment" hypothesis. Hypothesis 2: The reduction of a blindism in the laboratory sessions contributes to a simultaneous reduction of the blindism in classroom situations. Hereafter, this hypothesis has been referred to as the "Laboratory Generalization" hypothesis. 34 Hypothesis 3: The reduction of a blindism in one classroom setting contributes to a simultaneous reduction of the blindism in other classroom settings. Hereafter, this hypothesis has been referred to as the "Classroom Generalization" hypothesis. Hypothesis 4: The treatment and reduction of one blindism contributes to a simultaneous reduction in a second untreated blindism. Hereafter, this hypothesis has been referred to as the "Concomitant" hypothesis. Hyppthesis 5: A neutral stimulus (CS) paired with an aversive stimu- 1us (US) during initial treatment sessions serves as an effective aversive stimulus in repeated treatment sessions. Hereafter this hypothesis has been referred to as the "CS-US" hypothesis. Hypothesis 6: The reduction of a blindism in a laboratory setting contributes to a reduced manifestation of this blindism over a period of time without further treatment. Here- after this hypothesis has been referred to as the "Duration“ hypothesis. The hypotheses relative to the research questions were stated for each subject. In terms of the replication design, each subject represents an independent study and therefore the cases varied and hypotheses were tested subject by subject. The scope of the study permitted most of the research questions to be tested by several independent studies while two of the research questions were tested by only one independent study. Subjects The criteria for selecting the subjects were as follows: they were enrolled in grades 4 through 12 or equivalent; were of normal hearing; obtained I.Q.'s of not less than 80; were congenitally blind; showed visual acuity of not more than light perception: and exhibited two or more blindisms. Additionally it was necessary that permission to serve as a subject was granted by the subject, the 35 parents, and the administration. The subjects were all selected from the Western Pennsylvania School for Blind Children which offered a population of 146 students from which to choose. Only six students met the above criteria. These six students, two males and four females, ranged in age from 16-20 with 1.0. scores from 85 to 134. One of the students was in 9th grade, four were in 11th grade and one was in an ungraded program. Five of the students were reported to be congenitally blind due to retrolental fibroplasia and one student was congenitally blind due to optic atrophy. Finally, the visual acuities varied from no light perception in either eye to light perception in both eyes. Settings There were two types of settings in which treatment was administered and/or in which observations occurred. First, there was what is called hereafter the "laboratory" setting. It consisted of an unused normal classroom. The second setting was a regular class- room in which the subject was a member. The selection of classrooms for each subject met the following criteria: the class occurred in the morning; the experimental study in that room had the approval of the principal and teacher so as not to impose an imposition on the student, teacher or school; and it was the kind of class in which the subject remained in his seat and was not restricted from engaging in the specified blindism. So as not to disrupt the class, the teachers of the specific classes were informed as to the nature of the experiment, of the fact 36 that observers would be visiting classes and maintaining records of the students' blindisms, and of possible behavior modification tech- niques to be administered within the classroom. For T2 (see below) the classroom where the greatest incidence of the primary blindism was emitted was selected as the setting in which to administer the treatment. Phases of the Studies To unify discussion of the designs of the six individual case studies the notion of "phases" of observation and treatment was utilized. These phases are named and defined below. They gain added meaning as they are applied in the tables describing the design for each case in the last part of this chapter (Tables 2 through 7). Baseline: a pre-experimental period in which the operant level of the blindism was established. Baselines were established in each of the settings on days 1 and 2 (Case 4 was an exception). Treatment in Situation 1 (T1): the period during which the aversive stimulus, treatment (T), was first administered, in either the laboratory or classroom setting. Treatment in Situation 2p(T2): the period during which T was administered in a second setting, in each case a classroom. Observation (0b): the period during which T was not adminis- tered in any setting but observations (0) occurred in classrooms. Treatment in Situation 2 repeated (T2'): the period during which T was reinstated in the same setting as in T2. 37 The Behavior Modification Procedures (The Independent Variables) In this section the behavior modification techniques are described. A description of the procedures used is included. The Laboratory Setting Punishment The aversive stimulus that was used in the punishment pro- cedure was a recording of a screeching sound of chalk on a blackboard. This aversive stimulus was presented each time the subject emitted the primary blindism and was maintained until the blindism stopped. In the laboratory setting the subject was engaged in an activity which tended to maximize the probability of the subject exhibiting the blindisms. Therefore, subjects were engaged in the following activities: listening to pre-recorded stories: reading out loud from a braille book with soft background music, or typing via a braille writer with soft music in the background. Pre-recorded four track tapes were made with stories or music on two tracks and the aversive stimulus on the other two tracks. Each time the subject emitted the primary blindism a switch on the tape recorder was used to change from either the music or the story to the aversive stimulus. When the subject stopped emitting the blindism, the recording was again switched from the aversive stimulus back to the music or story. Positive Reinforcement The positive reinforcer was 75¢ given at the end of the laboratory session. This reinforcement was given for participation 38 in the session and was not contingent on the subject's behavior during the session. The Classroom Settings Punishment The "chalk screeching noise" was recorded on casette tape. The ear piece from the casette recorder was placed next to the student (on the individual's desk or ledge next to the student). When the student emitted the blindism the recorder was immediately turned on. Variations of this procedure are discussed for the specific case. Positive Reinforcement Positive reinforcement, as in the laboratory setting, in- volved the giving of 75¢, except that in the classroom settings this reward was contingent on the subject not exceeding a criterion level of the blindism. The frequency of the primary blindism on the last day in T1 was used as a base for the first treatment session in T2. The following formula was used to establish a criterion frequency for purposes of reinforcement in order that by day five of T2 the frequency would be zero: T2, day 1, Base-(.33 x Base) = X]; day 2, Base -(.66 x = 0. If at Base) = X2; day 3, Base-(.99 x Base) X3; and day 4, X 4 the end of day l, a frequency of less than X2 had been obtained, this frequency was used as a criterion rather than X2 and so on. Table l portrays these behavior modification techniques schematically. 39 TABLE l.--Definitions of Positive Reinforcement, Punishment and Negative Reinforcement. Presentation Withdrawal Positive Reinforcer Positive Reinforcement Punishment (money) Ex: (for not rocking) Ex: (for rocking) Aversive Stimulus Punishment Negative Reinforce- (Chalk-screeching noise) Ex: (for rocking) ment Ex: (for not rocking) Punishment, the presentation of an aversive stimulus, occurred in both the laboratory and classroom settings. Punishment, as with- drawal of a positive reinforcer, occurred only in classroom settings. Positive reinforcement contingent on the reduction of blindisms occurred only in classroom settings. Due to the abruptness of the withdrawal of the aversive stimulus (i.e., duration of less than a second), negative reinforce- ment was not considered to be used in this study. Measurement of Emission of Blindisms (The Dependent Variable) Definitions Three blindisms were observed in this study. Head-Rollipg: repetitive side to side rotation of the head in a figure eight pattern. Rocking: repetitive movement of the head and trunk in the frontal plane (with subject 5, for reliability, repetitive movement of the head only in the frontal plane was also scored as rocking). 40 Eye-Poking: contact of a finger or part of the hand with the eye. PrimarypBlindism When the subjects were selected, the frequency and duration of each of the subject's blindisms was established. The most preva- lent blindism was considered the primary blindism. Secondary Blindism After a primary blindism was established (see above) the second most prevalent blindism was considered the secondary blindism. Measures There were two measures of the occurrence of a blindism: frequency, the number of times a blindism was initiated; and the elapsed time, or duration, of the blindism. Each served as a sepa- rate dependent variable. The blindisms were measured in the following way: for eye poking, for example, a frequency of one was counted, and a stop watch started each time a finger or part of the hand made contact with the eye. When the subject removed his finger or part of his hand from his eye, the timing was discontinued. The subject had to remove his hand from his eye before another frequency could be scored. Therefore, it was possible to have a frequency of one and a time of 20 minutes. The scoring of head rolling and rocking was accomplished in a similar fashion. Each time the subject began to move his trunk in the frontal plane in a rocking fashion, or, in the case of head rolling, began to rotate his head literally in a figure eight pattern, the action was counted as a frequency of one. 41 At the same time, the stop watch was activated. The subject had to stop these movements of head rolling or rocking for a minimum of 3 seconds in order to permit a second frequency count. The data were collected in three (3) separate and independent twenty minute (20) sessions per day. Each of these three separate sessions represented a different stimulus situation. Observer Training and Inter-Observer Agreement The observers participated in a training session which in- volved a description of the study, a video tape of samples of rocking, head rolling and eye poking behavior and observations of students (other than subjects) that exemplified the specified mannerisms. The subjects were observed from within the experimental room or the particular classroom used. The observers used the observation forms (Appendix A), stop watches, a regular watch, clip boards and pens. Inter-observer agreement was analyzed by having a second observer periodically (at least once during each phase of the study) make a simultaneous but independent observation record. Reliability was calcu- lated by scoring each five-minute frequency count (i.e., first 5 minutes, second 5 minutes, third 5 minutes and fourth 5 minutes of the observation session) as agree or disagree, and dividing the total number of agreements by the number of agreements plus the number of disagreements (Bijou, Peterson and Ault, 1968). Inter-observer reliability of the observation procedure was analyzed during all phases of the study for each subject which totaled 21 such occasions (once during each phase of the study for each 42 subject). Agreement on the number of intervals scored (frequency) ranged from 79 percent to 100 percent, with a median of 97 percent. Experimental Procedures The experiment (baseline) was begun on the same day for all subjects except for one. However, the beginning of T1 was on the same day for all subjects. The reason for starting T1 the same for all subjects stems from the fact that since the students were living in a residential school there was a great deal of communication. Therefore, this eliminated contamination of the baseline data by one student sensitizing the other students to the observance of blindisms by the experimenter. From this point on, occurrence of the different phases for the different subjects varied, based on the time at which each subject met the specified criteria (T1) and on variations in the different independent experiments. The duration of the studies varied from a minimum of 13 days to a maximum of 16 school days. This length of time was based upon the number of times the individual was absent or called out of classes for school business. The Laboratory Sessions Two observers were present during the laboratory sessions. These sessions occurred in the morning prior to the subject's attending any morning classes. Immediately before the laboratory session on the first day of the experiment, verbal instructions containing the following points were given: 43 1. You have been asked to participate in an experiment at this time for the next several days. 2. You will receive 75 cents for listening to stories, typing braille, or reading out loud. 3. There will be other people in the room from time to time. 4. Do you have any questions? Immediately before the first session of T1, the additional instructions were given to the subject: I have noticed you have a habit of (Primary Blindism) ____, Today, each time you (Primary Blindism) _____the recording will go off. Instead you will hear this noise that sounds like chalk screeching on a blackboard (sound). As soon as you stop (Primary Blindism) _____the story (or music) will go on again. If there are no questions, we will begin. The Classroom Sessions Class sizes varied from 10 to 15 students. Since the students were visually impaired, observations were made within the classroom without the students' knowledge. Since class size was small, obser- vations were made from various locations within the classroom that allowed a satisfactory view of the subject. During each of the class- room sessions the observers were instructed to note any actions or attitudes on the part of the teacher which possibly elicited or rein- forced the "blindisms" of the subject (since some of the teachers were blind, it is doubtful if there were any reinforcers given by the teachers)- This information was recorded at the bottom of the observation sheet under comments (see Appendix A); however, there was no attempt to modify teacher behavior in this study. 44 Treatment of the Data The hypotheses were tested by inspection of the changes in the emission of blindisms as measured by changes in frequency and time. These changes were measured by what is called here "percentage scores." A percentage score is obtained by dividing the mean score for each phase by the mean of the baseline. Thus, a percentage greater than 100 indicates an increase in a blindism, a percentage less than 100 indicates a decrease in the blindism and a percentage of 100 indi- cates no change. There is no accepted method for estimating the statistical significance of such changes. Instead, the significance of the changes is evaluated by a reasonable interpretation based on the magnitude of the experimental effect. This treatment of the data represents a functional analysis of the behavior under study. A functional analysis requires a believa- ble demonstration of the relationship between the experimental variable and the dependent variable. As Sidman (1960) points out, an experi- menter has achieved an analysis of a type of behavior when he can exercise control over it. He further emphasizes that replication across successive individual subjects, each dealt with singly, is more powerful in testing reliability and generality than are repli- cations of group data. Further explication of the treatment of the data is included in the discussions of the individual cases that follow, and in the presentation of the findings. 45 The Individual Experiments Case 1 Special Consideration.--The primary blindism was rocking and the secondary blindism was eye poking. The three stimulus situations included a laboratory session, an English class during the third period and a biology class during the fourth period. The subject was listening to recorded stories during the laboratory sessions. The experimental design consisted of only a Baseline, T1 and 0b (refer to Page 36). Experimental Procedure.--This experiment (see Table 2) repre- sented intrasubject direct replication (T1) and the first treatment segment of the multiple baseline technique (see Page 33). TABLE 2.--Experimental Procedure for Case 1. Stimulus Base- Situation line Tl Ob Laboratory Session 8 B T T Classroom 1 B .B 0 0 O 0 0 O 0 0 0 0 0 0 Classroom 2 B B O 0 0 0 O O 0 0 0 0 O 0 Days 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 Note: 8 = Baseline; 0 = Observation; T = Treatment of primary blindism. 46 Case 2 Special Considerations.--This subject did not participate in a laboratory session but rather three different classroom situations. Instructions were not given before the baseline: however, immediately before the first period class on the third day of the experiment (T1), the following instructions were given to the subject: "You have been asked to participate in an experiment during your first period class for the next several days. Your teacher knows that you are participating in the experiment. You will receive 75 cents for each 20 minutes of the class you participate in the experiment. This earphone is placed on your desk. I have noticed you have a habit of rocking, so today each time you rock you will hear this noise that sounds like chalk screeching on a blackboard (sound). As soon as you stop rocking, the screeching noise will stop. 00 you have any ques- tions?" The aversive stimulus was administered via a casette tape recorder within the first classroom by placing the earphone on the subject's desk. The primary blindism was rocking and the secondary blindism was eye poking. The three stimulus situations were: a social studies class during the first period; a typing class during the second period; and an English class during the fourth period. In T2, the treatment was discontinued in Classroom 1; however, observations continued for the remainder of the experiment. The aversive stimulus was then administered via a headset controlled by a Dictaphone Class- master Transmitter during Classroom 2. 47 Experimental Procedure.--This experiment (see Table 3) repre- sented intrasubject direct replication, (T2 and T2'), systematic replication (Tl vs. T2), multiple baseline technique (see Page 33) and A-B-A-B reversal technique (Classroom 2). TABLE 3.--Experimental Procedure for Case 2. Stimulus Base- Situation line T1 T2 0b T2' Classroom 1 B B T T O O 0 O O O O O O 0 0 0 Classroom 2 B B 0 O T T T T 0 O O O O T T T Classroom 3 B B O * O O 0 * O 0 O O O 0 O 0 Days 12 34 5678 910111213141516 Note: 8 = Baseline; T = Treatment of primary blindism; O = Observation; *.= Absent from class. Case 3 Special Considerations.--The primary blindism was rocking and the secondary mannerism was eye poking. The three stimulus situations were: laboratory session before school; an algebra class during the first period; and a history class during the second period. The subject was listening to prerecorded stories via earphones during the laboratory sessions. In Ob the experimental sessions were dis- continued and the aversive stimulus was administered via casette tape recorder within Classroom 2, by placing the earphone on the student's desk. 48 Experimental Procedure.--This experiment (see Table 4) repre- sented systematic replication (T1 and T2) and multiple baseline technique (see Page 33). TABLE 4.--Experimental Procedure for Case 3. Stimulus Base- Situation line T1 T2 Ob Laboratory Session B B T T T T T Classroom 1 B B O 0 0 O O O O O O O 0 Classroom 2 B B O O 0 O , O T T T T O 0 Days 1 2 3 4 5 6 7 8 9 10 ll 12 13 Note: B = Baseline; T = Treatment of primary blindism; O = Observation. Case 4 Special Considerations.--The primary blindism was head rolling and the secondary blindism was eye poking. The three stimulus situ- ations were: a laboratory session before school; history class during the second period; and an English class furing the third period. The subject was reading aloud from a braille book while music was playing softly in the background during the laboratory session. In T2 the experimental laboratory sessions were discontinued and the aversive stimulus was administered via casette tape recorder within classroom 1, by placing the earphone on the desk. (This subject has only one base- line score due to her absence on the first day of the experiment.) 49 Experimental Procedure.--This experiment (see Table 5) repre- sented intrasubject (T1 and T2) direct replication and multiple base- line technique (see Page 33). TABLE 5.--Experimental Procedure for Case 4. Stimulus Base- Situation line T1 T2 0b Laboratory Session B T T Classroom 1 B O O T T T T O O O O 0 0 Classroom 2 B 0 0 O O O O O O 0 0 O 0 Days 1 2 3 4 5 6 7 8 9 10 ll 12 13 Note: B = Baseline; T = Treatment of primary blindism; O = Observation. Egg}. Special Considerations.--The primary blindism was rocking (due to the inability to differentiate head nodding from rocking--both were scored as rocking) and the secondary blindism was eye poking. The three stimulus situations were a laboratory session during the first half of the first period of the school day; an English class during the second period; and a typing class during the third period. The subject was using a braille writer while music was playing softly in the background during the laboratory sessions. In T2 the experimental laboratory sessions were discontinued and the aversive stimulus was administered 50 via a headset controlled by Dictaphone Classmaster Transmitter within Classroom 2. Experimental Procedure.--This experiment (see Table 6) repre- sented intrasubject (T2 and T2') direct replication, systematic replication (T1 and T2), multiple baseline technique (see Page 33) and the A-B-A-B reversal technique (baseline-~Tl-T2-Ob-T2'). TABLE 6.--Experimental Procedure for Case 5. Stimulus Base- Situations line 11 T2 Ob T2' Laboratory Session B B T T T T T' Classroom ‘ l B B 0 O 0 O 0 O 0 0 O O O O O 0 Classroom 2 B B 0 0 O O 0 T T T T 0 O T T T Days 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 15 16 Note: 8 = Baseline; T = Treatment of primary blindism; 0 = Observation. Case 6 Special Considerations.--The primary blindism was rocking and the secondary blindism was eye poking. The three stimulus situations were: a laboratory session before school; an English class during the third period and a biology class during the fourth period. The subject was listening to prerecorded stories during the laboratory sessions. There was a click (produced from a toy clicker or cricket) 51 immediately preceeding the aversive stimulus during T1. In T2 the experimental laboratory sessions before school were discontinued and the "click" only was administered in Classroom 1. Treatment was discontinued during sessions 8 and 10 (Ob) and the subject was observed in the two classrooms for these days. During the session in Classroom 2 on day 9, the "click" (CS) was reintroduced with no prior instructions given to the subject (refer to the reversal technique page 33). During the classroom sessions on days 11-14 (T2') the subject was instructed that during her two consecutive classes she would be observed for 20 minutes (she did not know which 20 minutes out of the possible 100 minutes). If she did not rock during that time she received one dollar. The "clicker“ was no longer used. The subject was observed for the total 100 minutes; however, data was only re- corded for 20 minutes out of each class or a total of 40 minutes. Experimental Procedure.--This experiment (see Table 7) repre- sented intrasubject (T2 and day 9--Ob) direct replication, systematic replication (T1, T2 and T2'), multiple baseline technique (see Page 33) and A-B-A-B reversal technique (Baseline--T2-T2'). 52 TABLE 7.--Experimental Procedure for Case 6. 1 Stimulus Base- Situation line T1 T2 0b T2' Laboratory Session B B T Classroom Classroom 2 B B O O 0 O O O T O T T T T Days 1 2 3 4 5 6 7 8 9 10 ll 12 l3 14 Note: B = Baseline; T = Treatment of primary blindism; O = Observation; CHAPTER IV RESULTS Introduction This study was undertaken to determine if the combination of punishment and positive reinforcement would be effective in the reduction of blindisms; if treatment of one blindism influences the rate of occurrence of concomitant blindisms; and if treatment effects can be generalized to non-treatment stimulus situations. The study consisted of six independent experiments which provided replication across six successive individual subjects, based upon the experi- mental style used by B. F. Skinner, for the purpose of providing in- creased support for either accepting or rejecting the hypotheses. Walker (1968) states: Producing a decrease in variability means increasing the quality of the experimenter's control so that factors other than the one under study cannot produce unsystematic variation in the scores. Skinner clearly chooses the latter procedure and undertakes to exert sufficient control over the behavior in question to make a statistical test unnecessary. Thus he chooses to exercise experimental rather than statistical control over his subjects. It should be obvious that this choice leads to an ultimate style of research in which a single organism is a sufficient 'group' for the establishment of a principle. Additional organisms are then tested only to determine that the principle works with all or most individuals. In Chapter III, the general design characteristics of all six studies were stated and the individual cases were described separately to point up the unique features of each experiment. 53 54 The six hypotheses tested were stated and each given a label. Data from each of the experiments were treated separately, case by case, and the hypotheses in the particular experiment were tested. Only the labels of the hypotheses have been utilized (see Page 33). ‘The general conclusions derived from the six experiments regarding the hypotheses are treated at the end of each case. Finally, a discussion of the results and a summary conclude this chapter. Case Study Form The order of presentation of information for each of the six experimental cases is as follows: the primary blindism (since eye poking was the secondary blindism for all six cases, it is not stated separately); a description of the laboratory session, classroom 1 and classroom 2: results and discussion; tables; graphs; and a summary. For each case, the first table contains the actual data for the primary blindism and the second table contains the actual data for the secondary blindism. These tables are constructed in the fol- lowing manner: each of the stimulus situations (i.e., Laboratory, Classroom 1 and Classroom 2) consists of three rows of information; schedule of the experimental procedure; frequency results in terms of the total number of occurrences of the blindism during the 20-minute session; and time results in terms of the total duration (minutes and seconds) of the blindism during the 20-minute session. (The symbols in the schedule rows represent the following: B = baseline; T = Treatment of the primary blindism; and 0 = observation. The columns represent the following information: Baseline, a pre-treatment record of the 55 operant level of the blindism; T1 the treatment was administered in one stimulus situation; T2, the treatment was discontinued in the first stimulus situation and administered in a second stimulus situation; Ob, all treatment was discontinued and; T2', treatment was reinstated in the same stimulus situation as in T2. The third table contains the mean and percentage scores for the primary blindism. The fourth table contains the mean and percentage scores for the secondary blindism. These mean scores were obtained for the baseline and each of the phases for each stimulus situation. The percentage scores were obtained by dividing the mean of each phase by the baseline of that particular stimulus situation. Therefore, a percentage greater than 100 indicates anincrease in the blindism, a percentage less than 100 indicates a decrease in the blindness and a percentage score of 100 indicates no change has occurred. The cumulative frequency is presented in the first graph and the cumulative time is presented in the second graph for the primary blindism for each case. The cumulative records presented represent the cumulation of responses (i.e., frequency or time) for each day of the experiment. Therefore, rapid responding (frequency) or a continued response (time) is described as a steep slope on the cumulative record. With a decrease of the frequency of responses or the time emitting the responses, there is a corresponding decrease in the slope of the cumulative record until the curve becomes essentially horizontal, indicating no further occurrences of the blindism. 56 9.51.1. Case 1, was an 18—year-old eleventh grade female whose primary blindism was rocking. She was involved in the following experimental situations: the laboratory session in which she listened to recorded stories; Classroom 1 which was an English class during the third period (10:20-11:10); and Classroom 2 which was a biology class during the fourth period (11:10-12:00). The experimental schedule represents the first treatment segment of the multiple baseline technique. Results and Discussion Case 1 tested the following hypotheses: the treatment; Laboratory Generalization; Concomitant; and the Duration hypothesis. The unique features of this experimental case included the appli- cation of treatment in only the one stimulus situation and the observation of these effects in two other stimulus situations over a period of time without further treatment. The Treatment hypothesis was tested in this experiment in T1 of the Laboratory Session (see Table 10). The percentage scores for both frequency and time were zero (see Table 10) indicating the blindism (rocking) did not occur when the treatment schedule was in effect. That is, after the instructions and sample presentation of the aversive stimulus, the subject did not emit the blindism during the two sessions of T1. Therefore, via the instructions the subject was made aware of the operant contingency (i.e., when you rock you will hear'this sound) and punishment was administered at once. Azrin and H012 (1966) state: How quickly does punishment reduce behavior? Virtually all studies of punishment have been in complete agreement that the reduction 57 up” Pm" o m” N¢"_ Noup o m” o m” o_ o o . F o _ o o o o o o monp woum mp Foam um no m coup mmuo op o qum mm um o Ppum Noun NN mp m m acne wmuw mm an m m omum Noupp em cu m m PH mcwpmmmm a mppammm L mppzmmm mpaumgum N soocmmmpo p mupzmmm e mppammm mpaumcom P soocmmmpu p mppsmam c mbpammm m_:umnum cowmmwm auopmcoam; coauaapsm m=P=e_pm ._ ammo com mcomumspwm ms—ssmum may we comm Low mommzm popcmswgmaxm as» we sumo cm vmcwmuno smwucwpm xgmarca on» com museum wave new aucmaamcm up mmwuun.w m4mwm an mmmpmpmo mcm mwgoom mmmucmocma 59 N .mmcoomm cw Auv ws_u mam va Aucmaamcw cw mommmgaxo mmcoom cmmzP AmN.NNV e.~e Amm.mmv Nm_ m.mm_ m AoF.NNV N.e Amm.va m, mp c N soosmmmpu Amm.m_v m.m~ Amm.mmv mNN N_e m Aom.omv m.m Aom.omv m.m~ mm c _ soocmmmpo on o FFm m on m Am C cowmmmm Pmpcmswcqum N z x z NN F2 cowpmsuwm szmswum no —» m:WPmmmm .mmmna pmucmswcmaxm mcm cowpmzpwm mzpmewum mm Emvmcwpm acmsmcm mcp cow mmcoom mmmpcmucma mcm mcmmz "_ mmmu--.o_ m4mwm an mmcwmumo mum museum mmmpcmucmm N .mmcoumm cm A»: mew» mam AN: zucmacmcw cw mommmcnxm mmcoUm cmmzp I: 28.: m. 2: o 2 a. 8.: N. 2: o m. c N accommmpu 2: o 2: o o m 2: o 2: o o c F soocmmmFu 2: o o m 2: o o c =o_mmmm Pmucmawgomxm N z N z NN .z :oNpmame , m=_:ewum so P» mcwpmmmm .mmmna Fmpcmsmcmaxm mcm comumzumm mmpzewum an smwmcwpm xgmmcoumm may com mmgoum mmmucmocma mcm mcmmz up ommouu._p m4mFm Ac mmcFmoco mom mmcoom mmmucmocmc .mmcoomm :F Fa: mer mcm AF: zocmscmcm cF mmmmmcmxm mmcoom cmmz N F FFm.mF: mm FNm.em: m.Nem FNm.mm: mm.eo~ am.~m mNN m.me N AFm.mN: m.N Fem: N.FN NNe.mN: ON FNo.NmF: mm m.m~ F m EoosmmmFo ANN.F: FF Fmo.mm: m.oom 2mm.: mN.N Fm.mm: mmm mom o 2mm.Fm: mm.m 2mm.FmF: m.mm 20F.mF: mN.e Fem.mFF: m.em m.aN e N Eoogmmmpu 2mm.mm: FoF ANN.NN: mN 2m.ee: mN.NFF ANF.: m emu o FNm.Fm: m FmF.Nm: m.oF 2mm: mN.NF FmN.F: m m.mN F F soogmmmFu N z N z N z N z NN F2 emwwmmwwm .NF no NF FF oeFFoaem .mmmca qucmEchaxu mcm coFszuFm maneFum xc EmchFFm Nngch mcF com mmcoom mumpcmocma mcm mcmmz HN mmmuut.¢F mcmFm ac mmcquco mcm mmcoom mumpcmocma 74 N .mmcoomm cF A»: mEFu mcm AN: zocmzcmcw cF mommmcqu mmcoom cmsz 2m.NmN: NNF FN.FN: m.NF FNN.ON: mm.m 2N.NNN: mm me o 2mm.Fm: m.m 20m: N.F FNN.ON: mN.F 2mm.mmF: OF N L m EoocmmmFo ANN.NN: NN.F 20: o 20: O 2m: 0 e o ANN.ON: mm. 20: o 20: o Fe: o e e N soocmmmFu 20: o 2am.N: N Fm.NF: mN.m 2Na.F: m. mN N Am: C 2mm.mF: N. 2N.NF: mN. ANN.N: m. m N F EoocmmmFu N z N z N z N z NN F: eaFomszm szeeFom .Nh co NF FF mchmmmm .mmmca FmpcmEchaxm mcm conmzme szzeFum ac EmchFFm acmmcoomm mcu cow mmcoom mmmpcmocmm mcm mcmmz “N mmmu--.mF mcmFu Ac mmcquco mcm mmcoom mmmpcmocma N .mucoomm cF FF: msFu ucm FF: FocmzamcF cF mmmmmcaxm mmcoom cmsz FN.oN: NNN FN: N. FN.NF: NFF NNNF F FNoF: NF FN.e: N. FF.NN: N.oF FF F N soocmmmFu FNN: NoF FN.NF: em FF.FN: eNF NNN F FN.FN: N.N FN.Ne: N FN.NN: N.NF NF F F EoocmmmFu FN.N: N NFF N FNN: N.N NF F comemm Fmpcmchmamm N z N z N z N z NN Fz eoFFezoFN mansFNN .NF No NF FF oeFFomNN .mmmca qucmEchaxm mcm :oFmeuFm szzeFum Ac EmchFFm Fgmcha mcu FoF mmcoom mmmucmocmm mcm mcmmz ”m mmmuut.mF mcmFm Fc mmcFcho mFm mmcoom mmmucmocma .mmcoomm cF F»: mEFF mcm FF: FocmmcmcF cF mmmmmcaxm mmcoom cmmzw FF.oNN: co FN.FNN: om F¢.NNN: NN NN N FF.FN: N FN.NNF: N.e FN.NF: N.N N.N F N EmogmmmFu FN.NNF: FNF FN.NoN: eNN FN.ooN: FNN oFF F FN.NNF: FF FN.NNF: N.FF FNNN: N.FF N.F F F soocmmmFu FN.NNN: NNN NNF F FN.NNF: N.N N.N F NNNmmmw Fmpcmchmmxm N z N z N z N z NN Fz emw#mflwwm .NF co NF FF mcFmemm .mmmcm chcmchmaxm ucm conmsme szzeFFm Ac EmchFFm Femocoomm mcp LOF mmcoom mmmucmocma ucm mcmmz ”m mmmuuu.mF mcmFm Fc mmcFcho mFm mmcoom mommcmoFmNN .mucoomm cF F»: mst mcm FF: FocmmcmFF cF ummmmcqu mmcoom cmsz 233 F: 8.2 N. 2.: NF NF F as eFF 23: NN. 2:: 2 N F N EoocmmmFu FN: NF 2: o 2: o o F FNF.NN Ne.N ‘20: o 20: o o F F soocmmmFo 2: o NNN F 2: o B F comemm qucmeFFmaxm N z N z N z N z NN Fz coFszuFm szmsFFm .NF co NF FF mcFmemm .mmmca FmpcmEFFmaxm mcm coFFmame szzeFum ac EmFucFFm FFmEFFN mcp FoF mmcoom mmmucmocma mcm mcmmz "v mmmuII.NN mcmFm ac mmcFcho mFm mmcoom mmmucmocma N .choomm cF FF: mEFp mcm FF: FocmmcmcF cF ummmmcaxm mmcoom cmsz Fm: 0 Fe: o Fe: o o o 20: 0 Fe: o Fe: o o F N soocmmmFu FNN: NN.F FNN: F FN.NF: N. e F FN.NF: FN.F ,FoN: F FNN: N. N F F soocmmmFu Fe: o o F Fe: m o F comemm Fmpcmchmaxm N z N z N z N z NN Fz coFszme msFaeFum .NF co NF FF mcFmemm .mmmcm FchmEFFmaxm ucm coFmeuFm szzerm Fc EmchFFm acmmcoomm mcp coF mmcoom mmmpcmocmm ucm mcmmz ”c mmmu-u.mN mcmFm ac mmcFcho mFm mmFoom mmmucmoFmNN .mmcoomm cF FF: mEFF mcm FF: FocmzcmFF cF mmmmmcaxm mmcoom cmsz FNF.: N FFN.NF: N.NN FFF.: N.N FNN.NFF: N.NFN NNN F FFN.F: NN.F FNe.oFF: N.FN FN.NF: NF.N FNN.NNF: N.NN NN F N soocmmmFo FNN.NN: NN.NNF FFN.NF: Ne ANN.NN: NF.NFN FNN.NN: N.FNF FNN o FFe.FN: NN FNF.eN: N.NF FNN.FN: NF FNN.NN: N.NF N.NF F F sooFNNmFu FNN.F: N.FF N.NNN F FNN: N.NF N.¢ F comemm FchmEFFomxm N z N z N z N z NN F: eoFmeamFN msteFFN .NF No NF FF oeFFoNeN .mmmca FchmEFFmaxm mcm coFmeuFm szstFm Fc EmFucFFN FngFFN mcu FoF mmcoom mmmpcmocma ucm mcmmz um mmmutt.mN mcmmu an umcwmpao mew mmcoom wmmucmugma .mvcoumm cw Auv as?» ten Rev xucmaamcw cw commmgnxm mmcoom cam: N F A-.~_V mm.um Amm.~mv N~_ Amw.mv m.mm Amm.m~v m.¢m m.-m b A-.-v m A¢_.mev m.o Aom.mpv m~.~ Amm.m¢v 8.0 m.m_ c N Eoogmmm 5 Am_.~mmv mm.mm~ Am_.om_v m._mm Aem.mppv m~.¢ep Aom.mm.v e._o_ m.mm_ p A¢o_v mm.o_ Aeo.mmpv m , Aoopv m.o Amm.mmv N.m m.m c F soocmmmFQ Aom.msv m.~ o p Amm.mmv e. m._ c mmflmmmw Pmucmswchxm & z e z s z x 2 N5 _2 cmwwmnwwm .NF no we PP a:_Fammm .mmmgm pavemewcmaxm ucm compmzuwm m=F=EFgm xa Emwncw_m zgmucoumm ms» cow museum mmmucmogmm ucm meow: um mmmuuu.nm m4mpc an nmcpmuao mew mmcoom mmmgcmogoa .mucoumm cw Any we?» use Amv aucmacmgw cm ummmmgaxm museum cam: .coppmzupm mapzspum mcpucoammggou mg» com N p Nov o Ao¢.mmpv mum ANN.V _ A~.mfiv m~.-~ “up.mev m~_ mom a on o AN~.Fm_V NN Nam.ov _ Amo._m_v m_ Amo.omv m m.¢. c N Eoogmmmpu A~_.~V mN.m flee.me_v 44m Afio.¢_v em APm.v mN._ “om.NmPv omm m._e~ a Am.NFV m~.N Amo.mo_v 4N Am.o¢v m Ame.mv m~.P Aem.omsv om mm c F Eoogmmm_u o c we a o o op c .mmmmmmw pancmewgqum & z z x z s 2 we 2 _z coppaspwm .NF ANVN» A_v~p .P a=_P m=P=E_pm -mmmm use coupasppm .mmmza Fmpcmewcwaxu ma—aewum An Emwccppm Newswca mg» com museum mmmuzmucma ucm mammz no wmmuuu.om w4mwu ma umcmmuao mgm mmcoum ammucmucma .mvcoumm cw any mama ucm Amv mucmacmcm cw ummmmsaxm mmgoom cam: N P oo.~ . m~.. A~.~mv m.~m_ Amm._v m “we.mmv No. Amm.MNv ~N_ mm.m m. Amm.mm_v m.o Amo.o_v _ Amm.o~Fv m~.op Amo.m~_v N m.vop u m m N Eoocmmmpu N~.m m.e~ Amo.m¢v m.~__ Amo.omv mm “84.4V mN._F Ao.mmv _mm «mm a m~.ee mu.m Amo.mmv a Ao¢.mmv m flam.wwv m~.m Am.N¢v N_ m_ c _ soogmmmpu Amm.ouv o m.w p Am.mov m.~ e c cowmmmm Faucmswcmqu R z s z x z s 2 N3 2 F2 compm=p_m m=_=smpm .NF no ANVNP AFVNF _h mcw_ -mmmm .mmmza Pmucmswgmaxm ucm comumzuwm m:_=ewpm An Emwucwpm Acmucoumm on» com museum mmmpcmucma ucm cam: no mmmoun.~m m¢m