Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Diagnostic indices of hyperactivity and an investigation of verbal mediation training as an alternative to pharmacotherapy
(USC Thesis Other)
Diagnostic indices of hyperactivity and an investigation of verbal mediation training as an alternative to pharmacotherapy
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS This malarial w a s produoad from a microfilm copy of tlw original documant. While the most advanced technological m e a n s to photograph and reproduce this document h ave be en used, the quality i s heavily dependant upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The s i g n or “target" for p e g e s apperently lacking from the document photographed is "Missing Page(s)". If it wes pos sible to obtain the miss ing page(s) or section, they are spliced into the film along with adjacent p a g e s . This may have necessitated cutting thru an image and duplicating adjacent p a g e s to insure you complete continuity. 2. W han a n image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may h ave moved during exposure and thus c a u s a a blurred image. You will find a good im ag o of the p e g s in the edjacent freme. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to be gin photoing at the upper left hand corner of a large sheet an d to continue photoing from left to right in equel sections with a small overlap. If necessary, sectioning » continued again — beginning below the first row and continuing on until complete. 4. The majority of u s e rs indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could b e made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific p a g a s you wish reproduced. 5. PLEASE NOTE: Soma p a g e s may have indistinct print. Filmed a s received. Xerox University Microfilms 300 Nort h 2Mb Road Ai m A l te r, Mt oMcan 4 S 1 0 S TROZZOLINO, Linda Ann, 1948- DIAGNOSTIC INDICES OF HYPERACTIVITY AND AN INVESTIGATION OF VERBAL NEDIATION TRAINING AS AN ALTERNATIVE TO PHARMACOTHERAPY. University of Southern California, Ph.D., 1975 Psychology, clinical Xerox University Microfilms , Ann Arbor, Michigan 48106 THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. DIAGNOSTIC INDICES OF HYPERACTIVITY AND AN INVESTIGATION OF VERBAL MEDIATION TRAINING AS AN ALTERNATIVE TO PHARMACOTHERAPY By Linda Ann Trozzolino A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) January 1975 UNIVERSITY O F SOUTHERN CALIFORNIA TH E GRADUATE SCHO OL UNIVERSITY PARK LOS AN G ELES. C A LIF O R N IA S 0007 This dissertation, written by Linda Ann Trozzolino under the direction of h.frJf... Dissertation Com mittee, and approved by a ll its members, has been presented to and accepted by The Graduate School, in partial fulfillm ent of requirements of the degree of D O C T O R O F P H IL O S O P H Y Dm DISSERTATION COMMITTEE To my parents whose love, guidance, and support have made this achievement possible. ii ACKNOWLEDGEMENTS To all who contributed to this dissertation, espec ially to my Chairman and friend, Dr. Norman Tiber, and to the members of my Committee, Dr. A. Steven Frankel and Dr. Sidney Russak, to Dr. James Kahan, who acted as my design and statistics consultant, to the women who helped run the project: Brook Otte, Jennifer Hansen, Maxine Townes, and Kathy Flynn, and to Robert, who wrote this beautiful expression of thanks for me.... A POEM OF GRATITUDE Here I came, and then I went for our paths had crossed the times, that love and learning with you spent. And my future must resign the goals, and sorrows that we met, To help give me what is mine. But how can one express these thoughts without a tear or two. For what you worked to give to me was a love expressed by few. I learned through you the art of love; the feelings of your soul, I learned through you the need to care; and the sadness to bear and toll. I learned to help, I learned to love, a life through you that's new. But I also learned to teach and be a person that stemmed from you. (Robert Mueller) iii TABLE OF CONTENTS Page DEDICATION........................................... ii ACKNOWLEDGEMENTS.....................................iii LIST OF TABLES.......................................vi LIST OF FIGURES.......................................viii Chapter I INTRODUCTION............................... 1 II REVIEW OF LITERATURE...................... 13 Stimulant Treatment of Hyperactive Children I: General Comparisons of Hyperactives vs. Normals.............. 14 Stimulant Treatment of Hyperactive Children II: Arousal, Differential Diagnosis, and Differential Treatment Response.............................. 22 Activity Level........................ 32 Operant Conditioning and Treatment of Hyperactivity................... 40 Conceptual Tempo...................... 48 Verbal Mediation...................... 53 Therapeutic Application of Verbal Mediation Training ................. 64 III METHOD..................................... 75 Subjects.............................. 75 Procedure............................ 78 Verbal Mediation Training ............ 83 Analysis.............................. 90 iv Chapter Page IV RESULTS................................... 93 V DISCUSSION............................... 115 Evaluation of Hypotheses ........... 115 Diagnostic and Treatment Implications.................. 122 Summary........................ 136 APPENDICES..................................... 139 Appendix A: Training Materials. 140 Appendix B: Forms and Diagrams of Equipment........... 15 7 Clinic Intake Form... 158 Social Worker Evaluation Form................ 172 Psychiatry Evaluation Form................ 173 Discharge Summary Form. . . 174 Informed Consent Form . . . 175 Subject Record Form .... 177 Diagram 1: Stabilimetric Chair................ 178 Diagram 2: Activity Recording Apparatus. . . . 179 REFERENCES...................................... 180 v LIST OF TABLES TABLE PAGE 1 Planned Comparison Analysis of Pre-Test Differences Between Hyperactive and Normal Subjects. . . 94 2 Pre-Test Mean Scores In Hyperactive and Normal Subjects................... 95 3 Planned Comparison Analysis of Pre-Test Differences Between High- Active and Low-Active Hyperactive Subjects.............................. 96 4 Pre-Test Mean Scores in Low-Active and High-Active Hyperactive Subjects............................ 98 5 Planned Comparison Analysis of Change in Scores from Pre-Test to Post-Test 1 Across all Groups................. 99 6 Planned Comparison Analysis of Inter action Between Change From Pre-Test to Post-Test 1 in Normal and Hyperactive Subjects ............... 1°1 7 Mean Pre-Test, Post-Test 1 and Change Scores in Normal and Hyper active Subjects...................... 102 8 Planned Comparison Analysis of Inter action Between Change From Pre-Test to Post-Test 1 in Low-Active and High- Active Hyperactive Subjects......... 108 9 Mean Pre-Test, Post-Test 1, and Change Scores In Low-Active and High-Active Hyperactive Subjects ............... 1°4 10 Planned Comparison Analysis of Change in Scores From Post-Test 1 to Post-Test 2 Across all Groups................. 106 vi TABLE PAGE 11 Planned Comparison Analysis of Interaction on Between Change From Post-Test 1 to Post-Test 2 in Normal and Hyperactive Subjects............. 107 12 Mean Post-Test 2 and Change Scores in Normal and Hyperactive Subjects................................ 108 13 Planned Comparison Analysis of Interaction Between Change From Post-Test 1 to Post-Test 2 In Low- Active and High-Active Hyperactive Subjects................................ 109 14 Mean Post-Test 2 and Change Scores in Low-Active and High-Active Hyperactive Subjects................... 110 15 Correlations Between Activity and Performance Measures................... 112 16 Correlations Between Activity Measures................................ H I 17 Correlations Between Performance Measures................................ 114 vii LIST OF FIGURES Figure Page 1 Change in DRL-Operant Performance Scores in Normal, Low-Active, and High-Active Hyperactive Subjects. . . . 120 2 Change in PR-F-Operant Performance Scores in Normal, Low-Active, and High-Active Hyperactive Subjects................................... 121 3 Change in MFF Activity in Normal, Low-Active and High-Active Hyperactive Subjects..................... 130 4 Change in Resting Activity in Normal Low-Active and High-Active Hyperactive Subjects................................... 131 viii CHAPTER I INTRODUCTION Hyperactivity, alternately referred to as "hyper kinetic syndrome," "hyperkinesis," "minimal brain dysfunction" and in a number of other terms, is a child hood disorder primarily defined by the apparently constant presence of excessive movement. Other symptoms which are commonly associated with this disorder include perceptual, motor and intellectual deficits, distractibility, short attention span and impulsivity (Clements, 1966; Werry, 1968). Any number of these symptoms may be present in any one case. The report from a conference held at the Office of Child Development in Washington, D.C. (Freedman, 1971) notes the lack of definitive diagnostic criteria in differentiating this disorder from others which may have similar symptoms but different etiology and different treatment requirements. Because the perceptual, motor and intellectual deficits are sometimes severe as in mental retardation, and are sometimes associated with a medical history such as anoxia at birth, a common view is to consider all of the symptoms as manifestations of mild cerebral dysfunction (Paine, 1962; Stevens et al., 1967). However, Werry and 1 2 his associates (1964) suggest that hyperactivity with asso ciated learning problems is a "final common pathway of behavior” having several possible causes including brain damage, delayed maturational development, and emotional disturbance. Hyperactivity is generally believed to be a highly prevalent disorder among elementary school children, with estimates as high as 15 percent (Rosen, 1969), yet there is substantial divergence of opinion concerning the diag nostic criteria, etiology and treatment. From a clinical point of view not only is the clinician faced with the problem of the vague, subjective diagnostic criteria which defined this disorder, but given a decision that the child is hyperactive, the clinician is then faced with the fact that the only widely accepted treatment, namely stimulant medication, is effective in only one third to one half of children diagnosed as hyperactive (Freedman, 1971). Furthermore, even with stimulant medication there are no objective accepted guidelines as to evaluation of treatment and dosage. Some studies suggest that one third to two fifths of hyperactive children respond to psychotherapy (Cytryn et al., 1960; Eisenberg, 1961) but again evaluation of treatment is very subjective. Dramatic differences among children diagnosed as hyperactive in response to therapeutic interventions, as well as extreme inter individual variability observed in laboratory studies on 3 physiological measures and task performance suggest that hyperactive children as currently diagnosed are not a homogeneous group as to nature of their disorder and therefore treatment response (Marwit and Stenner, 1972; Millichap and Johnson, 1974; Satterfield et al., 1971; 1972; 1973). Therefore, one purpose of this study is to investigate parameters which might objectively, reliably and economically differentiate the hyperactive child from the normal child, and possibly differentiate subgroups within the hyperactive group. Intrinsically related to the problem of finding parameters helpful in differential diagnosis is treatment choice and objective evaluation of that treatment. Because there is not yet a reliable criterion index for indicating whether a child should receive medication and if so which type, or psychotherapy, the current practice is generally to determine by trial and error the appropriate treatment. The problems of this approach are numerous: delay of effective treatment, possible exacerbation of symptoms if the drug is not the correct treatment, drug side effects and placebo effect. The latter refers to the possibility that parents and teachers may subjectively evaluate the treatment as effective, primarily because of positive expectancies, while in fact, the child is not improving. Treatment evaluation is usually based on subjective reports 4 by parents and teachers, which have other disadvantages. For example, parents often report improvement when the dosage has been increased to the point where the child's activity has been sufficiently decreased to meet their subjective expectations of normal activity. But often, this dosage may be so high as to depress the child's ability to learn. In this study by Sprague and co-workers (1969), the dose found optimal for learning was consider ably below that prescribed by many clinicians. Although hyperkinesis or hyperactivity is said to remit in adolescence, numerous follow-up studies have shown that if not effectively treated, hyperactive children develop a variety of social and psychiatric problems as adolescents and adults, (Eisenberg, 1973; Millichap, 1969; Wender, 1971). Therefore, a second purpose of this study is to examine those parameters which would most reliably predict differential treatment among hyperactive children. The third purpose of this study is to assess the effectiveness of verbal mediation training in treating hyperactive children. Presently a number of hyperactive children do not respond to medication or psychotherapy. Others may respond to stimulant medication or behavior modification procedures in terms of effectively reducing motoric overactivity in the classroom (out of seat behavior and moving about in the seat) without associated gains in 5 academic performance (Patterson, 1965) . The verbal media tion training procedure has as its main focus the improve ment of academic performance with associated decrease in motoric activity. If found to be a successful treatment intervention, it has the added benefit of no side effects. Although Ritalin, the most commonly prescribed medication for a hyperactive child is reasonably safe, there may be a number of associated side effects including loss of appetite, insomnia, headache, abdominal pain, and suppression of height and weight gains, with other more rare symptoms including increase in heart rate, blood pressure, and tachycardia (DiMascio and Shader, 1970; Safer et al., 1972; Sprague and Boileau, 1973). When these side effects do occur, if prolonged and/or severe, medication is usually discontinued. To summarize, the goals of this study are three fold (1) To investigate parameters which might differen tiate the hyperactive child from the normal child and possibly different subgroups within the hyperactive group. (2) To assess the effectiveness of verbal mediation training in treating hyperactive children. (3) To examine those parameters which would most reliably predict differential treatment response among hyperactive children given two treatments, verbal mediation training and pharmacotherapy (Ritalin). 6 Verbal mediation refers to the functional inter action between self-verbalizations (overt or covert) and nonverbal behavior (Lovaas, 1961, 1964; Luria, 1960, 1961; Piaget, 1926; Vygotsky, 1962). To summarize a large body of literature, there is a direct relationship between the child's development of voluntary control over his behavior and the development of his speech. The self-guiding function of the child's verbalizations follows an age- related developmental sequence in terms of area of control from initiating his own behavior, to inhibiting an ongoing behavior, and finally, complex planning. Another dimen sion in development of the self guidance function of speech is the gradual internalization of this speech: During the first few years of life only the external verbal commands of others control the child's behavior; by age 3 or 4 the child's behavior begins to respond to his own self-directing comments said aloud, finally beginning at 6 or 7 the child's self-directing verbalizations are inter nalized to linguistic thought. The ages of this develop ment are quite variable in the literature, depending on cultural background and mental maturity of the child. Brighter children use self-guiding speech in difficult task situations earlier than normal children do (Kohlberg et al., 1968). It has been pointed out that the existence of such a developmental sequence does not imply that maturational 7 factors would prevent acceleration of the development through learning techniques. A number of studies have shown that normal children can be trained in the use of verbal mediators to improve performance on a number of tasks (Bern, 1967; Keeney et al., 1967; Weir and Stephenson, 1959). In relation to hyperactivity, a few studies describing children with symptomatology similar to hyperactive chilren such as learning difficulties, im- pulsivity, and generally poor inhibitory control evidenced by overactivity and distractibility, suggest that these children have a delay or deficit in the development of the self-guiding function of speech (Dykman et al., 1970; Luria, 1961). Three studies have begun to investigate the therapeutic possibilities of providing specific training to develop this function of speech with children having the symptoms described above (Burns, 1972; Meichenbaum and Goodman, 1971; Palkes et al., 1968). The details of these studies will be presented later in the literature review but briefly the results were conflicting, ranging from no improvement in the Burns study to improvement on several test performance scores in the Meichenbaum study. However, there were several procedural differences and differences in the subject samples. None of these studies provided a normal comparison group as a basis for evaluation of task performance improvement, nor did they provide any 8 comparison between the effectiveness of this training vs. medication. The present study will include the normal compari son group in the design as well as a hyperactive group, and will include the addition of a medication trial. Also, to provide a more realistic and extensive evaluation of the treatment, changes in several parameters will be evaluated simultaneously, including activity level and a number of task performance variables. It is hoped that these parameters will reliably differentiate hyperactives from normal children before the treatments and predict differential treatment response among hyperactive children. Regarding activity level, some research suggests that hyperactive children may not be more active in a purely quantitative sense but there may be a qualitative dimension of "situational inappropriateness" (Werry, j.968) . Studies comparing activity level of normal vs. hyperactive children in play situations or simple task situations fail to find significant differences between them in terms of activity (Bell et al., 1969; McFarland et al., 1966), but at least two studies found that in difficult tasks situa tions requiring children to inhibit their movements and concentrate, hyperactives were significantly more active than normal children (Pope, 1970; Sykes et al., 1971). Unfortunately results are not clear cut, partially due to the different types of measuring instruments used, between 9 which correlations are generally low (Cromwell et al., 1963). Activity level has been used as an index of treat ment response in a number of studies (Burns,~ 1972; Sprague et al., 1969, 1970). Also several recent articles relating activity level to treatment response suggest that relatively more hyperactive subjects respond better to stimulants than the less active subjects (Fish, 1971; Howell et al., 1972; Millichap and Johnson, 1972; Post and Tiber, 1974). Considerable refinement and replication is needed before activity level can be clinically utilized as a diagnostic index and/or predictor of treatment response. The present study will further explore this possibility, measuring seat sactivity level during task performance with the electronic activity monitoring setup employed in several previous studies (Post and Tiber, 1974; Sprague et al., 1969, 1970). In addition, activity level has been related to performance on a variety of tasks. It appears high activity levels are correlated with poor performance (Dykman et al., 1970; Pope, 1974; Sprague and Werry, 1974; Sykes et al., 1971), though some investigators have observed that learning performance and activity level are not necessarily related (Douglas, 1974). It has been reasoned that to the extent that activity level and task performance relate to each other and to predicting differential treatment response, some of these takks may 10 be promising diagnostic indices and/or predictors of treatment response. The present study will investigate two tests of conceptual tempo and two operant learning tasks. The two operant learning tasks which will be in vestigated are both simple lever pressing tasks where the child receives points on a screen for lever pressing which is correctly timed. The tasks will differ only in terms of the rate of pressing required to trigger the points on the screen. Studies noting that hyperactive children do poorly on tasks requiring them to monitor a screen and press a lever in response to the stimuli, propose that their poor performance is due to (a) fast and/or irrelevant finger movements on the lever, that is, poor motor inhibi tion and (b) poor attention span so that the screen is not monitored (Cohen et al., 1971; Sykes et al., 1971). Both studies reported improvement following stimulant medica tion. In at least one recent study poor performance on the slow rate operant test similar to the one proposed for the present study, seemed to be predictive of later improvement on stimulant medication (Post and Tiber, 1974), but it is not clear if the poorer performance is due to poor attention to the screen, thus not connecting response rate to reinforcement, or to inability to inhibit fast finger movement. Therefore, the second operant task will be a high rate task. Comparing differential performance on both of these would provide a clearer statement as to the specific impaired function involved in poorer perfor mance on the DRL task, that is, on the slow rate response operaht task. Two tasks measuring conceptual tempo will also be included. Conceptual tempo is defined as the degree to which the child reflects in considering the solution to a problem (Kagan, 1966), and is generally described along a continuum from impulsivity to reflectivity. Hyperactive children are often described as impulsive, and research has shown that impulsivity is highly positively correlated with a tendency to make errors on a number of tasks (Kagan, 1965). The most widely used test for differentiating im pulsive from reflective normal children is the Matching Familiar Figures Test-MFF (Kagan et al., 1964) with impulsive subjects being defined in terms of a greater number of errors and shorter response latency. A number of studies suggested impulsive conceptual tempo in normal children can be modified (Debus, 1970; Kagan et al., 1966 B; Yando and Kagan, 1968), but only two treatment studies have employed the MFF test as a measure effect with hyper active children (Campbell et al., 1971; Meichenbaum and Goodman, 1971). In the Meichenbaum study improvement on the MFF following treatment was correlated to improvement on the other test of impulsivity, the Porteus maze test (Porteus, 1965). This is significant considering that the 12 Porteus Maze test has been well standardized and repeatedly shown to be a sensitive index of drug effect with hyper active children (Conners et al., 1964; 1969; Epstein et al. 1968; Sprague et al., 1973 A). However, neither study employed more than one treatment. Also the Meichenbaum study did not have a normal comparison group, while Campbell's studies suggested the MFF differentiated hyper actives from reflective normals but not from impulsive normal children. Therefore, the present study will analyze the relationship between activity level and per formance on the Porteus Maze, the MFF, and the operant tasks, in normal and hyperactive children to investigate whether these measures can differentiate hyperactive children from normals, and possibly subgroups within the hyperactive sample in terms of pre-treatment measures and correlated differential treatment response. CHAPTER II REVIEW OF LITERATURE In reviewing the literature relevant to the theo retical framework for this study, data concerning com parisons of hyperactive and normal children on the various parameters will be presented in the context of discussing treatment studies rather than in a separate chapter, since most of this comparison data was obtained as preliminary testing in treatment studies. It seems this form of pre sentation provides more depth of understanding for each study and less redundancy. The first section of this review will present some of the clinical and laboratory data on stimulant treatment of hyperactive children and will discuss possible reasons for conflicting results. Then, additional laboratory and clinical stimulant treatment studies which focus on dif ferential treatment response among hyperactive children will be presented. The concept of arousal and its relationship to diagnosis and differential treatment response will be discussed here. The third section will present data providing the theoretical basis for investi gating activity level as a diagnostic index and/or 13 14 predictor of treatment response. However, a good deal of data on activity level will be presented in other sections discussing treatment studies, since activity level has been used as index of treatment response. The next section will deal with clinical treatment studies of behavior modifi cation with hyperactive children, and more theoretically based laboratory studies of operant learning task perfor mance in hyperactive children. The concept of conceptual tempo will be defined in the following section including literature treating the modification of impulsivity in normal children. The final two sections will present relevant theoretical and experimental literature on the development and modification of verbal mediation. In particular, the focus will first be on the development of self-guiding private speech in normal children, and then on the therapeutic potential of verbal mediation training for impulsive and hyperactive children. Stimulant Treatment of Hyperactive Children: General Comparisons of Hyperactives vs. Norma3s The results of studies examining the efficacy of stimulant drugs with hyperactive children are contra dictory for nearly every criterion measure: learning tasks, standardized psychometric tests, rating scales, and activity level. Seemingly equally conflicting are the 15 results of comparing normal and hyperactive children on these measures. Several factors may account for the variability in results: variability in subject samples, type of stimulant (amphetamine or methylphenidate) dosage, context of measurement, and type of measuring instrument. The subject samples have been described as emo tionally disturbed, children with learning disorders, mentally retarded hyperactives, and hyperactives of normal intelligence. Some of the subject samples include children with other severe neurological symptoms, while some studies included subjects evidencing only overactivity, distrac- tibility and poor impulse control. As for context of measurement, there are differences in terms of laboratory measures vs. home and/or school environment. Even within the laboratory studies, differences in type of task situa tions, especially difficulty level, have been found to be salient variables. The type of measurement, ranging from various teacher and parent rating scales to sophisticated elec tronic physiological monitoring devices, has repeatedly been found to influence the results of a study. Large discrepancies exist between clinic, parent and teacher behavior ratings, when compared to objective laboratory measurement of activity level and task performance (Blackridge and Ekblad, 1971; Conners and Rothschild, 1968; Sprague etal., 1973 B). For example, children on 16 medication are often rated by teachers as improved irre spective of dosage level while concomitant laboratory measures of task performance show a decline in performance as the dose is raised above a certain level (Sprague and Werry, 1973). A recent survey of studies published between 1937 and 1971 observed that only 17% of the studies employing psychological tests reported significant results while 57% using behavior rating scales and 88% using clinical impressions reported significance (Sulzbacher, 1973). Conners (1970) has shown that weighted symptom scores based on ratings from teachers have been shown to decrease significantly before the drug was actually ad ministered, and to decrease significantly in both control and placebo group, as well as in the drug group. In general, it seems the more subjective the ratings the more likely the treatment will be found to be effective. With these variabilities in mind, the following results have been reported in the literature. Subjective behavior ratings by parents, teachers, or psychiatrists focusing on activity level clearly attest to the success of stimulant treatment (Conners et al., 1967; Conners et al., 1969; Conners and Rothschild, 1968; Knights and Hinton, 1969; Zrull et al., 1963). The following improvements have also been observed by raters: less distractibility at home (tfeiss et al., 1968), improved 17 contacts with teachers (Sprague et al., 1969; 1970), improved attitude toward authority (Conners and Roths child, 1968) and less impulsivity at home (Millichap and Boldrey, 1967). However, in one study where both ratings and objective tests were employed, ratings indicated that the stimulant medication was helpful but performance on tests did not improve (Blackridge and Ekblad, 1971). In general, the efficacy of stimulant treatment for hyperkinesis is less impressive when considering changes in objective criterion measures following treat ment. In general, no drug effect is reported on general intelligence test scores (Conners et al., 1964; Conners et al., 1969; Werry and Sprague, 1970; Zrull et al., 1963). However, a few investigators have reported bene ficial drug effects on some performance IQ subtests (Conners and Rothschild, 1968; Knight and Hinton, 1969) and some verbal subtests (Conners and Rothschild, 1968; Douglas et al., 1969). A variety of learning tasks including visual dis crimination (Conners, 1966; Conners and Rothschild, 1968; Sprague et al., 1969; 1970) paired associate learning (Conners et al., 1969; Conners and Rothschild, 1968) and automated recall (Barnes, 1968) have reflected improvement following stimulant medication. Most impressive has been the improvement occurring on the Porteus Maze test, a test 18 of impulsivity, Indicating that stimulant medication does help decrease impulsivity in hyperactive children (Conners et al., 1969; Conners and Rothschild, 1969; Eisenberg, 1968; Epstein et al., 1968). Werry (1970), summarizing the effects of stimulants on a wide variety of task per formance measures, postulates that the improvement is due to improvement in general functions such as ability to ignore distractions, response speed, and vigilance or attention. Attention has often been invoked as one of the most important factors in learning, particularly discrimi nation learning (Broadbent, 1962; Zeaman and House, 1963). Specific studies focusing on drug effects on these variables lend support to Werry's notion. For example, reaction time has been found to be significantly slower in hyperactive children (off medication) than normals (Dykman et al., 1970; Stevens et al., 1968 B) on a Luria- type conditioning task where the subject was required to press a button in response to a particular visual stimu lus. Further, administration of stimulant medication to twelve "emotionally disturbed" boys (nine of whom were hyperactive) resulted in a decrease in their reaction and simultaneous improvement of performance on a similar visual-motor task requiring attention and more complex stimulus discrimination (Sprague et al., 1969; 1970). Thus, it seems the stimulant enhanced attention as well as reaction time. In apparent contradiction to these studies, Campbell and her co-workers (1971) found reaction time of hyperactive children to be significantly faster than normal subjects on a visual matching test (with a signi ficantly greater number of errors than normals) and to increase following methylphenidate administration. Sig nificantly, the number of errors decreased following medi cation. One possible explanation for this difference may lie in the difficulty level of the task employed. Campbell's task involved discrimination among seven stimul simultaneously, while Sprague's task involved a maximum of three simultaneous stimuli. Conceivably, reaction time would depend on the level of difficulty of the task so that the faster reaction time of the hyperactives on the difficult discrimination task (and poorer performance) would suggest a deficiency in cognitive strategy. Such deficiencies have been correlated with impulsive concep tual tempo (Drake, 1970; Siegelman, 1969) and will be dis cussed in greater detail later. On the other hand, one would expect the faster motor reaction time in normals during the simple color-light discrimination tasks used by Dykman and Stevens, since the motor response depends less on cognitive strategies per se and more on sustained attention, the latter being frequently found deficient in hyperactive children. 20 Sykes and associates (1971) investigated several dimensions of attention in hyperactive children and the effects of methylphenidate on this variable. They com pared the initial performance of a normal control group to that of a group of hyperactive children on a visual vigilance-attention task, manipulating the duration of the interstimulus interval (1.0 seconds and 1.5 seconds). On the pre-drug test, the hyperactive group made signifi cantly fewer correct responses (detecting the signal stimulus) than the normals, but there was a nonsignificant difference in the number of errors (responding to other than the signal stimulus). This seems to indicate that the hyperactive children had lapses of attention, there fore missing some of the signal stimuli, while they did not have difficulties with the perceptual discrimination. Also, both groups made a similar number of errors during the fast interstimulus interval but the normal children made significantly fewer errors during the slow inter stimulus interval. This suggests that the hyperactive child's impulsivity may have prevented him from making good use of the extra time to evaluate the stimuli. Treat ment with methylphenidate resulted in a significantly better overall score on the task, and significantly fewer errors during the slow interstimulus interval. The authors conclude that the stimulant medication enhanced two 21 factors, attention and capacity to Inhibit responses, Conners and his associates (Conners et al,, 1969; Conners and Rothschild, 1968) administered Dexedrine (an amphetamine) to children with learning disorders (some of whom were hyperactive) and found the following changes in test performances: 1) significant improvement on visual and auditory attention tasks, on the Porteus Maze test (a test requiring capacity to inhibit and plan ahead), and on a paired associate learning test; 2) no improvement on tests of memory, intelligence or perception. The authors conclude that stimulant medication primarily improved two functions essential to the learning process— attention and the capacity to inhibit. However, the authors caution: It is recognized that no single psychological func tion exists independently of the operation of other functions, so that an attempt to specify isolated mechanisms which are influenced by drug action is extremely difficult. (1968, p. 200) It is known that the physiological arousal mecha nism plays an important role in perceptual and attentional processes, as well as in controlling autonomic and skeleto-motor functioning on the organism and amphetamines and methylphenidate are central nervous system stimulants known to operate on central arousal structures. Thus, there seems to be a consistent relationship in the hyper active child between stimulation of the arousal mechanism and ability to pay attention and inhibit responses. Most 22 of the theories concerning the physiological mechanisms underlying hyperactivity invoke the concept of arousal. Although there are a few studies which have directly inves tigated arousal in the hyperactive child using physiologi cal indices (Boydstun et al., 1968; Satterfield et al., 1971, 1972, 1973; Stevens et al., 1968A and B; Tiber and Trozzolino, 1972) most of the theorizing has resulted from observing the response to stimulant medication and a knowl edge of the pharmacological action of these drugs. The next section presents additional clinical and laboratory studies on stimulant treatment of hyperactive children. These studies however, focus on differential treatment response within the hyperactive sample, sug gesting diagnostic heterogeneity. Again the concept of arousal is invoked to elucidate this differentiation. Stimulant Treatment of Hyperactive Children II: Arousal, Differential Diagnosis, and Differential Treatment Response "Arousal" is defined as the overall level of ex citation of the individual as measured by a number of physiological indices including skin conductance (GSR), heart rate, blood pressure and EEG patterns (Duffy, 1962). The level of arousal at any given time is conceived as a result of the interaction between the significance of the environmental stimuli impinging on the organism and the organism's metabolic condition, which is determined by such 23 things as fatigue, hormone balance, muscular activity, and drugs. Duffy (1962) notes that "motivation" and "emotion" are concepts related to arousal, but they incorporate both the dimensions of direction and intensity of behavior, while arousal only refers to the intensity dimension. The relationship between arousal and performance on a given task has been described as an inverted U-shaped function (Duffy, 1962; Malmo, 1959). That is, as arousal level increases from a very low level, efficiency of per formance will increase up to a point which is the maximal activation level for that particular task. As arousal is further increased, there will be a decrement in perform ance. The optimum level of arousal depends on the task and the individual, but, in general, moderate levels of arousal result in the best performance. This relationship has been found between reaction time and arousal as well as quality of performance on a wide range of tasks (memory, attention, auditory tracking, perception). Since the performance of hyperactive children im proves with an increase in arousal (following stimulant medication), several authors conclude that they are under aroused physiologically (Conners, 1970; Satterfield and Dawson, 1971; Wender, 1971). Therefore, it is conceivable that too high a dose would result in poor performance if the inverted-U relationship exists. Sprague and his asso ciates (1969) substantiate this relationship in a study 24 investigating the effect of six dose levels on a learning task which had three difficulty levels. A significant dosage X difficulty level effect obtained. That is, dose level significantly affected accuracy on the most difficult tasks. Also, accuracy improved as the dosage was increased to .3 mg. per kg. of body weight, but declined with a dose of .4 mg. per kg. (a relatively small dose range, especial ly considering that clinicians commonly prescribe from 10 to 80 mg. methlyphenidate per day). The children in the drug groups were not rated as significantly less active, less noisy, or more attentive, behaviors which are often of great concern to parents. The dose levels used in this study were considerably below most levels prescribed clini cally which raises the interesting possibility that higher doses are usually prescribed in order to achieve more acceptable behavior, at the expense of the child's ability to learn. This is also suggested in another study (Epstein et al., 1968) which investigated differential responsive ness to dextro-amphetamine of two groups of hyperactive children, "organics" and "nonorganics," separated on the basis of history of cerebral trauma. The dose level was individually adjusted; on the average, the organic group tolerated a dose of 20 mg. per day while the nonorganics averaged 15 mg. per day. The results interestingly show that the nonorganics (lower dose) were not rated as 25 improved, since their motor activity did not appear to decrease, yet they did improve on all tasks (motor co ordination, Porteus Maze, WISC). On the other hand, the organic group (higher dose) were rated as significantly less active, but did worse on the verbal subtests. (Of course, the findings are confounded by differential or- ganicity in the groups.) Differential responsiveness to stimulants within groups of hyperactive children has often been found (Cutts and Jasper, 1939; Freedman, 1971; Ginn and Hohman, 1953; Post and Tiber, 1974; Satterfield et al., 1971, 1972, 1973), leading some researchers to search for indices which might better characterize subgroups of children sub sumed under the same diagnostic category— hyperactivity. The EEG has been the most frequently investigated physio logical index in hyperactive children. Normally, the frequency and amplitude of the wave pattern of the EEG varies with arousal level on a continuum from deep sleep to excited conditions. High amplitude low frequency (slow) waves predominate in deep sleep. There is a progressive decrease in amplitude, increase in fre quency, and reduction of synchronization of the wave pat tern as the individual moves from deep sleep through relaxed wakefulness (characterized by the alpha wave of 8 to 12 cycles per second), alert wakefulness, to excited states. Beta waves, about 18 to 30 cycles per second, 26 become increasingly predominant as the person becomes more alert (Jasper, 1941). The activation pattern (desynchro nization of alpha) is elicited by the presenting of a novel stimulus and is subject to habituation so that as the stim ulus is repeatedly presented, it is less effective in blocking alpha. Orienting responses and habituation are also reflected in the galvanic skin response (GSR), another index of arousal (Lindsley, 1951). It seems reasonable to suggest that if a group of individuals are consistently found to have slower EEG pat terns, their level of arousal is lower. This seems to be the case with hyperactive children. Unfortunately, many of the studies of EEG patterns in hyperactive children are clinical evaluations rather than systematic laboratory in vestigations. As early as 1939, Cutts and Jasper examined the EEG's of twelve "behavior problem" children described as hyperactive, impulsive, and of average intelligence. The chief abnormal patterns they noticed were two types of slowing, one of which was found in seven out of twelve children. Coincidentally or not, after administration of benzedrine, seven out of twelve cases improved (none im proved following a sedative), the same seven who exhibited the first slowing pattern. "Improvement" was defined as an increase in cooperativeness and a decrease in activity and impulsivity. Burks (1964) studied the differential amphetamine response of two groups of hyperkinetic children, separated on the basis of normal versus abnormal EEG patterns. (He did not specify the type of abnormality.) Both groups im proved (based on teacher rating scales), but the normal EEG group improved significantly more. Knights and Hinton (1969) studied EEG records of thirty-three children diag nosed as minimally brain damaged and found that one-third were "abnormal." One-half of the abnormalities were in creased slow activity. An additional six records demon strated a minor increase in slow activity. Stevens and his colleagues (1968A) found a significant correlation (but not perfect) between slowing of EEG frequency and hyperactivity. The foregoing studies all suggest hetero geneity in at least one index of arousal (pre-treatment) among children with the same diagnosis, and some of the studies also report correlated differential treatment re sponse . Researchers, utilizing other indices of arousal have obtained similar results. Satterfield and Dawson (1971) found three arousal indices— basal skin conductance (SCL), nonspecific GSR, and specific GSR (GSR response to an auditory stimulus)— differentiated hyperactives from normals. On the pretesting, the indices indicated sig nificantly lower median resting arousal level and signifi cantly smaller orienting response (OR) in the hyperactive group as a whole. However, there was wide interindividual variability in the hyperactive group with the following results suggesting the existence of distinct diagnostic subgroups: 1) on the pretest, only twelve hyperactives out of 24 demonstrated significantly lower arousal (SCL), ten were within the range of the normalfiTT and two had higher arousal; 2) following stimulant medication, 75% of the hyperactives demonstrated an increase in arousal level. Unfortunately the authors do not specify whether those that showed the increase in arousal after the medi cation were those who were underaroused on the pre-drug testing. They do speculate that those hyperactives with low SCL might respond best to stimulants while the other hyperactives would require a different treatment. Two subsequent studies by Satterfield and his co workers similarly found some subjects improved with stimulant medication, while some deteriorated (1972; 1973). In comparing the "good response" subgroup and the "poor response" subgroup to a normal comparison group, pre treatment indices of arousal (skin conductance and EEG amplitude) indicated lower than normal arousal in the good response group and a higher than normal arousal level in the poor response group. Following treatment, however, neither of these indices showed significant changes cor related to the significant behavior rating changes reported. Comparing normal children to minimally brain 29 damaged (MBD) children revealed significantly wider interindividual variability in the latter group on several indices of arousal in at least two other studies. The first one (Stevens et al., 1968B) failed to obtain signif icance in comparing the two groups on any single index, including heart rate, muscle potential or skin resistance, possibly due to the extreme variability in the clinical group. Similarly, Boydstun and his associates (1968) , comparing 26 MBD children to 26 normals on a tone dis crimination and conditioning task, found that resting arousal indices (heart rate, skin resistance, and muscle potential) did not significantly differentiate the two groups but the controls tended to be more homogeneous. However, the MBD group were, on the average, less reactive in skin resistance and muscle potential, significantly so on the first tone, indicating a weaker orienting response (OR) . Cohen and co-workers (1971) found a group of hyperactive children to be deficient in the orienting re sponse in comparison to a normal group of children, using a skin conductance index and auditory signals. The hyper actives also habituated faster to the stimuli and were unable to inhibit irrelevant motor responses. That is, they continued to press a response button after having made the response. Administration of methylphenidate re sulted in an increase in reaction time, better ability to 30 inhibit irrelevant motor responses, and increases in resting arousal level and reactivity. Luria (1963) regards continual orienting during early stages of learning to be important in facilitating learning and performance. A relative lack of responsivity and rapid habituation causes the child to stop actively analyzing the stimuli. This would explain why hyperactive children appear to have high distractibility and short attention spans. As noted earlier, structures involved in the arousal system play important roles in perceptual and attentional processes, as well as in controlling autonomic and skeleto-motor functioning. Several studies have investigated performance on various tests as possible predictors of drug response. A number of other studies finding differences in response to stimulants among supposedly homogeneous groups of hyper active children reported that a large number of the vari ables employed were not predictive of drug response (Knights and Hinton, 1969; Sprague and Werry, 1973; Weiss et al., 1971). Among the nonpredictive variables were WISC performance, reading achievement test, birth history, neurological exam, IQ, parent and teacher behavior ratings, and socioeconomic status. Conners (1971), using a sta tistical technique called cluster analysis, concluded that tasks measuring attention are most predictive of drug re sponse. That is, poorer performance on tasks requiring 31 continuous attention to a visual stimulus, and poorer performance on paired associate learning tasks, seemed most predictive of positive responses to stimulant medi cation. Reviewing some of the literature on hyperactivity, Marwit and Stenner (1972) described two types of hyper active children, the "hyperactive pattern" and "hyper reactive pattern." The first was associated with imma turity, early onset and inability to inhibit impulses and activity. Stimulants would be highly effective for this t . oup, they predict. The second pattern is characterized by some inhibitory defect, but emotional factors are be lieved causal in this pattern, wic.ii a later onset (mid to late childhood). They predict that for this group, stimu lant medication may be ineffective or less effective than in the former group. The problem still lies in finding a simple index to reliably differentiate these two types, if their predictions are correct. The only measures thus far reviewed which show some consistency across studies in differentiating sub groups of hyperactive children are arousal indices though the consistency is far from perfect. Summarizing the studies presented thus far and invoking the inverted-U re lationship between arousal and performance, it seems there is a subgroup of children diagnosed as hyperactive who fall at the lower end of the arousal continuum and for whom 32 stimulant medication is indicated since it raises their arousal enough to improve performance and learning (as suming the dose is not too high). There seems to be anoth er subgroup of children subsumed under the diagnostic label of hyperactivity whose arousal level lies in the normal range (or higher) and for whom stimulant medication is not indicated. However, measurement of arousal is generally a technically complex, expensive procedure and therefore does not promise to be a practical index for clinical situations. It is essential to develop a simple objective criterion index for deciding a priority whether or not a given hyperactive child is a candidate for stimu lant medication. One promising predictor is activity level. Activity Level The term "hyperactivity" seems to imply a disorder characterized by a quantitative excess of movement of all kinds evident in all situations. Research suggests, how ever, that hyperactive children may not be more active in a purely quantitative sense, but there may be a qualitative dimension of "situational inappropriateness" (Werry, 1968). Other investigators (Bindra, 1961; Morgan and Stellar, 1950) have suggested analyzing activity level in terms of specific dimensions such as locomotor versus dif fuse, relevant versus irrelevant, and goal-directed versus 33 non-goal-directed. Consideration of these dimensions is certainly critical in reviewing the literature on activity level in hyperactive children. Another important consider ation is the type of measuring instrument used since the correlations between the various instruments are generally poor (Cromwell et al., 1963). Several studies measuring the activity level of hyperactive children directly with mechanical devices have found that hyperactives are not more active than a normal matched sample. In a free play situation, Bell and his colleagues (1969), employing pedometers (instruments at tached to the subject's leg, measuring number of leg move ments) which primarily measure gross motor activity, did not find hyperactive nursery school children to be signif icantly more active than normals. McFarland and his associates (1966) similarly found that 30 retarded chil dren (9% hyperactive) were less active than normal chil dren when playing alone in a small room. Hutt and his co workers (1963) used an experimental room with a grid marked floor and a camera filming the children's activity. They observed that the hyperactive epileptic children spent less time in any one activity than the normal children, again suggesting a differentiating qualitative dimension of activity. There are studies, however, which have found sig nificant quantitative differences. Sykes and associates 34 (1971) measured activity using a stabilimetric cushion (placed on child's seat, it measures movement in three di mensions) while the child performed a vigilance-attention task. They found significantly more activity in the hyperactive group. Pope's study (1970) suggests an ex planation for the discrepancy between the studies men tioned. Comparing activity levels of normal and hyper active children in four types of situations, significant differences obtained during the difficult task situation and during a voluntary inhibition task (child asked to sit quietly in full view of toys), but not during a play situation or during a simple task situation. Thus it appears that hyperactive children are more active than normals in situations requiring voluntary inhibition and sustained attention and concentration. This relationship between activity level and at tention is also suggested in a study by Gardner and his co-workers (1959). They investigated the effects of reduced versus increased visual stimulation (black screen surrounding the subjects versus a screen with multicolored lights and trinkets) on four groups of children: hyper active, hypoactive, organic retarded and familial retarded. All groups were more active under the reduced stimulation condition. It seems reasonable to postulate that atten tion responses, occurring under increased visual stimu lation inhibited motor activity. Alabiso (1972) has 35 similarly suggested that attention functions to reduce motor activity. Deutsch and Deutsch (1963) describe the highly active subject as paying attention to everything but selective attention to nothing. Sprague and Toppe (1966) conducted a study which further supports the inverse relationship between activity level and attention. They measured activity level (using a stabilimetric chair) of 30 retarded subjects during a simple discrimination task under a twelve-second delay of reinforcement condition. The performance of the most ac tive subjects (upper quartile of activity) was signifi cantly poorer than the low active group (bottom quartile) throughout the task. The authors propose that the high active group did not attend long enough to associate nega tive reinforcement (delivered twelve seconds after the response) with the correct or incorrect response. At least three studies found stimulants had no effect on activity level. Davis and his co-workers (1969) were unable to reach any significant conclusions concern ing activity response to methylphenidate, using a tele metric record of overall activity, because of extremely wide variability in response. McConnell and his associ ates (1964) found no drug effect in either the high or low active group of retardates (median split on actives) measuring activity using the ballistograph while the sub jects sat alone in a room for four minutes. They were measured tinder three drug conditions— placebo, 7> 5 mg, and 15 mg. When the ten most active subjects were compared to the ten least actives, there still was no significant drug effect. In a third study, stabilimetric chair activity during performance on a simple motor task in a group of retardates did not decrease with stimulant medication, al though task performance improved (Schnickedanze, 1967). The results of these studies are not contrary to expec tation. It was mentioned earlier that activity differen tiated hyperactives from normals only in situations re quiring sustained attention and ability to inhibit. Neither the Davis study nor the McConnell study employed task situations during the measurement and it is question able how much sustained attention and inhibition was required for the marble dropping task (Schickedanze, 1967). By comparison, at least four studies finding sig nificant drug effects on activity measured the activity level during performance of cognitive tasks. Breitmeyer (1969) found a significant decrease in seat activity following a dose of .45 mg. per kg. body weight of methyl phenidate. The subjects were male retardates, ages 6 to 20. Another study (Millichap et al., 1968) measured body activity of subjects during performance on psychological and learning tests, pre-drug and on drug (1.5 mg. per kg. body weight). The trend (.05<p<Al) was as expected— 37 activity decreased with stimulant medication. Testing behavior problem boys of normal IQ (ages 6 to 11), Barnes and co-workers (1968) found that administration of methyl phenidate (.25 to .35 mg. per kg.) resulted in a decrease in seat activity and increase in accuracy and faster re action time on an automated recall task. Sprague and co-workers (1970) using the same medication and dosage range with twelve emotionally disturbed boys (nine hyper active) also found decreased seat activity (using stabilimetric chair) and improved accuracy and reaction time on a discrimination learning task. Summarizing studies which examine the effect of medication on seat activity during performance of a task, Werry (1970) proposes that the decrease in seat activity is a secondary effect, secondary to improved task perform ance, since the decrease is usually found in studies where performance improves. Similarly, Sykes and associates made the following comment: In the light of the Cromwell et al. (1963) suggestion that the restlessness of hyperactive children is not mere activity per se, but continued shifting from task to task because of short attention span, it may be that the effect of methylphenidate is to improve the ability of hyperactive children to pay attention . . . such that they shift less from task to task and thus appear less restless or active. (p. 138, 1971) It was mentioned earlier that the central nervous system arousal mechanism plays an important role in atten tion. One might surmise then, that if poor attention poor 38 task performance, and high activity level are correlated, arousal indices would also be correlated with activity level. In fact, Cromwell and his associates (1963) did find such a relationship. They simultaneously measured activity level (using the ballistographic chair which is suspended on cables), reaction time and alpha blocking latency (an arousal index tapping the time lapse between presentation of a stimulus and the EEG shift to an acti vation pattern), finding that highly active subjects demonstrated the slower reaction time and alpha blocking latency. These results are typical of the relationship of these variables found in one subgroup of hyperactive chil dren: high activity, low arousal, and slower reaction time. It seems likely that the hyperactive child who demonstrates this pattern would be responsive to stimulant medication. Two recent studies have found relatively higher activity levels amount hyperactives to be correlated with positive response to stimulants (Millichap and Johnson, 1974; Post and Tiber, 1974). Millichap and Johnson found that the 14 more active subjects (according to actameter measuring device) tended to have a greater number of neurological abnormalities and were more likely to show a positive response to the stimulant medication (Ritalin) than the low active subjects. However, the re sults certainly were not clear-cut: in the high active 39 group, 13 of the 14 subjects demonstrated, as expected, decreased activity level following medication, but in the lower active group, 6 subjects showed a decrease and 8 subjects showed an increase in activity level following medication. Post and Tiber (1974) measured activity level (using the stabilimetric chair) of hyperactive children on placebo versus Ritalin (methylphenidate), while they per formed a variety of tasks and obtained the following results: 1) two subgroups of hyperactives were signifi cantly different from each other on activity in the place bo condition; 2) the activity level of the high active subgroup decreased significantly in the drug condition while the activity level of the other subgroup actually increased (although nonsignificantly). They also found that tasks performance on a directed activity task and performance on a differential low rate operant task (lever-pressing) differentiated these two groups, sug gesting the possibility that, given normative data, these tasks may be useful clinical predictors of treatment re sponse. The results of these two studies suggest that ap parently contradictory results on studies of stimulant treatment in hyperactive children may in part be accounted for by the subgroup differences which mask significant main effects. Also, neither of these studies employed a 40 normal control group so that it is not clear how the low active subjects compared to normal children in terms of activity level. It may be there was no difference at all, or, they may have been below normal activity level, in which case, an increase in activity level would be inter preted as improvement. It seems activity level, recorded during difficult task performance merits further investigation as a possible diagnostic index for differentiating hyperactive from normal children, and possibly subtypes of hyperactive children. Finally, if activity level is found to have diagnostic value in terms of prediction of treatment re sponse, performance on some brief task which consistently correlates significantly with activity level could serve as a practical clinical index. Operant Conditioning and Treatment of Hyperactivity A number of studies have shown that many of the behaviors associated with hyperactivity, ranging from ex cessive motor behavior to poor attention span, can be improved with operant conditioning techniques. With regard to motor behavior, Anderson (1964) was the first to demonstrate that out-of-seat behavior in moderately re tarded hyperactive children could be effectively reduced by making teacher attention contingent on sitting behavior. Later, Pihl (1967) reported increasing sitting behavior 41 in hyperactive boys of normal intelligence who had failed to benefit from drug therapy. Most recently, two re searchers (Christensen and Sprague, 1973) report that stimulant medication plus reinforcement for reduced seat activity resulted in more dramatic and sustained decreases in seat activity than reinforcement alone. A light flashed for each minute duration that the subject's activ ity level was below baseline, while a point counter cumulated all the light flashes. Later the points were exchanged for money. The criterion for reinforcement was adjusted every three sessions. Patterson and his colleagues (1965) reinforced a ten-year-old hyperactive brain injured boy for each 10 second time interval in which he remained seated and working on his assignments in the laboratory. Reinforce ment was a buzzer signal accompanied by candy. After several days of laboratory training, the boy was observed to be significantly less active in the classroom compared to his baseline activity ratings and compared to a hyper active control subject. Other investigators replicated Patterson's results in non-retarded hyperactive subjects using primary (Grindee, 1970a; Nixon, 1966) and secondary reinforcement (Allen et al., 1967; Daniels, 1973; Grindee, 1970b). Daniels (1973) demonstrates that parental social reinforcement for sedentary non-hyperactive behaviors in a boy suffering from ulcerative colitis and hyperactivity 42 produced not only a decrease in overactivity but a concomitant significant decrease in two psychosomatic symptoms, vomiting and excessive bowel movements. These results occurred after only two weeks of reinforcement therapy and continued at the time of the three months follow-up. One shortcoming in the studies mentioned thus far is that there is no measure of correlated change in aca demic performance along with changes in activity. An experimenter in another study observed that although training in sitting behavior was effective, there was not an improvement in academic performance (Patterson, 1965). Another important finding of this study was that his pro cedure resulted in better peer relations; poor peer re lationships is a common problem for hyperactive children. He achieved this by telling the class to ignore the sub jects hyperactive behaviors and that they would also receive a piece of candy each time the subject received one. A reinforcement light signaled the subject and his classmates each 10 second sitting response. Thus the light flash became a cue for the subject's classmates to give him praise and other social reinforcements. In order to achieve better learning as well as de creased activity, some researchers reason that operant reinforcement of attention responses and/or actual task performance would be preferable to reinforcing sitting behavior. Quay and his colleagues (1967) reinforced orienting responses in a group of 5 hyperactive intel lectually average subjects during a classroom period. The reinforcement, on a variable ratio schedule, consisted of a light flash controlled by an observer for each 10-second interval of attention. The light, mounted on each child's desk, signaled him that he earned candy, but he did not receive the candy until after the class ended. Later, re inforcement was switched to praise alone. There was a highly significant increase in the number of orienting re sponses under both kinds of reinforcement, but during extinction period there was a significant decrease in the number of orienting responses. Perhaps some form of self reinforcement training would have prevented this decrease. There was no reported change in actual academic perform ance, again, a short-coming in interpreting the signifi cance of their results. Walker and Buckley (1968) reported that classroom distractibility was substantially decreased after laboratory reinforcement training for intervals of at tention to a teaching machine. Points were given to the subject for each 10-seconds of attention. These points were later exchanged for toys. Previously the boy had been placed on a special education token-economy program which resulted in an increase in his social skills and academic performance, but he had still remained very distractible. At least three other studies report improved aca demic performance as a result of operant conditioning. One study (Dubrose and Daniels, 1966) reinforced six over- active mentally retarded boys, age 8 to 13, in a laboratory playroom setting (one boy at a time), for engaging in puzzle-solving, drawing, and writing activities. The boys received tokens dispensed by remote control on a fixed schedule. After 12-days of training, hyperactive motor behaviors decreased by one-third and there was a signifi cant increase in appropriate use of play media. The positive results continued after extinction and at the 6-day follow-up observation. The other two studies differed in that the hyperactive subjects were of normal intelligence. Kennedy and Thompson (1967) reinforced one boy for each 60-second interval during which he observed the experimenter and followed his instructions. After several days of training in the laboratory, the subject was observed in the class room. A significant increase in visual attention span obtained. The teacher also reported concomitant increases in arithmetic scores and assignment completion. Knowles and Associates (1968) reports results with a 7 year old hyperactive boy, who also reversed letters in writing. He went to a special school once a week where the teacher re inforced him with candy when he walked slowly and sat 45 quietly at his desk. After 5-weeks, his rewards were transferred to correct letter copyings four candies for each correct letter turned in. The paper did not specify how many weeks the boy was trained in letter copying, but the authors report that 6-weeks after the training ses sions ended at the special school, he was found to be significantly less active and to consistently write letters correctly in his regular class. Not all studies of operant conditioning with hyperactives report improvement. Stevens and Associates (1970) report that a group of hyperactive subjects did not increase their finger-tapping speed when offered re inforcement, while a normal control group did demonstrate a significant increase. However, the hyperactives were significantly faster than the normals in the non-reinforce ment condition, so the possibility of ceiling effect exists. Freibergs and Douglas (1969) report that hyper actives were significantly poorer on a concept learning task than normals under conditions of partial reinforce ment, whereas, no significant differences were obtained between these two groups under continuous reinforcement and delayed reinforcement conditions. The authors inter pret the poorer performance under the partial reinforce ment condition as evidence of lower frustration tolerance. In other laboratory studies, finding that hyperactive children perform poorly on operant learning tasks (Boydstun et al., 1968; Douglas et al., 1965; Post and Tiber, 1974), there are so many other psychological factors involved in the task performance that the results cannot necessarily be attributed to lack of sensitivity to incentives. For example, in the Boydstun study (1968), pennies were offered to hyperactive and control subjects as reinforcement for correct responses on a tone dis crimination task. Yet, only 62% of the hyperactive chil dren could reach the criterion of five consecutive correct responses, while 95% of the controlled group did. These results could indicate an auditory attention deficit un responsive to incentives. The others report that the subjects' impulsivity resulted in so much indiscriminate responding that the penny machine was emptied. Post and Tiber (1974) found that performance on a differential low-rate operant task differentiated two sub groups of hyperactive children: one sub-group responded beneficially to stimulant treatment (Ritalin) and the other group did not. These sub-groups also differed significantly on activity level: poorer operant perform ance was associated with higher activity levels and positive drug response. The authors explained that per formance on this task requires both attention to the reinforcement screen and capacity to inhibit speed of finger movements. In general, the clinical studies report that 47 target behaviors in hyperactive children can be brought under operant control. Yet, many of the laboratory stud ies either report the opposite, or report general deficiencies in operant task performance. One possibility is that the reinforcement in the clinical situation was primarily individually determined, that is, the reinforce ment used was known to be reinforcing that particular child, whereas, the laboratory studies did not do this. Also, in most of the clinical studies, the hyperactive child's performance in the incentive condition was com pared to his earlier performance, while the laboratory studies compared the hyperactive child's performance with that of normal children. Finally, the laboratory studies offered reinforcement for correct responses occurring in one session, thus, presupposing the child possessed the necessary cogment of skills, whereas, the clinical studies gradually built-up a desirable response over a number of sessions. This response, such as attending to the in structor, later was instrumental in better performance. Admittedly, few clinical studies offer objective evidence of specific improved task performance as a result of behavior modification. Generally, the evidence consists of teacher ratings which, as mentioned before, are highly susceptible to placebo effect. The most significant im plications of the research presented are: 1) that laboratory operant task performance may be a useful 48 diagnostic index; 2) reinforcement principles may be used to develop behaviors which are instrumental in improving academic performance. One problem at hand is to determine which behaviors are most relevant to academic performance, and deficient in hyperactive children. The literature on conceptual-*tempo and verbal mediation are especially pertinent to this question. Conceptual Tempo Conceptual tempo has been defined as "the degree to which the child reflects in considering the solution to a problem." Conceptual tempo, along with motivational variables and conceptual skills is considered to be an important factor contributing to the quality of problem solving (Kagan, 1966). The tendency to be reflective or impulsive has intraindividual stability over time and generality across a number of problem situations (Kagan et al., 1964). Impulsiveness has been highly correlated (positively) with the tendency to make errors on a number of tasks, including reading (Kagan, 1965), serial learning (Kagan, 1966), matching to sample tests (Kagan, 1965), and inductive reasoning tests (Kagan, Pearson and Welch, 1966). The impulsivity is defined by both the greater number of errors and the faster response time. That is, impulsive subjects, by definition, show faster response time and more errors on a variety of tests. The most widely used 49 test of Impulsivity is the Porteus Maze Test. It is highly validated as a measure of planning ability, fore sight, and capacity to inhibit impulsive responding (Porteus, (1965). It has also been repeatedly found to be a sensitive index of drug response in hyperactive children (Conners and Eisenberg, 1963; Conners et al., 1964, 1969; Sprague et al., 1973A). Generally these studies found Porteus Maze scores improved following stimulant treatment of hyperactive children, although opposite results have also been recorded (Blackridge and Ekblad, 1971; Freston, 1972). A newer test which is being widely used for selecting impulsive and reflective subjects is one vali dated by Kagan and his associates (1964) , the Matching Familiar Figures Test (MFF) consisting of twelve problems, each one picturing a standard picture and six variants of it. The subject'8 task is to select the variant which exactly matches the standard picture. Typically, if the child falls above the median in number of errors and below the median in response time, he is considered "impulsive." The median refers to the performance of the group from which he is being selected. Normative data has not yet been published. To date, this test has not been used as an index of drug effect with hyperactive children but it has been used as an index of effort for another type of treatment with hyperactive children, verbal mediation 50 training (Meichenbaum and Goodman, 1971) . Significantly, Meichenbauxn and Goodman reported correlated changes in the MFF and Porteus Naze Test following their treatment, thus adding to the validity of the MFF as an index of im pulsivity. It has been repeatedly used to differentiate reflective and impulsive individuals in a non-clinical population. Reflectivity on the MFF has been correlated with motor impulse control and both show parallel developmental trends (Bucky et al., 1972). Specifically, between the ages of five and fifteen years, there is an increase in reflectivity and ability to inhibit motor activity. The impulsive reflective dimension is not merely based on dif ferences in amount of time given to consideration but involve specific differences in cognitive scanning strate gies. Studies of 8 to 10 year old children utilizing filming of eye movements during performance on the MFF revealed the following results: 1) reflectives observed more alternatives, in greater detail, and distributed their attention more evenly among the variants while the impulsives ignored many alternatives and distributed their attention among one or two; 2) the reflectives searched for explicit differences, made more comparisons, were more systematic in eliminating alternatives, while the impul sives selected a variant which only globally resembled the standard (Ault et al., 1972; Drake, 1970; Siegelman, 51 1969). Thus the underlying cognitive strategy seems deficient in impulsive children. The impulsive and reflective children who served as subjects in these studies were not a clinical popu lation but rather were selected from normal classrooms. When these children are compared to hyperactive children in terms of MFF performance, the normal reflectives demonstrate significantly longer reaction time and fewer errors than the hyperactive children (Campbell, 1973). Attempts to modify impulsive conceptual tempo have been made by a number of researchers with varying success. Kagan and his associates (1966B) trained impulsive first- graders (based on MFF performance), in three fifty-minute sessions, to defer their answers for ten to fifteen sec onds during which time they were encouraged to study the stimuli and think about their answers. The MFF was not used in training, but an alternative form was used in the posttest six to eight weeks later. The results revealed an increase in response time but no change in errors. Similarly, Yando and Kagan (1968) obtained an increase in response time and no change in errors using a modelling paradigm. They had preselected reflective teachers and then measured MFF performance of their students at the beginning and end of the school year. The authors of both studies conclude that a modification technique must in clude teaching better cognitive strategies to achieve 52 improvement in number of errors. At least one study using a modelling paradigm in corporated teaching cognitive strategies in the design and was partially successful in improving error scores (Debus, 1970). The subject watched the model execute the MFF test, while he talked to himself about the strategy he was using. Therefore, the model provided both 1) behavioral cues of attention and deliberation and 2) explicit verbalizations as to the best strategy for performing the task. The results indicated that the subjects who watched the reflective model increased their reaction time on the MFF (second test compared to Pre-test) but only those sub jects who showed above median increases in reaction time also improved their error score. The author believes that the behavioral cues of slower reaction time were responsible for the increased latency in the subjects and the improved performance was an indirect effect arising from the increased latency. This does not seem likely in view of the other studies mentioned which did not obtain improved performance nor did they offer verbal cues as to scanning strategy. It seems more plausible to suggest that some subjects modelled the model's cognitive self-guiding verbalizations, thereby improving their performance. Their significantly increased latency could have resulted indirectly, being necessitated by the new cognitive strategy. There is a 53 body of literature which attests to the efficacy of self- guiding verbalizations in mediating behavior and specifi cally, in modifying impulsiveness. Verbal Mediation The development of the functional interaction be tween self-verbalizations and nonverbal behavior has been observed and investigated from the standpoint of a wide variety of disciplines within psychology (Lovaas, 1960, 1961; Luria, 1960, 1961; Piaget, 1926; Reese, 1962; Vygotsky, 1962). Vygotsky's theoretical formulations of private speech (1962), further elaborated and documented by another Russian, Luria (1960, 1961) are central to the verbal mediation literature and can be summarized as follows. There is a direct relationship between the child's development of voluntary control over his behavior and the development of his speech. Developmentally, the nominative function of speech develops first, then the communicative function and finally the planning and self-guidance func tion, which is of particular relevance to this paper. One dimension in the development of the self-guidance function of speech is the gradual internalization of the speech which proceeds from responding to the verbal commands of others, to responding to overt self-directing comments, to responding to covert (silent) self-directives. Thus the 54 child who is seen talking to himself aloud while per forming a task is probably in stage two and his speech is referred to as private (or egocentric) speech. There are several types of private speech which also follow a developmental sequence proceeding from the earliest forms such as talking to objects, to self answered questions, direct self-guiding comments, and finally muttering (varied versus partially internalized) self-guiding verbalizations. The internalization of these verbalizations represents the beginning of linguistic thought, which in its more advanced stages consists of highly condensed verbal symbolizations referred to as inner speech. Vygotsky says the "next plane" of linguistic or verbal thought, more "inward" than inner speech, is thought itself. This is not to say that the development of speech and thought are synonymous. Rather they are seen as developing independently of each other up to a point so that there is a prelinguistic phase in the development of thought and a preintellectual phase in the development of speech. But eventually their developmental processes interact. The self-guiding function of a child's self verbalizations also follows an age-related sequence in terms of area control, from initiating his own behavior to inhibiting an ongoing behavior, and finally complex 55 planning. And within the initiating phase, there is another developmental differentiation— that at first the child is responding only to the motoric component of his own speech and later he responds to the semantic content. It is difficult to critically evaluate Luria's research reports since the work is presented only in sum mary found in English translation. No significance levels or detailed methodologies are presented. To illustrate some of these sequences, Luria (1961) relates the following experimental observations. A two-year-old child was told "press the bulb once when the light goes on but don't press after that." The child presses to the light several times— that is, the speech of an outsider initiated his behavior but was not able to inhibit it once it was on going. It is not until the child is three years old that the Experimenter was able to use his self-verbalizations to help him inhibit. At this age, the child still could not respond to the inhibitory component of the Experi menter's commands. Like the two-year-old, he presses the bulb when given the same command but when the Experi menter told the child to verbalize the word "press" only once to the light, this was accomplished. Then the Experimenter told him to respond both verbally ("Say press”) and motorically ("and do it") only once when the light went on. This too was successful so that the in hibitory control over the motor response was achieved by 56 the simultaneous verbalizing. That it was not simply the result of practice was shown by the next phase, where the child was given the same command but told not to verbal ize. Again, he pressed several times instead of once. The two-year-old child was not able to control his motor behavior by the use of his own verbalizations as the three- year-old could. In a similar example illustrating the age sequence of the use of self-verbalizations for inhibitory control of motor behavior, Luria relates the following: The com mand was "press to the red light, don't press to the green." The three-year-old child verbalized the instruc tion correctly but did not respond motorically correctly. The four-year-old demonstrated a number of correct trials but not until five years of age could the child use his own verbalizations efficiently to inhibit his response discriminatively. The apparent age discrepancy between the first example and the second is probably due to the relatively greater difficulty of the second task, since level of task difficulty has been found to correlate with the interaction between age and the use of self-guiding verbalizations (Kohlberg et al., 1968). Another study looking at the developmental se quence of verbal inhibition of motor behavior in response to an external command ("Press to green light, don't press to red”) found that all subjects in the 5% to 6h year age 57 group met the criterion of 90% correct responding, while 75% did in the 4% to 5*s year range, and only 37% in the 3% to 4 year age range (Birch, 1967). To demonstrate the theory that children first respond to the motoric component of speech rather than the semantic, Luria relates the following: a three-year-old child was told to press twice to a signal but pressed several times. Then he was told to say, "I press twice" when the signal occurred, and to do so simultaneously. He uttered the correct verbalization but again pressed sever al times. Finally, he was told to say "toot toot" to the signal and to press. This was successful— the child pressed twice, in apparent response to the two mono syllabic verbalizations. This phenomenon has been docu mented by American researchers as well (Lovaas, 1964; Meichenbaum and Goodman, 1969B). Lovaas (1964) reported that when four and five- year-old children counted aloud fast, their simultaneous lever pressing was faster than when they counted aloud slowly and the children emitted one lever press for each verbalization in both conditions. That is, they seemed to be responding to the motor component of their speech. Then, he asked subjects to say the word "slow" to one light and "fast" to another, while simultaneously pressing the lever. The younger group (six-year-olds) demonstrated no reaction to the semantic content of the words while the 58 older groups (nine and eleven-year-olds) demonstrated significantly faster reaction time when they were saying "faster." It is interesting that Meichenbaum and Goodman (1969B) found that impulsive kindergarten children re sponded on a finger tapping task to the motoric component of their speech, whereas the reflectives relied on the semantic content of their self-instructions, suggesting that impulsive children are developmentally less mature. Also, reflective kindergarten subjects performed as well as the average first-grader on the MFF while the impulsive kindergarten children performed significantly poorer. Lovaas (1964) conceptualizes verbal behavior as operant behavior and proposes that since certain verbal and nonverbal operants occur so frequently together in every day behavior, they begin to share reinforcements such that the verbal operant acquires discriminative stim ulus control over nonverbal operants. For example, the parent may emit verbal behavior (e.g., a command) and then reinforce the child's subsequent behavior so that the com mand becomes associated with the nonverbal behavior and the reinforcement. Then when the child utters the command to himself he essentially reinforces himself for doing the action. Thus, as the Russians postulate, the origin of the self-instructional private speech is in the external commands of the parent (or other authority figure). A number of studies have examined Vygotsky's 59 proposal that early In life external verbal commands control behavior, then overt self-verbalizations and fi nally covert self-verbalizations. Meichenbaum and Goodman (1969) observed the differential speed of finger tapping in response to the words "faster" and "slower" under three conditions: external command (experimenter said the words); overt self-verbalizations; covert self-verbal izations (subject told to whisper the words to himself while tapping). The results partially supported the de velopmental sequence, for the kindergarten children's motor performance responded to the overt self-verbal izations while the covert condition had a minimal effect on the motor performance. The reverse was true for the first-graders. However, both groups demonstrated the most motor responses in the external condition. Similarly, Kendler, Kendler and Carrick (1966), using an inferential problem solution task, found that overt labels helped kindergarten children but interfered with the performance of third-graders. At least four other research studies (Flavell et al., 1966; Gratch, 1966; Klein, 1963; Kohlberg et al., 1968) support the postulated age increase in cognitive self-guiding private speech, the postulated increase in its internalization with age, and the postulated function al role of private speech in task performance. Klein's study employed an observation paradigm where children of 60 ages three to seven were left alone in an observation room with puzzles and drawing task. The older children demon strated more task-relevant private speech while task- irrelevant private speech decreased with age. The older children demonstrated more muttering and lip movement than the younger subjects, who demonstrated more audible pri vate speech (except for audible task-relevant speech, which was more prevalent in older groups). An interesting finding which Luria (1960) reports concerning the proposed internalization of cognitive self-guiding speech is that EMG (muscle tension) reactions were registered in the tongue of a child who was silently performing a difficult task. These muscle reactions are only evoked in a diffi cult task situation. Further, he reports that during the "thinking" of a deaf-mute child, EMG reactions are regis tered in the hand instead of the tongue. Klein also reported a significant correlation be tween amount of task relevant speech and successful puzzle completion. Using a pictorial serial recall task to elic it private speech in children of three age groups (five, seven and ten years), Flavell and co-workers noted a significant increase in spontaneous rehearsing with age and an apparently better performance (although a signifi cant practice effect may have confounded this result). Flavell proposes two possible explanations of his finding a significant age increase in spontaneous 61 rehearsing. First of all, since he had ascertained that all the children knew the names of the pictures in the task, perhaps the failure to use them in rehearsing is a "linguistic immaturity" factor; that is, the child does not produce the relevant mediators in the task situation although he knows the words. This has been termed the "production deficiency" hypothesis, in contrast to Reese's (1962) "mediation deficiency" hypothesis. The latter hypothesis states that there is a stage in development where children tend not to mediate their behavior verbally, despite the fact that he is able to understand and cor rectly employ the verbal responses necessary for such mediation. The crucial point is, can the child mediate if he is given the verbal mediators, or is he unable to; that is, is it really a mediation deficiency or a production deficiency? Keeney and associates (1967) offer evidence to support the production deficiency hypothesis. They re quired nonrehearsers to rehearse the names of the stimuli in a recall task and found that their performance improved significantly; that is, they were able to mediate their task performance verbally but when subsequently given the option of rehearsing or not, they reverted to nonre hearsing. The authors conclude that the unmediated be havior on the task is due to a failure to produce the mediators. Similarly, Weir and Stevenson (1959) found 62 training their subjects to name the stimuli before pressing the response lever (discrimination task) resulted in improved performance. Flavell's second explanation of the age increase in spontaneous rehearsing involves a "general cognitive immaturity" factor, since rehearsing demonstrates an ac tive, systematic approach to the problem and sustained at tention. The relationship between cognitive maturity and the use of private speech was very systematically and ex tensively investigated by Kohlberg and his colleagues (1968). Specifically, they investigated the relationship between various types of private speech, difficulty level of the task, and age sequences versus cognitive maturity sequences. Their results are as follows: 1) There is a curvilinear relationship between amount of private speech and cognitive development (defined as the interaction age and IQ or MA). That is, private speech, in general, in creases with age and then decreases but bright children use more private speech at an early age than average chil dren and the decline in its use is earlier among bright children. The peak amount of private speech found in bright children occurred at about age four, while average children peaked at age seven; 2) Bright children use more private speech in a task situation than a free play situ ation, presumably this represents a relatively better development of one category of private speech; 3) 63 Investigating the relationship between age and categories of private speech, they found that private speech which simply describes the child's ongoing activity steadily declines after age five, while the self-guiding comments increased between ages five and six, and then began a steady decline. Finally, muttering, presumed to be the highest form of private speech (partially internalized) began to increase at the same age (six) that the audible self-guiding comments began to decrease, suggesting the possibility that in the older subjects, the self-guiding comments had become partially internalized, represented by the muttering. Unfortunately, the youngest group was five years of age so there is no data as to trends before age five. Cognitive tasks elicit more private speech than motor tasks; difficulty cognitive tasks elicit more than simple tasks. The categories of private speech which in crease with task difficulty are the self-guiding comments. Interestingly, a simultaneous decrease in muttering oc curred during the difficult task suggesting that perhaps the child is reverting to an earlier method of self guidance. Two studies mentioned previously (Keeney et al., 1967; Weir and Stevenson, 1959) demonstrated that training subjects in the use of verbal mediators enhances per formance on discrimination and serial recall tasks. Bern (1967) also demonstrated improved performance on a visual 64 motor lever pressing task in a group of three-year-olds following specific verbal mediation training in con junction with practice on the task. Bern proposes that the absence of verbal self-control is not a "developmental deficiency" but a "learning deficit" and therefore a learning procedure can establish verbal self-control; in fact, a learning paradigm probably underlies the "natural emergence of verbal self-control." It seems that the extensive developmental literature supporting a develop mental sequence based upon mental age or cognitive maturi ty cannot be ignored. Accepting the existence of the sequence does not necessarily imply that it is so fixed by maturation factors such that learning techniques would be useless in accelerating or remedying this development. Therapeutic Applications of Verbal Mediation Training Several writers have investigated the utility of self-instructional training as a method of treating chil dren with a variety of learning problems (Jensen, 1963; Luria, 1960, 1961; Meichenbaum and Goodman, 1971; Palkes, 1968). The use of this training procedure presupposes the existence of some deficit in verbal mediation. Dykmsn (1970) observed that children with learning difficulties (ages eight to eleven), especially the hyperactives under age ten, tend to "think aloud" in problem situations more 65 so than normal children. At these ages (eight to ten) this behavior might represent a developmental lag since Kohlberg's (1968) data indicates a sharp decline in pri vate speech after age seven. Luria (1961) reports positive results in training children with a disorder the Russians call "cerebral asthenia." This disorder sounds identical to the hyper kinetic syndrome, being characterized by poor concentra tion, distractibility, excessive impulsiveness and poor school performance in spite of normal intelligence. He reports work with these children employing the visual motor discrimination task ("push to red light, don't push to green") described earlier. Normal nine-year-old chil dren perform the task well but the cerebral asthenics deteriorate when the light flashes and interstimulus in tervals become shorter. Their failure is in response to the inhibitory signals; they press to both red and green lights. If they are trained to verbalize the words, "press" and "don’t press” to the appropriate lights while simultaneously responding motorically, their inhibitory responses improve markedly. The improvement is explained in terms of the orienting reflex. That is, Luria proposes that the child's speech helps to enhance the child's orientation to the task stimuli. Such an effect would certainly be helpful in hyperactive children who have been found to be deficient in attention, based on behavioral 66 data, testing and physiological arousal indices. Jensen (1963) compared the performance of retarded children to average and gifted children (all ages twelve to fifteen) on a visual motor task requiring new learning ability. Initially, the retardeds performed much more poorly than the other groups. Then they underwent a training procedure involving three elements: a) explicit differential verbal reinforcement; b) stimulus naming (the verbal mediator) in conjunction with the task; and c) forced delay of response time. The retarded group showed significant improvement on the post-test, approximating (though not equalling) the performance of the normal group. It is interesting that the two fastest learners had IQ's of 147 and 65. Jensen states: School learning probably depends upon transfer from previously learned habits of verbalization and other symbolically mediated behavior. The habit of making verbal responses either overtly or covertly to events in the environment seems to be one of the major ingredients of the kind of intelligence that shows itself in school achieve ment and in performance on intelligence tests. Jensen notes the spontaneous vocal or subvocal be havior during the learning trials was much more frequent in the average and gifted groups. He also raises an interesting question which is particularly relevant to this paper— why did the training procedure result in dra matic improvement in some cases while others showed no improvement at all. Apparently, the nature of the 67 learning deficits must vary from case to case. Luria (1960) proposes that if the speech mechanism is intact, then such a training method can be used in compensating for learning defects. Meichenbaum and Goodman (1969B) first established that there is a relationship between private speech and conceptual tempo and then (1971) tested the efficacy of a self-instruction training paradigm in children labelled as hyperactive or simply demonstrating poor self-control. In their first study, they observed the differential control (exerted on lever pressing) of overt self-commands versus covert (whispered) self-commands in impulsive and re flective kindergarten children. The commands were simply "don't push" and "push" in response to two different colored lights. The two groups of children did not differ in response to the initiating command ("push”) in either overt or covert condition but the impulsive subjects made significantly more errors in response to the inhibitory command ("don't push") under the covert condition. Under the covert condition, there was a nonsignificant differ ence in number of errors between the two groups, but the magnitude of errors (longer and harder lever depression) was greater for impulsives. The authors conclude that the impulsive children demonstrate poor inhibitory capacity when relying on internal control. In their subsequent study, Meichenbaum and Goodman (1971) found that they could improve performance of second-grade children who are characteristically impulsive (hyperactive and poor impulse control children) using a training procedure involving the following: First, an ex perimenter modelled appropriate task performance with simultaneous overt self-guiding verbalizations. Then the child was asked to perform the task while the experimenter again verbalized aloud. Then the child performed the task again and was asked to verbalize aloud himself. Subse quently, the task was performed with whispered verbal izations by the child and finally totally covertly, with out lip movement. Thus the authors incorporated the supposed natural developmental sequence of internalization of verbal commands into the training procedure. The content of the verbalizations was aimed at compensating for any possible comprehension, production, or mediation deficiency. For example, included were self-asked ques tions and answers about the demands of the task, cognitive rehearsal and planning strategies for solution to the task. Also included in the verbalizations was self-reinforcement, thus compensating for a possible production deficiency. There were four training sessions of one-half hour each. The results were as follows: 1) improved WISC performance IQ; 2) increased latency on MFF; 3) decreased errors on MFF (trend); and 4) improved Porteus Maze performance— no results were reported with regard to concomitant changes 69 in activity level. In order to determine whether the modelling component of the training procedure was responsible for the improvement rather than the self-instructional training, they then ran three groups: modelling, model ling plus self-instruction, and a control. There was only one training session. Another difference is that the sub jects were normal five and six-year-olds who were classi fied as impulsive on the basis of MFF performance. As in the modelling studies mentioned earlier, only the decision time was increased, with no concurrent effect on error scores. However, the modelling plus self-instruction training produced significant improvement in error scores as well as increased latency. (It should be mentioned that the training materials were different from the cri terion tests.) The authors conclude that self-instruction training can bring an impulsive subject's overt behavior under his own verbal control. Palkes and co-workers (1968) also demonstrated im provement in hyperactive boys (eight to ten years) fol lowing training in self-directing verbal commands. The verbalizations employed in their training procedure were much more general, and therefore possibly applied to a wider variety of situations. Examples of some of the verbalizations include: "Stop, listen I look and think before I answer." The subject was required to say certain 70 of these commands aloud, depending on the particular test, before each response. Social reinforcement was contingent on his remembering to say them. Again, the training mate rial was different from the criterion test, the Porteus Maze. The children who received the training improved their Porteus Maze performance significantly, and were significantly better on the post-test compared to the con trol group of hyperactives. Again, it would be interesting to know if the training effect generalized to other areas such as activity level. At least one study has included observations of changes on academic performance and activity level follow ing training in self-directed verbal commands (Burns, 1972). The researchers employed two 30-minute training sessions on two consecutive days during which the hyper active children received training in self-directed verbal commands and arithmetic practice. The content of the com mands focused on inhibition of impulsive responding. A control group received only the arithmetic practice. Motor activity was observed and recorded on an activity data sheet. The results of this study were very disappointing: there was no improvement in arithmetic performance and no decrease in activity level following training in either group. Pearson product-moment correlation revealed no systematic correlation between activity level and arithmetic 71 performance. The authors suggest three faults in the study which might account for the negative results: 1) the training sessions did not incorporate tasks in in creasing difficulty; 2) the verbal commands only focused on inhibition; some focus on planning or cognitive strate gy is suggested; 3) there was no systematic manipulation of motivational factors. That is, they failed to incorporate self-reinforcement training, thus not compensating for a possible production deficiency. All of these factors are accounted for in the design of the present study. As discussed earlier, Keeney and co-workers (1967) found that though you could easily train children to use verbal mediators, when given the option, the original non rehearsers chose not to. Their procedure did not com pensate for production deficiency, a serious failing, since one would want the effects to generalize outside the laboratory. Also these researchers failed to provide any feedback to the subjects concerning their improved per formance while verbalizing and in general failed to pro vide any reinforcement for the self-instructions. The use of reinforcement in the training procedure is essential, since the performance of the verbal behavior outside the laboratory will ultimately be self-reinforced; that is, it seems there roust be a transition from external to internal reinforcement. Thus incorporating reinforcement into a training procedure should overcome any production 72 deficiency. The actual training in self-guiding verbal izations should be aimed at overcoming comprehension and mediation deficiencies in relation to task and test per formance, as Meichenbaum and Goodman did (1971). Based on the literature, hyperactive children seem to demonstrate a more impulsive conceptual tempo, poor at tention, and less motor inhibitory capacity than normal children. It seems likely that a training program in corporating all the factors mentioned above would ef fectively reduce impulsivity, increase attention and re sult in a concomitant decrease in activity level especial ly the sub-group of hyperactive children who seem to respond minimally (if at all) to stimulants. Based on the literature on activity levels and drug response in hyper active children, seat activity level in a cognitive task situation should differentiate two diagnostically separate sub-groups of hyperactive children. Possibly, task per formance will also differentiate these groups. It is also of interest to observe the relative effectiveness of stimulant medication versus this training program in treating the high active sub-group of hyperactives who would be expected to respond beneficially to the stimu lants. Hypotheses 1. a) Hyperactive children will demonstrate higher 73 activity levels than normal children in low stimulation situations requiring voluntary motor inhibition. b) Hyperactive children will perform more poorly than normal children on tasks requiring re sponse inhibition (motor and verbal) and attention. 2. Two sub-groups of hyperactive children will be differentiated by activity level and will demonstrate dif ferential task performance. Specifically, High-Active hyperactive subjects will perform more poorly than the Low-Active hyperactive subjects on tasks requiring re sponse inhibition (motor and verbal) and attention. 3. Verbal Mediation Training is an effective treat ment intervention for hyperactive children. a) Activity levels of hyperactive children will decline following a training program in verbal mediation. b) Hyperactive children's performance on tasks re quiring response inhibition and attention will also improve following verbal mediation training. 4. Verbal Mediation Training will be more effective with the Low-Active hyperactive subjects. a) The Low-Active hyperactive subjects will demonstrate a greater improvement in task 74 performance than the High-Active subjects following Verbal Mediation Training. b) The Low-Active group will demonstrate a greater decline in activity level following the Verbal Mediation Training than the High-Active group. 5. Stimulant medication administered to hyperactive subjects after the verbal mediation training will result in additional improvement. a) Activity level for the hyperactive children will continue to decline. b) Task performance of hyperactive children will continue to improve. 6. Stimulant medication administered after the verbal mediation training will be significantly more effective for the High-Active hyperactive subjects. a) The High-Active subjects will demonstrate a further decline in activity level while the Low-Active subjects will not show any further decline. b) The High-Active subjects will demonstrate greater improvement than the Low-Active sub jects. In other words, stimulant medication will result in additional improvement over that obtained on verbal mediation training only in the High-Active hyperactive subjects. CHAPTER III METHOD Subjects Sixteen hyperactive male subjects and eight normal male subjects participated in this study. The mean age of the hyperactive subjects was eight years five months, with a range of six years five months to nine years ten months. The mean age of the normal group was eight years seven months, with a range of seven years eight months to nine years ten months. All the hyperactive subjects were drawn from children referred to the Child Outpatient Psychiatric Clinic at Los Angeles County-U.S.C. Medical Center. All underwent the Clinic's standard intake procedure including a social worker evaluation and a psychiatric evaluation. Copies of the Clinic's Intake Form, Social Worker Evalua tion Form, and Psychiatry Evaluation Form are included in Appendix B. Criteria for referral to the study included the following: 1. Age six to nine years. 2. Initial referral to the Clinic by an M.D. or a Clinical Psychologist with a referral diagnosis of hyperkinetic syndrome. 75 76 3. Concordance with this diagnosis by the psychiatrist or clinical psychologist performing the psy chiatric intake evaluation. Those children referred to the study were then included if all of the following criteria were satisfied: evidence of overactivity, short attention span, distractibility, and poor academic performance. Children were excluded from the study if they evidenced psychotic or neurotic sympto matology, or severe neurological impairment such as mental retardation or epilepsy. Because of the strict criteria and diagnostic counter-checks involved in including a child in this study, obtaining the sample was a difficult problem, necessitating certain changes in the originally proposed design of the study. Specifically, the design of the study was changed from a separate groups design to a crossover design. Secondly, the Experimenter ideally preferred that all children be new referrals who had never received treatment, particularly medication, for the hyperactivity. However, this so reduced the available subjects that it was decided to specify a medication wash-out period of at least one week for those children who were taking medication at the time of referral to the Clinic. The final hyperactive sample therefore included seven subjects who had never been treated for the hyperactivity, four subjects who had a history of treatment including medication but who had 77 not taken medication for several months to several years, and five subjects who were currently taking medication (Ritalin). In all cases, the behavior complaints con tinued. As for the normal group, four of the subjects were obtained by speaking to parents whose sons were in a Cub Scout group located in a Mexican-American neighborhood since the majority of hyperactive boys were first genera tion U.S. born of Mexican extraction. The remainder of the boys were obtained by simply approaching parents in a lower middle class neighborhood. In all cases, the study was explained and parents volunteered their sons' participa tion. There was no money incentive offered as this may have constituted a biasing effect as well as an expense for the Experimenter. The boys were included in the normal sample if they were not taking medication for any medical or psycholo gical problem, if there was no neurotic or psychotic symptomatology, no history of neurological problems, and average or better academic performance. None of the parents of the normal subjects had ever sought professional help for their children regarding behavior problems of any kind. They all stated that their boys got along well with their friends and did well in school. The racial-ethnic composition of the normal group was seven Mexican-American and one Black. The hyperactive 78 group included eight Mexican-American children, four Blacks and three Caucasians. All subjects were lower middle class socio-economically. All subjects spoke English fluently, though all the Mexican-American children also spoke Spanish. In four cases, parents spoke only Spanish, at which time interpreters conveyed the necessary information. The mean IQ of the normal group was 97.5 with a range from 81 to 110, while the mean IQ of the hyperactive subjects was 90 with a range from 71 to 132. Procedure The study was conducted at the Child Outpatient Psychiatric Clinic of Los Angeles County-U.S.C. Medical Center. All subjects, normal and hyperactive, participated in three experimental testing sessions conducted by the Experimenter. Before the first experimental session, the mothers of the hyperactive children were shown and read the informed consent form which contained details about the experimental procedures (see Appendix B). Their signatures were required in order for their children to receive the treatments which were to be evaluated in the study. Each experimental testing session followed the same sequence of tasks, lasting approximately thirty minutes. Each subject was tested individually during the hours from 9:00 A.M. to 11:00 A.M. At the start of session one, the Experimenter assured each child that the tasks he would 79 perform did not test how much he learned in school and that his performance would not be reported to his parents. The subjects were told that their help was needed in order to determine how children their age would perform on certain tasks. The subject sat at a table opposite the Experimen ter. The subject's chair was specially designed to monitor movement (see Diagrams 1 and 2 in Appendix B). Specifi cally, the chair was equipped with strain gauges which were sensitive to movement in three planes (forward and backward, up and down, side to side). The gauges were attached to a recording apparatus which cumulated activity and translated the output onto a dial with a scale from 0 to 100 (See Diagram 2 in Appendix B). Rarely, the sub ject's activity exceeded 100, at which time the reset button was pressed and cumulation was continued beginning at 0 again. The reset button was pressed before the sub ject began each new task, and cumulated activity was recorded on the subject's record sheet (see Appendix B) as soon as he completed the task. After the Experimenter's introductory comments to the subject, the Experimenter told him he would return to the room in a short while and instructed the subject to remain seated. The activity recorder was set and the Ex perimenter left the room for three minutes, observing the subject through a one-way mirror to be sure he remained 80 seated. When the Experimenter returned, she recorded the cumulated activity. This is referred to as the resting activity level. Then the Experimenter told the subject that he could choose some of the candy or baseball cards, which were displayed in full sight, after the session was over if he tried his best on the tasks. Next the Peabody Picture Vocabulary Test was administered, to insure that all the subjects were of normal intelligence and to deter mine whether there were significant IQ differences between groups. Although activity was recorded for this task, it was not later used as a dependent measure because the task was not timed and therefore a "per minute” activity score could not be computed. During each of the subsequent tasks, activity was recorded. Also each task was timed so that activity per minute on each task could be later com puted and serve as separate dependent measures. The next task was the Porteus Mazes. Each subject was given three mazes complete, first a maze at a level below his chronological age, then one at his own age level, and third the maze level one year above his age. During the first testing session, the Pre-test, the Vineland Revision Form of the test was used. Number and type of errors were recorded. Later, the error scores were weighted as follows: entering blind alleys and back tracking received a weight of two points; cutting corners, lifting the pencil off the paper, and crossing the lines 81 received a weight of one point. These weights are very similar to those prescribed by Porteus (1965) . The depen dent measure, Maze Error, was the sum of the weighted error scores on the three mazes. Next, six of the twelve problems on the Matching Familiar Figures (MFF) test were administered following the two sample items. As described in Chapter 2, the MFF test requires the child to select the one picture out of six variants which is identical to the standard (see Appendix C). If his first response is incorrect, subject continues to select responses until he answers correctly, or until he has used up six guesses. The latency for each response and the number of errors were recorded, constituting two other dependent variables. Finally, the two operant tasks were administered; first the differential low rate (DRL) and then the fast rate - partial reinforcement (PR-F). The operant tasks required the child to determine on his own how to maximize the number of points he received. The points appeared as lighted numbers above the response lever (see Diagram 3 Appendix B). He was told only that the points he earned depended on whether he found the best speed for pressing the lever. An electronic counter controlling the number of lever presses required to advance the number by one in one second was first set at one. That is, reinforcement was on a one to one differential low rate schedule, so that a 82 point was earned only after a lever press timed no sooner than one second after the previous lever press. The lever could be pressed several times in one second but no re inforcement or number increase would result. After two minutes at this setting, the number of points on the counter was recorded, as well as activity level. Then the operant counter was set for the PR-F operant task. This time fifteen lever presses occurring in one or more seconds would result in an increase of one point. Thus, unlike the first task, the ratio of reinforcement was one to fifteen. Also, faster pressing was required to achieve higher scores at the end of the two minutes. This task concluded each experimental testing session. The subject was then allowed to reinforce himself with the candy or baseball cards unless he forgot, in which case he would be reminded to do so by the Experimenter. Based on the activity scores recorded during the first testing session (Pre-test), the hyperactive children were divided into two groups, the High-Active (HA) group and the Low-Active (LA) group. Specifically, the division was based on the combined activity scores recorded during two of the tasks, the MFF and the Porteus Maze. These con stituted the basis for the separation rather than combining all of the activity scores for two reasons: 1. Based on a visual inspection of scores, acti vity during the operant tasks do not seem to differentiate 83 normals from hyperactive nor subgroups within the hyper active sample, while this is not the case with the other activity scores. 2. The activity scores recorded during the MFF and Porteus Maze tasks correlated highly with each other (r * 0.796; p < 0.000004). These two subgroups were treated identically in the study so that the division was made for purposes of a statistical analysis of hypothesized differences. The mean age of the Low-Active group was eight years nine months with a range from six years five months to nine years ten months. The mean age of the High- Active group was eight years with a range from seven years to nine years ten months. Ethnic composition was very similarly distributed; the Low-Active group was composed of five Mexican-Americans, two Caucasians and one Black, while the High-Active group was composed of three Mexican- Americans, three Blacks and two Caucasians. As for IQ, the mean IQ in the Low-Active group was 95.3 with a range from 73 to 132 while the mean IQ of the High-Active group was 89.6 with a range from 71 to 97. Verbal Mediation Training During the two and half to three weeks between the Pre-test session and the second testing (Post-Test 1), hyperactive subjects participated in four sessions of Verbal Mediation Training conducted by a female college 84 student who had been trained by the Experimenter. Her task was not only to work with the hyperactive boys individ ually, but also to motivate their mothers to continue to bring their sons. Therefore, before each training session with the child, the Instructor spoke to the mother, rein forcing her for coming, expressing concern and offering help concerning problems such as transportation, scheduling visits, getting time off from work, etc. In one case the Instructor wrote a letter to the mother's employer verify ing her visits to the Clinic. These efforts at establish ing and maintaining rapport with the mothers seemed well rewarded in that all of the subjects who began the training completed it and completed their participation in the study. The only attrition of subjects occurred before the first experimental testing sessions two subjects who had been referred and accepted into the study phoned saying they were going on vacation some time during the next six weeks and therefore could not attend the sessions as scheduled. This low attrition was surprising especially considering that the mothers made eight visits to the Clinic including the intake evaluation session, testing, and training sessions, all on workday mornings. The training sessions with the boys lasted an average of forty minutes, ranging from twenty minutes to one and one-half hours. The variability in time is due to the fact that training ended when the subjects reached 85 criterion on the verbalizations required for the tasks assigned for that session, unless this did not occur after one and one-half hours. Failure to reach criterion during any of the training sessions occurred in only two subjects. The materials used in training were different from the criterion tests used in the testing sessions. The training tasks and the required verbalizations increased in diffi culty over the four sessions. Specifically, the training tasks included: 1. Two Trail Making Tasks, an easy one requiring the subject to connect ten numbers in order, and a more difficult one requiring the subject to connect numbers and letters alternately in order. These tasks were adapted and simplified from Reitan's Trail Making Test (1955); 2. Ten pictorial visual discrimination problems taken from the Primary Mental Abilities Test (Thurstone, 1962); 3. Two design visual discrimination problems taken from Raven's Progressive Matrices (1956). The training tasks are reproduced in Appendix A and are grouped according to the order they were presented to the subject. The Instructor introduced the training program to the subject as follows: "Hi Johnny, I'm Miss Smith. I'll be seeing you four times and we will be doing different kinds of tasks together. I'll be teaching you to say things to yourself while you are doing these tasks. The things I will teach you to say will help you to do the tasks better and some of these things will be fun. You know, most children and adults do say things to themselves to help them with their work and sometimes to help them win games. Each time you speak the words I teach you, you may take a piece of candy which you can bring home with you, or you can trade in your candy after we're finished for the base ball cards.” This introduction was given only on the first day of train ing. Immediately following, the verbal mediation training began. The Instructor pointed to a card on the table saying the following: "This card will remind you to say some of the most important words to yourself. Can you read this one? (Allow subject to read card.) Yea, it says 'atop, look, and think.' That meana before we start a new task you must always aay to yourself 'I must stop everything I'm doing, look at the task, and think before I answer, stop, look, and think.' You must stop swinging your legs, stop turning your 87 head, stop all moving and pay attention to the task. Let's hear you say these words and stop moving at the same time. (Coach if necessary. After successful trial, give one M&M and say good). The essence of the training program consisted of teaching the subjects appropriate self-verbalizations aimed at focusing their attention on the task, reducing their motor behavior, and completing the task correctly. A second essential aspect of the training program is sys tematic development of self-reinforcement. The program was very similar to that employed Meichenbaum and Goodman (1971) with two additions: 1. Systematic use of tangible reinforcers (candy) paired with the social reinforcement. 2. Inclusion of verbalizations aimed at reducing motor activity. Each session followed a fixed sequence: First, the Instructor performed the training task, while instructing and reinforcing herself aloud, thereby modeling the appro priate verbalizations. The self-reinforcement was con tingent on the appropriate verbalizations, not necessarily on correct answers. Self-reinforcement consisted of taking candy from the candy dish and saying "good." Secondly, the child is asked to perform the task while the Instructor still verbalizes aloud. Next, the 88 child is asked to do the task and verbalize aloud, rein forcing himself as modeled. Then the child again performs the task with self-reinforcement but whispers the verbali zation. Finally, the child is told to do the task and say the words to himself (no lip movement) reinforcing himself as he did previously. Thus the sequence in each session is intended to follow the theoretical developmental sequence: a gradual internalization of verbal commands in mediating behavior. The Instructor's instructions and the exact verbalizations required for each task are presented in Appendix A. The content of the verbalizations are intended to compensate for any possible comprehension, mediation, and/ or production deficiencies. To illustrate, following are the verbalizations for the first task in session one: "I stop, look, and think before I move (take M&M). What is it I have to do? I have to draw a path starting with number one and ending with number ten (M&M). One- two-three-I must go very slowly-four-five- six-seven-eight-nine-ten . Good I finished. (M&M) I earned three M&M's!" Thus, first the subject asks himself what the demands of the task are and then he answers this, compensating for a possible comprehension deficiency. Then the subject verbalizes am appropriate task strategy while doing the 89 task, compensating for any possible mediation deficiency. He simultaneously reinforces himself, thus increasing motivation for using the mediators, and compensating for possible production deficiencies. The first Post-Test (PT-1) occurred two and a half to three weeks after the Pre-Test for all subjects, normal and hyperactive. As stated earlier the exact sequence of testing occurred. An alternate form of the Porteus Maze test was used, the Extension Series. Also six different MFF problems were presented. The second Post-Test (PT-2) occurred two days to one week following PT-1. One hour before PT-2, the Clinic nurse administered ten milligrams of Ritalin to the hyperactive subjects. The subject was told only that the Experimenter was interested in seeing whether taking the pill would affect his scores. Since additional forms of the MFF and Porteus Maze could not be obtained, the Experimenter administered the Porteus Vineland revision series as in the Pre-Test and the first six MFF problems as in the Pre-Test. However, the Experimenter changed the order of the six variants for each of the six MFF problems so that the subject could not choose on the basis of recalling the location of the correct choice. Analysis The statistical analysis consisted of Planned Com parisons with repeated measures (Hays, 1963) utilizing the Multivariate Analysis of Variance program (MANOVA). It was decided to use this method since specific comparison ques tions were intended from the start and since the proba bility of detecting the true effect, that is, the power of the test, is greater with a Planned Comparison Analysis than with an Unplanned Comparison Analysis such as an overall F test. Also, the Multivariate program was used because it analyzes the effect of the independent factor on all the dependent criteria and variables simultaneously, talcing into account the correlations among the variables. This program uses the Wilks Lambda Criterion and Canonical Correlations. Three factors were specified: 1) Group; 2) Treat ment, and 3) Subjects. There were three levels of factor one (Normal, High-Active Hyperactive, Low-Active Hyper active); three levels of factor two (Pre-Test, Post-Test 1, Post-Test 2, Verbal Mediation Training - Ritalin) and eight levels of factor three (8 subjects per group). That is, each subject was essentially analyzed as a separate level of a factor. Because there were a few very extreme scores, square root transformations were employed in the Planned Comparisons. However, the group means and overall corre lations will be presented in terms of untransformed scores. 91 This program was also used to obtain overall correlations between all dependent variables. Specifically the comparisons were designed to ans wer the following questions: 1. Do hyperactive subjects differ significantly from the normal subjects on the Pre-Test variables? 2. Do the High-Active Hyperactive subjects differ significantly from the Low-Active Hyperactive subjects on the Pre-Test measures? 3. Do the dependent variable scores of the hyper active subjects change significantly more than those of the normal subjects from Pre-Test to Post-Test 1? In other words, does the verbal mediation training result in improved performance above any possible practice effect? 4. Do the High-Active hyperactives as compared to the Low-Active hyperactives demonstrate differential response to the verbal mediation treatment? 5. Do the dependent variable scores of hyperactive subjects change significantly more than the normal subjects from Post-Test 1 to Post-Test 2? In other words, does the Ritalin treatment result in improved performance beyond any possible improvement which occurred on Post-Test 1, and beyond practice effect? 6. Do the High-Active hyperactive subjects and the Low-Active hyperactive subjects demonstrate differential responsivity to the Ritalin treatment? 92 7. Based on the statistical comparisons, which dependent variables consistently differentiated the hyper active subjects from the normal subjects in terms of pre treatment measurement and treatment response? That is, which variables seem to be the best diagnostic indices? 8. Which dependent variables consistently differ entiated the high active subjects and the low active sub jects in terms of pre-treatment measurement and treatment response? That is, which of the variables seem to be the best diagnostic/treatment prediction indices? 9. How do the various dependent measures correlate with each other? CHAPTER IV RESULTS Pre-Test Data The multivariate Planned Comparisons analysis comparing hyperactive subjects with the normal control subjects on the initial testing (Pre-Test) yielded overall significance (p < 0.02) with significant differences on 6 dependent variables: MFF Activity (p < 0.002), Haze Activity (p < 0.001), Resting Activity (p < 0.001), PR-F Operant Activity (p < 0.02), MFF Error (p < 0.01) and MFF Time (p < 0.02) (See Table 1). The mean scores indicate that the hyperactive subjects were more active, made more errors and responded faster than the normal subjects. The untransformed mean scores are presented in Table 2. As mentioned in Chapter III, visual inspection of the Activity scores suggested that combining MFF and Maze Activity scores would provide the basis for differen tiating the High-Active and Low-Active subgroups. The correlation between the pre-test activity scores on these two tests was 0.796 (p < 0.000004). The mean pre-test MFF activity score for the Low-Active group was 3.5 while the High-Active group mean was 26.7. As for the Maze Activity scores, the Low-Active mean was 9.5, the High-Active mean, 93 TABLE 1 PLANNED COMPARISON ANALYSIS OF PRE-TEST DIFFERENCES BETWEEN HYPERACTIVE AND NORMAL SUBJECTS D.F. Hyp. D.F. Error F P< Multivariate 10 13 3.6 0.02 Univariate F Tests Variable MS F (1.22) P< Resting Activity* 19.15 13.38 0.001 Maze Activity* 43.16 23.77 0.001 MFF Activity* 19.98 12.05 0.002 DRL Operant Activity 2.75 2.17 0.155 PR-F Operant Activity* 6.77 5.97 0.023 Maze Error 3.58 1.62 0.216 MFF Error♦ 6.10 7.77 0.011 MFF Time* 16.25 6.29 0.020 DRL Operant 13.06 1.85 0.188 PR-F Operant 0.00 0.00 0.954 ♦Significant atoC = .05 95 TABLE 2 PRE-TEST MEAN SCORES IN HYPERACTIVE NORMAL SUBJECTS AND Mean of Mean of Hyperactive Group Normal Group Variables N * 16 N - 8 Resting Activity* 11.6 2.8 Maze Activity* 19.4 4.1 MFF Activity** 15.1 3.4 DRL Operant Activity 3.7 1.4 PR-F Operant Activity*** 4.5 1.2 Maze Error 10.1 4.6 MFF Error+ 14.2 7.6 MFF Time++ 10.3 25.0 DRL Operant* 120.2 157.9 PR-F Operant 15.8 16.1 *Significant p < .001 ‘♦Significant p < .002 ♦♦♦Significant p < +P < .01 ++P < .02 .02 96 TABLE 3 PLANNED COMPARISON ANALYSIS OF PRE-TEST DIFFERENCES BETWEEN HIGH-ACTIVE AND LOW-ACTIVE HYPERACTIVE SUBJECTS Multivariate D.F. Hyp. D.F. 8.0 14 Error .0 F 1.02 P < 0.465 Univariate F Tests Variable F (1, 21) MS P < Resting Activity 3.22 6.79 0.09 DRL Operant Activity 0.45 0.87 0.51 PR-F Operant Activity 1.57 3.05 0.22 Maze Error 0.05 0.15 0.82 MFF Error 0.03 0.02 0.86 MFF Time 0.01 0.04 0.92 DRL Operant 0.71 4.13 0.41 PR-F Operant 0.99 1.23 0.33 97 29.3. Summing the "per-minute" activity scores on the MFF and Maze Tests, the Low-Active group mean score is 13.0 with a range from 2.0 to 22.6, while the High-Activo group mean is 55.2 with a range from 39.2 to 80.7. Thus, there is a 16.8 point gap between the most active subject in the Low-Active group and the least active subject in the High- Active group. Since MFF activity and Maze Activity were used to differentiate the two hyperactive subgroups, only 8 dependent variables were analyzed in the Planned Comparison analysis comparing the Low-Active and High-Active groups on pre-test measures. The analysis yielded a nonsignifi cant multivariate F (p < 0.465) with only one univariate F, Resting Activity, approaching significance (p < 0.09) (See Table 3). The mean scores are presented in Table 4. VerbalMediation Treatment Effect, Interactions and Practice Effect on Post Test 1 The analysis yielded a significant main effect in comparing the pre-test scores to post-test scores across all subjects (p < 0.001), raising the possibility that a practice effect is operating. The univariate F tests in dicated that seven variables changed significantly - Maze Activity (p < 0.002), MFF Activity (p < 0.008), Maze Error (p < 0.002), DRL Operant (p < 0.015), and PR-F Operant (p < 0.001) (See Table 5). However, when the analyses for interaction effects were performed, they both TABLE 4 PRE-TEST MEAN SCORES IN LOW-ACTIVE AND HIGH-ACTIVE SUBJECTS Variables Means in Low-Active Subjects Means in High-Active Subjects Resting Activity 7.0 16.2 Maze Activity* 9.5 29.3 MFF Activity* 3.5 26.7 DRL Operant Activity 4.8 2.6 PR-F Operant Activity 5.8 3.1 Maze Error 9.5 10.6 MFF Error 14.4 13.9 MFF Time 9.8 10.8 DRL Operant 133.5 106.8 PR-F Operant 17.9 13.6 *Not included in Planned Comparison Analysis. 99 TABLE 5 PLANNED COMPARISON ANALYSIS OF CHANGE IN SCORES FROM PRE-TEST TO POST-TEST 1 ACROSS ALL GROUPS Multivariate D.F. Hyp. 10 D.F. Error F 35 7.9 P < .001 Univariate F Tests Variable MS F (1, 44) p< Resting Activity 0.01 0.00 0.951 Maze Activity* 11.80 10.54 0.002 MFF Activity* 9.34 7.66 0.008 DRL Operant Activity 1.39 2.21 0.144 PR-F Operant 0.58 1.00 0.322 Maze Error* 7.09 10.71 0.002 MFF Error* 7.10 16.10 0.001 MFF Time* 33.37 21.95 0.001 DRL Operant* 23.53 6.39 0.015 PR-F Operant* 3.68 17.90 0.001 *Signifleant at flt * .05 100 yielded significance, indicating that the hyperactive groups changed significantly more than the normal group on several variables. The first interaction analysis compared the changes from Pre-Test to Post-Test 1 between normal and hyperactive subjects. The results, presented in Table 6, indicate that there was significant overall differential change (p < 0.008). Specifically, 4 of the dependent variables attained significance in this analysis. Maze Activity (p < 0.001), MFF Activity (p < 0.02), DRL Operant Activity (p < 0.02) and PR-F Operant Activity (p < 0.01). One other variable, DRL operant, approached significance (p < 0.095). The mean Pre-Test, Post-Test 1 and change scores for hyperactive and normal subjects on all variables are presented in Table 7. It is interesting to note that while all of the activity scores decreased on PT-1 in the hyperactive group, there was a consistent increase in all activity scores in the normal subjects. The second analysis yielding a significant inter action effect tested the differential change from Pre-Test to Post-Test 1 between the High-Active and Low-Active Hyperactive subjects. The multivariate F was significant (p < 0.02) as well as 2 univariate F tests: Maze Activity (p < 0.02), and MFF Activity (p < 0.006) (See Table 8). Mean Pre-Test, Post-Test 1 and change scores for the Low- Active and High-Active groups are presented in Table 9. 101 TABLE 6 PLANNED COMPARISON ANALYSIS OF INTERACTION BETWEEN CHANGE FROM PRE-TEST TO POST-TEST 1 IN NORMAL AND HYPERACTIVE SUBJECTS Multivariate D.F. Hyp. 10 D.F. Error 35 F 2.97 P < 0.008 Univariate F Tests Variable MS F (1, 44) P < Resting Activity 1.72 1.24 0.269 Maze Activity* 14.62 13.06 0.001 MFF Activity* 7.34 6.02 0.018 DRL Operant Activity* 3.75 5.96 0.019 PR-F Operant Activity* 4.20 7.28 0.010 Maze Error 0.43 0.65 0.424 MFF Error 0.75 1.70 0.200 MFF Time 3.16 2.08 0.156 DRL Operant 10.72 2.91 0.095 PR-F Operant 0.02 0.10 0.752 *Slgnifleant at 01 * .05 TABLE 7 MEAN PRE-TEST, POST-TEST 1 AND CHANGE SCORES IN NORMAL AND HYPERACTIVE SUBJECTS Hyperactive S's Normal S's Variable Pre-Test Post-Test 1 Change Pre-Test Post-Test 1 . Change Resting Activity 11.6 11.0 - 0.6 2.8 4.3 + 1.5 Maze Activity* 19.4 9.8 - 9.6 4.1 6.6 + 2.5 MPP Activity+ 15.1 9.3 - 5.8 3.4 5.3 + 1.9 DRL Operant Activity+ 3.7 2.9 - 0.8 1.4 2.5 + 1.1 PR-F Operant Activity** 4.5 3.3 - 1.2 1.2 2.9 + 1.7 Maze Error 10.1 8.7 - 1.4 4.6 5.3 + 0.7 MFF Error 14.2 9.2 - 5.0 7.6 6.1 - 1.5 MFF Time 10.3 21.3 + 11.0 25.0 36.3 +11.3 DRL Operant++ 120.2 151.7 + 31.5 157.9 133.6 -24.3 PR-F Operant 15.8 19.8 + 4.0 16.1 19.1 + 3.0 *p < 0.001 **p < 0.01 +p < 0.02 ++p < 6.095 103 TABLE 8 PLANNED COMPARISON ANALYSIS OF INTERACTION BETWEEN CHANGE FROM PRE-TEST TO POST-TEST 1 IN LOW-ACTIVE AND HIGH-ACTIVE HYPERACTIVE SUBJECTS D.F. Hyp. D.F. Error F P< Multivariate 10 12 3.47 0.02 Variable Univariate MS F Tests F (1, 21) P< Resting Activity 3.84 1.82 0.192 Maze Activity* 11.67 5.88 0.024 MFF Activity* 15.02 9.43 0.006 DRL Operant Activity 0.27 0.14 0.711 PR-F Operant Activity 1.18 0.61 0.444 Maze Error 0.00 0.00 0.981 MFF Error 0.51 0.93 0.347 MFF Time 3.17 0.72 0.406 DRL Operant 7.03 1.20 0.285 PR-F Operant 0.35 0.27 0.608 ♦Slgnifleant at Of * .05 TABLE 9 MEAN PRE-TEST, POST-TEST 1, AND CHANGE SCORES IN LOW-ACTIVE AND HIGH-ACTIVE HYPERACTIVE SUBJECTS Low-Active Subjects High-Active Subjects Variable Pre-Test Post-Test 1 Change Pre-Test Post-Test 1 Change Resting Activity 7.0 9.5 + 2.5 16.2 12.4 - 3.8 Maze Activity* 9.5 7.2 - 2.3 29.3 12.4 - 16.9 MFF Activity** 3.5 4.3 + 0.8 26.7 14.3 - 12.4 DRL Operant Activity 4.8 4.4 - 0.4 2.6 1.3 - 1.3 PR-F Operant Activity 5.8 4.3 - 1.3 3.1 2.3 - 0.8 Maze Error 9.5 8.9 - 0.6 10.6 8.4 - 2.2 MFF Error 14.4 8.4 - 6.0 13.9 10.0 - 3.9 MFF Time 9.8 23.7 + 13.9 10.8 18.8 + 8.0 DRL Operant 133.6 183.3 + 49.7 106.8 120.0 + 13.2 PR-F Operant 17.9 23.6 + 5.7 13.6 16.0 + 2.4 *p < 0.02 **p < 0.006 i 104 105 These mean scores Indicate that It Is the High-Active group which demonstrated the significantly greater reduction in activity during Maze testing and MFF testing on Post-Test 1 as compared to their Pre-Test scores. Ritalin Treatment Effect. Interactions and Practice Effect on Post-Test 2 Analyzing the changes occurring on Post-Test 2 as compared to Post-Test 1 across all groups yielded a signi ficant multivariate F (p < 0.005) suggestive of a practice effect. Three dependent variables changed significantly. MFF Activity (p < 0.01) Maze Error (p < 0.001) and DRL Operant (p < 0.02). The data relevant to this analysis is presented in Table 10. The subsequent Planned Comparison, testing the interaction between Ritalin Treatment effect (change from Post-Test 1 to Post-Test 2) and group (Normal versus Hyperactive) yielded nonsignificant results in both the multivariate test (p < 0.879) and all univariate tests. (See Table 11) Mean scores for both groups on Post-Test 2 and change scores are presented in Table 12. The results of the final Planned Comparison analysis testing the differential change from Post-test 1 to Post-Test 2 in Low-Active and High-Active hyperactive. Subjects were also not significant for both the multi variate test (p < 0.805) and all univariate tests, (See Tables 13 and 14). Thus the data suggest the Ritalin treatment did not result in significant changes in activity 106 TABLE 10 PLANNED COMPARISON ANALYSIS OF CHANGE IN SCORES FROM POST-TEST 1 TO POST-TEST 2 ACROSS ALL GROUPS Multivariate D.F. Hyp. 10 D.F. Error 35 F 3.23 P < 0.005 Variable Univariate F Tests MS F P < Resting Activity 0.01 0.01 0.937 Maze Activity 0.86 0.77 0.385 MFF Activity* 8.07 6.62 0.014 DRL Operant Activity 1.02 1.63 0.209 PR-F Operant Activity 0.90 1.56 0.218 Maze Error* 10.91 16.50 0.001 MFF Error 0.00 0.01 0.924 MFF Time 3.34 2.20 0.146 DRL Operant* 20.66 5.61 0.022 PR-F Operant 0.20 0.98 0.328 *Signifleant at * .05 107 TABLE 11 PLANNED COMPARISON ANALYSIS OF INTERACTION BETWEEN CHANGE FROM POST-TEST 1 TO POST-TEST 2 IN NORMAL AND HYPERACTIVE SUBJECTS D.F. Hyp. D.F. Error F P < Multivariate 10 35 0.50 0.879 Variable Univariate F Tests MS F P< Resting Activity 0.80 0.58 0.449 Maze Activity 0.02 0.01 0.907 MFF Activity 0.23 0.19 0.667 DRL Operant Activity 0.81 1.29 0.262 PR-F Operant Activity 0.20 0.34 0.560 Maze Error 0.00 0.00 0.982 MFF Error 0.00 0.01 0.931 MFF Time 1.61 1.06 0.309 DRL Operant 3.28 0.89 0.351 PR-F Operant 0.41 1.98 0.166 TABLE 12 MEAN POST-TEST 2 AND CHANGE SCORES IN NORMAL AND HYPERACTIVE SUBJECTS Hyperactive S1s Normal S's Variable Change Post-Test 2 (PT 1 to PT 2) Change Post-Test 2 (PT 1 to PT 2) Resting Activity 9.5 - 1.5 5.0 + 0.7 Maze Activity 8.9 - 0.9 4.9 - 1.7 MFF Activity 5.1 - 4.2 3.5 - 1.8 DRL Operant Activity 1.3 - 1.6 2.8 - 0.3 PR-F Operant Activity 3.1 - 0.2 2.0 - 0.9 Maze Error 4.5 - 4.2 1.6 - 3.7 MFF Error 5.0 - 4.2 6.6 - 0.5 MFF Time 34.0 + 12.9 33.0 + 0.3 DRL Operant 175.2 + 23.5 187.0 + 54.0 PR-F Operant 24.8 + 5.0 22.9 + 3.8 109 TABLE 13 PLANNED COMPARISON ANALYSIS OF INTERACTION BETWEEN CHANGE FROM POST-TEST 1 TO POST-TEST 2 IN LOW-ACTIVE AND HIGH-ACTIVE HYPERACTIVE SUBJECTS D.F. Hyp. D.F. Error F P < Multivariate 10 12 0.575 0.805 Variable Univariate F Tests MS F (1, 21) P < Resting Activity 1.22 0.58 0.456 Maze Activity 2.26 1.14 0.298 MFF Activity 1.27 0.79 0.383 DRL Operant Activity 0.43 0.22 0.643 PR-F Operant Activity 0.00 0.00 0.994 Maze Error 1.46 0.52 0.478 MFF Error 0.05 0.10 0.760 MFF Time 1.59 0.36 0.554 DRL Operant 0.01 0.00 0.966 PR-F Operant 0.01 0.01 0.433 TABLE 14 MEAN POST-TEST 2 AND CHANGE SCORES IN LOW-ACTIVE AND HIGH-ACTIVE HYPERACTIVE SUBJECTS Low-Active Hyperactive S's High-Active Hyperactive S's Variable Post-Test Change 2 (PT 1 to PT 2) Post-Test Change 2 (PT 1 to PT 2) Resting Activity 10.7 + 1.2 8.2 - 4.2 Maze Activity 10.0 + 2.8 7.7 - 4.7 MPF Activity 3.7 - 0.8 6.4 - 7.9 DRL Operant Activity 1.6 - 2.8 1.0 - 0.3 PR-F Operant Activity 4.1 - 0.2 2.0 - 0.3 Maze Error 2.9 - 7.0 6.1 - 2.3 MFF Error 7.4 - 1.0 10.0 0.0 MFF Time 45.3 + 21.6 23.1 + 4.3 DRL Operant 211.4 + 28.1 139.0 + 19.0 PR-F Operant 23.5 - 0.1 16.1 + 0.1 , Ill or performance, and that the significant changes occurring across groups is attributable only to a practice effect. Correlations The correlations between activity measures and per formance measures are presented in Table 15. Only one significant correlation obtained; Maze Error with MFF, Activity (r - 0.467; p< 0.033). Two significant correla tions among performance measures obtained. MFF error and MFF time were negatively correlated (r * 0.474jp < 0.019) and performance on the 2 operant tasks were correlated (r - 0.426; p < 0.038) (See Table 17). Finally correla tions among the activity measures,presented in Table 16, yielded 3 significant coefficients. PR-F Operant activity was highly correlated to DRL operant activity (r - 0.792; p < 0.000005) and MFF Activity was correlated with Resting Activity (r ■ 0.418; p<.043) and with Maze Activity (r - 0.538; p < 0.007). TABLE 15 CORRELATIONS BETWEEN ACTIVITY AND PERFORMANCE MEASURES Variables Resting Activity Maze Activity MFF Activity DRL Operant Activity PR-F Operant Activity Maze Error 0.226 0.110 0.467* - 0.074 - 0.133 MFF Error 0.038 - 0.256 - 0.064 - 0.161 - 0.092 MFF Time 0.187 - 0.024 - 0.090 0.221 0.279 DRL Operant - 0.010 - 0.115 - 0.161 - 0.042 - 0.066 PR-F Operant 0.078 0.157 0.049 0.009 - 0.045 *P < .033 112 TABLE 16 CORRELATIONS BETWEEN ACTIVITY MEASURES Variables Resting Activity Maze Activity MFF Activity DRL Operant Activity Maze Activity 0.189 MFF Activity 0.418** 0.538* DRL Operant Activity 0.085 0.225 0.213 PR-F Operant Activity 0.100 0.327 0.199 0.7924- +p < 0.000005 *p < 0.007 **p < 0.043 113 TABLE 17 CORRELATIONS BETWEEN PERFORMANCE MEASURES Variables Maze Error MFF Error MFF Time DRL Operant MFF Error 0.186 MFF Time - 0.159 - 0.474* DRL Operant - 0.192 - 0.046 - 0.060 PR-F Operant - 0.242 - 0.167 - 0.074 0.426** *p < 0.019 **p < 0.038 114 CHAPTER V DISCUSSION Evaluation of Hypotheses The present study was designed to further investigate and to refine diagnostic indices of hyper activity and treatment methods. Specifically, the goals of this study were threefold: 1) To investigate para meters which would differentiate the hyperactive child from normal children and possibly differentiate subtypes of hyperactive children. 2) To evaluate the effectiveness of the verbal mediation training program in reducing over activity and improving task performance of hyperactive children. 3) To examine those parameters which would pre dict differential treatment response among hyperactive children given two treatments— verbal mediation training and stimulant medication (Ritalin). Hypothesis One Hyperactive children will demonstrate higher activity levels in situations requiring motor inhibition and will perform more poorly on tasks requiring response inhibition (motor and verbal) and attention than normal children. This hypothesis was supported. The hyperactive subjects were found to be significantly more active when asked to sit and wait before testing began, and during 115 116 three out of the four tasks. These results are consistent with the studies by Pope (1970) and Sykes and co-workers (1971) which found activity level differentiated hyper active and normal children during voluntary inhibition tasks (asking the child to sit still), during moderately difficult task situations, and during tasks requiring con tinuous attention. Only activity during the DRL-Operant task failed to differentiate the groups. As predicted, on task performance, the hyperactive subjects demonstrated poorer response inhibition than the normal subjects. Spe cifically, they evidenced a more impulsive conceptual tempo than the normal subjects, characterized by signifi cantly shorter response latency on the MFF Test. The Porteus Maze test of impulsivity and both operant tasks failed to differentiate the two groups. Hypothesis Two Two separate groups of hyperactive children will be differentiated by activity level and will demonstrate differential task performance. Specifically, the High- Active hyperactive subjects will perform more poorly than the Low-Active hyperactive subjects on tasks requiring re sponse inhibition (motor and verbal) and attention. This hypothesis was partially supported, by the activity vari ables but not by the performance variables. Subjects clustered in two groups, differentiated by MFF activity, 117 Maze activity, and Resting activity. Activity during the operant tasks failed to differentiate these groups, how ever. Thus it appears that while children diagnosed as hyperactive show significant differences in amount of activity, they uniformly demonstrate some learning dis abilities, characterized by an impulsive conceptual tempo, resulting in less accuracy on tasks. Hypothesis Three Verbal mediation training will be an effective treatment intervention for hyperactive children, resulting in decreased activity level and improved task performance. This hypothesis was partially supported. Hyperactive subjects' activity scores during all of the tasks were significantly lower following the training as compared to the pretest scores, while the normal subjects actually evidenced a slight increase in activity. It was under standable that the Resting activity did not show this sig nificant differential change since the verbal mediation training taught subjects self-verbalizations, with associ ated reinforcement, in relation to task performance, so that they were not taught to mediate their behavior verbally in a resting situation. As for performance scores, only performance on the DRL-Operant task approached significance with the hyper active subjects' mean score increasing by 31.5 points (from 120.2 to 151.7). This is more impressive considering the normal control subjects showed a decline of twenty- four points from 157.9 to 133.6. Thus the hyperactive group actually surpassed the normals, suggesting that the training was moderately effective in increasing motor inhibition and attention. As for the tasks measuring con ceptual tempo, improvement on the MFF test in the hyper active group compared to the change in the normal group failed to reach significance, but observing the mean error scores and reaction times suggests that there was moderate ly more improvement at the hyperactive group (see Table 7). Their mean error score on the MFF test declined five points from 14.2 to 9.2 while the controls only declined 1.5 points from 7.6 to 6.1. In general, the training resulted in (1) large decline in activity, (2) improvement in operant task performance indicative of better attention and capacity to inhibit motor responses, and (3) moderate ly less impulsivity. Hypothesis Four Verbal mediation training will be more effective with the Low-Active hyperactive subjects. This hypothesis was not supported. The results indicated the reverse. The High-Active subjects demonstrated the significantly greater improvement following this treatment. Specifi cally, there were significantly greater declines in 119 activity level on the MFF and Maze tests in the High- Active group as compared to the Low-Active group. As for task performance, there was no significant differential change in any of the variables following verbal mediation training. It was interesting however that the Low-Active subjects surpassed the normal subjects on both operant tasks while the High-Active subjects, although having shown improvement, were still far below both the normal and Low-Active groups. (See figures 1 and 2) Thus the results suggest that verbal mediation training is a highly effective intervention for reducing activity in hyper active children, irrespective of their degree of activity, and is moderately effective in improving task performance. However, the relatively more active subjects show more dramatic declines in activity following this treatment. Hypothesis Five Stimulant medication administered to hyperactive subjects after the verbal mediation training will result in additional improvement, including decrease in activity level and improved task performance. This hypothesis was not supported. The data indicated that there was a sig nificant practice effect operating and that the hyper active subjects failed to change any more than the normal group on any of the variables. Thus it appears that stimulant medication (Ritalin) may be an unnecessary, if Figure 1 DRL- Operant Score 220 .ow-Active S 210 200 190 180 170 160 150 gh-Active S 140 130 120 110 100 Post- POSt- Pre- Test Test 1 Test 2 Change in DRL-Operant Performance Scores in Normal, Low-Active, and High-Active Hyperactive Subjects PR-F Operant Score Figure 2 121 26 24 22 20 18 16 14 12 10 8 6 4 2 Low-Active S's No High-Active S's Pre- Test Post- Test 1 Post- Test 2 Change in PR-F-Operant Performance Scores in Normal, Low-Active, and High-Active Hyperactive Subjects 122 not detrimental pharmacological Intervention if a child has the opportunity for the behavioral treatment investi gated in the present study. The design of this study does not permit a statement as to the relative effectiveness of Ritalin alone versus verbal mediation training alone since all subjects had been trained first. Hypothesis Six Stimulant medication administered after the verbal mediation training will be significantly more effective with High-Active hyperactive subjects. This hypothesis was not supported. There was no evidence of differential re sponsiveness to stimulant medication in contrast to pre vious research suggesting that the relatively more active hyperactive subjects respond better to stimulants (Fish, 1971; Howell et al., 1972; Millichap and Johnson, 1973; Post and Tiber, 1974). However, results are not directly comparable to these studies since the High-Active subjects in the present study had already received the verbal medi ation treatment which resulted in a significant decrease in activity and some moderate task improvement. Diagnostic and Treatment Implications The results of this study are consistent with the literature suggesting activity level, in certain situ ations, as a diagnostic index of hyperactivity. Specifi cally, this study found that amount of seat activity 123 differentiated hyperactive from normal children during moderately difficult tasks (MFF, Porteus Mazes, and PR-F Operant) requiring attention and response inhibition and during a low stimulation-voluntary motor inhibition situ ation where the child is asked to remain seated waiting for the Experimenter to return. It was interesting that activity during the DRL- Operant task did not differentiate these groups, while activity on the PR-F Operant task did. A possible ex planation for this, consistent with other studies on ac tivity level and task performance, invokes the relation ship between attention, environmental cues, and activity level, as follows: In general, there is a great deal of experimental and theoretical support for the proposed inverse relationship between attention and motor activity (Alabiso, 1972; Deutsch and Deutsch, 1970; Gardner et al., 1959) . Also, it is known that environmental cues can function to enhance the orienting response and continued attention to task stimuli. The DRL-Operant task was a continuous reinforcement task where the child received continuous novel stimuli in the form of lights flashing a higher number each time he pressed the lever at a slow enough rate. The required rate was not as slow as in the Post and Tiber study (1974) so that it was much easier for the hyperactive child to receive the reinforcement, that is, to cause the points to change. Therefore, it seems 124 conceivable that the frequently changing lighted numbers functioned to enhance their attention to the task. Conse quently there was a decline in motor activity, so that their activity level was closer to that of the normal group. Whereas, during the PR-F operant task, the chil dren experienced numbers changing much less frequently, only after fifteen lever presses, probably resulting in less task orientation and therefore higher activity. In fact, the Experimenter observed that when some of the hyperactive subjects began the PR-F task and saw that the numbers did not increase as rapidly as on the previous DRL task, they frequently stopped pressing the lever, and ceased attending to the screen. Some children spontane ously made comments such as "Look, the machine is broken." This difference in the nature of the task stimuli is par ticularly salient in relation to hyperactive children, for they have been found to have deficient orienting responses and faster habituation than normal children, as measured by physiological-arousal indices (Cohen et al., 1971; Satterfield and Dawson, 1971). This interpretation of the results, invoking attention, is supported by Sprague and Toppe (1966), who propose that their hyperactive subjects did poorly on a twelve-second delayed reinforcement- discrimination task because they did not attend long enough to associate the reinforcement with the response. The results of this study are also consistent with 125 the literature in suggesting activity level as a diag nostic index in differentiating two distinct groups within a group of subjects diagnosed hyperactive. That is, hyperactive subjects seem to cluster into a High-Active or a Low-Active group, based on the amount of activity during the MFF task and Maze task, as well as during the low stimulation-voluntary motor inhibition situation (Resting activity). However, these subgroups were not differenti ated by PR-F Operant activity, although this measure did differentiate the hyperactive group as a whole from the normal children. Given that hyperactive subjects seem to demonstrate an attention deficit, and given the inverse relationship between attention and activity, it may be that activity on the PR-F task failed to differentiate the High and Low Active group because the task provides fewer cues to focus attention than the other tasks, and therefore re sulted in less attention and more activity across all hyperactive subjects. The normals, having better attention responses, demonstrate lower activity level on this task. The results pertaining to the task performance variables lend additional support to the observation of poor at tention in hyperactive subjects irrespective of their ac tivity level. The High-Active versus the Low-Active hyperactive subjects did not demonstrate differential performance on any of the task performance measures while two of the 126 variables, accuracy and reaction time on the MFF test did differentiate hyperactive group as a whole from the normal subjects. Also, performance scores on the other tasks, with the exception of the PR-F Operant task, demonstrated a trend in the expected direction. For example, the mean average scores on the Mazes was 4.6 in the normal group and 10.1 in the hyperactive group. The mean number of points earned on the DRL Operant task was 157.9 in the normal group and 120.2 in the hyperactive group. Another question which may be raised is, if the hyperactive did demonstrate poor attention on the PR-F task suggested by their high activity, why did task per formance scores fail to differentiate them from the normal subjects? A possible explanation is that points could be scored irrespective of the attention to the screen, just on the basis of fast lever pressing. In fact, other re search has found that hyperactive children demonstrate faster finger tapping speeds than normal children (Stevens et al., 1970) in a non-incentive condition, but normal children surpass hyperactive children when incentives were offered. The point counter on the PR-F Operant task is a form of incentive, but evidently not strong enough to re sult in significantly faster lever pressing in the normal groups as compared to the hyperactive subjects, who naturally respond at a faster rate irrespective of atten tion to the point counter, that is, irrespective of reward 127 contingencies. To summarize thus far, the results suggest that activity level monitored during three cognitive tasks (MFF, Porteus Maze, and PR-F Operant) and during a low stimulation-motor inhibition situation, is a useful diag nostic index in differentiating hyperactive children from normal children, and, with the exception of the PR-F Oper ant activity, useful in differentiating two subgroups of hyperactive children, referred to as the High-Active sub jects and the Low-Active subjects. Also, tasks measuring conceptual tempo and requiring attention and voluntary motor inhibition differentiate hyperactive from normal subjects, but fail to differentiate the High-Active and Low-Active group. Thus it is proposed here that children currently diagnosed as hyperactive are primarily differen tiated from normal children by the presence of certain learning disabilities characterized by impulsive conceptu al tempo, attention deficits, and deficits in motor in hibitory capacity. However, the presence of common learning disabilities or deficits does not necessarily justify or validate a singular diagnosis of hyperactivity in children who demonstrate significant differences in amount of activity, and as will be further discussed, dif ferences in responsivity to treatment interventions. Rather, it seems plausible to suggest that the High-Active subjects are a separate diagnostic entity from the 128 Low-Active subjects. Considering these studies finding that relatively more active hyperactive subjects respond better to stimulant medication and considering the physio logical studies of hyperactive children, it may be that the High-Active subjects represent the "organic-neuro logical" syndrome while the Low-Active subjects represent a "behavioral-functional" syndrome. The results of the verbal mediation training fur ther support a differentiation based on activity level for while both the High-Active and Low-Active groups demon strated improvement following the training, the High- Actives demonstrated significantly more improvement in terms of greater activity declines during the MFF and Maze tests. In fact, the Low-Active group did not decline on MFF activity, but increased slightly. It is possible that this significantly greater improvement in the High-Active group on MFF and Maze ac tivity is due to a statistical regression to the mean effect, but it does not seem likely for two reasons: 1) On Maze activity, both groups decreased? the Low-Active subjects decreased by 2.3 points while the High-Active subjects by 16.9 points whereas the regression effect would predict that the Low-Active subjects would increase. 2) On MFF activity, though the Low-Active group mean did increase, it was only a 0.8 increase whereas the High- Active group declined 12.4 points. This disproportion in 129 amount of change seems inconsistent with a regression effect. Also, the basic question at hand is whether the High-Active and Low-Active subgroups really represent separate diagnostic entities. Looking at MFF activity scores across all sessions and groups (see figure 3), the Low-Actives objects seem distinctly different from the High-Active subjects, and interestingly behave much like the normal subjects. This is also true on resting activ ity (see figure 4). Also, looking at operant performance changes (see figures 1 and 2) the Low-Actives again seem distinctly different from the High-Actives in that their performance is above that of the normal group after treat ment. Looking at all the hyperactive subjects irrespec tive of pretest activity level, the data indicates that the verbal mediation training was highly successful in reducing activity level on all four tasks but not in re ducing Resting activity. This is understandable since the essence of the verbal mediation training consisted in teaching self-verbalization to mediate motor inhibition and attention in a task situation. The content of the verbalizations was directly applicable to task situations (see Appendix A) and not to a resting situation. As Luria proposed (1964) self-verbalizations function to orient the child to task at hand. In the present study the decline in task activity levels and the 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Figure 3 130 High-Active S's Low-Active S's Normal S's Pre- Post- Post- Test Test 1 Test 2 Change in MFF Activity in Normal, Low-Active and High-Active Hyperactive Subjects 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Figure 4 131 High-Active S's Low-Active S’s Normal S1s Pre- Test Post- Test 1 Post- Test 2 Change in Resting Activity in Normal, Low-Active, and High Active Hyperactive Subjects 132 nearly significant increase in DRL performance suggests that the verbal mediation training does function to in crease the child's orientation to the task stimuli, con sistent with Luria's proposition and consistent with the proposed inverse relationship between activity and atten tion. That is, the training provided the child with self generated attention cues, in the form of task-related verbalizations. It is not determinable whether the decreased activ ity resulted directly from the learning of mediators aimed at reducing activity ("I must stop all moving"), or in directly from the increased attention, given the inverse relationship between attention and activity level, or from both. The improved DRL-Operant performance may also be due to improved motor inhibitory capacity mediated by the verbalizations "I must go very slowly," but first the child would have had to attend to the screen to learn that points were contingent on relatively slow pressing. To better evaluate whether the training does in fact operate primarily to provide the child with attention cues, a purer test of attention should be included in a research design. Although task performance on the MFF test did not improve significantly, the mean scores did improve moderately after training. Possibly further training would have resulted in more dramatic improvement since the MFF task requires a more complex cognitive strategy than 133 the DRL-Operant task. That Is, It seems plausible to sug gest that the mediators were learned well enough to inprove the child's orientation to the task and to reduce his activity, but correct responses on the MFF also required a methodical cognitive strategy which four training sessions could not provide for the children. The number of training sessions in a clinical setting would obviously have to be adjusted to the degree of the child's deficit. A few sub jects in the present study demonstrated nearly perfect performance on the MFF test after only four training trials while others demonstrated little or no improvement. Meichenbaum and Goodman (1971) , using a similar procedure, found only a trend for improved MFF scores in a sample of hyperactive children after four training sessions but found significant improvement after only one training ses sion with non-reflective normal subjects. In summary, the effective application of the verbal mediation training was wider than expected, being beneficial to both the High-Active and Low-Active hyper active subjects. Further, the addition of Ritalin to the treatment regimen failed to significantly produce any more improvement in either group. It appears that though ac tivity level differentiated two distinct subgroups within the hyperactive sample, this index did not serve to pre dict differential treatment response to Ritalin within the limitations of the design of the present study. It was 134 surprising that there was so little improvement with the stimulant medication among the hyperactive subjects, given the large body of literature supporting this kind of treatment for hyperactivity. There are several possible explanations for the present results. First of all, while most of the other studies com pared post-drug performance with performance before any treatment intervention, the present study compared post drug performance with another post-treatment measure serving as the new base-line. In many of the other clini cal and laboratory studies, the "drug effect" is observed over a period of weeks during which time various learning and placebo effects contribute to the child's improvement as compared to pre-treatment. The question asked here was whether the addition of Ritalin would further add to the improvement, given beneficial results with the mediation treatment. Possibly a separate group's design would sug gest that certain hyperactive children could benefit as much from Ritalin alone as from the training alone. An other possibility is that combining the training with the stimulant medication would result in the most dramatic and sustained improvement, as suggested by Christensen and Sprague's study (1973). Another possible interpretation of the results is that the "organic-neurological" hyperactivity, highly responsive to stimulant medication, which was hypothesized 135 to be represented by the High-Active group, is a relatively rare syndrome, as suggested by Bax (1972), and is not represented by subjects ir this sample. This interpreta tion was further suggested by the Experimenter's recent clinical experience in treating a six-year old hyperactive boy not included in the study. He was far more active than any of the subjects in the study and demonstrated im mediate and dramatic improvement following stimulant medication in terms of activity decrease and behavioral- academic improvement. Several clinicians have reported immediate and dramatic improvement within one half-hour of administering stimulant medication to a hyperactive child, but again they report this is a relatively rare occurrence. A final interpretation is simply that hyperactive children, with either a behavioral or neurological etiol ogy can be taught, using behavioral principles, to bring their overt behavior under their own verbal discriminative control, resulting in improvement equal to or better than that obtainable with pharmacotherapy. Future research should investigate the effect of more training sessions and hopefully will add other mediators to the training to extend the application to other tasks and other situations. Also, considering the success of this training with the Low-Active subjects, new research might investigate the potential application of this treatment in improving academic performance in non- clinical populations. As Meichenbaum and Goodman (1971) suggest, this training may well be beneficial to the culturally deprived lower SES child, who characteristically fails to spontaneously use verbal mediators. These and other children with mediation deficits may not be diag nosed as hyperactive but may also respond well to this type of training. As for general education programs, it was mentioned in the literature review that the develop ment of verbal mediation follows a developmental sequence determined by chronological age and I.Q. It is interesting to speculate that this type of training program, adminis tered before the child spontaneously learns to use verbal mediators, could accelerate certain types of learning in normal children. The laboratory studies of Weir and Stevenson (1957), Bern (1967), and Keeney and his co-workers (1967) attest to the potential of verbal mediation training in improving task performance in normal children. What is needed now it seems are more long-term broadly-based re search studies, incorporating this kind of a training procedure in ongoing educational programs, with in-class performance serving as dependent variables. Summary In summary, the present study found that children diagnosed as hyperactive according to current clinical practice are highly differentiated from normal children by 137 two indices: 1) Activity level monitored during cognitive tasks and during a low stimulation situation requiring voluntary motor inhibition. 2) MFF performance indicative of an impulsive conceptual tempo. Activity level also highly differentiated two sub groups within the hyperactive sample, referred to as High-Active and Low-Active subjects, but task performance measures failed to differentiate them at a significant level, although there were trends in expected directions. Thus the results suggest that current diagnosis of hyper activity is primarily dependent on the presence of learn ing disabilities irrespective of amount of activity. How ever the substantial differences in amount of activity suggest the need to recognize more than one disorder. Secondly, the results support verbal mediation training as an effective intervention for children with learning disabilities, regardless of activity level. In fact, it appears that the relatively more active children demonstrate the most dramatic declines in activity. It is proposed that this training procedure functions to pro vide attention cues to task stimuli and to provide effect ive cognitive strategies. Since the training resulted in significant declines in activity and only moderate task performance improvement, it was suggested that the four training sessions accomplished primarily the orienting function while only some children learned the cognitive strategy well enough to demonstrate significant task im provement. This interpretation could be tested by the in clusion of a purer test of attention, requiring a simple cognitive approach, along with the cognitively more diffi cult MFF task. Finally the present experiment suggests that stimulant medication does not result in further im provement, above that achieved by verbal mediation train ing. Possibly, combining the medication with the training would provide more dramatic improvement in some cases. However, based on the results of the present study, it seems clear that, contrary to present clinical practice, the Clinician should first thoroughly instigate non-phar- macological treatments for the hyperactive child, re serving medication "trials" for those cases clearly unre sponsive to other more benign treatments. APPENDICES APPENDIX A TRAINING MATERIALS 140 141 TRAINING MATERIALS: INSTRUCTIONS, VERBAL MEDIATORS AND TASK MATERIALS Session 1 - Task 1: Trail Making I "Okay, now we're ready for our first task. Listen to the words I say to myself while I am doing the task. I will give myself a reward for saying the words." 1. "I stop look and think before I move” (takes M&M). "What is it I have to do? I have to draw a path starting with number one and ending with number ten." (M&M) "One-two-three— I must go very slowly— four-five-six-seven-eight-nine- ten. Good, I finished," (M&M) "and I earned three M&M's." 2. "Now I want you to try it." (Give subject a test sheet.) "I will say the words for you while you do the task. Listen closely because you will have to say the words on your own next." 3. "Okay, now you say the words aloud while you do the task. Remember to take the M&M's when you remember to say the words." (Coaching will probably be necessary at this time. Do not pro ceed to the next repetition until subject verbal izes correctly. At this time the Instructor gives the subject the bowl of M&M's.) 4. "Now we will do the same thing except I want you to whisper the words." (Do not proceed to next repetition until subject rewards himself appropriately.) "We will do this one more time. This time I want you to say the words to yourself. Do not move your lips." TRAIL-MAKING I © © Eho © ( D < D © @ © 144 Session 1 - Task 2: Matrix I "Let's try the same thing with a new task. Listen care fully to the words." 1. "I stop, look, and think before I answer." (M&M) "What is it I have to do? I have to find the picture," (pointing to alternatives) "that best fits in here" (pointing to space in the standard picture). "First I will look closely at the top picture. What is missing? One, two, three designs across and one, two designs down." (M&M) "Next I will look at every choice before I pick." (M&M) "It's not this one because there's only one design. It's not this one or this one" (pointing to two and three) "because the designs are not the same. It's not four or five because there are not enough. It's number six. Only number six has two rows of three— two down and three across. Good, I found it" (M&M) "and I earned four M&M's." 2-5. (Same sequence as first task.) MATRIX I ^ 4*" 4* < 4 > :*§**& < 4 > '4> < $ > < $ > < $> < $ > < $ >< ^ > ,4> < ^'4> < ^' > t < & > < & • < £ > < & > <&>.<&>• < & . < & • < & > oooooc ooooooc o : o : o ; o : o : o : c o o o o : d c 146 Session 2 - Task 1; Trail Making II (Check to see if the boy knows his ABC's) "Listen closely to what I say to myself again today be cause you will be saying these words later like we did last time." 1. "I stop, look and think before I move." (M&M) "What is it I have to do? I have to draw a path starting with number one and ending with number six. First the number, then a letter, then the next number, then the next letter." (M&M) (Start task) "One, then A, the next number is two, then a letter— B comes after A, I must go very slowly and think before I move. The next number is three, the next letter is C... Good, I finished." (M&M) "I earned three M&M's." 2-5. (Same sequence as previous session.) Session 2 - Task 2; Picture Match I "Now we will do something different. We have four matching problems, so we will have to say the words before each problem. Listen carefully.” 1. "What is it I have to do? 1 must find the one picture on the bottom that exactly matches the one on top. First I must look carefully at the top picture and then slowly look at each and 147 every other picture before I answer." (M&M) "I stop, look, and think before I answer." (M&M) "It's not this one because .... this one looks right but let me check the others before I choose. It's not this one because This is the best one." (M&M) (Use fingers to point to each one while considering it.) (At the end of the four problems say, "Good, I'm finished and I earned 12 M&M's.") 2-5. (Same sequence) TRAIL-MAKING II 148 PICTURE-MATCH I 149 150 Session 3 - Task 1: Matrix II "Are you ready for another task? Listen closely to the words." 1. "I stop, look and think before I answer." (M&M) "What is it I have to do? I have to find the picture," (pointing to alternative) "that best fits in here" (pointing to standard). "First I will look closely at the top picture. What's missing? One" (point), "two lines down and one" (point with finger), "two lines across." (M&M) "Next I will look at every choice before I pick." (M&M) "It's not this one because... this one looks good— one-two across and one- two down, but I must look at the others. It's not this one, because...." (give reason for each one). "Yes, this" (pointing) "is the best one. Good," (M&M) "I finished and I earned four M&M's." 2+3. (Same sequence.) 4+5. (Same sequence but this time, say .... "I think it will be easier to take all your M&M's after you've finished the task, don't you? If you remember to say all the words to yourself, you may take the four M&M's.”) (Be sure S says "Good" when he takes the candy.) 151 Session 3 - Task 2: Picture Match II "We're going to do a test much like one we did last time. Remember to listen closely." 1. "What is it I have to do? I must find the one picture on the bottom that exactly matches the one on top. First I must look carefully at the top picture and then slowly look at each and every other picture before I answer." (M&M) "I stop, look, and think before I answer." (M&M) "It's not this one because...." (pointing), "this one looks right but I must look at all of them before I choose." (M&M) "....Yes, this is the best one."(M&M) "Good." (After problem four say, "Good I finished and I earned 16 M&M's.") 2-5. (Same sequence as previous task.) MATRIX II 152 PICTURE-MATCH II 1 . ^ $ • ^ X - - -v J. (ry ' T Z ^ ' J Z ^ p f > ^ s d £ ) E ^ y. 6 = 2 ^ V / 0 ^ 3 > j < ^ 3> C^=J| £ - ^ p l 154 Session 4 - Task 1: Picture Match III "This task is like some of the ones we did before but we will only do two of them instead of four. Listen care fully." 1. "That is it I have to do? I must find the one picture on the bottom that exactly matches the one on top. First I must look carefully at top picture and then slowly look at each and every other picture before I answer." (M&M) "I stop, look, and think before I answer." (M&M) "It's not this one because....” (pointing) ...."This one looks right but I must look at all of them before I choose .... Yes, this is the best one." (M&M) "Good." (After problem two say, "Good, I finished and I earned six M&M's.") 2-5. (Same sequence as previous session.) Session 4 - Task 2: Trail-Making I "Now we're ready for our last task. We've done this one before so the words should be easier this time.” 1. "I stop, look, and think before I move.” (M&M) "What is it I have to do? I have to draw a path starting from number one and ending with number ten." (M&M) "One, two, three .... I must go very slowly .... four, five, six, seven, eight, nine, ten. Good, I finished," "and I earned three M&M*s." (Same sequence as previous session.) 156 PICTURE - MATCH III APPENDIX B FORMS AND DIAGRAMS OF EQUIPMENT 157 CLINIC INTAKE FORM 158 Los Angeles County - USC Medical Center Child/Adolescent Psychiatry INTAKE QUESTIONNAIRE DO NOT WRITE IN THIS SECTION Child's Name _________________ PF.# _________________________ Referral Source _______________ PSW __________________________ Date I. IDENTIFYING DATA ON CHILD A. Child's Name ____________________ Age Birthdate ____________ Sex ______ Race Address _____________________ Religion Home Phone School School District _________________ Grade II. IDENTIFYING DATA ON PARENTS, FOSTER PARENTS, STEPPARENTS OR GUARDIAN A. Name of person giving information _______ Relationship _____________________________ Mother's Name Maiden Name Age ______ Birthdate Race Address Home Phone Occupation __________________________________ Place of Employment Do you receive any public assistance? 159 Work Phone _________ Work Hours _________ Education conqpleted Religious Preference __________________ Number of marriages ________________________ Does mother have any significant medical problems at present? If yes, describe: ___________________ Has mother had any serious illnesses or accidents or surgeries in the past? ________________________ Have you ever been in psychiatric treatment? Yes _______ No______ B. Father's Name __________________________ Age ______ Birthdate ____________________ Race_________________ Address __________________ Home Phone ________ Occupation ___________________________________ Place of Employment ____________________________ Work Phone _____________ Work Hours ______________ Education completed ______________________________ Religious Preference ___________________________ Number of marriages _______________________________ Does father have any significant medical problems at present? If yes, describe: ___________________ 160 Has father had any serious illnesses, accidents, or surgeries in the past? _____________________ Have you ever been in psychiatric treatment? Yes ______ No_______ C. Stepparents or Guardian information. Name of Stepmother or Guardian (Specify) Age _____ Birthdate Race________ Address _______________________ Home Phone Occupation ______________________________ Place of Employment _____________________ Work Phone _________________ Work Hours _ Do you receive any public assistance? Education completed ________________________ Religious Preference ______________________ Number of marriages _____________ Do you have any significant medical problems at present? If yes, describe: ____________________ Have you had any serious illnesses or accidents or surgeries in the past? ______________________ 161 Have you ever been in psychiatric treatment? Yes ______ No_______ Name of Stepfather or Guardian (Specify) Age _________ Birthdate _________ Race____ Address _________________________ Home Phone Occupation _____________________________ Place of Employment ________________________ Work Phone __________ Work Hours __________ Do you receive any public assistance? Education completed _____________________ Religious Preference ______________________ Number of marriages _______________ Do you have any significant medical problems at present? If yes, describe: ____________________ Have you had any serious illnesses or accidents or surgeries in the past? __________________ Have you ever been in psychiatric treatment? Yes ______ No_______ D. Marital Status of Parents: Are parents presently together? _ _ _ _ _ _ _ _ 162 Separated? _________________ Divorced? ____________ Widowed? ___________________ Single? _______________ Date of present marriage Date of separation _________________________________ Date of divorce E. Family of Patient and Others Living in Household: List others living at above address (Parents, foster parents, brothers, sisters, grandparents, aunts, uncles, and non-relatives). Relationship Age Birthdate to child 1. __________________________ 2. _________________________ 3. _________________ _______________________________ 4. _________________ _______________________________ 5. _________________ _______________________________ 6. __________________________ 7. _________________ _______________________________ 8. ________________ 9. _________________ _______________________________ 10. __________________________ 11. __________________________ 12. __________________________ 13. ______ ____________________________________ Who has legal custody of the child? 163 If child is not with natural parents, when did separation occur and why? ___________________ III. SCHOOL AND AGENCY INFORMATION Name of teacher _________________________________ Does child have behavior problems in the school? Describe: ________________________________ ______ Does the child have learning problems in school? Describe: ________________________________ ______ If the child has ever been kept back or put ahead in school, explain: If the child has been in special classes, what were reasons? __________________________________ Since what age? ______ If the child has ever been excluded from school, explain when and why: _______________________ If the child is on probation, who is the Probation Officer (Name and Phone): ____ 164 Are any other agencies involved with the family (DPSS, Child Welfare, etc.)? ____________________ IV. MEDICAL HISTORY OF THE CHILD Has the child ever had any serious illness, accidents, or operations? _____________________ Please describe each incident and specify child's age (Include any present illnesses): Has the child ever had psychiatric treatment? Yes ______ No ______ V. HISTORY OF CHILD'S PROBLEM A. In your own words describe the child's present problem or condition as you see it: B. What methods were used in trying to help with these difficulties? Describe: 165 C. In what way do you think we can help you? VI. CHILD'S DEVELOPMENT HISTORY A. PERIOD DURING PREGNANCY Was the child planned? Sex preference? _______________ How did mother feel about having the child? Did the mother have any medical or emotional problems during pregnancy? (For example: convulsions, hemorrhages, infections, unusual nervousness) How did father feel about having this child? Sex preference ____________ Did the mother work during pregnancy? How long? _____________ 166 B. DETAILS OF DELIVERY, QUESTIONS ABOUT LABOR Were there any complications of labor and delivery? Please describe: _____________________ Did the mother have any "Blues" after baby's birth? C. POSTNATAL Weight of baby at birth? ____________ Was the baby full term (9 months) Yes ________ No _____ Were there any complications after the baby was born? (For example, difficulty breathing, baby cyanotic (blue), R.H. Factor, baby jaundiced?) Did the mother have any help in home after delivery? Yes _______ No If yes, how long? __________________________________ During the baby's first year of life, was there anything (even if it had nothing to do with the baby) that caused the mother unhappiness or anxiety or that placed her under special strain? Describe: _________________________________________ After baby's birth, how soon did mother return to work? _______________________ Was the child ever separated from both parents? Yes _____ No _____ One Parent? Yes No___ Describe the circumstances (reason, child's age at time, and how long?): _______________________ Did the father take an active part in the baby's care (such as changing diapers, bathing, feeding, etc.)? Yes ______ No_______ FEEDING Breast fed _______ How long _____ Bottle fed_ How long ______ Were there any feeding problems? (Colic, diarrhea, or food allergies) If so, explain: When was the child weaned? _____________ Why did the weaning occur at that time? How was child's discomfort handled? Any thumb-sucking? Describe: SLEEPING PATTERNS 1. Were there sleeping problems? Describe: _____________ Has the Yes child ever slept with No Describe the parents? circumstances: Present sleeping arrangements: MOTOR DEVELOPMENT Was your child ever too active or too quiet? Please Describe: _____________________________ At what age did your child start: Sitting Standing ____________ Walking___________ At what age was bowel training begun? ____ Completed? Method used: ________________________ ______ At what age was bladder training begun? Completed for day? __ _________ Method used: 169 Completed for night? Was your child's toilet training ever a problem? ________ Describe how:_______________________ Is this a problem at present? _____ Describe: Is the child primarily right handed? _________ Left handed? ______________ G. SPEECH DEVELOPMENT At what age did child first begin to speak in short sentences? ____________________________ If there have been any of the following speech difficulties, please check: Does not talk _____ Lisping _____ Delayed speech ____ Repeating Syllables _____ Mispronouncing words _____ Stuttering _______ Other, describe: _______________________________ Has the child ever had any speech therapy? H. SEXUAL DEVELOPMENT Has the child expressed curiosity about any 170 sexual matters to a parent? ______________________ About what? ___________________________________ Has the child been given information by a parent in any of the following areas? If yes, please check: The difference between boys and girls ____________ Birth Control ______ Menstruation _______________ How a woman becomes pregnant ______ Wet dreams _____ How the baby develops and is born _______ Intercourse ________________________ Masturbation ______ Other concerns of the parent: ____________________ PEERS AND INTERESTS Does your child have any difficulty making friends? _______ Describe: ____ Does he make friends primarily with children his own age ______ Children younger_______ Older children ______ Adults_______ Describe any special interests or hobbies: /------------------------------------ J. Do you have any questions, comments on the questionnaire, or additional information? 171 Best days and time for Clinic appointments: Mother Father 172 SOCIAL WORKER EVALUATION FORM Los Angeles County - USC Medical Center PSYCHIATRIC SERVICES FOR CHILDREN AND ADOLESCENTS Date Each Entry in This Column SUMMARY OF REFERRAL AND INTAKE Name of PSW: Date of Intake: Source of Referral: Presenting Problems and onset: (Include age, sex, grade, ethnicity, medical problems) Person seen for intake appointment: (Name and relationship to patient) Intake Recommendations: Date of Evaluation etc.: Reports Requested: (check) School Physician Agency Other Family Constellation: Marital status of parents: Parents availability for treatment: Availability for clinic appointments (days and hours) Name of last school (include city): Problem areas (please check one) (1) Emotionally Disturbed (2) Special Problems (3) Alcoholic (4) Drug Abuse (5) Family Crisis Name P.F.# PSYCHIATRY CLINIC RECORD INDEX: PSYCHIATRY CLINIC 173 PSYCHIATRIC EVALUATION FORM Los Angeles County - USC Medical Center PSYCHIATRY CLINIC RECORD Date Each Entry in This Column IDENTIFYING INFORMATION NAME: PF#: DATE OF REFERRAL SOURCE: BIRTH: RACE: RELIGION: SEX: TELEPHONE NUMBER: PARENTS NAME: SCHOOL: GRADE LEVEL: ADDRESS: FAMILY CONSTELLATION: PRESENTING PROBLEM: DEVELOPMENTAL DATA: MENTAL STATUS: FORMULATION AND DIAGNOSIS: RECOMMENDATIONS: SIGNATURE: Name P.P.# PSYCHIATRY CLINIC RECORD INDEX: PSYCHIATRY CLINIC 174 DISCHARGE SUMMARY FORM Loa Angeles County - USC Medical Center DISCHARGE SUMMARY - CHILD/ADOLESCENT OUTPATIENT CLINIC PSYCHIATRY CLINIC RECORD Date Each Entry in This Column I. II. III. IV. DATE OF DISCHARGE SUMMARY: PATIENT INFORMATION: Birthdate: Sex: Race/Ethnic Origin: TREATMENT: U T _ (2) Be (3) (4) Referral Source: ginning and~Endinq Date of Treatment: Frequency of Treatment ' 'or Number of (5) Sessions] Type of Services Provided: Admission Service or Hospital Care ) Evaluation, Diagnosis or Testing Only Individual Interview ( ) Group ) Conjoint Family Therapy ) Drug Therapy ( ) Application Only Teat Results (Lab or Psychological): ( ( ) ( ( (6) Presenting Problems and Outcome; (7) Reasons For Closing: FINAL DIAGNOSIS (DSM II CODE NO. AND RELEVANT MEDICAL CONDITION: RECOMMENDATION AT DISCHARGE: CONDITION AT DISCHARGE: ) Not Treated ) Treated and Not Improved ) Treated and Improved SIGNATURE MID TITLE: Name P.F.# INDEX: 175 INFORMED CONSENT FORM Los Angeles County University of Southern California Medical Center Title of Research Project; Department: Principal Investigator: Diagnostic Indices of Hyper activity and an Investigation of Verbal Mediation Training as an Alternative to Pharmaco therapy Psychology Norman Tiber, Ph.D. In order to determine the best treatment for your child, we would like to compare the effects of a verbal learning procedure with the effects of a drug called Ritalin. Both have been helpful in treating children with problems similar to your child's. The evaluation will involve six clinic visits of approximately one hour each during the next three weeks. During the first session, we will do some testing which re lates to his work in school. Then, during the next four sessions an instructor will practice certain tasks with your child, teaching him to think about the best way to perform the tasks. Your child will then be retested to see if there has been any change as a result of the treatment. Finally, during the last visit, the clinic nurse will ad minister the Ritalin to your child approximately one hour before he is tested again. Ritalin sometimes causes prob lems with sleep and appetite but the dose we will adminis ter is commonly prescribed for children. We hope that the treatments will improve your child's ability to learn and decrease his activity level. The results we obtain from this evaluation will not only help your child but will help us learn more about which treatments are most helpful. If you agree to this evalu ation for your child, sign where indicated. You may, of course, withdraw consent at any time. If you do not agree, your child will receive the usual treatment. We will be glad to supply any additional information you may wish. 176 I have read the above and give my consent for his partici pation. Signature Relationship Date Witness Date 177 SUBJECT RECORD SHEET Subject's Name Date Session # Group IQ Resting Activity Porteus Maze MFF DRL Operant PR-F Operant CUMULATED ACTIVITY SCORES Hyperactivity Study Linda Trozzolino Norman Tiber PERFORMANCE SCORES # errors # errors average latency in # on counter seconds # on counter LATENCY OF MFF RESPONSES IN SECONDS (maximum of six r's per problem) Problem # 1 _______ 2 ______ 3 ______ ______ ______ ______ ______ 4 ______ ______ ______ ______ ______ 5 ______ ______ ____ 6 178 Diagram 1.— Stabilimetric Chair. 179 FRONT 30 100 TMrr6*Ar»<t T/avca ihtsomnm* ri#** I 1 , V . a«. 9 »►* @ A.C. Roue I T iMrirr BACK *CCO*»fa I O-Joo M-V. o-a«®MI frttT *ta xNTftftm DATA I o (ku) data O a<ro«w4i. M«T«a G> O iwpur CDfcO ATTA6HFO To Diagram 2.— Activity Recorder, REFERENCES 180 181 Alabiso, F. Inhibitory functions of attention in reducing hyperactive behavior, Amer. J. Ment. Def., 1972, 77, 259-282. Ault, R., Crawford, D., and Jeffrey, W. Visual scanning strategies of reflective, impulsive, fast accurate, and slow inaccurate children on the Matching Familiar Figures Test. Child Development, 1972, 43, 1412-1417. Allen, K. H., Henke, L. R., Harris, F. A., Reynolds, N., and Baer, R. Control of hyperactivity through social reinforcement of attending behavior. Journal of Educational Psychology, 1967, 5£, 231-75TI Anderson, D B. Application of behavior modification techniques to the control of hyperactive child. Un published master thesis, University of Oregon, 1964. Banta, T. Tests for the evaluation of early childhood education: The Cincinnati Autonomy Test Battery (CATB). In J. Hellmuth (Ed.) Cognitive Studies, N. Y., Brunner-Mazel, 1970, 423-491. Barnes, K. Effects of methylphenidate and thioridazine on learning, reaction time, and activity level in hyperactive, and emotionally disturbed children. Unpublished Bachelors Thesis, University of Illinois, 1968. Bell, R., Waldrop, M., and Beller, G. A rating system for the assessment of hyperactive and withdrawn children in preschool samples. Unpublished manuscript, 1969. Bern, S. Verbal self-control: The establishment of effec tive self-instruction. J. Exper. Psychol., 1967, 74 (4) , 485-491. Bindra, D. Components of general activity and the analysis of behavior. Psychol. Rev., 1961, 68, 205- 215. Birch, D. Some effects of the verbal system on the non verbal behavior of pre-school children. Paper pre sented at the meeting of the Society for Research in Child Development, 1967. Blackridge, V., and Ekblad, R. The effectiveness of methylphenidate hydrochloride (ritalin) on learning and behavior in public school educable mentally re tarded children. Pediatrics, 1971, 47(5), 923-926. 182 Boydstun, J. A., Ackerman, P. T., Stevens, D. A., Clements, S. D., Peters, J. E., and Dykman, R. A. Physiologic and motor conditioning and generalization in children with minimal brain dysfunction. Conditional Reflex, 1968, 3, 81-104. Breitmeyer, J. Effects of thioridazine and methylpheni date on learning and retention in retardates. Un published Masters Thesis, University of Illinois, 1969. Broadbent, D. E. The approach to information processing through behavior. In Information Processing in the Nervous System, III, In Garard, R. W., and Duyff, J. W. (Eds.), Proc. Intl. Union Physiol. Sciences, Experta Medica Foundation, 1962. Bucky, S. F., Banta, T. J., and Gross, R. B. Development of motor impulse control and reflectivity. Perceptual and Motor Skills, 1972, 3£, 813-814. Burks, H. F. Effects of amphetamine therapy on hyper kinetic children. Arch. Gen. Psych., 1964, 11, 604-609. Burns, B. J. The effect of self-directed verbal commands on arithmetic performance and activity level of urban hyperactive children. Dissertation Abstracts International, 1972, 33(4-B), 1782-1783. Campbell, S. B., Douglas, V. I., and Morgenstern, G. Cognitive styles in hyperactive: children and the effect of methylphenidate. J. Child Psychol, and Psychi., 1971, 13, 55-67. Campbell, S. B. Cognitive styles in reflective, impul sive, and hyperactive boys and their mothers. Perceptual and Motor Skills, 1973, 36, 747-752. Christensen, D. E., and Sprague, R. L. Reduction of hyperactive behavior by conditioning procedures alone and combined with methylphenidate (Ritalin). Behavior Research and Therapy, 1973, 11, 331-334. Clements, S. D. Minimal Brain Dysfunction in Children. National Institute of Neurological biseases and Blindness Monograph No. 3, Washington, D.C.: U. S. Public Health Service, 1966. 183 Cohen, N. J., Douglas, V. I. and Morgenstern, 6. The effect of methylphenidate on attentive behavior and autonomic activity in hyperactive children. Psychopharmacologla, 22, 1971, 282-294. Conners, C. K. The effect of edxedrine on rapid dis crimination and motor control of hyperkinetic children under mild stress. J. Nerv. Ment. Dis., 142(5), 429-433, 1966. Conners, C. K. The use of stimulant drugs in enhancing performance and learning. In Smith, W. L., (Ed.) Symposium on Higher Cortical Function, Springfield, Conner8, C. K. Recent drug studies with hyperkinetic children. Journal of Learning Disabilities, 1971, 4, 476-483. Conners, C. K., and Eisenberg, L. The effects of methylphenidate on symptomatology and learning in disturbed children. American Journal of Psychiatry, 1963, 120, 458-464. Conners, C. K., and Rothschild, G. H. Drugs and learning in children. Learning Disorders, 1968, 3, 193-223. Conners, C. K., Eisenberg, L., Sharpe, L. Effects of methylphenidate (ritalin) on paired-associate learning and Porteus maze performance in emotionally disturbed children. J. Consulting Psychol., 1964, 28(1) , 14-22. Conners, C. K., Eisenberg, L., and Barcai, A. Affect of dextroamphetamine on children. Arch. Gen. Psychiat., 1967, 17, 478-485. Conners, C. K., Rothschild, G. H., Eisenberg, L., Schwartz, L., and Robinson, E. Dextroamphetamine sulfate in children with learning disorders: Effects on perception, learning, and achievment. Arch. Gen. Psychiat., 1969, 21, 182-190. Cromwell, R. L., Baumeister, A., and Hawkins, W. F. Research in activity level. In N. R. Ellis (Ed.) Handbook of Mental Deficiency, McGraw-Hill Co., N.Y., 184 Cutts, K. K., and Jasper, H. H. Effect of benzedrine sulfate and phenobarbital on behavior problem children with abnormal electroencephalograms. Arch. Neurol. Psychiat., 1939, 41, 1138-1145. Cytryn, L., Gilbert, A., and Eisenberg, L. The effective ness of tranquilizing drugs plus supportive psycho therapy in treating behavior disorders of children: A double-blind study of 80 outpatients. Amer. J. OrthopBychiat., 1960, 30, 113-129. Daniels, L. K. Parental treatment of hyperactivity in a child with ulcerative colitis. J. of Behavior Therag^ and Experimental Psychiatry, 1973, 4(2) , Davis, K. V., Sprague, R. L., and Werry, J. S. Stereo typed behavior and activity level in severe retardates: The effect of drugs. Amer. J. Ment. Defic., 1969, 75, 721-727. Debus, R. L. Effects of brief observation of model behavior on conceptual tempo of impulsive children. Devel. Psychol., 1970, 2, 22-32. Deutsch, J. D., and Deutsch, D. R. Attention: Some theoretical considerations. Psychol. Rev., 1963, 70, 80-90. Dews, P. B., and Morris, W. H. Behavioral pharmacology. Annual Review of Pharmacology, 1961, 1, 145-174. DiMascio, A., and Shader, R. E. (Eds.). Clinical Hand- book of Psychopharmacology. New YorJcl Science House, i970, 72-M. Douglas, V. I. Sustained attention and impulsive control: Implications for the handicapped child. In Sivets, J. A., and Elliott, L. L. (Eds.). Psychology and the Handicapped Child. Washington, D. C.: U. S. Office of Education, in press, 1974. Douglas, V. I., Werry, J. S., and Wteiss, G. Hyperactive behavior in children: Some findings regarding aetiology and treatment. Canadian Psychologist, 1965, 6, 219. 185 Douglas, V. I., Weiss, G., and Mlnde, K. Learning dis abilities in hyperactive children and the effect of methylphenidate. Paper read at Canadian Psycholo gical Association Annual Meeting, Toronto, 1969. Drake, D. M. Perceptual correlates of impulsive and reflective behavior. Devel. Psychol., 1970, 2, 202-214. Dubrose, S. C., and Daniels, G. T. An experimental approach to the reduction of overactive behavior. Behavior Research and Therapy, 1966, £, 251-258. Duffy, E. Activation and Behavior. New York: J. Wiley and Sons, 1962. Dykman, R. A., Walls, R. C., Suzuki, T., Ackerman, P. T., and Peters, J. E. Children with learning dis abilities: Conditioning, differentiation, and the effects of distraction. Amer. J. Ortho. Psychiat., 1970, 40, 766-782. Eisenberg, L. Behavioral manifestations of cerebral damage in childhood. In Birth, H. G. (Ed.), Brain Damage in Children - The Biological and Sociar Aspects. New York: Williams and Wilkins, Co., r m . Eisenberg, L. Psychopharmacology in childhood: A critique. In Miller, E. (Ed.). Foundations of Child Psychiatry. Oxford: Pergamon Press, 1968, 655-641. Eisenberg, L. The overactive child. Hospital Practice, September, 1973, 151-160. Eisenberg, L., Gilbert, A., Cytryn, L., and Moiling, P. The effectiveness of psychotherapy alone and in conjunction with perhpenazine or placebo in treatment of neurotic and hyperkinetic children. Amer. J. Psychiat., 1961, 117, 1088-1093. Epstein, L. C., Lasagna, L., Conners, C. K., and Rodriquez, A. Correlation of dextroamphetamine excretion and drug response in hyperkinetic children. J. Nerv. Ment. Pis., 1968, 146, 136-146. Fish, B. The "one child, one drug" myth of stimulants in hyperkinesis. Archives of General Psychiatry, 1971, 25, 193-203. 186 Flavell, J. H., Beach, D. R., and Chinsky, J. M. Spon taneous verbal rehearsal in a memory task as a function of age. Child Development, 1966, 37, 283-299. Freedman, D. X., Chairman. U. S. Department of Health, Education and Welfare. Office of Child Development Panel: Report of the Conference on the use of stimulant drugs in the treatment of behaviorally disturbed young school children. Psychopharmacology Bulletin, 1971, 1_, 23-29. Freibergs, V., Douglas, V. I., and Weiss, G. The effect of chlorpromazine on concept learning in hyper active children under two conditions of reinforcement Psychopharmacologia, 1968, 13^, 299-310. Freston, C. W. Verbal and Porteus Maze Performance of learning disabled children: Effects of methylpheni date and input organization. Dissertation Abstracts International, 1972, 32(11-A), 6246-6247. Gardner, W. I., Cromwell, R. L., and Foshee, J. G. Studies in activity level II: Effects of distal visual stimulation in organics, familials, hyper actives, and hypoactives. Araerl J. Ment. Defic., 1959, 63, 1028-1033. Ginn, S. A., and Hohman, L. B. The use of dextro amphetamine in severe behavior problems in children. Southern Med. J., 1953, 46, 1124-1127. Gratch, G. The use of private speech by Headstart and middle class pre-schoolers. Paper presented at the meeting of the Southwest Psychological Association, Arlington, Texas, 1966. Grindee, K. D. Operant conditioning of attending behavior in the classroom: A case study. In F. A. Benson (Ed.), Modifying deviant behaviors in various class room settings. Eugene, Oregon, Department of Special Education, College of Education, Monograph No. 1, 1970. Grindee, K. D. Operant conditioning of "attending behaviors" in the classroom for two hyperactive Negro children. In F. A. Benson (Ed.), Modifying deviant behaviors in various classroom settings. Eugene, Oregon, Department of Special Education, College of Education, Monograph No. 1, 1970. 187 Hays, W. L. Statistics. Holt, Rinehart and Winston, N. Y., 1573T Howell, M. C., Rever, G. W., Scholl, M. L., Trowbrige, F., and Rutledge, A. Hyperactivity in children. Types diagnosis, drug therapy, approaches to management. Clinical Pediatrics, 1972, 1, 30-39. Jensen, A. Learning ability in retarded, average, and gifted children. Merrill Palmer Quartering, 1963, 9J 123-140. Kagan, J. Reflection-impulsivity and reading ability in primary grade children. Child Development, 1965, 36, 609-628. Kagan, J. Reflection-impulsivity; the generality and dynamics of conceptual tempo. J. Abn. Psychol., 1966, 71(1), 17-24. Kagan, J., Pearson, L., and Welch, L. The modifiability of an impulsive tempo. J. Ed. Psychol., 1966 B, 57(6), 359-365. Kagan, J., Pearson, L., and Welch, L. Conceptual impul- sivity and inductive reasoning. Child Development, 1966, 37, 583-594. Kagan, J., Rosman, B. L., Day, D., Albert, J., and Phillips, W. Information processing in the child; significance of analytic and reflective attitudes. Psychol. Monographs, 1964, 78(1), Whole No. 578. Keeney, T. J., Canizzo, S. R., and Flavell, J. H. Spon taneous and induced verbal rehearsal in a recall task. Child Development, 1967, 38, 953-966. Kennedy, D. A., and Thompson, I. J. Use of reinforcement techniques with a first grade boy. Personnel and Guidance Journal, 1967, 4£, 366-370. Klein, W. L. An investigation of the spontaneous speech of children during problem solving. Unpublished doctoral dissertation. University of Rochester, N.Y., 1963. Knights, R. N., and Hinton, G. The effects of methylphen idate (ritalin) on the motor skills and behavior of children with learning problems. J. Nerv. Ment. Dis., 1969, 148, 643-53. 188 Knowles, P. K., and Prutsman, T. D. Behavior modification of simple hyperkinetic behavior and letter discrim ination in a hyperactive child. Journal of School Psychology, 1958, £, 157-160. Kohlberg, L., Yaeger, J., and Hjertholm, E. Private speech: Four studies and a review of theories. Child Development, 1968, 31), 691-736. Lindsley, D. B. Nonspecific reticulo-thalamo-cortical system in emotion. In Black, P. (Ed.) Physiological Correlates of Emotion. New York: Academic Press, 1970. Lovaas, O. I. Interaction between verbal and non-verbal behavior. Child Development, 1961, 3£, 329-336. Lovaas, 0. I. Cue properties of words: The control of operant responding by rate and content of verbal operants. Child Development, 1964, 35^, 245-256. Luria, A. Verbal regulation of behavior. In M. Brazier, The Central Nervous System and Behavior. Josiah Macy, Jr., Foundation, N.Y., 1960. Luria, A. The Role of Speech in the Regulation of Normal and Abnormal Behavior., Liveright Publishing Corpo ration, N.Y., 19(>1. Maccoby, E. E., Dowley, E. M., Hagen, J. W., and Degerman, R. Activity level and intellectual functioning in normal pre-school children. Child Development, 1965, 761-770. Malmo, R. B. Activation: A neuropsychological dimension. Psychol. Rev., 1959, 6£, 367-386. Marwit, S. J., and Stenner, A. J. Hyperkinesis: Deli neation of two patterns. Exceptional Children, 1972, 38, 401-406. McConnell, T. R., Cromwell, R. L., Bialer, I., and Son, C. D. Studies in activity level VIII: Effects of amphetamine drug administration on the activity level of retarded children. Amer. J. Ment. Defic., 1964. 68, 647-665. McFarland, J. N., Peacock, L. J., and Watson, J. A. Men tal retardation and activity level in rats and children. Amer. J. Ment. Defic., 1966, 71, 376-380. 189 Meichenbaum, D., and Goodman, J. The developmental con trol of operant motor responding by verbal operants. J. Exper. Child Psychol., 1969, 7, 553-565. Meichenbaum, D., and Goodman, J. Reflection-impulsivity and verbal control of motor behavior. Child Development, 1969 B, 40^ 785-797. Meichenbaum, D. H., and Goodman, J. Training impulsive children to talk to themselves: A means of develop ing self-control. J. Abn. Psychol., 1971, 77(2), 115-126. Millichap, J. G. Management of hyperkinetic behavior in children with epilepsy. Modern Treatment, 1969, 6, 1233-1246. Millichap, J. G., and Boldrey, E. E. Studies in hyper kinetic behavior II. Laboratory and clinical evaluation of drug treatments. Neurol., 1967, 17, 467-472. Millichap, J. G., and Johnson, F. H. Methylphenidate in hyperkinetic behavior: relation of response to degree of activity and brain damage. Pediatrics Supplement, in press, 1974. Millichap, J. G., Aymat, F., Sturgis, L. H., Larsen, K.W., and Egan, R. A. Hyperkinetic behavior and learning disorders. Amer. J. Dis. Child., 1968, 116, 235- 255. Morgan, C. T. and Stellar, E. Physiological Psychology. New York, McGraw-Hill, 1950. Nixon, S. J. Ways by which overly active students can be taught to concentrate on study activity. Cooperative Research Project No. 5-379, Office of Health, Edu cation and Welfare, 1966. Paine, R. S. Minimal chronic brain syndromes in children. Develop. Med. Child Neur., 1962, £, 21-27. Palkes, H., Stewart, M., and Kahana, B. Porteus maze performance of hyperactive boys after training in self-directed verbal commands. Child Development, 1968, 39, 817-826. 190 Patterson, G. R. An application of conditioning tech niques to the control of a hyperactive child. In L. N. Uliman and L. H. Krasner (Eds.), Case studies in behavior modification. New York: Holt, Rinehart and Winston, 1965. Patterson, G. R., Jones, R. W., Whittier, J. E., and Wright, M. A. A behavior modification technique for the hyperactive child. Behavior Research and Ther apy, 1965, 2, 217-226. Piage' ~ “ - - - ■ ■ “* * of the Child, Pihl, R. F. Conditioning procedures with hyperactive children. Neurology, 1967, £7, 421-423. Pope, L. Motor activity in brain injured children. Amer. J. Orthopsychiat., 1970, £0(5), 783-794. Porteus, S. Porteus Maze Tests; Fifty Years Application. Palo Alto! Pacific Books, 1965. Post, G., and Tiber, N. "The differentiation of Two Types of Hyperactive Children By Activity Level and Task Performance with Methylphenidate (Ritalin) and Placebo." Unpublished Doctoral Dissertation, U.S.C., Quay, H. F., Spague, R. D., Werry, J. S., and McQueen, M. W. Conditioning visual orientation to conduct problem children in the classroom. Journal of The Exceptional Child, 1967, 5, 512-517. Ravens, J. C. Standard Progressive Matrices, (Sets A, B, C, D and E), H. K. Lewis and Co., London, 1956. Reese, H. Verbal mediation as a function of age level. Psychol. Bull., 1962, 52# 502-509. Reitan, R. M. Trail making test. J. Consul. Psychol., 1955, 19' 393-4. Rosen, E. J. Behavioral and emotional disturbances asso ciated with cerebral dysfunction. Applied Thera- peutics, 1969, 11, 531-543. Rosenbloom, L. Learning disabilities and hyperkinesis. Devel. Med. Child Neur., 1972, 14, 394-395. 1974 191 Safer, D., Allen, R., and Barr, E. Depression of growth in hyperactive children on stimulant drugs. New England Journal of Medicine, 1972, 287, 217-226. Santostefano, S., and Paley, E. Development of cognitive controls in children. J. Clin. Psychol., 20, 213- 218. Satterfield, J. H., and Dawson, M. E. Electrodermal correlates of hyperactivity in children. Psycho- physiology, 1971, 8, 191-197. Satterfield, J. H., Cantwell, D. P., Lesser, L. I., and Podosin, R. L. Physiological studies of the hyper kinetic child: I. American Journal of Psychiatry, 1972, 128, 103-109. Satterfield, J. H., Lesser, L. I., Saul, R. E., and Cantwell, D. P. EEG Aspects in the diagnosis and treatment of minimal brain dysfunction. Annals of the New York Academy of Sciences, 1973, 205, 274- t s t . --------------------------- — Schickedanze, D. I. Effects of thioridazine and methyl- phenidate on performance of a motor task and con current motor activity in retarded boys. Unpub lished Bachelors Thesis, University of Illinois, 1967. Siegelman, E. Reflective and impulsive observing behav ior. Child Development, 1969, 4£, 1213-1222. Sprague, R. L., and Toppe, L. K. Relationship between activity level and delay of reinforcement in the retarded. J. Exper. Child Psychol., 1966, 3, 390- 397. Sprague, R. L., and Boileau, R. A. Are Drugs Safe? In Sprague, R. L., Chairman. Psychopharmacology of children with learning disabilities. Symposium presented at the American Psychological Association Montreal, Canada, August, 1973. Sprague, R. L. Detailed Progress Report of Grant MH 18909 for period from September 1, 1970 to March 31, 1973. Unpublished Manuscript, University of Illinois, 1973. 192 Sprague, R. L., Werry, J. S., and Scott, K. G. Dosgge effects of methylphenidate on activity level and learning in children. Paper presented At the meeting of the Psychonomic Society, St. Louis, November, 1969. Sprague, R. L., Barnes, K. R., and Werry, J. S. Methyl phenidate and thioridazine: Learning, reaction time, activity and classroom behavior in disturbed children. Amer. J. Orthopsychiat., 1970, 40, 615- 628. Sprague, R. L., Sleator, E. K., McNutt, B., Cohen, M., and von Neumann, A. Dosage effects of methylpheni date. In Sprague, R. L., and Werry, J. S. Detailed Progress Report of Grant MH 18909 for period from September 1, 1970 to March 31, 1973. Unpublished Manuscript, University of Illinois, 1973 A, 120- 146. Sprague, R. L., Sleator, E. K., von Neumann, A., Cohen, M., and McNutt, B. A direct comparison of dextro amphetamine and methylphenidate. In Sprague, R. L., and Werry, J. S. Detailed Progress Report of Grant MH 18909 for period from September 1, 1970 to March 31, 1973. Unpublished Manuscript, Univer sity of Illinois, 1973 B, 156-172. Sprague, R. L., and Werry, J. S. Psychotropic drugs and handicapped children. Unpublished manuscript; to be published in the Second Review of Special Education, 1974. Stevens, D. A., Sachdev, K., and Milstein, V. Behavior disorders of childhood and the electorencephalogram. Arch. Neur., 1968 A, 18, 160-177. Stevens, D. A., Boydstun, J. A., Ackerman, P. T. and Dykman, R. A. Reaction time, impulsivity and auto nomic lability in children with minimal brain dys function. Proceedings of the 76th Annual Convention of the American Psychological Association, 1968 B, 367-368. Stevens, D. A., Boydstun, J. A., Dykman, R. A. Peters, J. E., and Senton, D. W. Presumed minimal brain dys function in children. Arch. Gen. Psychiat., 1967, 16, 281-285. 193 Stevens, D. A., Stover, C. E., and Backus, J. T. The hyperkinetic child: Effect of incentives on the speed of rapid tapping. Journal of Consulting and Clinical Psychology, 1970, 34, 56-571 Sulzbacher, S. I. Psychotropic medication with children: An Evaluation of procedural biases in results of reported studies. Pediatrics, 1973, 51(3), 513- 517. Sykes, D. A., Douglas, V. I., Weiss, G., and Minde, K. K. Attention in hyperactive children and the effect of methylphenidate (ritalin). J. Child Psychol. Psy- chiat., 1971, 12, 129-139. Tiber, N. and Trozzolino, L. Arousal level and hyper activity. Unpublished Masters Thesis, University of Southern California, 1972. Thurstone, J. Primary Mental Abilities Test, Science Research Associates, Inc., 1962. Vygotsky, L. S. Thought and Language. New York: Wiley and Sons, 1962. Walker, H. A., and Buckley, N. S. The use of positive reinforcement in conditioning attending behavior. Journal of Applied Behavior Analysis, 1968, 1, 245- 7W. Weir, M. W. and Stevenson, H. W. The effect of verbali zation in children's learning as a function of chronological age. Child Development, 1959, 30, 143-149. Weiss, G., Werry, J. S., Minde, K., Douglas, V. and Sykes, D. Studies on the hyperactive child V. J. Child Psychol. Psychiat., 1968, 9_, 145-156. Weiss, G., Minde, K., Werry, J. S., Douglas, V. I., and Sykes, D. A comparison of the effects of chlor- promazine, dextroamphetamine, and methylphenidate on the behavior and intellectual function of hyper active children. Canadian Medical Association Journal, 1971, 104, 20-25. Wender, P. H. Minimal Brain Dysfunction in Children. Wiley-Interscience: N.Y., 1971. 194 Werry, J. S. The diagnosis, etiology, and treatment of hyperactivity in children. Learning Disorders, Vol. #3, Special Child Publications. Seattle, Wash., 1966 A, 173-19oT Werry, J. S. Some clinical and laboratory studies of psychotropic drugs in children: An overview. In W. L. Smith (Ed.), Symposium on Higher Cortical Function. Springfield, Illinois: Charles T. Thomas, 1670. Werry, M. B., Weiss, G., Douglas, V. Studies on the hyperactive child I: Some preliminary findings. Canad. Psychiat. Assoc. J., 1964, 9^(2), 120-130. Werry, J. S., and Sprague, R. L. Hyperactivity. In Costello, C. (Ed.). Symptoms of Psychopathology, New York: John Wiley and Sons, 1970, 146-157. Yando, R. M. and Kagan, J. The effect of teacher tempo on the child. Child Development, 1968, 39, 27-34. Seaman, D., and House, B. J. The role of attention in retardate discrimination learning. In N. R. Ellis (Ed.), Handbook of Mental Deficiency, 1963, 159-223. Zrull, J. P., Westman, J. C., Arthur, B., and Rice, D. L. A comparison of diazepan, d-ampheramine, and placebo in the treatment of hyperkinetic syndromes in children. Amer. J. Psychiat., 1964, 121, 388-389.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
A Behavioral Assessment Of The Reinforcement Contingencies Associated With The Occurrence Of Suicidal Behaviors
PDF
Death Anxiety In Leukemic Children
PDF
Weight Reduction As A Function Of The Timing Of Reinforcement In A Covertaversive Conditioning Paradigm
PDF
Role Playing Ability And Social Adjustment In Children
PDF
Aggression As A Function Of Arousal And Friendship Ties
PDF
The Effect Of Personalized Emotional Stimuli On Asthmatic Reactions
PDF
Adolescent drug attitudes: a seven-year study on marihuana and LSD
PDF
The Psychology Of Self-Actualization Involving The Communication Of Therapeutic Counseling Skills Between Beginning And Experienced Counselors
PDF
Relationship Between Mourner Characteristics And Factors Affecting Grief Work
PDF
Attention, retention, and incentive processes in observational learning
PDF
Sex Differences In The Organization Of Spatial Abilities In Older Men And Women
PDF
The Effects Of Repression-Sensitization, Race, And Levels Of Threat On Extensions Of Personal Space
PDF
Affective Information And Affect-Cognition Interaction: Mood And Emotion
PDF
The Development And Evaluation Of Three Therapeutic Group Interventions For Widows
PDF
A Temporal Approach-Avoidance Conflict In An Academic Test Situation
PDF
The effects of a behavior modification treatment program on a neurologically handicapped population
PDF
Experimenter Expectancy Effect Examined As A Function Of Task Ambiguity And Internal Versus External Control Of Reinforcement
PDF
Dementia And The Perception Of Experimental Pain
PDF
The Effect Of Conditions Of Risk, Internal Versus External Control Of Reinforcement, And Sex On Binary Choice Probability Learning
PDF
The effects of physical proximity and body boundary size on the self-disclosure interview
Asset Metadata
Creator
Trozzolino, Linda Ann
(author)
Core Title
Diagnostic indices of hyperactivity and an investigation of verbal mediation training as an alternative to pharmacotherapy
Degree
Doctor of Philosophy
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Psychology, clinical
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Frankel, Andrew Steven (
committee chair
), Kahan, James Paul (
committee member
), Russak, Sidney (
committee member
), Tiber, Norman (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c20-539770
Unique identifier
UC11225886
Identifier
7515588.pdf (filename),usctheses-c20-539770 (legacy record id)
Legacy Identifier
7515588.pdf
Dmrecord
539770
Document Type
Dissertation
Rights
Trozzolino, Linda Ann
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA