University of Southern California Dissertations and Theses

Psychological Test Changes In Schizophrenic Patients Under Brief Stimuluselectroconvulsive Therapy

(USC Thesis Other)

Psychological Test Changes In Schizophrenic Patients Under Brief Stimuluselectroconvulsive Therapy

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content PSYCHOLOGICAL TEST CHANGES IN SCHIZOPHRENIC PATIENTS UNDER BRIEF STIMULUS ELECTROCONVULSIVE THERAPY by James Henderson Sharp 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) August 1959 UNIVERSITY OF SO UTHER N CALIFORNIA GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES 7, CALIFORNIA This dissertation, written by ........ Jame.a..He&<aer.3Qn..Siiarj3........ under the direction of /ELS...Dissertation C o m mittee, and a ppro ved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R O F P H I L O S O P H Y ..............■ < £ ? ; • Dean" Date J . u n . e . . . 9 . , . . . .1959......... DISSERTATION COMMITTEE TABLE OF CONTENTS Chapter I. The Problem .......................... History of Electroconvulsive Therapy Brief Stimulus Therapy Comparison of Brief Stimulus and Conventional Therapies Previous Experimental Studies The Present Study II. Method ................................. Subjects Testing Instruments III. Results ............................... Hypothesis A Hypothesis B Hypothesis C Corollary Hypothesis IV. Discussion ............................ Interpretation of Results Limitations of the Study Suggested Improvements Suggestions for Further Research V. Summary and Conclusions ............. Problem Method Results Conclusions APPENDIXES Appendix A: .Difference Scores, Experimental Group Appendix B: Difference Scores, Control Group . . Appendix C: Raw Scores, Experimental Group . . . Appendix D: Raw Scores, Control Group ........... Appendix E: Sample Mosaics .................... . Page 1 49 68 86 136 147 152 157 164 171 ii iii Chapter Page Appendix F: Illustration of the Use of the Peak Rating Scale in Evaluating Mosaic Designs............................ 178 Appendix G: Instructions to the Raters in the Use of the Peak Rating S c a l e .......... 180 Appendix H: Instructions for the Lowenfeld Mosaic Test........................ 183 Appendix I: Illustration of the Use of the Rorschach Prognostic Rating Scale . 187 REFERENCES............................................ 190 LIST OF TABLES Table Page 1. Significance of Differences in Wechsler- Bellevue Scores ............................... 70 2. Significance of Differences in Prognostic Rating Scale Scores .......................... J6 3. Significance of Differences in Mosaic Rating Scale Scores................................... 81 4. Significance of Differences in Measures of Deterioration ................................. 84 iv LIST OF FIGURES Figure Page 1. Current Wave Forms Used in E C T .................. 9 2. Pre-treatment Mosaic of Patient A, an Experimental Subject .......................... 171 3. Post-treatment Mosaic of Patient A ............ 172 4. Pre-treatment Mosaic of Patient B, an Experimental Subject .......................... 173 5. Post-treatment Mosaic of Patient B.............. 174 6. Initial Mosaic Production of Patient C, a Control Subject ............................... 175 7. Second or Retest Production of Patient C, a Control Subject ............................... 176 v CHAPTER I The Problem Few if any therapeutic procedures in the long history of men's efforts to deal with the bewildering problems pre sented by mental or emotional illness have aroused the con tinuing controversy that still surrounds the use of the electroconvulsive therapies. Few, indeed, have been in use so long as discrete treatment forms with clearly identifi able elements persisting into the present. Despite the so phistication in contemporary treatment techniques, the basic concept of electric shock therapy is the same now as it was in its crude, eighteenth century beginnings. Then as now it was hoped that the application of electric currents to the brain and the production of clonic-tonic, grand mal type seizures would result in desirable changes in the emotional make-up of the treated individual. Pursuant to the attainment of these goals a vast liter ature has grown up, comforting in its volume but deeply dis quieting in its contentiousness and in the equivocal and often directly contradictory nature of the results and con clusions presented. It would seem that virtually the entire spectrum of criticism that can be leveled at medical and psychological research applies to one part or another of this body of work. In the writer's opinion good and suffi cient reasons can be found in the majority of cases for the 1 2 failure to obtain clear and consistent results. Appropriate criticisms would seem to range from basic defects in design to procedural faults, such as errors in the administration and scoring of tests. Other factors that loom large as sources of error in many of these studies are samples of in sufficient size, inadvertently biased samples, improper or unjustified use of certain statistical procedures, failure to obtain reasonably homogeneous samples, failure to provide adequate control groups where their use is clearly indi cated, absence of necessary measures to standardize treat ment procedures, and the use of inadequately validated test ing devices. There has been a heavy over-emphasis, particularly in the studies done by psychologists, on the presumed damaging effects of electroconvulsive therapy to the relative exclu sion of other parameters of major importance. The present study is concerned with the problem of brain damage result ing from shock treatment but only to the extent of consider ing it secondary to certain observed facilitating effects of this form of treatment upon intellectual functioning, upon ego-integrative capacity, and upon effective-cognitive inte gration as related to the mastery of reality situations. It is believed that the present study may serve to direct at tention to the essential qualities and characteristics of this treatment method about which we really want to know. Hopefully, it may aid us in approaching an answer to the two 3 basic questions that must ultimately be answered: Does electroconvulsive therapy cure or relieve certain kinds of emotional illness or distress? If so, is it then worth the risks Involved in its application? The present study is concerned with four defined as pects of a particular form of electroconvulsive therapy used in the treatment of schizophrenics. It is hoped that the results will add something of significance to our knowledge of the efficacy of this form of treatment in these pre scribed areas, not in the sense of proclaiming dramatic new findings or otherwise pleading the cause of electroshock therapy, but rather to set forth some relatively modest changes occurring in psychological test performances presum ably resulting from a specified treatment administered in a sufficiently well-controlled experimental design as to lend reasonable confidence in their validity. History of Electroconvulsive Therapy Though the deliberate production of artificially in duced grand mal convulsive phenomena with a therapeutic goal in view is generally regarded as a strictly modern treatment concept, its first known use occurred circa 1740 in France and the first formal reports on the earliest known work in this field were published in 1744 in the Histolre de L'aca- demie Royale des Sciences. Every two years thereafter a section, "Electricity and Medicine," published near- miraculous "cures" reported from all over Europe by enthusi astic and audacious experimenters, many of them lay persons. They seem to have been concerned mainly with cases of paral ysis and epilepsy. In the same journal in 1755, the French man, Dr. J. B. Le Roy, describes the treatment of what may be called hysterical blindness with sufficient clarity and detail as to leave no doubt that a true convulsive episode occurred in the last of a series of three shocks and was followed by remission of the presenting symptom. In the first known textbook on psychosomatic medicine published in 1797 in Germany, its author, F. C. G. Scheidemantel, devoted two chapters to a discussion of various types of physical and psychological shocks as a means of cure in a wide varie ty of emotional afflictions. In 1801, the German electro therapist, Friedrich L. Augustin, reported a case almost identical to that of Le Roy. During the first half of the nineteenth century the electroconvulsive technique was Increasingly widely adopted by European physicians so that by 1850 the topic was fre quently and extensively discussed in all major medical pub lications. This period of development saw the first dawn- ings of sophistication in the field in the change from a completely undifferentiated use of ECT to the establishment of differentiations between galvanic (unidirectional) and faradic (alternating) electricity, between various current intensities, duration of application, number and spacing of 5 treatments, and the like. By 1874 this work had provided sufficient data for the Austrian neurologist Moriz Benedikt to write two major works, giving specific prescriptions in terms of the then established differentiations in treatment for all known neuro- and psychopathology. A year later the Bavarian medical writer Johan B. Ullersperger (1875) re ceived an academy prize for a synoptic description of elec trotherapy. The Frenchman, G. B. C. Duchenne (l855)> who is sometimes referred to as the father of modern electrotherapy and who wrote an entire set of works on various aspects of the subject, stated unequivocally that no competent neurolo gist could practice without using electrotherapy. Perhaps the most important of all contributors during this period was the German psychiatrist Rudolf Arndt. In 1869 he pub lished a highly regarded study in which he attempted to es tablish fundamental theoretical concepts underlying the psy chological and neurological effects of electrotherapy. Though these concepts were based upon outmoded physiological and anatomical knowledge it is generally conceded that the basic method of approach employed in his study is still a valid one in today's research. The concept of shock as a potential curative agency in nervous and emotional disorders has always been a general one unrelated to any particular form or method of producing convulsion. Apparently the focus of attention from the out set has been upon the phenomenon of the convulsion itself, whether idiopathically determined or externally contrived, whether casual or purposeful, whether physical or psychic stimulation were the means employed. The occurrence of a grand mal seizure has always been regarded as the sine qua non in any therapeutically oriented shock technique, a view which is strongly supported by the most modern research. In all probability, pharmacological production of con vulsive seizures for therapeutic purposes predates the use of electroconvulsive methods starting in the mid-eighteenth century soon after the discovery of electricity itself. However, the first reported use of a chemical substance for this purpose was in 1798 by the German physician Weickhardt (Sargant and Slater, 195^) who recommended the use of cam phor to the point of producing vertigo and seizures and his method was widely accepted by other physicians of his day. The treatment was revived briefly by Meduna in 1933 but was quickly followed by his introduction two years later of the more effective drug, metrazol (Meduna, 1935)- This has also largely fallen into disuse in modern practice, supplanted by more efficacious and less toxic drugs, such as cardiazol, triazol, picrotoxin, and especially insulin. The develop ment of pharmacological methods has closely paralleled that of electroshock throughout, determined in good part by the commonality of purpose underlying their use. In recent years there has been a growing tendency to use them as al ternative treatments or in association with each other as more specific indications of their discriminated effective ness in different syndromes have been established. It is toward this more discriminated and conservative use of all the shock therapies that contemporary research effort is be ing directed. In view of the wide use of electrotherapy in Europe during the eighteenth and nineteenth centuries and the large volume of literature which appeared on the subject, it is surprising that American psychiatric literature is so com pletely lacking in any mention of these methods until the middle of the present century. It was not until 19^-0, fol lowing the appearance of a paper by Drs. Ugo Cerletti and L. Bini (1938)j that any real interest in electroshock was dem onstrated and its invention is generally ascribed to these far from pioneer workers in the field. This very brief re port originated in the clinic of which Cerletti is director, and detailed an experiment done by Bini who is introduced as an expert electro-technician. It contains an account of the successful treatment by means of a high voltage shock of an epileptic male and was the first of many conducted by this clinic mainly devoted to epilepsy. Following these early trials with epileptic patients, Cerletti and Bini, as well as numerous other European physicians, turned their atten tion to the use of electroshock techniques in the psychoses and psychoneuroses. The publications of Cerletti and Bini apparently aroused the dormant interest of American physicians and stimulated them to the intensive and fruitful research efforts which were soon to place this country in a position of unquestioned supremacy in the field of convul sive therapy. What has come to be called the "classical" technique of Cerletti and Bini, though with many modern refinements, is still in wide use today despite the existence of a large and growing body of evidence which seems to indicate the superi ority of the most recent departure in the field of convul sive therapy, the brief stimulus methods. This technique employs a 50 or 60 cycle alternating current (see Pig. 1A) with a stimulation time of 0.15 to 0.50 second which is con trolled by a preset timer. The strength of the stimulus em ployed varies between 70 and 135 volts or 400 to 1200 ma. The electrode placements used are bitemporofrontal combina tions. With this electrical wave form the above values have been found to be minimal ones for the production of a grand mal seizure without occasioning the deeply disturbing and traumatic "shock mancato" or missed shock. At least one contemporary worker, Delmas-Marsalet (1942), still advocates the use of a high voltage continuous current rather than the alternating current described above, though the modern consensus is that much greater danger of irreversible damage to the central nervous is incurred with the use of such currents (Kalinowsky and Hoch, 1946). The glissando onset or approach in which the current is built up A fX/X/X/Xy B J 4 - / 6 Y 'EAh I PEAK CwRREMT AVERA6S current L Some > C S ^ . P u ls e i n t e r v a i ? ^ ^ D u r a t i o n D iuimuifuuinjimumji E AAAAAAAAAAAAAAA. F j u i m u i IUUL G fi i J i J W l f l l L i f M 25 Fig. 1. Current wave forms used in ECT. A. Conven tional alternating current. B. Square waves of long dura tion and low frequency. C. "Courant chevauchant." D. Square waves of brief duration and high frequency. E. Brief stimuli of different form. F. Square waves showing second ary interruption. G. Brief stimuli with "glissando" onset. H. Brief stimuli with glissando onset and secondary interrup tion of approximate intensity and duration used in this study. gradually to its peak potential (see Fig. 1G) with a conse quent lessening of the danger of compression fractures and cardiovascular and respiratory difficulties has become standard operating procedure in conventional EOT. Various non-convulsive electro-stimulation techniques have been used adjunctively with ECT including electronarcosis, in which a grand mal seizure is induced conventionally followed by five to ten minutes of electrically maintained unconsciousness with continuing strong flexor tone and massive autonomic ac tivity; nonconvulsive transcerebral electrostimulation em ploying currents below the convulsive rheobase at all times with the patient under anesthesia; and counter shock, in which the convulsive episode is followed by a period of very light stimulation. Nearly every conceivable permutation and combination of current strength, frequency and duration, number and frequency of treatments, electrode placement, and association with various ancillary or modifying techniques have been used with generally conflicting reports of success and failure. Numerous modifying or "convulsion-softening" drugs such as curare, succinylcholine, thiamylal sodium, syncurin, intocostrin and many others, as well as many anxi ety-reducing and anesthetic agents continue to find wide use. The range of patients successfully treated with ECT has been extended to include children, the aged, and pa tients suffering from a wide variety of cardio-vascular, or thopedic, respiratory and other difficulties formerly 11 thought to be contraindicative of shock treatment. Perhaps of greater importance than all the technical advances has been the growing discrimination and sophistica tion in the use of the convulsive therapies and in the in creasing awareness of difficulties and dangers in their use as well as apparently favorable results in large numbers of cases. The remarks of Parsons, Bonnet, Keating, and Lawrence (1955) in their recent survey of the field of shock therapy provide us with an aptly stated consensus of contemporary thought on this subject: "Complications" resulting from the use of EST . . . no longer find adequate defense in modern treatment. Judi ciously applied to selected cases for particular purposes, EST has reached a state of therapeutic acceptance. Its application as a panacea for all psychotic syndromes in a "routine" manner is injudicious and may be hazardous. Modern EST implies the utilization of both technical skills and clinical acumen in making the treatment less hazardous than the disease. EST is a valuable tool; it can be a dangerous weapon. (p. 219) Brief Stimulus Therapy General Description The first to question the wisdom of the direct use of domestic alternating current as employed in the Cerletti and Bini "classical" technique of electroconvulsive therapy was Delmas-Marsalet (1942) and then, independently and almost simultaneously, Friedman and Wilcox (1942) who considered that the main disadvantage of these currents was the di phasic oscillation of the electrical potential. To correct the deleterious side-effects they believed ascribable to the basic nature of the alternating current they advocated the use of a rectified pulsating or "unidirectional" current. They did not, however, submit their theoretical considera tions to a comparative physiological analysis. The funda mental problem to be resolved was whether or not the conven tional alternating current supplied much more than an ade quate stimulus for the production of the seizure, or whether or not a more appropriate and therefore substantially small er current might induce the desired phenomenon with a re sultant smaller risk of irreversible tissue damage. Impor tant experimental studies aimed at the resolution of these and related problems were done by Hemphill and Walter (1941) and Toman and Goodman (1947) who studied the relationship between the neural threshold for 60-cycle current and the total time of its application, generally referred to as "treatment time." Liberson (1944) submitted different time factors in electrostimulation (pulse duration, pulse fre quency, treatment duration) to various methods of physiolog ical analysis and abandoned any further empirical attempts at reforming domestic current for electrotherapeutic pur poses. Offner (1946) shortly thereafter contributed a basi cally important theoretical study of the form of the pulsat ing current. The complex phenomena of summation in the ner vous structures of the brain and the differentiation between "self-sustained" or "self-propagated" reactions and the di rect effects of electric stimulation were studied by 13 Rosenbleuth and Cannon (1942), and others. It has been well established (Liberson, 1953) that the fundamentally important parameters of stimulation are the following: 1. Individual pulse of the stimulating current: its peak intensity, its duration, its form, and its uni- or bi directionality . 2. Frequency of repetition of the individual stimulus, with or without secondary interruption of the stimulating current. 3. The total duration of the stimulation, known as the "treatment time." As regards the first of these dimensions, a series of basic research papers by Liberson (1944, 1948a, 1948b, 1945) appear to have substantially corrected the serious deficien cy in exploration in the area of adequate stimulation and summative effects and to have filled out the fragmentary and often contradictory data obtained prior to that time so that fairly precise conclusions could be reached. The studies suggest that optimal conditions of stimulation as regards pulse duration are realized below 0.5 millisecond. The dur ation of each cycle of alternating current used in conven tional ECT is 16 milliseconds and that used in the original unidirectional method of Friedman and Wilcox is approximate ly the same; that is to say that these techniques employed currents having pulse durations 20 to 30 times as long as 14 the optimal values with which tonic-clonic convulsions could be elicited in both animal and human subjects. Ample con firmation of these findings is provided by the later correl ative work of Arieff (1948) and Toman and Goodman (1947). Conflicting findings were likewise reported in regard to optimal frequency of repetition of stimuli for maximal summative effect. Frequencies as low as 10 to 30 per second were reported as eliciting the phenomena of facilitation by Adrian (1936-1937) working with animals. Boynton and Hines (1933) and Dusser de Barenne and McCulloch (1936) reported that convulsive responses could be obtained with an optimum frequency of 60 per second. Wyss and Obrador (1937)* work ing on monkeys under anesthesia, found that the optimum fre quency is variable according to the region of the cortex stimulated and averages 25 per second. Liberson (1948a) and Arieff(1948) reported considerably higher values of optimal frequencies than those given above. Their findings have been corroborated and are now generally accepted. Using a pulse duration of 0.3 millisecond, Liber son reported an optimal value of 250 per second while Arieff found the lowest threshold current for 300 per second stimu lation with a pulse duration of 1 millisecond and for 200 per second with a pulse duration of 0.2 millisecond. Since the computed energy values remain relatively constant for frequencies above 100 per second, the use of relatively high frequencies in brief stimulus therapy as compared to 15 conventional methods does not contravene the general purpose and design of these newer methods to secure the desired mo tor responses with the smallest possible energy totals. Research on the third important parameter of stimula tion has shown convincingly that the convulsive threshold (and therefore the current necessary to produce convulsion) decreases markedly as stimulation time increases. Hemphill and Walter (1941) first showed this to be true with 60 cycle current within rather narrow limits and these limits were extended in subsequent animal studies by Toman and Goodman (1947). Liberson (1948a) reported strikingly similar re sults in research on human subjects using 120 per second stimuli of 0.5 to 0.7 milliseconds pulse duration. Total energy outputs in stimulation times varying from 0.5 to 1.5 seconds did not exceed 15 per cent. The voltage employed with a total duration of 1.5 seconds was about half that re quired to produce convulsion when the stimulation lasted only 0.5 seconds. Here again a basic dimension of unidirec tional currents in relation to the all-important capacity to produce summation effects, which might seem superficially to oppose the intent of brief stimulus methods, is shown rather to aid in its accomplishment. As is the case in the conventional electroshock meth ods, there are numerous modifications of the basic current variables and combinations of these modifications in dis tinctive treatment patterns. The treatment used in the present study represents one such modified technique. In addition to variations in pulse duration, frequency of repe tition, and total duration, some of the most important cur rent modifications are: (l) the "courant chevauchant" or overlapping current (see Pig. 1C, p. 9) in which the point of zero potential lies above the point of origin of the spike, (2) the "wave-and-spikes" current in which a brief period of alternating current stimulation, producing petit mal and unconsciousness, is followed by unidirectional cur rent stimulation, and (3) the "secondary interruption" of any unidirectional current (see Fig. IP, p. 9) which reduces total current flow by interspersing brief periods of stimu lation with "silent" periods of varying lengths. Through these and other modifications of the basic parameters of the current to be employed, the total pattern of electroshock may be dissociated to produce at will states of unconscious ness, of prolonged narcosis, of petit mal or grand mal seiz ures, as well as several forms of direct stimulation with and without the retention of the state of consciousness. Method Used in the Present Study The management program, which is the subject of evalua tion in this study, was devised by James S. L. Jacobs, M.D., formerly chief of the Division of Psychiatry at Long Beach Veterans Administration Hospital, Long Beach, California. All patients in both experimental and control groups were 17 seen and tested at this hospital by the writer and all con ditions relating to their admission, physical, psychological and psychiatric examination, general hospital management, and treatment were held strictly constant throughout the pe riod of the experiment. All patients were under Dr. Jacobs' direct care or supervision and all treatments were adminis tered by him. The major parameters of the program were as follows: 1. The apparatus used was that designed by Offner (19^6) at the General Electric Laboratories at Schenectady, New York. The basic design of this device, with subsequent modifications, is the most widely used in brief stimulus therapy today. 2. The current employed varied from 170 to 250 cycles per second with a current strength of 30 to 50 milliamperes. 3. Secondary interruption of the current was accom plished with twenty-five cycles of stimulation alternating with a silent period of equal duration. 4. The total treatment time varied between 0.5 and 2.5 seconds. 5. Parieto-temporal electrode placement was used throughout. 6. Patients were placed under general sodium amytal anesthesia, in no case reporting any conscious phenomena in any phase of the treatment. 7. The glissando onset, a gradual build-up of current 18 to Its peak intensity, was used throughout. 8. No mechanical restraints were used at any time, manual restraint being provided during the brief (10-15 sec ond) interval during which actual muscle spasms occurred. It will be noted that the current used in this treat ment procedure represents a relatively high frequency, low amplitude current with relatively long total treatment time as compared with the brief stimulus methods generally. Secondary interruption of the current greatly reduces the total amount used, theoretically producing less heat and consequent tissue damage with no reduction in the effective ness of the stimulus condition to produce the desired con vulsion. The glissando approach is now used almost univer sally in both conventional and brief stimulus methods. It provides a much gentler beginning to the period of stimula tion and thus greatly reduces the stress unavoidably placed upon the muscular and skeletal systems. No compression fractures or other orthopedic difficulties occurred with any of these patients and there were very few reports of muscu lar soreness or similar distress. Both of these factors also entered into the decision to provide no mechanical re straint during treatment. Manual control permits pressures to be exerted in amount and in direction as the patient's movements demand and provides almost automatic relaxation of -J-Minor variations in these current parameters are es sential in view of patients' differing physiological charac teristics, hence these values must be expressed as ranges. 19 pressure when the counter-pressure becomes unduly great. The selection of the parieto- or vertex-temporal elec trode placement is one of the most significant characterist ics of the present treatment procedure. Other placements either fail to produce convulsions except with much larger currents or they (notably the bi-frontal and bi-temporal) produce sensations of extreme pain, blinding flashes of light, intense anxiety, and momentary arousal into con sciousness even when the patient has been anesthetized. Un derstandably, these unfortunate side-effects arouse intense fear and resistance to treatment, psychological side-effects which cannot be ignored in the discriminating use of any physical procedure. When the physical procedure is designed to ameliorate a psychiatric disorder the presence of these factors may well profoundly alter the whole course and out come of treatment. The almost euphoric response of the great majority of these patients to their treatment experi ences and the minimal amount of resistance encountered at test to the absence of these factors in this particular method. Of even greater importance than the foregoing in the selection of the vertex-temporal placement was the desire to stimulate more or less selectively the midbrain and thalamus. It may be noted in passing that the unidirectional currents generally are capable of being focused in fairly discrete beams, avoiding the widespread changes in potential throughout the brain that occur in the use of alternating currents. These changes have been described as the "onion skin" effect to indicate the many-layered gradients radiat ing outward from the area of maximum stimulation. Thus un necessary and even harmful stimulation of extraneous areas of the brain is avoided and the full threshold current can be directed toward the midbrain and thalamus and the vitally important centers of emotional control in the thalamic nu clei. At the same time it may be mentioned that the pre sumed concomitant stimulation of other nuclei which lie within these structures, regulating cardiac and respiratory rates, temperature control, and the like, produced no dis cernible changes in these functions other than those gener ally associated with the convulsive therapies. The use of sodium amytal anesthesia permits the patient undergoing this particular form of therapy to experience his treatment consciously as a relatively trifling incident in volving an intravenous injection administered in his own bed on the ward. Thus he is spared the psychologically traumat ic sights and sounds usually experienced by patients in elec tro-convulsive treatment and is also protected against any painful or anxiety-provoking sensations induced by the treat ment itself. These factors undoubtedly figured largely in reducing to a very minor degree the resistance to treatment encountered in these patients. Comparison of Brief Stimulus and Conventional Therapies 21 Complete unanimity of opinion as to the pre-eminent de sirability of the brief stimulus over the conventional meth ods whenever electroconvulsive therapy is indicated cannot be claimed for these newer procedures. Baruk (1955) takes the position that electroshock is rarely, if ever, indi cated; that "less radical methods can lead to real recovery rather than the exchange of one set of symptoms for another." He does not, however, clearly distinguish between the two methods of treatment and most present-day clinicians feel that, while his remarks may have some validity as to conven tional electroshock, no "set of symptoms" ascribable to the brief stimulus treatment can be identified or described. Sargant and Slater (195^) feel that the case for BST has not been proven and they prefer the use of various forms of al ternating current therapy. Wilcox, Funderburk, and Wilcox (1955) point out that modern drug therapies have tended to supplant EST in both forms and will continue to encroach upon this field in specific areas but they hold the view that electroshock methods will retain an important place in the treatment of mental illness for an Indefinite period. Most clinicians who continue to use the conventional tech niques, of whom there are a great number, merely ignore the claims put forth by the advocates of brief stimulus therapy. No clear-cut defense of the use of alternating current 22 methods in preference to brief stimulus therapy can be found in the literature. We may therefore list the following as well-substanti ated and widely accepted advantages of brief stimulus thera py over the conventional treatments in the opinion of its advocates,, advantages which apparently have not been specif ically challenged: 1. The proponents of the brief stimulus methods have from the outset advocated an adaptable and, hopefully, dis criminating approach, varying treatment characteristics to suit the problem, as opposed to a set, routine concept of treatment--an approach that recommends itself as inherently more sophisticated and potentially more effective. 2. The total amount of current used in a brief stimu lus treatment is only from one-fifth to one-twentieth of that used in conventional treatment. There is an almost unanimous consensus that these smaller currents produce less irreversible histological change in the tissues of the brain than the alternating currents. 3. Ancillary to the above, it has been shown by a num ber of workers, including Liberson (1953) and Bayles (1950) that BST produces fewer electroencephalographic changes and these of shorter duration. 4. It is generally agreed that BST produces a milder convulsion with a shorter latency period without the initial stunning effect of alternating current. Among the important 23 practical consequences of this fact are the markedly de creased incidence of compression fractures and other ortho pedic and muscular symptoms, and fewer cardiac and respira tory difficulties. These unfavorable side-effects were al most completely absent in the present group of patients. 5. Memory disorders appear to be greatly reduced in brief stimulus therapy. The studies of Scherer (1951) and Medlicott (19^8) are typical of a number of recent studies reporting this finding. 6. The one seemingly clear contraindication to the use of brief stimulus therapy is the fact that it, to a greater extent than is true of conventional ECT, was characterized by brief though intense sensations of pain, of anxiety and other dysphoric feelings, and of incomplete retrograde am nesia for events surrounding the treatment itself. In fact, Liberson (1953) states that the problem of fear is the real shortcoming of all the new forms of electric convulsive therapy. Although it may be minimized by psychotherapy, preliminary sedation, or other procedures, it remains of great theoretical and practical significance. (p. 217) These same reservations are voiced by Medlicott (19^8) and by Miles (19^9)> among others. However, the particular method of treatment employed in the present study appears to have obviated virtually all criticism on these grounds. Though none of these individuals was committed as a patient and though initial acceptance and continuation of treatment were entirely voluntary, no patient was discontinued at his 24 own request. Patients were completely amnestic to all phases of the treatment proper. They regained consciousness about two hours after treatment-relaxed, well-oriented and well-co-ordinated. Near-euphoric expressions of confidence in the efficacy of the just-completed treatment were the rule rather than the exception. 7. Because of the virtual absence of disorientation, confusion and amnesia (beyond the period of anesthesia), better continuing contact can be maintained for psychothera peutic purposes (Summerskill, Seeman, and Meals, 1951; Li- berson and Wilcox, 1945). In the opinion of the writer this fact is of fundamental significance because many clinicians, including Jacobs (1954) and Arieti (1955) among others, be lieve that psychotherapy is a critically important adjunct if gains made in electroconvulsive therapy are to be held or extended. 8. Particular areas of the brain can be stimulated se lectively by brief stimulus currents so that, while all ner vous structures are involved in the eventual massive dis charge, extrinsic stress is imposed only upon certain re stricted structures or areas (Hirschfeld and Bell, 1951)- It is generally felt that this should reduce the likelihood of brain damage, should permit convulsion with smaller cur rents, and should produce differential emotional or psycho dynamic sequellae. Previous Experimental Studies 25 Conventional Therapy In dealing with the vast literature that has by now been amassed on the electroconvulsive therapies in general, it is felt that our present purposes may best be served by selecting individual studies that are thought to be repre sentative of a type or class of research in this area and examining them in some detail. All too frequently critical examination of a particular study with respect to exact methodology or the statistical procedures employed may lead the reader to quite different conclusions from those set forth by the author. No attempt will be made to present an exhaustive review of this literature but the intention is, rather, to make a selection of particularly important or representative stud ies so as to provide an overview of the research that has been done in this field and to point out whatever consensus has been achieved in the major problem areas. The limited scope of this paper, focused as it is on one of the newer treatment techniques, permits only brief mention of the re search on the conventional methods. A number of papers which have a particular bearing upon the present study in regard to method or findings has been reserved for a more detailed, specific comparison in the chapter devoted to a discussion of the results of this study. (See Chapter IV.) From the outset the possibility that irreversible 26 trauma to the central nervous system might result from shock treatment has been a matter of concern to everyone familiar with it. This concern has been reflected, particularly in the work of psychologists, in a very large number of studies devoted to an examination of this one problem. In the opin ion of many, this effort has been disproportionately large and has precluded adequate study of other, equally important parameters of electroshock therapy. Steiper, et aJL. (Toman and Goodman, 19^7) and Janis (1950) have attacked the problems of memory loss and change in affective status with ECT in particularly well-controlled and detailed studies. Their conclusions are that unequivo cal and significant memory defects do occur and that even favorable changes in emotional integration are related to such losses. Aral and Obonai (195^) suggest a "memory- stratum" theory in an effort to explain the differential ef fects of ECT upon recent and remote memories. Pabing (1955) has presented a sophisticated neurophysiologically based theory in an attack upon the baffling problem of the pecul iar lacunar nature of much of the memory loss associated with ECT. While his theory cannot be reviewed in detail here, it may be stated in passing that its implications would support the view that fewer memory defects and less tissue damage should occur with the use of brief stimulus therapy. Heatherington (1952) has contributed an important study of psychological changes during as well as after ECT showing significant interruptions in normal learning curves involv ing intellectual activity and retention on the one hand, and probably significant increases in the speed of motor learn ing on the other. A great number of additional studies chiefly concerned with the problems of supposed intellectual deterioration, post-shock confusion, and amnestic phenomena in conventional ECT are presented in the literature, whose conclusions would seem generally to support the contention that conventional electroconvulsive methods produce disturb ances in intellectual functioning of several kinds as func tions of underlying tissue damage. Of the more significant ones whose findings, where applicable, will be related spe cifically to the results of the present investigation are those of Korngold (1953); Marilyn (1952), Krasner (1952), Williams (1950), Brower and Oppenheim (1951); Janis and As- trachan (1951); Rabin (1947); Stone (1947), Flynn (1954), Levit (1955), Salzman (1947), and Worchel (1950). However, there is much continuing controversy over the implications of the research on ECT and wide divergences in attitude toward its use persist. There are enthusiastic ad vocates as well as bitter opponents. A representative spokesman for the latter school of thought is Morrow (1951) who feels that ECT is an expedient at best whose principal effect is fear hidden in numerous ways. He ventures the opinion that whatever restitution is gained does not exceed 28 the premorbid, often very tenuous, homeostatic balance and that it is never a good substitute for psychotherapy. He is joined in these sentiments by Baruk (1955) and others, and solidly opposed by such writers as Gross (1952), Fergus (1952), Parsons, et al. (1955), Wolff (1955), Brill, Crump ton, Eiderson and Grayson (1957), and Alexander (1952). Brief Stimulus Therapy Scherer (1951) has done one of the outstanding studies on brief stimulus methods as well as one of the earliest on this relatively new treatment form. In view of the large amount of research on conventional shock treatment which purports to show significant memory losses, electroencephal- ographic changes, deterioration in intellectual functioning, and other evidence of cerebral trauma, he attempts to defend the position that similar damage does not occur with brief stimulus therapy and he appears to do so successfully. An experimental group of 41 mixed psychotic and psychoneurotic patients was given a series of 10-31 treatments. These and a group of 17 "mental patients" excluded from treatment be cause of physical and other complications were given a lengthy battery of tests in a group situation approximately three weeks before the treatment series began. The battery included several memory scales, the Shipley-Hartford, the Object Sorting Test, parts of the Wechsler-Bellevue Intelli gence Scale, and the Rorschach Multiple Choice Test. The post-shock battery was given from two to six weeks after the conclusion of the treatment series. Among the most signifi cant specific findings were improvement in form level with a marked diminution in minus responses, interpreted as im proved reality contact; a decrease in CF and C responses, interpreted as indications of less regressed affectivity; an increase in word-pair memory, interpreted as improved atten tive capacity. There were also widespread though minor ten dencies toward Improvement on a number of the tests of vari ous aspects of intellectual functioning. No significant de terioration was found in any area. The author concludes that brief stimulus therapy has no adverse effects on memory, abstractive capacity, attention and concentration, certain motor abilities, and some important personality attributes including the individual's emotional orientation toward the world about him. In a further study which may be included here because of its direct bearing upon the present work, Scherer (1955) extended the findings of Heath, Munroe and Michel (1955; Heath and Norman, 1946) concerning the differential effects of various electrode placements. The latter had established that stimulation of Broadman's Area 4 permitted exceptional ly rapid post-convulsive recovery; that when Areas 9 and 11 were stimulated subjects complained of anxiety and other acute, dysphoric feelings; and that stimulation of Area 38 produced markedly increased amnesia. In Scherer's study bi temporal and vertex-temporal placements were used with the 30 following results: no differences were noted between the two placements in recovery time to the point of recall or recognition of objects presented just before treatment; short recovery time correlated well (.49) with patient im provement; there were no differences in the amount of retro grade amnesia; there was a non-significant tendency for hand dynamometer readings to be lowered by the vertex-temporal placement but a marked and significant decrease (1 per cent level) with the bitemporal placement; there were no signifi cant changes in either group in recall time as the treatment series progressed; there was a tendency with the vertex-tem poral placement toward shorter confusional periods in the more chronically ill patients. A study done by Wilcox (1954) has done much to illumi nate the transient nature of such defects in intellectual functioning as have been reported in the literature on brief stimulus therapy. It was predicted that a temporary decre ment would occur tending to be reversible within an hour following a single ECT; that the decrement would tend to be cumulative with a series of treatments; and that this cumu lative decrement would be temporary and would be reversible within two weeks following the end of a series of ten treat ments. A short battery of five brief tests was administered at 15 minute intervals to 23 women patients as soon as each had recovered from the first, fifth, and tenth treatments. This battery consisted of the Niver block, serial sevens, 31 Stroop color naming, and the copying of Bender designs 6 and 7. Results showed a significant decrement for the first quarter-hour, an equally significant improvement for the second, then a more gradual improvement for the third. At this time three of the five measures showed essential recov ery to the pre-shock level. A longer battery consisting of all the first battery plus the Shipley-Hartford Scale, the Fairfield block substi tution scale, and two arithmetic tests was administered be fore treatment, and one day, two weeks, and three months after treatment. From pre-ECT to one day after the series every measure showed a decrement in performance, significant for five of ten. Two weeks later, nine of the ten measures showed more improvement than could be attributed to practice. Three months later there was a further slight improvement. These score changes were reported to have been accompanied by clinical improvement and it will be noted that all three directional predictions were supported. An ancillary find ing was that performance on measures of both old and new learning were affected, but the latter to a greater degree than the former, a conclusion that recalls the work of Arai and Obonai (1954), Michael (1954), and others working with conventional electrotherapy. Holzberg and Cohen (1952) have contributed an important study concerned with the relationship between psychiatric improvement in patients under brief stimulus therapy and 32 certain pathologic changes in the Rorschach during the course of treatment. Twenty mixed psychotic patients were given a series of 14-24 BST and Rorschachs were administered once per week throughout the course of treatment. A rating scale was made out by the psychiatrist to rate them "Im proved" or "Unimproved," and the Rorschachs were scored and rated on the Munroe check list. The hypothesis was advanced that patients who improved in treatment would first show pathologic Rorschach changes during treatment. It was found (5 per cent level) that more such changes did actually occur in the improved group. Perhaps of even more importance is the finding that a significantly (3 per cent level) greater number of fluctuations or reversals in trend occurred in the improved group. This was interpreted as indicative of "less personal rigidity." Specific Munroe factors significant at or below the 5 per cent level were C, m, and c. It was also shown graphically that a steady rise in fluctuation occurred in the improved group for approximately 13 treatments as op posed to a steady fall in the unimproved group from the first treatment on. Bayles (1950) has contributed an important comparative study contrasting EEG changes in brief stimulus therapy with those found in conventional treatments. Using the glissando approach and vertex-temporal electrode placement, he found that 15 per cent of the post-shock EEG's were normal, 31 per cent showed moderate disturbance, and 5^ per cent severe 33 disturbance with brief stimulus therapy as contrasted with 4 per cent normal, 12 per cent moderate, and 84 per cent se vere when conventional electrotherapy was used. These dif ferences are significant at the 5 per cent level. Clinical observation tended to refute the hypothesis that results would be better in cases in which depression was prominent, a finding with which Kallnowsky (1951)* Jacobs (1954), Malz- berg (1952), Lopez-Ibor (1950), Heatherington (1952) and many others categorically disagree. He found great differ ences in the clinically observed confusion and disorienta tion in favor of the brief stimulus treatment. Another paper by Liberson (1948b) is characteristic of the work done by psychiatrists on the brief stimulus methods. Using clinical ratings done by the hospital staff on a group of 46 patients, he found that 80 per cent of those classi fied as suffering from an "affective disturbance" were rated as recovered and the remaining 20 per cent improved; 45 per cent of the schizophrenics were rated recovered, 45 per cent improved, and 10 per cent unchanged. Both of the patients who comprise the 10 per cent unchanged were hebephrenics. He concurs with Bayles (1950) and others in the finding that EEG changes were much smaller. He found a shorter post-con vulsive recovery period with BST and he states that compres sion fractures were much less frequent. He concurs with Jacobs (1954) and Arieti (1955) in an important finding that patients who have received a series of brief stimulus 34 treatments are often rendered amenable to psychotherapy who would otherwise be completely unreceptlve. The chief disad vantages in BST as he sees it are fear of treatment, unob literated minor reactions, and "dissociated" shocks in which there is a partial preservation of consciousness. It may be added in passing that the method employed in the present study obviates all these criticisms. He relates the advant ages of BST, as do the great majority of those who have done basic physical research in the field, to the much smaller (of the order of one to two per cent as much) total energy requirement in these methods. Summerskill (1951) has made an important contribution to the literature on brief stimulus therapy in a study de signed to test the hypothesis that the confusional effects following the use of the Reiter Electrostimulator were neg ligible. His subjects were 12 men and 12 women diagnosed as psychotic or severely psychoneurotic. They were adminis tered an Orientation Questionnaire, the Whipple-Healy Tap ping Test and the Wechsler-Bellevue prior to and immediately after the first treatment in a course of therapy. The pre shock testing was done the week before treatment began and the post-shock battery was given within thirty minutes of the treatment. Alternate forms of the Wechsler-Bellevue were used so that half the group would get Form I initially and Form II finally, with the other half just reversed. (This is the method used in the present study.) His 35 conclusions are that The differences between the pre- and post-treatment Wechs- ler scores are negligible and indicate that post-electro- shock confusion with the Reiter apparatus is not suffi cient to interfere with the subjects' ability to deal with the variety of complex tasks comprising this test. A com parison of results with the other tests also indicates minimal confusion in most respects. The claimed absence of confusional effects from the use of the Reiter Electro stimulator is substantiated by these results which have implications for the practical questions of managing and testing patients following electroshock treatment. (1951, p. 344) Despite the unequivocal assertions of the superiority of the brief stimulus methods over conventional shock treat ment found in the above study and in many others like it, Kendall, Mills and Thale (1956) published a recent study questioning the enthusiastic claims made for the former methods and pointing out the need for actual comparative studies if such claims were to receive any credible support. It is, in fact, the only study specifically devised for a direct comparison of the two methods within a controlled ex perimental design to come to the writer's attention. Kendall advances three major hypotheses: (l) Brief stimulus therapy causes less impairment of the cognitive functions for the same therapeutic gains as conventional methods. (2) Patients receiving brief stimulus therapy show less residual cognitive disturbance two weeks after treat ment. (3) Post-treatment gains in tests involving cognitive tasks are greater and the losses are less than with conven tional methods. Bi-temporal electrode placements were used and no pre-treatment sedation was given. Brief stimulus 36 patients received 20 treatments and the conventional group 12, which in the opinion of the authors equated the series as regards therapeutic efficacy. Two groups of 20 each, containing both male and female, mixed psychotic patients were employed. To evaluate immediate post-shock disturbance a battery consisting of picture absurdities, substitution, cancelation, and cube imitation tests was administered 30 minutes after the first, the middle, and the final treat ments. Three hours after the first battery, a second bat tery was given consisting of naming months in reverse, sub tracting serial sevens and serial threes, color naming, dup lication of block designs from memory, and a recognition test. Form I of the Wechsler Memory Scale was administered after the first treatment "to measure temporary intellectual inefficiency," and Form II after the final treatment. Fi nally, the full Wechsler-Bellevue, Form I was given before the series began and Form II two weeks after it had been completed. Negative changes on the first two batteries were noted for both groups in eight of the ten tests significant at or beyond the five per cent level. The brief stimulus group, however, showed a significant gain (5 per cent level) in the memory for designs test. There were no other significant differences between the two groups. The brief stimulus group showed far less cognitive disturbance as measured by the Wechsler Memory Scale, Form I but this difference, significant at the one per cent level, had dropped just be low the point of significance when Form II was administered two weeks after treatment. There was a slight decrement in Wechsler IQ scores in both groups, significant at the 5 Per cent level for the control group, just below significance for the brief stimulus group. The authors feel that this indicates a residual organic damage since improved scores should be expected in a "patient with a remitting psychosis." They conclude that the hypotheses concerned with smaller im pairment and greater post-treatment cognitive gains in the brief stimulus group are not sustained, but that good con firmation was obtained for the hypothesis that these patients would show less residual disturbance two weeks after treat ment . Probably the most extensive, if not the best controlled, study yet done on the brief stimulus methods is that done by Wilcox (1947);, reporting on 23,000 cases at the Traverse City (Michigan) State hospital covering a five year period of observation. His attitude is summed up in these words, There and elsewhere therapeutic results, superior to those obtained with the more commonly used Cerletti-Bini tech nique, have been secured with much less evidence of even temporary functional impairment of the brain beyond the immediate reorientation period of from 20 to 60 minutes following the convulsion. We have not needed to soften our convulsions with curare because the technique produces a smoother, less violent convulsion. We have been able to carry out psychotherapy concurrently with the treatments because the patient does not become disoriented or con fused and his memory remains clear as far as the effects of treatment are concerned. (p. 123) As evidence in support of this fulsome praise of the brief stimulus methods he cites marked improvement in cer tain specified kinds of ward behavior in a selection of 500 of the most chronic patients at the hospital. Most of these had been treated previously with one or more physical thera pies without success. Thirty-three per cent were markedly better in four months though there were many relapses. In such cases the author stresses the point that intensive ini tial treatment should be followed by intensive treatment of relapses until rehabilitation has been accomplished. Approximately 65 per cent of all first admissions who were treated early and adequately, in his view, were "social ly rehabilitated" at the end of one year. These patients were reviewed at intervals of four, eight, and twelve months and it was found that 52, 63, and 65 per cent respectively were able to leave the hospital. Of all patients treated early in the first psychotic episode, 76 per cent were symp tom-free by the time the two-year review was made. The au thor feels that important benefits from this form of treat ment accrue to patients with a wide variety of emotional problems including alcoholism, CNS lues, psychoses with in fectious diseases, organic syndromes in the aged, involu tional problems, manic-depressive psychosis, and all forms of schizophrenic disorder. Among the latter, he found that the largest number of treatments necessary to produce the desired changes occurred in the catatonic group. Psychoneu rotics, most of whom had pronounced depressive or- obsessive 39 features, responded well with a 75 per cent rate of remis sion within four months. Though the above study is based on an almost unpreced ented number of cases observed over a considerable period of time, its claims and purported findings cannot but leave the reader with a multitude of unanswered questions. Its crite ria of Improvement are admittedly loosely defined parameters of ward behavior and "social rehabilitation." As is charac teristic of so much work in this field, clinical judgment is the sole measure of pre-treatment condition, of Improvement, and of final status. There have been increasingly numerous attempts of late to bridge the gap between neurophysiological theory on the one hand and psychodynamic theory on the other. Whatever their limitations or deficiencies at the present time, it is certainly to be hoped that these attempts will ultimately give us the unified psychological theory without which we cannot hope to emulate the advances in the other basic dis ciplines . Jeans and Toman (1956), and Alexander (1950? 1952, 1956, 1953)? among others, have studied the phenomena of cortical excitability levels and their relationship to sys temic sub-cortical influences operating through the autonom ic nervous system. This work has extended that of Gellhorn (1956, 1957) and Funkenstein, Greenblatt and Solomon (19^-9) who have pioneered much of the research on autonomic and 40 hypothalamic functions. Alexander's theory of the effect of electroconvulsive therapy appears to be representative of present-day thought. In his words, Electroshock therapy relieves the secondary traumatic state of the ego (panic and/or depression) by reducing the excitability of the nervous system, especially the cortex. Unconscious warning anxiety (subcortical, epinephrine-pre- clpitable) can then no longer over-stimulate the cortical ego and thus frighten it into panic or paralyze it into depression. (1952, p. 293) The most recent research trends in this field appear to « be toward the study of brief stimulus treatment in associa tion with other major therapeutic agents. Since the advent of the ataractic drugs, in particular, a large number of ar ticles have appeared in the literature devoted to study of the combined effects of the two agents. These studies at the present time, however, consist of mere observational re ports with little or no attempt at standardization of treat ment procedures or objectification of findings. A final, recent study may be mentioned here which reports some inter esting observations on the supposed reciprocal action of electroconvulsive and electrostimulatory therapy. Sweel (1958) advances the hypothesis that there is a reciprocal relationship between anxiety and depression and that patients in whom either factor is prominent are fre quently subject, in or out of therapy, to a disturbing "shifting of affect" from the one to the other. On the ba sis of his own treatment experience he believes that elec troconvulsive therapy relieves depression and potentiates 41 anxiety while electrostimulatory treatment is usually quite effective in relieving anxiety but at the risk of potentiat ing depression. In a general comment on the brief stimulus methods, he states that not only do they fail to produce the unfavorable side effects of conventional treatment, but they in fact relieve the confusion and amnesia produced by these latter methods and are used for this specific purpose in his own clinical practice. His statement concerning the major dynamic effects of electroconvulsive therapy, with those at tributed to electrostimulatory treatment an almost exact re ciprocal, are worth inclusion here because of their direct bearing on some of the major premises of the present study. Brief stimulus therapy is said to: (l) restore, raise, and strengthen ego defenses and to make them more flexible, (2) remove depressive features, (3) potentiate anxiety features, (4) inhibit awareness of disturbing mental content, (5) re lieve repressed material, and (6) activate the defenses of isolation and denial. Summary It is hoped that the foregoing has provided a reason ably complete, general survey of representative research in the field of electroconvulsive therapy. By far the major portion of it has been done on the conventional methods which are still used much more widely than the brief stimu lus methods. If the picture seems somewhat confused as re gards what has actually been learned or proven or generally 42 accepted about electroconvulsive therapy, it may truly be said that this confusion accurately mirrors contemporary thought in the field. The greatest amount of effort has been expended in the research on both treatment forms on attempts at clarifying the problem of tissue damage in the central nervous system. While agreement is far from complete, and though dissenting voices will undoubtedly continue to be raised, it seems that the weight of evidence now at hand supports the view that appreciable and apparently irreversible brain damage is as sociated with the use of alternating currents. There remains a serious question as to whether it also occurs with the brief stimulus methods but the consensus appears to be that, if present at all, it is substantially smaller or more eas ily reversible than in conventional shock therapy. While admittedly a serious problem and one which cer tainly merits the time and attention given to it, it is nonetheless to be regretted that the problem of brain damage has to such a degree pre-empted the efforts of research workers and left so few of always limited facilities for the investigation of the many other and equally crucial problems concerned with the psychodynamic changes that apparently take place as a result of these treatments. Such research as has been done in this area has yielded even more con flicting results and opinions than that on the problem of tissue damage. 43 It would be foolish indeed to ascribe all the continu ing controversy in this field to mere opinionation or rigid ity on the part of those concerned with it. There are genu ine doubts based upon real inconsistencies in the reported findings of the many researches presented in the literature. An attempt has been made in this survey, and further at tempts will be made in later sections, to call attention to a number of characteristic faults which have no doubt been responsible In good part for these contradictory findings. The two most serious and most obvious defects, in the opinion of the writer, are the failure to provide coherent experimental designs and the use of sensitive statistical devices based upon assumptions which cannot logically be made about the data under consideration. Too much reliance has been placed upon unaided clinical judgment in loosely planned, observational situations. Obviously significant variables are often left completely uncontrolled. Therapeu tic and evaluatory procedures are permitted to vary over wide ranges. Tests which have been neither standardized nor validated are often employed. Subject populations frequent ly lack necessary homogeneity and control groups, if used at all, are inadequately matched. These and other design and procedural faults have added unnecessarily to the unusual difficulties inherent in this kind of investigation. We cannot limit ourselves to the laboratory study of meaningless bits of behavior in the interest of 44 methodological refinement but this does not imply abandoning all effort at the conduct of controlled experiments within the framework of adequate experimental designs, and at eval uating the results by means of statistical procedures which can be shown logically and mathematically to fit the data ob tained. Whatever contribution, if any, the present study may make to our knowledge in this field will be due in large measure to the avoidance or minimizing of this kind of meth odological error. The Present Study Rationale It is felt that these two major considerations point up the need for a study such as the present one: (l) no study on precisely this form of treatment program can be found in the existing literature; (2) the results of the investiga tion of most of the important aspects of the electroshock methods in general have been incomplete and inconsistent. As to the first of these considerations, it may be ar gued that treatment forms which appear to be highly similar, if not identical, have been studied in some detail and that no great differences are to be expected between this and seemingly closely related methods. This argument may, per haps, best be answered by referring to the gestaltist con cept that any significant change in any important structur ing element in a field must result in significant changes in the field as a whole. Whac might appear to be a minor change in the physical properties of the current used, for Instance, may and usually does result in profound altera tions in its physiological effects. One of the major con traindications for the use of BST, the deeply disturbing fears that usually accompanied it, has been obviated by the use of anesthesia and by the particular electrode placement employed. The validity of the second consideration would seem to need little defense. Aside from the achievement of a reason able consensus in regard to certain aspects of the deteriora tion in intellectual function ascrlbable to the use of ECT, little is actually known of the effects of shock therapy, little agreement has actually been reached as to what kinds of emotional changes do occur, if any, and as to what their significance might be when they occur. If the present study can illuminate the nature of these changes, if in only a few of the unknown total number of possible changes, it will have been worth the effort. The writer's clinical observation of a number of pa tients undergoing this form of treatment first stimulated his interest in it. At this most superficial level it ap peared to offer much that other electroshock methods did not, and to lack several of their most serious disadvantages. The decision was then made to conduct a pilot study using the Wechsler-Bellevue Intelligence Scale, the Rorschach, and the Mosaic Test to ascertain whether actual test changes might serve to confirm the initial clinical impressions. A group of fifteen patients was given pre- and post-treatment batteries and, while no elaborate statistical analysis was attempted, it was felt that significant trends were indi cated in the expected directions. Important changes ap peared to have taken place in these patients' emotional bal ance, in the effectiveness of their intellectual function ing, and in their ability to perceive and to deal construc tively with concrete reality problems. It was also felt that the pilot study helped to answer affirmatively any question as to the relevance or appropriateness of the test ing instruments used. In the light of the experience gained through conducting the pilot study and in view of the tenta tive findings resulting from it, four specific hypotheses were advanced predicting the nature and direction of the ex pected changes and the experimental design was completed. Hypotheses The present study is proposed to test the following hy potheses : Hypothesis A. It is hypothesized that patients receiv ing brief stimulus electroconvulsive therapy will demon strate an improvement in the general level of intellectual functioning. It is felt that the intended meaning of the term "intellectual functioning" is implied by the nature of the instruments selected to measure it, Forms I and II of the Wechsler-Bellevue Intelligence Scale, and that no further definition is required. Improvement is expected to be gen eral and no reason is seen to exist for predicting selective gains in particular areas or functions. Hypothesis B. It is hypothesized that patients receiv ing brief stimulus electroconvulsive therapy will give evi dence of increased ego integrative capacity. This phrase is taken to denote the total adjustive cap acity or basic ego strength which is derived from the inte grated functioning of the personality variables of reality testing, emotional integration, self-realization, and mas tery of reality situations (Klopfer, Ainsworth, Klopfer and Holt, 1954). Hypothesis C. It is hypothesized that patients receiv ing brief stimulus electroconvulsive therapy will demon strate improved affective-cognitive integration as related to the mastery of reality situations. This concept is intended to apply to the complex of perceptual-cognitive functions and the integrated emotional responses which motivate them to function effectively in contemporary, concrete, reality conditions. The actuality of present functional ability is emphasized here, whereas in Hypothesis B potentiality is stressed, whether or not the 48 inferred resources are immediately available for use (Lowen- feld, 1949). Corollary Hypothesis. It is hypothesized that patients receiving brief stimulus electroconvulsive therapy will dem onstrate no decrement in the specific areas of intellectual functioning that have been found to be most sensitive to the effects of organic brain damage or deterioration. Since some selection has to be made from among the very large number of supposed "organic indicators" provided by the testing instruments used in this study, the substantiat ing evidence for the above hypothesis will be taken from a few of the best known and accepted of these signs. It will immediately be evident that the concepts de fined above have many areas of commonality, that they do not represent mutually exclusive categories of "pure" functions. However, it is usually found that such functions, when they can be isolated and identified with any confidence, have on ly very limited and qualified applicability to the actual life adjustments which must be our ultimate concern. They are not so discovered in vivo and they function invariably as elements in an integrated complex with many other such elements. They never function in isolation. The attempt here is to define and evaluate several of these complexes Under certain special conditions as they operate in actual life situations. CHAPTER II Method Subjects Experimental Group Twenty-five patients were selected from among those ad mitted to the Psychiatric Section of the Long Beach Veterans' Administration Hospital who met the following criteria: 1. Patients were white males between the ages of twen ty and forty. 2. The present hospitalization was required to be the first incidence of mental or emotional breakdown. 3. Unanimous agreement was obtained from the psychiat ric staff and the psychologist that patient's illness was properly classified as a form of schizophrenic disorder. In case of any dissenting opinions, patient was excluded. Di agnoses were made on the basis of psychiatric examination and the pre-shock psychological test battery. Patients must have completed a minimum of eight years' formal schooling in order to minimize biasing factors primarily in intelligence test results. 5. No members of strikingly different subcultural background were accepted with a particular view to minimiz ing fortuitously deviant Rorschach and Mosaic responses. 6. An intelligence quotient of 85 or above on the Wechsler-Bellevue was required. 49 50 7- No patients with complicating physical illnesses or disabilities were accepted. 8. Patients must have had no previous treatment for any form of emotional illness, including intensive psycho therapy, and must have had no previous experience with any of the testing instruments used in this study. Control Group A control group of twenty-five patients was selected on exactly the same basis as the experimental group described above. All patients were subjected to identical admission, processing, and examination procedures. The control pa tients were given the same diagnoses and ultimately the same treatment. Therapy was, however, suspended for a period equal to that occupied by treatment in the experimental group so that the control patients merely resided on the same wards of the hospital for the period of time covered by the study. No formal therapy of any kind could be given during this period. Any necessary deviation from the normal admitting or evaluatory procedures led to the exclusion of such patients. From the date of the beginning of the formal study all newly admitted patients who met the criteria de scribed above were accepted. Exclusions were on the basis of these criteria only. Testing Instruments The Wechsler-Bellevue Adult Intelligence Scale was____ selected as an appropriate instrument for the testing of Hypothesis A which postulates improvement in the general level of intellectual functioning in the experimental group. These scales are probably the most widely used and most thoroughly standardized now in use for the testing of general intelligence. They have been designed for use with adults, hence many of the criticisms leveled at such instruments as the Stanford-Binet with its emphasis on the testing of chil dren and adolescents have been obviated. The tasks pre sented are such as can be expected to be of some interest and to some degree challenging to most adults. The over emphasis in many tests upon verbal skills and other products of formal education is not present in these tests. They sample a wide variety of mental functions and some of the tests are almost culture-free. Standardization of the scales has been uniquely thorough, particularly in regard to the careful stratification of the samples employed. The Wechsler-Bellevue scale is also to be used in test ing the corollary hypothesis which proposes that no decre ment will be found in the experimental group in selected areas which have been found to be highly sensitive to the effects of organic deterioration. The various subtests com prising the "Don't Hold" group will be examined individually together with the calculated Deterioration Quotient. Though much has appeared in the literature questioning the validity of these items as adequate measures of organic deterioration, 52 at least as much has appeared in support of them. It may reasonably be claimed that their use here is justified on the grounds that they are to be appraised individually as well as collectively, that other specific measures are also to be employed for this purpose, and that additional informal evidence is available in the results of the other tests car ried out. The question of organic deterioration will be considered in this context. The Rorschach Prognostic Rating Scale was chosen to test Hypothesis B since it is so directly and specifically concerned with the problem of evaluating basic ego strength or ego integrative capacity as here defined (Klopfer, et al., 1954; Greene, 1954). Derived from the standard, individual ly administered Rorschach test, its six separate scales are designed to reflect as many basic elements which are thought by its authors to comprise the individual's basic ego strength, and to sum them up in a Final Prognostic Score. It possesses the unique advantages of combining considera tions of content with those of formal scoring. It permits, to a greater degree than any other known device, a global approach to the Rorschach, an approach which is relatively free of the often excessive rigidity of formal scores, while it ultimately provides meaningful scores in numerical terms. It seems the best attempt yet made to furnish the clinician with statistically manipulable quantities and at the same time to preserve the more important combinations and 53 interrelationships so vital in Rorschach data. The Mosaic Test (Lowenfeld, 1952) is to be used in the testing of Hypothesis C. In the context of this study it is considered an adjunctive measure to the Rorschach, reflect ing many of the same areas of personality function, with the important difference that it purports to measure resources immediately available to the individual in a contemporary, reality situation. Peak (1953)* who constructed the rating scale and did the ratings on the mosaics used in this study, states that the Mosaic Test was originally conceived by Low- enfeld as a means of demonstrating the functioning of cog nitive and emotional processes in a concrete situation. It is contrasted with such instruments as the Rorschach and Thematic Apperception Tests which are commonly thought to reveal the individual's potentials having their origin in the deeper strata of personality. The Mosaic Test provides a measure of what he can actually do in a contemporary, con crete problem situation. Physical Management Program A brief recapitulation may be appropriate at this point in regard to the physical details of the treatment method employed. (For a full physical description and rationale for the use of this method, see Chapter I, pp. 16-21.) A unidirectional, square wave current of 170-250 cps and cur rent strength of 30-50 ma was used, with secondary interrup tion and a total treatment time of 0.5~2.5 seconds. The 54 glissando approach was used with parieto-temporal electrode placement. A l l t r e a t m e n t s w e r e c a r r i e d o u t u n d e r s o d iu m a m y t a l a n e s t h e s i a s o t h a t t h e p a t i e n t ’ s own c o n s c i o u s e x p e r i e n c e o f h i s t r e a t m e n t was t h a t o f an i n t r a v e n o u s i n j e c t i o n i n h i s own bed on t h e w a rd . No d i s s o c i a t e d s h o c k s o r o t h e r c o n s c i o u s phenom ena w e r e r e p o r t e d b y a n y o f t h e p a t i e n t s t e s t e d . I m m e d i a t e l y f o l l o w i n g t r e a t m e n t p a t i e n t s w e r e r e t u r n e d to t h e i r b e d s by w ard p e r s o n n e l , r e g a i n i n g c o n s c i o u s n e s s i n o n e o r tw o h o u r s i n t h e i r p r e - t r e a t m e n t s u r r o u n d i n g s . All patients were given a series of six treatments. Jacobs (1954) considers this the standard, basic course of treatment for all patients. It is felt that no fewer treat ments can really show whether a given patient will or will not improve under this form of therapy. In clinical prac tice, treatment may be terminated after a series of this length if he has shown no improvement or if he has reached an optimal point of recovery. If it is thought that further gains can be registered the series may be extended to ten, but not beyond this point except for an occasional "mainte nance" treatment. In this Jacobs is among the most conserv ative of clinicians using any form of shock therapy. A l l p a t i e n t s , e x p e r i m e n t a l and c o n t r o l g r o u p s a l i k e , w e n t t h r o u g h t h e same a d m i s s i o n and d i a g n o s t i c p r o c e d u r e s an d r e s i d e d on t h e same w ard i n t h e n e u r o p s y c h i a t r i c s e c t i o n o f t h e h o s p i t a l . T r e a tm e n t w as c a r r i e d o u t on T u e s d a y and 55 Friday mornings for three successive weeks. Control pa tients merely resided on the ward for the period of time oc cupied by the treatment series with no formal therapy of any kind. Every effort was made to keep patients' general hos pital experience the same and, when this was not possible, patients were dropped from the study. All patients were first admission, voluntary patients who accepted treatment voluntarily and who could have withdrawn at any time upon request. None were lost on this account. Testing Procedures The test battery was administered in one testing ses sion two days before the first shock treatment and two days after the last. This interval was chosen for two reasons: (l) It was felt that it provided sufficient time for patient to overcome any residual drug effects and to return to what ever homeostatic balance he could achieve following treat ment. (2) It was considered too short a period to recover from any marked or serious deteriorative effects resulting from treatment. Tuesday and Friday were designated as treat ment days in order that testing could be accomplished on Wednesday for the initial battery and on Thursday for the final battery. The purpose in view was to avoid the dis turbance in ward routine occasioned by the week-ends and to provide a "normal," work-a-day milieu in which to place the testing periods. The effect upon a hospitalized individual of being required to take a rather lengthy battery of psychological tests while a loved one waits in the wings needs no clarification. Other disturbances at this time are common and unavoidable. Standard administration of all tests obtained through out with the single exception that testing the limits on the Rorschach was omitted in both initial and final testing. The administration of the Wechsler-Bellevue and Rorschach tests is too well known to require review here but a brief summary of the procedure employed in the use of the Mosaic Test may be in order. The full instructions supplied by the Psychological Corporation to users of this test will be found in Appendix H. The opened box of mosaic pieces is placed in front of the testee and the five different shapes and five different colors supplied by the set of materials are carefully pointed out to him. The tray in which a closely fitted piece of white paper has been laid is placed in front of him and he is asked to make something with the pieces. It is pointed out that he has complete freedom to construct any thing he likes and that he may use as many or as few pieces as he wishes. Any structuring on the part of the examiner as to the nature of the design must be carefully avoided. When the design is finished the examiner should ask for a brief description of the idea represented and should gain some understanding of its meaning to the maker. A time rec ord should be kept beginning with the conclusion of 57 instructions and ending when the subject indicates his pro duction is finished. Ideally the mosaic should be photo graphed in color as was done in the present study, but if this is impracticable, the designs may be traced on the un derlying paper with each piece labeled as to color. Forms I and II of the Wechsler-Bellevue scales were used alternatively with half of each group receiving Form I for the initial testing and Form II for the final testing, and the other half of both groups receiving Form II initial ly and Form I at the conclusion of the treatment series. The designation as to which patients were to receive Form I first and which Form II first was done through the use of a table of random numbers. It has long been the writer's view that the personality of the examiner and personal idiosyncracies which remain un controlled, despite the most rigorous efforts toward stand ardizing administrative procedures, can and do give rise to appreciable differences in the scores or other findings of a great many psychological tests. It is felt that this is true even in the case of the more objective tests such as the Wechsler-Bellevue scales, but it would seem to be of much more fundamental importance as a source of error or distortion with the administratively complex and sensitive projective tests like the Rorschach. It does not seem at all unreasonable to suppose that possibly significant dif ferences that might have appeared in many studies have been 58 canceled out by the uncontrolled, chance variability intro duced by the varying personal attributes and small procedural vagaries of different examiners. This impression was given real substance by Baughman (1951) who showed that very wide differences indeed could occur in Rorschach scores as a func tion of these examiner differences. It was decided early in planning the experimental de sign for the present study that chance variability arising from examiner differences could not be tolerated and that, therefore, all tests used in the study would be given per sonally by the writer. It may then be claimed that the per sonal impact of the examiner upon the subjects was standard throughout and that administrative biases were systematic, presumably affecting all tests and all patients alike. Par ticularly since we are dealing with difference scores, it may be claimed that extraneous sources of error arising in the examiner have been minimized. Methods of Evaluation The Wechsler-Bellevue scales were scored by the examiner according to the standard criteria established for their use by the author. In view of the relatively objective nature of these tests and of the carefully defined scoring rules pre sented with them, no reason was seen for requiring that these particular scores be verified by other examiners. The Rorschach records were initially scored by the writ er and then reviewed by Dr. Gertrude Baker, a staff 59 psychologist at the Long Beach Veterans' Hospital. Although the writer was necessarily aware of the identity of the rec ords, Baker was not informed as to whether the particular record under consideration was a pre- or a post-treatment protocol or as to whether it had been obtained from a member of the experimental or of the control group. Scoring was carried out as nearly as possible according to the method of Klopfer and Kelley (19^2) recently revised and extended (Klopfer, et. al., 195^)* Any particularly dif ficult or ambiguous responses or scorings upon which the raters could not agree were taken to Klopfer himself whose decision was regarded as final. When the scoring had been completed for both groups a further accuracy check was pro vided. The second, the middle, and the next-to-last pre- and post-treatment sets from both experimental and control groups were selected to provide a reasonable sampling of records throughout the series, and these records were sub mitted to an outside rater. He was ignorant of the condi tions of the experiment and all identifying marks had been removed from the records. In the protocols presented to him there were 555 scoring decisions to be made and disagreements were found in only 12 of them, or a scoring agreement of al most 98 per cent. The preparation of the Prognostic Rating Scales was done by the writer in accordance with the rules established for the use of this evaluatory method by its authors, 6o Klopfer, et al. (1951* 195^)- Again it was thought that the rules for the preparation of this scale were sufficiently well defined that, given an agreed-upon set of scores upon which to base it, there would be little likelihood of sig nificant discrepancies between raters. Therefore, no out side rater was thought necessary. An illustration of the use of the Rorschach Prognostic Rating Scale is given in Ap pendix I. The Mosaic productions were photographed by the writer in 35mm • Kodachrome and were presented to Dr. Horace M. Peak, author of the Peak Mosaic Rating Scale (1953)* for evalua tion. Peak had no knowledge of the experimental conditions of the study or of any of the patients in it. The mosaics were presented in pairs representing the before- and after testing of each individual patient, and control and experi mental sets were intermingled. The only knowledge he had aside from that imparted to him by the photographed mosaic itself was the time required for the subject to complete the production and the brief description obtained as part of the standard procedure for the administration of this test. Peak was told that an intervening variable existed in all cases which we had reason to believe should have brought about some improvement in the later productions. He was asked merely to judge which was the better of each of the paired mosaics on the basis of his rating scale. A scale was thus made out for each mosaic. An illustration of the 6l use of this scale will be found in Appendix P. Sample mosaic productions are found in Appendix E. Three months after the initial scoring of the mosaics, they were again presented to the same rater who was asked to separate them again into "Improved" and "Unimproved" groups as a check on the reliability of the scale in the hands of this particular rater. The agreement obtained between first and second ratings was 97 per cent, yielding a Spearman rho of .92. Only three of the 25 sets of mosaics in the experi mental group were judged wrongly on each trial. In two of these cases the formal ratings showed them to be unimproved and they were so judged both times. In the remaining case a mosaic that had been judged correctly on the first trial was judged incorrectly on the second, and one that had been judged incorrectly the first time was judged correctly the second. The judgments on the control group with respect to designating the second as the supposedly "improved" mosaic were approximately at the level of chance. On the first trial 14 of the second mosaics were judged unimproved, with 11 improved, and on the second trial 13 of the second mosaics were judged improved, with 12 unimproved. Statistical Procedures The problems of statistical analysis posed by the data furnished by such instruments as the Rorschach and Mosaic Tests are formidable indeed. If one wishes to preserve any semblance of scientific rigor certain admissions must be 62 made at the outset as to the basic nature of these data, ad missions which quickly remove them from the purlieus of par ametric statistical methods. First and foremost is the impossibility of establishing the claim that equal interval scales can be obtained in the use of these instruments. The assumption of equal interval scales is a fundamental one in all parametric methods and the failure to establish the appropriateness of this assump tion with Rorschach and Mosaic scores, and with terms de rived from them, must cause us to reject as inapplicable all such methods. As regards the assumption of normality of distribution, it must be said that Rorschach and Mosaic scores are rarely distributed normally in practice and it cannot be shown logically that they could be expected to be so distributed. Another basic criterion in the use of parametric sta tistics is the assumed independence of scores, an assumption which cannot be made in the case of Rorschach and Mosaic scores and which, if actually justified, would deprive these measures of a great part of their value in the eyes of those most convinced of their clinical validity. It Is precisely because these measures are believed to reflect molar, inter related, clinically meaningful complexes of psychological functions that they have become so widely regarded as almost indispensable clinical tools. Cronbach (19^9) has contributed what many consider 63 a definitive opinion on this and related problems. He be gins with the premise that quantitative research is not only desirable but completely necessary in the further develop ment of the projective techniques. He feels also that many of them lend themselves well to quantitative procedures if only the basic nature of the data they provide is well under stood and if proper logical and mathematical safeguards are established. Improper or unjustified selection of a partic ular statistical device can lead not only to false claims of significance but also to failure to establish an actual sig nificance . In a specific reference to work with the Rorschach, Cronbach states, "Of the differences reported in the Ror schach literature as significant at the 5 per cent level of confidence, probably the majority are due to chance." He feels that three special problems exist in research with in struments of this kind: (l) the skewness almost universally found in the distributions of scores, (2) complications in troduced by ratio and composite scores, and the like, and (3) dependence upon the total number of responses. The read er is reminded that an indication of significance could oc cur by chance once in every 20 significance tests as well as once in every 20 cases in a population for which significant differences are claimed. Because of the great number of separate scores and the large number of subgroups of subjects characteristically involved in Rorschach studies, they are thought to be more prone to inflation as regards claimed significances than many other forms of research. Monroe (19^5)* Rapaport (19^-9)* and Stephenson (1953* 1952) all maintain much the same position in regard to the unjustified use of parametric methods in the evaluation of Rorschach and other similar scores. The latter even goes so far as to advocate the use of a wholly different approach as a means of avoiding the peculiar difficulties encountered in attempting to establish significances with such measures. (Stephenson's Q Technique approach will be discussed more fully in a later section.) The above mentioned writers specifically recommend the use of the chi square test in a wide variety of applications, with the single admonition that it must not be allowed to take advantage of chance variations. It can deal, at no risk of inflating significances, with ratio scores, differ ence scores, proportions, and with composite scores such as are used in the various check lists and in summed rating scale scores, two of which are used in this study. Chi square makes no assumptions about equal scale units and it must be admitted, in the light of the foregoing discussion, that this assumption cannot be made with such data as are provided by the Rorschach Prognostic Rating Scale and the Peak Mosaic Rating Scale used in this study. Normality of distribution is not assumed in the use of the chi square test so that the skewness found in the distributions of 65 virtually all measures of this sort will not be a source of inflation or other distortion in the findings. Chi square makes, in fact, fewer assumptions than any other standard test of the significance of differences. It assumes only (l) that the variable being measured is a con tinuous one which has been artificially dichotomized, and (2) that the measures are independent of each other. The as sumption of the continuity of distributions of measures of psychodynamic and intellective factors is universally made and accepted and the assumption of independence implies lit tle more, in the context of this study, than that the meas ures were obtained from different people, i.e., that varia tion in the scores of one individual did not affect varia tion in the scores of another. It may then be said that the chi square test is the most conservative and least sensitive significance test that could be employed, that it makes the fewest possible assumptions about the nature of the distri bution under study, and that it is particularly appropriate for use with small samples. In the present study the chi square test is used as a two-tailed significance test against the null hypothesis. This, again, is regarded as the most conservative approach possible since a good case might be made for the use of a more sensitive, one-tailed test, i.e., positive increments in these scores always signify improvement and change in this one direction only might be hypothesized. The 66 distributions of difference scores from both experimental and control groups in all variables are lumped together and the point of dichotomy between improved and unimproved groups is chosen as near as possible to the grand median. Yates's correction for continuity (1937) is applied in all cases in which cell frequencies are too small. Though all primary claims of significant differences between experimental and control groups of the nature pre dicted in the hypotheses will be based on the use of chi square as the test of significance, t-tests are to be used in an ancillary manner in the interpretation of the data ob tained with the Wechsler-Bellevue scales. Precedents are everywhere at hand in the literature concerned with these instruments for the use of parametric methods of analysis. There appears to have been very little doubt in the minds of the great majority of those who have published research em ploying these scales of the appropriateness of the t-test as a test of significance, and it must be admitted that the writer has encountered no criticism of this practice. While the position of these workers may well be justi fied, and without intent to impugn the validity of their findings, one possibly cogent point may be made against the use of the t-test under these circumstances. It is felt that there is serious question as to whether or not the es sential criterion of an equal interval scale is met by these tests. It is difficult to argue, for example, that an 67 increment of 10 IQ points from an IQ of 75 to one of 85 has exactly the same meaning and implications as a like increment from 135 to 145. This argument would seem to be even more appropriately made with reference to the individual subtest scores. A further criticism of the use of the t-test in this study specifically may be made regarding the normality of distribution. While most of the distributions of scores for the various Wechsler subscales were within reasonable limits of normality, some were not. Many of them tended to be some what leptokurtic in form and consequently, to some small de gree at least, the values of t must be inflated. Whatever final decisions may be made as to the appropri ateness of this use of the t-test, it was felt in the plan ning of the present study that since at least some question of its validity existed, a consistently conservative approach would demand that it be used in an ancillary fashion only. The t-test is also to be used as a check on the equiva lence of the experimental and control groups on the initial measures of intelligence and to compare the improvement scores of the 10 highest experimental subjects with those of the 10 lowest in initial intelligence ratings. These uses, however, are again in the conservative direction for the rea son that a sensitive statistical tool is employed to detect differences which, if found, would tend to argue against substantiation of the hypotheses. CHAPTER III Results The present chapter is devoted essentially to a presen tation of the results of the chi square test of significance as applied to the distributions of difference scores, i.e., the differences between initial and final test results, for each of the testing instruments used. The individual, sub scale components of each test are examined, as well as the composite scores for the scales as a whole, the Wechsler- Bellevue quotients and the final weighted scores for the Rorschach Prognostic and Mosaic Rating Scales. The presen tation is arranged according to the hypotheses advanced for the study, each instrument intended to test one hypothesis specifically. The corollary hypothesis, in addition, is to be tested by portions of two instruments. Preceding each specific presentation of findings will appear the results of various tests for initial equivalence of the two groups in all the parameters under consideration. The chi square tests of significance are based on the combined difference or "improvement" scores of the experi mental and control groups with the grand median of this dis tribution taken as the point of dichotomy (see Table l). We are attempting by this means to detect whether a significant differential improvement in favor of the experimental group has occurred. Stated another way, the chi square test applied as it is in the present experimental design will answer the question, "Does a group of patients treated in the manner described improve more with respect to the par ticular variables being measured than an equivalent group of patients who merely reside in the hospital for the same length of time?" The concept of differential improvement is crucial. We have every reason to believe that hospitaliza tion alone, without formal treatment of any kind, can result in improvement of many kinds in emotionally disturbed indi viduals. We therefore cannot be concerned with any attempt to measure absolute improvement in an experimental group of patients; but having subtracted, as it were, the effects of hospitalization by providing a closely equivalent group of control subjects who had no treatment, and by combining the distributions of improvement scores and applying conserva tive significance tests, we can fairly say that the residual or additional improvement shown by the experimental group is attributable to the treatment given. The N in all distributions for all tests is 50. Chi squares are reported for 1 degree of freedom applicable in a 2x2 table (Walker, 1940). Columns 2 and 3 of Table 1 list the numbers in each group which fall above the point of di chotomy, i.e., those whose scores reflect more improvement than the median case in the combined distribution. Since the N for each separate group is 25 the number of cases fall ing in the lower or less improved group can easily be 70 Table 1 Significance of Differences in Wechsler-Bellevue Scores Item Point of Dichotomy Exp . Gp. Improved Con. Gp. Improved Chi Square0 VQ 2,3 19 6 13*520 PQ 5,6 15 9 2.885 IQ 3,4 18 7 9.68ob Information 0,1 18 12 3.000 Comprehension 0,1 20 5 l8.000b Digit Span 0,1 12 6 3*125 Arithmetic i —1 •\ o 11 14 0.720 Similarities 0,1 15 12 0.725 Vocabulary 0,1 16 7 6.522a Pic. Arrangement 0,1 15 12 0.725 Pic. Completion 0,1 19 8 9*742b Block Design 0,1 15 17 0.347 Object Assembly 0,1 15 11 1.282 Digit Symbol 0,1 15 16 0.085 Deterioration Quotient -2,-1 14 11 0.720 a Significant at the 5 per cent level. Significant at the 1 per cent level. CA chi square value of 3*841 is required for signifi cance at the 5 per cent level, a value of 6.635 at the 1 per cent level. 71 determined by subtraction. Hypothesis A Forms I and II of the Wechsler-Bellevue Intelligence Scale, used alternatively with each subject, have been se lected as a specific test of the hypothesis that experimen tal patients would demonstrate significant improvement in the general level of intellectual functioning. In studies such as this in which comparisons are made of changes occurring in two separate groups, it is important to know that they are as nearly equivalent as possible in all ways that might reasonably be expected to affect the fi nal results or their interpretation. This applies not only to the general characteristics that are governed by the strict controls established for these groups, but also to Initial performance on the specific tests employed. The use of difference scores as in this study minimizes the distort ing effects of unequal initial or basal scores, since we are interested not in absolute improvement but in relative change. It is nonetheless desirable and appropriate to dem onstrate reasonable equality in the two groups at the start of the experimental period. With this purpose in mind, the following comparisons were made between experimental and control groups with re gard to initial IQ scores: 72 Total Range Mean Standard Deviation Exp. Gp. 40 Control Gp. 5i 105.60 10.65 110.36 12.12 Standard Error of Difference Between Means t score ......... 3.621 1.315 It will be noted that while the total ranges are almost iden tical, means and standard deviations appear to be appreciably different. The t ratio, however, is well below the level of significance at the 5 per cent level. In this important parameter, then, the two groups are not significantly dif ferent . The theory has been advanced by some psychometricians that greater improvement in such factors as intelligence might be expected in those individuals who start out with a greater potential. We would prefer to think that whatever value electroconvulsive therapy may have would accrue as well to individuals of average or below average intelligence as to those of superior endowment, rather than selectively ben efiting those who already possess greater assets. In an at tempt to clarify this problem the following comparisons were made between the 10 highest and the 10 lowest IQ scores in the experimental group: Total Improvement, High Group............... 31 Total Improvement, Low G r o u p ............... 89 Standard Error of Difference Between Means . O.769 t score....................................... 7-5^2 It is immediately evident that our question is answered in the negative. Much greater improvement is shown by the 10 73 subjects who scored lowest on the initial measures of intel ligence. The t-test applied as a test of the significance of differences between means of the distributions of differ ence (improvement) scores is significant at the 1 per cent level of confidence. A suggested alternative to the above view may be that patients in the low group were there principally because of a deterioration in intellectual functioning as a result of their illness; that these more disturbed patients gained more from electroconvulsive therapy than did patients in the higher group; and that therefore they were more nearly able to realize their full potential on the final tests. The qualification was made earlier in regard to Hypoth esis A that the anticipated improvement in intellectual func tioning was expected to be general and that no reason was seen for predicting selective gains in particular areas or functions. Reference to Table 1 (p. 70) reveals scattered significances which, though important in themselves, lend only conditional support to the premise of general improve ment. However, two of the three important composite meas ures designed to sum up the individual's functioning over relatively wide areas, the VQ and IQ scores, show improve ment in excess of whatever may have been made by the control group significant at the 1 per cent level. Of the individu al subtest items differential improvement significant at the 1 per cent level is shown by the Comprehension and Picture 74 Completion tests, and at the 5 per cent level by the Vocabu lary test. It will be noted that when significances are found in these distributions through the use of the chi square test they are in almost all cases very highly significant, where as the remaining, non-significant items are far below the level of significance. Reference to the table of Wechsler- Bellevue difference scores for the experimental group in Ap pendix A shows that the results obtained with the chi square test tend to be confirmed by the t-test. Significant differences between pre- and post-treatment scores denoting improvement are found in 10 of the 14 sepa rate measures including all three of the important composite scores (VQ, PQ, and IQ,), all but one of which are signifi cant at the 1 per cent level. This may be compared with the results obtained for the control group (see Appendix B) in which not a single significant improvement was found. In fact, the Comprehension subtest which improved so markedly in the experimental group as shown by both significance tests is very significantly poorer (l per cent level) in this t-test comparison of initial and final scores for the control group. It is interesting to note further that all the very significant improvements indicated by the chi square test in the comparison of experimental and control groups also have unusually high t-scores in the comparison of initial and final test results with the experimental 75 group. Hypothesis B This hypothesis, concerned with improvement in ego in tegrative capacity or basic ego strength, is to be tested by means of the Rorschach Prognostic Rating Scale. In all the subscales of this instrument, as well as in the composite, weighted score given for the test as a whole, increases in scores are theoretically associated with improvement. The results of the chi square test as a test of the significance of differences in improvement scores (pre- and post-test difference scores) are found in Table 2. The dis tributions of difference scores on which these findings are based will be found in Appendix A for the experimental group and in Appendix B for the control group. Raw scores of all tests for the experimental group will be found in Appendix C, and for the control group in Appendix D. The chi square test was also applied to the distribu tions of initial scores to determine the pre-test equiva lence of the two groups. The mean weighted scores were ex actly the same (6.97). Chi square was also non-significant for this measure as well as all other variables on this scale with the single exception of m. The mean scale value of m for the experimental group was .84 and for the control group .44. Chi square shows a very significant difference between the groups in this variable. We may conclude then 76 Table 2 Significance of Differences in Prognostic Rating Scale Scores Item Point of Dichotomy Exp. Gp. Improved Con. Gp. Improved Chi c Square M 0,1 20 22 0.l49d PM -1,0 22 21 0.000d m 0,1 3 4 0.000d Shading 0, .1 16 9 3.920a Color 0, .1 14 10 1.282 Form Level 0, .1 17 6 . b 9-742 Weighted Score • 4,.5 15 10 2.000 Significant at the 5 per cent level Significant at the 1 per cent level. CA chi square value of 3*841 is required for signifi cance at the 5 per cent level, a value of 6.635 at the 1 per cent level. Sates's correction for continuity applied. 77 that there is probably significantly more anxiety of the particular sort reflected in the m response in the experi mental group initially than in the control group. However, since no significant differences as regards comparative im provement in this variable can be reported, this finding is of little immediate importance either in substantiating or in rejecting Hypothesis B. Yates's correction for continuity has been applied to the first three variables appearing in Table 2, M, FM, and m, for the reason that one or more of the cell frequencies ob tained in computing chi square were too small. Though none of these variables approaches significance even without the correction, it is felt that the basic reason which necessi tated the use of the correction is that the scale units of the instrument are too large. The M variable, for example, is scored on a scale containing only five intervals. The FM and m scales are still more restricted with only four inter vals. This implies a serious loss in the potential discrim- inability of the scales. It leads to the occurrence of many zero difference scores and consequently to insufficient cell frequencies in all too many cases. This same characteristic of the Prognostic Rating Scale may also have masked possibly significant differences which smaller scale units, and hence greater discriminability, would have revealed. This and other limitations of the scale will be discussed in more de tail in the following chapter. 78 Of the four remaining variables on the Prognostic Rat ing Scale only the form level rating shows very significant improvement over that of the control group. Improvement in the use of shading is significant at the 5 per cent level. Both the color variable and the final weighted score are well below the level of significance. Thus it may be said that such differential improvement as has occurred appears to be related to the handling of affectional needs and expecta tions and to intellectual effectiveness or efficiency. This latter finding, in addition to what it may enable us to in fer about basic ego strength, also has important implica tions for the premise of absence of organic signs adopted in the corollary hypothesis. Hypothesis C The third major hypothesis is concerned with improve ment in what has been termed affective-cognitive integration as reflected in the individual's capacity to achieve mastery in a contemporary, reality-based situation. With its empha sis on the actual, functional ability immediately available to the individual, rather than upon potentiality which is the special concern of the Prognostic Rating Scale, the Mo saic Test seemed to be a particularly appropriate instrument for the testing of the present hypothesis. In brief review of the procedure employed with this test, the rater was presented with colored photographs of the mosaic productions in pairs representing the pre- and post-tests of each subject. The designs of experimental and control patients were intermingled. He was asked to rate them on the basis of the Peak Mosaic Rating Scale and to designate one of each pair as improved. Scales were thus made out on each mosaic, the results of which will be found in Appendix C for the experimental group and in Appendix D for the control group. Three months later the same rater was again asked to separate the mosaics into improved and un improved groups and a 97 per cent agreement with the first selections was obtained. Twenty-two out of the 25 post treatment designs for the experimental group were selected as improved while the selection for the control group was exactly at the level of chance, 13 judged improved and 12 unimproved. Since, even with the use of difference scores, reason able equivalence of experimental and control groups on the initial tests in respect to the particular variables under consideration is to be desired, the distributions of initial mosaic scale scores were compared by means of the chi square test. No differences approaching significance were found in any of the six variables measured by this scale or in the final weighted scores. Though the use of parametric methods cannot be justified in this case, it was noted that mean scale scores in each variable were all within a few points of each other, as were the means of final weighted scores. 80 We may assume then that no substantial differences exist be tween the two groups in regard to initial performance on the Mosaic Test, bearing in mind also that the effects of what ever small differences are present are minimized by the use of pre- and post-test difference scores. The results of the comparison of difference or improve ment scores in the two groups by means of the chi square test are found in Table 3- It will be noted that signifi cant differences are shown at or beyond the 5 per* cent level in all the subscales which comprise this test and in the fi nal weighted score or sum, with the single exception of the color variable. The generality of differential improvement thus shown, distributed as it is throughout the scale with the exception just noted, the extremely high value of chi square obtained for the composite score summing up the scale as a whole, and the very high validity and reliability dem onstrated by the rater in discriminating between treated and untreated individuals, all add up to a convincing argument for the efficacy of the treatment in effecting changes in the areas or functions with which the present hypothesis is concerned. The specific implications of these changes as related to this hypothesis will be discussed in the follow ing chapter. Corollary Hypothesis The intent of the corollary hypothesis is to predict 81 Table 3 Significance of Differences in Mosaic Rating Scale Scores Item Point of Dichotomy Exp. Gp. Improved Con. Gp. Improved Chi Square0 Closure I —1 o 16 2 17.0l4b Articulation 0,1 16 2 l?.0l4b Complexity 0,1 9 2 5.7Ha Color i —1 o 8 8 0.000 Spoiling 0,1 13 5 5•556a Variety 0,1 12 4 5.882a Sum 1,2 19 3 20.779b aSignificant at the 5 per cent level. T_ Significant at the 1 per cent level. °A chi square value of 3-841 is required for signifi cance at the 5 per cent level, a value of 6.635 at the 1 per cent level. 82 that no decrement will be found in the experimental group following brief stimulus electroconvulsive therapy in certain selected areas of intellectual functioning which have been shown to be particularly sensitive to the effects of organic insult. The specific measures selected to test this hypoth esis are: (l) the "Don't Hold" items of the Wechsler-Belle- vue scales, (2) the Deterioration Quotient provided by these same scales, and (3) the Form Level rating provided by the Rorschach Prognostic Rating Scale. Though a large number of supposed "organic indicators" are furnished by the tests used in this study, it is felt that most of them have not shown sufficient reliability to justify their use here. This is not to imply that no ques tion exists as to the validity of the measures selected for the present purpose. The deterioration quotient has been seriously challenged by several writers and alternative uses of the subscales for the purpose of evaluating deterioration have been proposed. It is doubtful, however, that these al ternatives have received as much substantiation as the meas ure they were intended to replace. The deterioration quo tient is used here only as one of a series of measures and is intended to be interpreted only in the context of these other measures. The Form Level rating was selected for three principal reasons: 1. It represents a nearly culture-free measure of 83 intellectual capacity, depending little, if any, upon previ ously acquired knowledge. 2. It constitutes an entirely different approach to the problem in that it estimates intelligence as a function of the degree of differentiation and of precision in the ba sic perceptual-integrative processes without the aid of for mal control or structuring. It thus provides a measure hav ing a desirable degree of independence from the others which are all more or less interrelated. 3. It purports to measure or reflect ultimate potential rather than immediately available resources and so should be less affected by transient situational factors or even by deeper psychodynamic disturbances than the more conventional tests. The results of the chi square test of the significance of differences between the two distributions of difference scores for these variables will be found in Table 4. The pre-test equivalence of the two groups in these variables has been discussed previously (see pp. 71-72, 74-77) and need not be repeated here. It will be noted that only the Form Level rating shows improvement significant at the 1 per cent level over that made by the control group, though the Digit Span test approaches significance. Reference to the table of Wechsler-Bellevue difference scores for the experimental group in Appendix A shows that 84 Table 4 Significance of Differences in Measures of Deterioration Item Point of Dichotomy Exp. Gp. Improved Con. Gp. Improved Chi Square Digit Span 0,1 12 6 3-125 Arithmetic 0,1 11 14 0.720 Block Design 0,1 15 17 0.347 Digit Symbol 0,1 15 16 O.O85 Deterioration Quotient -2,-1 14 11 0.720 Form Level 0, .1 17 6 9-742a cl Significant at the 1 per cent level. chi square value of 3-841 is required for signifi cance at the 5 per cent level, a value of 6.635 at the 1 per cent level. 85 even when the t-test"*" is applied to the difference scores for this group no significant pre- and post-test differences are found in four of the five Wechsler variables, with only the Block Designs test significantly improved (1 per cent level). All scores improved at least slightly except the Digit Symbol test which showed an extremely small decline. There was an improvement in the mean DQ score though it was well below the level of significance. The weight of evidence in substantiation of the corol lary hypothesis appears to be considerable. The chi square test indicates that the treated group improved at least as much in all variables as did the untreated group and, in one case, Form Level, very significantly more. The means of all scores, with the exception of Digit Symbol, show at least small improvement following the series of treatments, with the Block Designs test showing very significant improvement. The implications of these findings will be discussed further in the next chapter. ^The use of the more sensitive t-test is probably jus tified in this case since it is used against the hypothesis. CHAPTER IV Discussion Interpretation of Results To recapitulate briefly our basic purposes, the present study was designed to test the validity of hypotheses propos ing that an experimental group of patients who received a standard series of brief stimulus electroconvulsive treat ments would show significant improvement in: (l) the general level of intellectual functioning as measured by the V/echs- ler-Bellevue Intelligence Scales, (2) ego Integrative capaci ty or basic ego potential as reflected In the Rorschach Prognostic Rating Scale, and (3) the perceptual-integrative functions and their affective components as reflected by the Mosaic Test. It is assumed in all cases that such gains, If any, must be shown to be over and above those that might be made by a control group of patients who received hospitali zation only for a similar period of time. A corollary pos tulate that treated patients would reveal no signs of organ ic deterioration as determined by a selected group of organ ic indicators was also tested. Considerable evidence has been obtained that the ex pected gains have been realized in certain areas, while in others the absence of such supporting evidence argues as strongly against the occurrence of such favorable changes. Inasmuch as four discrete hypotheses have been advanced 86 concerned with a like number of more or less distinct aspects of emotional and intellectual functioning, the discussion which follows will be arranged in these major divisions. Hypothesis A As outlined in the preceding chapter, a number of sig nificant differences were obtained indicating differential improvement in the experimental group in various parameters of the Wechsler-Bellevue scales. A detailed description of these scales, which need not be repeated here, and of the rationale underlying their use may be obtained in Wechsler's text, The Measurement of Adult Intelligence (1944). It would seem the course of poor logic, unless one wishes to adopt a "unique traits" theory of intelligence, to insist that the gains demonstrated in the experimental group are entirely specific to the tasks measured and that no im provement could have occurred in other areas with known in terrelationships, even though such presumed improvement is not shown statistically. We cannot measure intelligence per se; we can only infer its existence from certain selected behavioral end-products. There are many errors of measure ment in the best of tests, errors which, operating as chance factors, tend more strongly to cancel out actual differences than to create spurious ones. Whatever differences survive the relatively insensitive measures which conservative sta tistical analysis demands, and all the distortions inherent in the testing process, must be regarded as minimal with a 88 strong probability that additional, concomittant changes have occurred elsewhere in the matrix of interlocking func tions . It is concluded, therefore, that the hypothesis of gen eral improvement in intellectual functioning can be accepted if, in addition to the gains demonstrated by the chi square test, the significant differences between pre- and post- treatrnent performances established by means of the t-test are accepted as indicative at least of significant trends toward improvement. We do not, of course, refer to any gains in intellectual potential, but rather to improvement in the ability to realize such potential as exists due to the facilitating effects of the treatment. Results similar to these are not without precedent in the literature. Widespread though minor tendencies toward improvement on a number of the Wechsler subtests led Scherer (1951) to conclude that appreciable gains in several impor tant aspects of intellectual functioning were made by pa tients after brief stimulus therapy. Similar tendencies following transient decrements in performance were reported by Wilcox (195^0 using tests similar to a number of the Wechsler items. Summerskill (1951) reported no significant differences between pre- and post-treatment scores using the Wechsler-Bellevue Forms I and II before and within thirty minutes after treatment. Kendall's (1956) important compar ative study failed to show positive gains in the brief stimulus group as reflected in Wechsler-Bellevue scores, but it did establish that there was appreciably less cognitive disturbance than occurred with conventional electroshock and that such disturbance was more quickly reversible. Levit (1955) reported t-test significances at the 1 per cent level for improvement on the Information, Comprehension, Digit Span, and Similarities subtests of the Wechsler-Belle vue Verbal Scale in a group of 30 paranoid schizophrenics one month after the final brief stimulus treatment. In a group of 35 mixed psychotics, Brower (1951) found all Wechs ler-Bellevue subtests improved to some degree with the ex ception of Information, though no tests of statistical sig nificance were reported. Smykol (1950) gave the Wechsler-Bellevue, Form I, after the fifth and tenth brief stimulus treatments in a series to 32 female schizophrenic patients. He noted that the pre treatment scatter patterns were very much like those pre dicted for schizophrenics by Rapaport (1949). Following treatment the distribution of IQ's was more nearly normal. This finding can also be reported of the present study. The distributions of IQ's in the experimental group tended to become less widely dispersed with a decrease in the total range from 40 to 30 and in the standard deviation from 12.12 to IO.36, thus more closely approximating a normal distribu tion . A further interesting point of similarity between the present and other recent studies is provided by Howard's (1956) report on 50 schizophrenic patients who were given the Wechsler-Bellevue before brief stimulus treatment, the results of which were compared to psychiatric ratings of im provement after treatment. There was a statistically sig nificant positive relationship between Verbal Scale scores and the degree and maintenance of recovery. This may be compared to the finding in the present study (Table 3* P* 8l) that the 10 lowest IQ's in the experimental group improved very significantly more than the 10 highest as regards post treatment IQ scores. Hypothesis B As we have seen in the preceding chapter, the indica tions of favorable changes in the experimental group as re flected in the Rorschach Prognostic Rating Scale are consid ered to be far less than sufficient to confirm the hypothe sis of improved ego strength. While the majority of varia bles measured by this scale, as well as the composite score for the test as a whole, showed no changes approaching sig nificance, two of them did reveal significant changes in the predicted direction. It might be well at this point to ex amine these particular variables and to discuss the implica tions of the changes that appear to have occurred in them. The first of these, the shading variable, showed dif ferential improvement in the experimental group significant at the 5 per cent level. The basic hypothesis underlying 91 the use of shading is that the nature and quality of the subject's reactions to this physical aspect of the blots il lustrates the manner in which he manages his own affectional needs (Klopfer, et al., 195^)- It is thought to be a re sponse indicating the subject's desire or willingness to re late to objects in the environment in a more or less inti mate way, to approach the world beyond the self in the search for gratification of affectional needs. The ultimate dispo sition of these feelings, the degree of success attained by the individual in his attempts to relate, vitally affect his basic security needs. Not only the actual responses to shading, but also the subject's efforts to defend himself against the emotional impact of these areas by evasion, de nial, or other means is revealing of his characteristic ways of protecting his vital security interests. Thus these re sponses may comment upon the integrity of the basic ego de fenses and in this way provide a measure of one important aspect or constituent of ego strength. The second of these variables, the Form Level Rating, showed differential improvement in the experimental group significant at the 1 per cent level. Form Level is consid ered the most important of the various Rorschach parameters used in estimating the intellectual potential. The highest level achieved in the record is believed to give an indica tion of ultimate capacity, while the prevailing level of ef ficiency of intellectual functioning or the degree to which 92 this potential is realized is provided by the average form level. Though it is not suggested that the Rorschach is an adequate substitute for standard tests of intelligence, it is nonetheless felt that it can reveal, through the form level and other variables, the existence of unrealized po tential which cannot be detected by such tests. This is particularly true in cases of serious emotional disturbance in which the individual can realize only a small fraction of his actual capacity in any context of application. The significance of the Form Level Rating as a measure of important determinants or dimensions of basic ego strength is not limited to the assessment of Intelligence alone. It reflects the individual's ability to structure with some creativity and imagination a segment of unstructured reality and then to describe his production with reasonable care and precision. The efficacy of intellectual controls over these imaginal processes and the affects which drive them is also put to the test. These and many more affective as well as intellective ego resources contribute to this rating which, in sum, may be said to be a reasonably valid representative of the reality testing functions generally. It may be stated parenthetically that the originators (Klopfer, et al., 1951) of the Prognostic Rating Scale make no claims that the validity of the suggested interpretations of the various subscale items has been verified by direct experimental evidence. Some evidence has been obtained 93 which is thought to lend support to a few of the interpreta tions hut, for the most part, they are derived from the main body of Rorschach theory and so are supported experimentally only indirectly and in so far as this theory has been con firmed . We have only to turn our attention briefly to the areas in which no significant differences were established between experimental and control groups to appreciate the critical importance of the functions involved. These include M, per haps the most highly significant single Rorschach variable in the minds of most of its clinical users; FM, a reflection of the basic instinctual life of the individual; m as an in dicator of anxiety, as a warning signal of tension between ego and instinctual forces; and C as an indication of the mode of reaction to the whole spectrum of emotional chal lenge from the environment. It is abundantly evident that the absence of significant change in these extensive and crucially Important areas renders it impossible to accept the hypothesis that general or fundamental changes have oc curred in basic ego strength or ego integrative capacity as evaluated by this instrument. A technical criticism was made earlier (see p. 78) which may in part account for the absence of significant findings where there are seemingly adequate theoretical grounds for expecting them. The M variable is scored on a scale containing five intervals, FM and m on scales 94 containing only four. The effect of these relatively non discriminating measures may well be to minimize, if not to erase completely, actual differences that might have oc curred. Reference to the table of difference scores for the experimental group in Appendix A will show that l6 zero dif ference scores were obtained for the M variable, 21 for PM, and 18 for m. This applies to three of the four variables in which no significant differences were found. The theoretical implication of these zero difference scores must be that the responses upon which the scores are based were repeated exactly as given originally in so far as the scale is able to evaluate them. However, both logical inference and our previous experience with such phenomena deny that these responses could have been exact replicas, precisely the same in all details after a three-week period of time during which the subjects experienced six electro- convulsive treatments. This is not to say that the presumed changes that must have occurred, if only in small and unimportant details, would have either statistical significance or clinical mean ing. Statistical theory would predict that such scores would produce some variant of the normal distribution, how ever skewed or otherwise atypical. The appearance of these scores is that of a few chance variations from a strong cen tral tendency of zero. Scores that actually reflect natural phenomena should be distributed somehow, they should not fall 95 at a point, as in this case. If a group of scores were ob tained that were sufficiently dispersed to give us what might fairly be termed a statistically manipulable distribution, and if the obtained variance was then merely chance variance, appropriate statistical analysis would so indicate. The point is that the differences should have appeared. The findings reported by Sheehan, Frederick, Rosewar and Spiegelman (195^) provide at least suggestive evidence that when there is greater freedom for the M, FM, and m scores to vary, quite different results may be obtained. Using merely the number of responses in each category as a means of predicting the successful outcome of psychotherapy in a group of stutterers, these workers found that M, FM, and m were significantly associated with improvement at the , 2 per cent level. Though the P values for the other meas ures were all lower than this, the Prognostic Rating Scale scores calculated in the manner prescribed for the scale yielded a t-score significant at the 1 per cent level. Here is at least one instance of the successful use of this in strument in predicting improvement without the artificial strictures imposed by four- and five-interval scales. The list of studies purporting to question the validity or the predictive value of the Rorschach itself is a long one and there is no intention of reviewing here either the results of these studies or the cogent criticisms that have been made of most of them. However, since it so closely parallels the results obtained in the present study, both in the intellective and personality areas, a further reference to that of Levit (1955) mentioned earlier (see p. 89) may be relevant. He reported substantial changes in intellectual functioning as reflected in the Wechsler-Bellevue Verbal Scale but he found no such improvement in his patients' Ror schach performances. He concludes that there is no evidence that electroconvulsive therapy produced changes of statisti cal significance in personality traits. Zubin, Eron, and Sultan (1956) devised a system of per ceptual and content scales for the Rorschach similar in many respects to the PRS. Statistical studies of validity and reliability showed only those concerned with content to be of value. The author concludes that the Rorschach should be used as a systematic, controlled interview whose principal value lies in content analysis, a conclusion which has gained some currency in the field. Though Piotrowski1s work (1950, 1955) is open to some criticism on statistical grounds (Cronbach, 19^9), it is still widely accepted by users of the Rorschach. He reported that the use of his alpha formula enabled him to predict im provement in a group of schizophrenic patients with 87 per cent accuracy. Among the more significant changes observed were improvement in the number and quality of movement re sponses, increase in the number and percentage of FC re sponses, and a rise in Form Level. He interprets these as 97 suggestive of generally "improved integrating capacity," a claim which cannot be substantiated in the present study. Mindess (1953) reported high (Pearson r = .81) correla tions between Prognostic Rating Scale scores and indices of improvement in a mixed group of psychotic and neurotic pa tients in psychotherapy. In this study the Form Level Rat ing was the most accurately predictive item with the M scale next in importance. Kirkner, Wisham, and Geidt (1953) found the M, m, and Shading scales significant in predicting im provement in psychotherapy, but it is interesting to note that when raw scores only were used, the FM variable was al so found to be significant in addition to these three. Us ing the phi coefficient, they reported a correlation of .67 between PRS weighted score and the criteria of improvement. Another successful use of the PRS was reported by Johnson (1953) who obtained a chi square significant at the 1 per cent level between the PRS and indices of improvement in a group of children in play therapy. She noted also that the PRS scores correlated well with the Binet-Raven z-difference scores. Both Symonds (1955)* and Lundin and Schpoont (1953) agree with the observation made by Zubin, et al. (1956) that content may be of as great or even greater importance in making judgments based on the Rorschach than the formal de terminants. In the former study psychologists were asked to make certain judgments and predictions which were then 98 checked against a very thorough and extensive psychiatric report. When determinants alone were used 59 per cent of correct judgments were obtained, which rose to 74 per cent when the judgments were based on content. Lundin and Schpoont also support the concept apparently first suggested by Reiman (1950) that the PRS basically is a measure of "un used ego strength." They feel that it is "more a predictor than a reflector," and continue, "The scale appears to be more sensitive to latent or unconscious potential than to overt and more transient behavior." In direct contrast to the present study, they found the C scale to have the great est predictive value and they noted no significant changes in Form Level. In a later study, Zubin (1954) extended his analysis of the efficacy of the various interpretive approaches to the Rorschach and reached the following conclusions: (l) Global evaluations of the Rorschach seem to work well when the Ror schach worker and the clinician work closely together; (2) Atomistic evaluation, as well as global, of the content of Rorschach protocols seem to work; (3) Atomistic analysis of perceptual factors is a failure; (4) Factor analysis of atomistic scores of both the perceptual and content variety seem to work. He concludes with the suggestion that the best hypothesis to explain these four facts is that the Ror schach is essentially an interview whose evaluation must be accomplished primarily in terms of its content, with 99 appropriate rating scales as important aid in the process of analysis. Despite the continuing appearance of such studies as the recent one of Mintz, Schmeidler, and Bristol (1956);, in which judges were able to discriminate between improved and unimproved patients in psychoanalysis at the .003 level of confidence using only the Rorschach test, there appears to be growing doubt that any form of treatment can effect truly fundamental changes in the basic personality variables which this test presumably measures. In reporting her highly suc cessful predictive results, Mintz ascribes most of the change to improved defenses and other adjustive mechanisms including repression, to better tolerance of stress, im proved intellectual control, and the like, rather than to alterations in the basic character structure. In his work on the evaluation of the effects of insulin shock therapy, Piotrowski (1938* 19^1) some years ago made the comment that the treatments had ". . .a curative effect on the secondary or compensatory personality changes but not on the primary personality disorder." More recently, Brower (1951) has stated a similar view. He enumerated a number of gains following electroconvulsive therapy in both intellec tual control and efficiency, less constriction, greater sen sitivity to and more effective management of external stimuli generally. Despite these gains, however, of great impor tance in their own right, he concluded that the composite 100 structure of the personality in the post-shock state ap peared to be essentially the same as in the pre-shock condi tion . In the light of the present results, the writer is con strained to accept this view. It is, perhaps, merely the course of naivete to expect a crude, physical stimulus to effect the incomprehensibly complex and delicate restructur ing which must underlie any major changes in personality. It may not be so surprising then that this instrument, which purports to reflect only the more constant, unvarying par ameters of personality, those that are relatively unaffected by contemporary events, should register none or little change in treatment. In addition to whatever specific ex perimental evidence we now have, this view gains further support from the radically different results that are ob tained when an evaluatory instrument is employed whose chief purpose is to assess just those immediate changes in effec tiveness of application which the Prognostic Rating Scale is designed to discount. Hypothesis C It will be recalled that significant differential im provement was shown for the experimental group in all vari ables of the Peak Mosaic Rating Scale except that evaluating the use of color. Some idea as to the specific meaning of the individual subscales may be obtained from the instruc tions to the raters in the use of the scale as formulated by 101 the author reproduced in Appendix G. For a more detailed description, reference is made to Peak's original work on the subject (1953). An illustration of the use of the scale with an experimental subject is given in Appendix F. Sample mosaics made by both experimental and control subjects will be found in Appendix E. The author made no attempt in devising the scale to ob tain independent measures of discrete functions in the pro duction of mosaics and no assumption of independence among these variables is made. In fact, they have been shown to be quite highly correlated. The product moment coefficients of intercorrelation ranged from .79 between Spoiling and Va riety to -94 between Closure and Articulation (1953* P> 48). Something approaching this degree of intercorrelation was, in fact, anticipated when the variables were chosen. The fi nal selection was made because, in the author's experience, these particular parameters seemed to cover most completely the several important aspects of gestalt quality as re flected in mosaic designs. In view of the very high discriminability between treated and -untreated patients demonstrated in the present study by this particular rater, the question may well be raised as to whether or not this may represent a highly de veloped, idiosyncratic skill which might be difficult or im possible to communicate to others. The answer to this ques tion is found in the intercorrelations which were obtained 102 among several independent raters who had little or no previ ous knowledge of the test and who were given only a reason able minimum of instruction in the use of the scales. These correlations ranged from -71 to .81 and, using the Spearman- Brown formula, an over-all reliability coefficient of .92 was obtained. The discriminability of the scale between schizophren ics and controls, and between controls and organics, was well established with all subscale scores and the summated scores yielding t ratios significant at or beyond the 1 per cent level. Discriminability between organics and schizophrenics was poor, neither subscale scores nor summated scores ap proaching significance (1953* PP• 50-51)- The high reliabil ity of the rating scale as a whole was also confirmed and compared to that of a number of other suggested variables which have been used in the evaluation of mosaic productions. It was found to be among the highest tested in this impor tant characteristic. P values of t ratios obtained for the differences in rating two matched groups of psychotics, or ganics, and normal controls did not approach significance, the highest being .30 (1953# P- 63). Considering the rather impressive reliability and valid ity of the scale established as outlined above, the rater's achievement in the present study may be seen as not incon sistent with his past performances. Nor is this the only such scale to receive favorable mention in the literature on the Mosaic Test. Such pioneers in the use of the test as Himmelweit and Eysenck (19^5)* Diamond and Schmale (1944), and Wertham and Golden (1941) all reported varying degrees of success in their early attempts to introduce formal scor ing or rating procedures in the use of the test. Prom the outset, Lowenfeld herself (1949, 1952) and other recognized authorities such as Colm (1948) have been and continue to be skeptical of any efforts aimed at the establishment of formal scoring categories or signs. The principal cause of their concern is that the significant molar relationships and ges talt properties of the designs may be obscured or lost if undue emphasis were to be placed on the more atomistic meas ures. Despite these fears, rating scales or formalized rat ing procedures continue to be devised, most of them obvious ly seeking to avoid the pitfalls inherent in atomistic eval uations . Bowen (1954) has developed a serial testing procedure which she feels shares some of the advantages of testing the limits on the Rorschach, and through which she has been able to make a number of valid predictions as to progress in therapy and reaction in social situations. A less formal ized sort of procedure is used by Stewart and Streiter (1957) in the assessment of developmental level in children and in making predictions of behavior of a general, qualita tive nature. Using the scorned atomistic approach, Wideman (1953, 1955) reported a validation study in which 32 of 39 104 discrete scoring categories showed significant differences between a normal and one or more pathological groups. Of particular interest is the fact that all the color variables tested were significant at or beyond the 5 per cent level, a finding in direct contrast to the results obtained in the present study. As is to be expected, the recent literature provides at least one example of a study purporting to show strongly negative results. Lewin (1956) reported almost no signifi cant discriminations among normal and pathological groups in a long list of selected indicators, and from this concluded that the test was of no value in its present form. Typical of the indicators he tested are the use of 40 per cent or more red pieces, the construction of a well defined cross, the construction of a downward-pointing arrow, and the use of super-imposed pieces. It is doubtful that anyone would now claim any great discriminability for these or any other atomistic measures as such, devoid of context or known rele vance, but he represents them as being in common use today. Each discrete sign was treated individually and no attempt was made to combine them into clinically meaningful groups or patterns. Apart from certain statistical and methodolog ical criticisms that may be made, it is obvious that a sweep ing generalization has been made on the basis of dolefully inadequate evidence. It is ardently to be hoped that the regrettable early history of the attempts at validating the 105 Rorschach by counting responses is not to be repeated. The failure of the Peak scale to discriminate between schizophrenic and organic groups is somewhat surprising in view of both its fine discriminability in other applications, and of the successful use of the test almost from its incep tion for just such a diagnostic purpose. Soon after its in troduction into general use in this country, such workers as Colm (1948), Wertham and Golden (1941), Diamond and Schmale (1944), Kerr (1939)? and others reported favorably on this aspect of the test, often, in fact, singling it out as a particularly valuable and unique asset. More recently, Wide- man (1953? 1955)? Rioch (1954), Maher and Martin (1954), Rei man (1950), Bowen (1954), and Stewart and Streiter (1957) have all confirmed these earlier reports on its value in the diagnosis of brain damage in a wide variety of applications, ranging from work with school children (Stewart and Streiter, 1957) to evaluating the effects of cerebro-arteriosclerosis in the aged (Maher and Martin, 1954). This failure of the Peak scale is, of course, of no direct significance in the present study but it is to be hoped that this one obvious defect in an otherwise sensitive and discriminating evalua- tory method can be remedied. As is also true to a considerable extent of the Ror schach Prognostic Rating Scale, the specific clinical or theoretical Implications of the various subscales of this instrument are as yet unconfirmed by experimental evidence. 106 The high degree of intereorrelation among them would seem to mitigate against establishing defined, specific meanings in any case. The common factors or areas of commonality shared by many of the subscales evidently contribute most of the total variance ascribed to them, with specific factors of relatively minor importance. The instrument might be sche matized as a group of closely spaced, converging lines rep resenting slight variations in approach to a common goal. It would seem, therefore, that recourse must be made to the few principles of theory now extant that relate particu larly to the Mosaic Test, and to the larger body of general projective theory in the effort to gain insight into the na ture of the functions reflected in mosaic productions. One of the first such principles to be formulated is attributed to Reiman (1950) and was early subscribed to by Lowenfeld herself {±9^9, 1952). The general thesis is that the Mosaic Test measures, not basic potential or other relatively con stant parameters of personality as the Rorschach is thought to do, but only those ego resources that are immediately available for use by the subject in a concrete, reality- oriented situation. Thus we have at the outset established a basic limitation as well as a special utility in this de vice. It presents the subject with a novel situation which demands some ingenuity and enterprise if a successful solu tion is to be brought about. It requires effort in the im mediate present and only those resources which the individual 107 can readily summon to practical use are of consequence. The test is virtually culture-free in the usual sense, though Stewart and Leland (1952) have shown some interesting and suggestive cultural variations. It depends little, if any, upon acquired knowledge or prior experience of any sort be yond the most elementary perception and retention of famil iar gestalts. It requires the subject to deal with concrete materials of a defined nature, to set a goal, and it then tests his ability to realize that goal. The fact that the test materials are presented in a limited range of colors and in five angular, geometric shapes is cited by Peak (1953) as a limitation of the test. In the writer's opinion, these characteristics might also be re garded as important elements in the challenge provided by this instrument, as one of its tests of ingenuity and of the subject's ability to cope with reality. This bit of reality is presented to him, as reality always is, on its own terms and not as the subject might wish it to be. The alert, adaptable individual should be expected to modify both goals and methods to fit the circumstances. The added difficulty in constructing particular kinds of designs with these limi tations as to color and shape is seen, therefore, as con tributing to the test's discriminability. As mentioned earlier, the interpretation of mosaics leans heavily on general projective theory, and in particu lar upon Rorschach theory. A number of writers have 108 elaborated upon the similarities between the two tests in basic rationale and in interpretive implications, among them Bowen (1954), Reiman (1950)? Zucker (1950), and Colm (1948). The latter1s comments appear to be representative: Rorschach and Mosaic findings correspond in most cases. One test usually confirms the findings of the other. . . . Mosaic and Rorschach interpretations are based on common elements in the test projections which express the inner structure of the personality difficulties . . . [but whereas] . . . the Rorschach provides an opportunity to see the personality in reaction to complex internal and emotional stimuli, the Mosaic provides a greater opportun ity to observe in a quick and direct way the personality in spontaneous action. The writer1s personal experience with the instrument confirms this view of the two tests as related, adjunctive approaches to the evaluation of personality functioning. As their end-results differ in emphasis upon various aspects of this functioning, a useful reciprocity in interpretive im plication is retained. We have seen that one of the earliest and best substan tiated uses of the Mosaic Test was in the diagnosis of schiz ophrenia. It has long been regarded as of special, if not unique, value in this regard, which is one of the principal reasons for its use here. In effect the task of the rater in the present study was to decide which one of a pair of de signs presented to him was more characteristic of the pro ductions of schizophrenic subjects as he conceived them to be, and which was closer to the normative or typical produc tion. However small the differences might be, he was asked to select the better mosaic. The results strongly confirm 109 the value of the test in making such discriminations between two often closely spaced points on the continuum of schizo phrenic productions. We have, however, no indication at all of the degree or amount of improvement in these patients. The nature of these data does not justify the use of quanti tative measures and it was this fact, of course, which neces sitated the use of the relatively insensitive chi square test. What we do know at a high level of confidence is that the treated patients made appreciable gains over and above any that might have been made by the untreated patients as a result of hospitalization only; that these changes were re flected in their mosaic productions; and that this particu lar rater was able to discriminate between improved and unim proved productions with a high degree of accuracy and reli ability. We have now to consider the fact that strongly positive changes have been demonstrated as regards contemporary lev els of functioning, as opposed to substantially negative re sults when the attempt was made to show gains in basic ego resources. As was mentioned earlier (see pp. 99~100), the view that electroconvulsive therapy may bring about impor tant functional improvement of many kinds without change in basic character structure or ego potential is not a new one. It would seem that the striking contrast between Mosaic and Rorschach results provides strong evidence in support of this view. It may be proposed, then, that the net effects 110 of this form of treatment are to facilitate the use of such intellectual and characterological resources as the patient has, to make them more readily accessible when needed, to render him more effective in his day-to-day functioning. The writer's own experience with these patients suggests that this increased effectiveness of functioning applies as well to the psychotherapeutic as to the practical life situ ation . Corollary Hypothesis The direct evidence in support of the corollary hypoth esis is derived from the four "Don't Hold" scales and the Deterioration Quotient provided by the Wechsler-Bellevue, and the Form Level Rating from the Rorschach Prognostic Rat ing Scale. It is also felt that supportive evidence is fur nished by the context in which these particular scores were obtained, a context of extensive and statistically signifi cant improvement in a wide variety of intellectual and per sonality functions. The specific implications of the individual subscales which make up the index of deterioration may be found else where (Wechsler, 1944). The final selection of tests was made by Wechsler on the basis of their discriminability among the various age levels and between known organic and normal groups. Those which discriminated best were placed in the "Don't Hold" group and those which discriminated least, i.e., which demonstrated the least sensitivity to the Ill deteriorative effects of age or trauma, were retained in the "Hold" group as measures of former or maximal capacity. Though this use of the Wechsler-Bellevue has been criticized on both theoretical and empirical grounds (Marks, 1953; Jackson, 1955; Cohen, 1955; Jastak, 1953; Gervitz, 1952), it still finds wide clinical acceptance particularly when per formance on the scales is interpreted qualitatively and in association with other measures of personality as well as intellectual functioning. In the present study the attempt is made to approximate this sort of clinical approach to the interpretation of these findings. In addition to the absence of any significant differ ences between treated and untreated patients (see Table 4, p. 84), reference to the table of Wechsler-Bellevue differ ence scores for the experimental group in Appendix A shows that no significant negative differences are indicated by the t-test between pre- and post-treatment scores. In fact, one of the tests generally regarded as among the most sen sitive and informative in the assessment of brain damage, the Block Design test, was very significantly improved. One small decrement was observed, that occurring on the Digit Symbol test, but it was so small as obviously to be a chance variation. These findings may be compared to the differences between pre- and post-test scores for the control group (see Appendix B). Three such minor, non-significant decrements occurred in this group on the Digit Span and Vocabulary 112 tests, and on the VQ. A fourth decrement on the Comprehen sion test was significant at the 5 per cent level. Summing up the evidence supporting the postulate of ab sence of intellectual deficit resulting from shock treatment, we may say that no post-treatment differences occurred be tween experimental and control groups; no post-treatment decrements in performance were shown in the experimental group and, in fact, one very significant improvement oc curred. These findings are contrasted with three minor dec rements in the control group and a fourth which was signifi cant at the 5 pe** cent level. Finally, these results are seen in the context of extensive, statistically significant improvement in the intellective area generally, as well as in some critically important parameters of personality func tioning. Viewed in this light, it seems highly improbable, unless one wishes to adopt some form of unique traits theory regarding these functions, that any appreciable deficit can have occurred. A number of studies whose implications appear to be generally consistent with those reported here have been re viewed in some detail earlier (see pp. 28-42, 101-102). These studies, for the most part, report varying degrees of improvement in the intellective functions under brief stimu lus therapy or, at worst, no indications of appreciable im pairment beyond the immediate, post-convulsive recovery pe riod. Typical of such studies are those of Scherer (1955)* 113 Wilcox (1954), Summerskill (1951), Levlt (1955), Brower (1951), Smykol (1950), and Perlson (1945). The latter re ported no indications of intellectual or emotional trauma in a schizophrenic patient who had received 248 electroconvul- sive treatments I Zirkle (1956) reported profound interfer ence with learning ability in a variety of tasks immediately following the convulsion. He found it to be greater for ab stract than for concrete materials. However, after a recov ery period lasting from two to three hours, patients ap peared to have regained their former levels without deficit. The Rorschach Form Level Rating was the one measure of those selected to test the hypothesis of absence of impair ment to show differential improvement significant at the 1 per cent level. Of all the commonly used Rorschach variables, it is thought to be the most important index of the level and quality of intellectual functioning. It is relatively culture-free, depending in only very minor degree upon prior learning. In common with the other variables which comprise the Prognostic Rating Scale, it is thought to reflect basic ego resources of certain kinds but it can also be used to obtain a valid index of the subject's present level of func tioning. The highest Form Level Rating achieved by the sub ject in a given record is thought to be an indication of the ultimate capacity he possesses, whether functional or not, whereas the average of the ratings is believed to reflect what he can do with these resources in the immediate present. 114 The Form Level Rating used herein is the average of the in dividual ratings corrected in a few cases for certain char acteristics of the record (extreme discrepancies, weakening specifications). In the present application it would seem that our at tention should be directed to the second of these implica tions in regard to improvement in Form Level. We have been unable to show general improvement in basic ego resources as postulated in Hypothesis B, and in the discussion of results relevant to this hypothesis (see pp. 101-114), a rationale was suggested for this absence of positive findings. The correlative evidence furnished by the results obtained with the Mosaic Test indicates that we should view this rise in Form Level as evidence of the facilitating or potentiating effects of treatment upon such resources as the patient has in this area, and not as gains in the fundamental capacities themselves. We may then say that the treated patients not only failed to show a decrement, but also revealed a very signifi cant differential improvement in the ability with which they were able to function in the critically important areas of perceptual-cognitive integration and effective control of imaginal processes in the interest of competent reality test ing. Such a finding would appear to be quite inconsistent with one of cerebral trauma. The work of several writers, the majority of whom have 115 reported favorably on the use of Form Level ratings in con junction with other measures either as prognostic indices or as measures of improvement under various therapeutic condi tions, has been reviewed in previous sections (see pp. 28-33, 95-98). The recent work of Spiegelman (1956) is of particu lar relevance to the present discussion. He found a good correlation (Pearson r = -55) significant at the 1 per cent level between the Wechsler-Bellevue VQ and average Form Level which he used as an indication of the general level of in tellectual functioning. He also reported a correlation of .52 significant at the 1 per cent level between the highest Form Level in the record, which was used as an indication of potential intelligence, and a composite Wechsler-Bellevue measure designed to reflect the same basic characteristic. In this we have direct experimental evidence of the validity of the twin concepts of basic potentiality and present func tional level as represented in various aspects of the Form Level Rating. Beck (1932), Ford (1946), and more recently, Motoaki, Tomita, and Yumato (1957) have all supported the concept of Form Level as a reliable index of the functional level of intelligence in a wide spectrum of patient populations rang ing from mentally defective children (Beck), through normal children (Ford), to adults in a cultural context widely dis parate from our own (Motoaki, et al.). Malish, Hanlon, and Kurland (1957), using a Q-sort technique, recently reported 116 that the only Rorschach variable to improve following insulin coma therapy was the F+ per cent, closely related to the Form Level Rating as used in this study. To end our discussion in a minor key, Taulber (1955) reported in a recent study that F+ per cent specifically was not validly correlated with intellectual functioning as as sessed by the Wechsler-Bellevue. However, he suggested in explanation of this finding that the degree of emotional dis turbance present in the group of schizophrenic patients he studied affected Rorschach performance more than it did that on the formal test of intelligence. This finding is not in consistent with the writer's own observations with similar patients. Patients in obviously poor reality contact often did surprisingly well in the structured situation provided by the Wechsler-Bellevue and as poorly when the structuring elements were no longer present as in the Rorschach inter view. Summary Evidence has been presented which is believed to con firm the hypothesis of general improvement in the level of intellectual functioning. Chi square tests show very sig nificant differential improvement in the experimental group on the Wechsler-Bellevue VQ, IQ, Comprehension, and Picture Completion scales, and significant improvement on the Vocab ulary scale. Comparisons of the pre- and post-treatment scores for the experimental group by means of the t-test 117 show very significant improvement in eight of the fourteen measures, including all three of the important composite scores, and significant improvement in two additional vari ables. These findings are contrasted with the performance of the control group in which not a single post-test improve ment was found, with one variable significantly poorer. It is felt that, while the use of t-tests with these distribu tions cannot be fully justified, they may at least be ac cepted as confirming significant trends toward improvement. In view of the generally accepted concept of the global na ture of intelligence itself, and of the high intercorrela tions among these tests indicating large areas of commonali ty, it is felt that the results justifying the claim of gen eral improvement in the level of functioning as a facilitat ing or potentiating effect of the treatment. The evidence in support of Hypothesis B, postulating gains in basic ego strength, is limited to the demonstration of very significant improvement in the Form Level Rating and significant improvement in the shading variable of the Ror schach Prognostic Rating Scale. Though these measures are believed to constitute important indices of improvement in several fundamental aspects of personality functioning, the supportive evidence they furnish is considered to be insuffi cient to confirm the hypothesis. A technical limitation of the rating scale, in that it provides too few scale inter vals for several of the variables, is discussed as a possible 118 reason for the failure to obtain more significant differ- x ences. Interpreted in the light of results obtained with the Mosaic Test, with its emphasis on contemporary levels of functioning rather than on potentiality, it is concluded that electroconvulsive treatment does not effect actual gains in the basic ego resources. The results of the analysis of Mosaic productions by means of the Peak Mosaic Rating Scale, which demonstrated significant or very significant differential improvement in the experimental group in all but one of the seven variables of this instrument, strongly confirm the hypothesis of im proved affective-cognitive integration as related to the mastery of contemporary reality situations. The rater's performance with this test showed both highly accurate dis crimination between treated and untreated patients and very high re-test reliability after an interval of three months. It is concluded that, while electroconvulsive treatment adds nothing to the basic store of ego resources, it does render them more readily available to the subject for practical use in the immediate present. It is believed that this princi ple applies as well to the psychotherapeutic as to the test or practical life situations. The evidence offered in support of the corollary hypoth esis of no intellectual deterioration in the treated patients was derived from the four "Don't Hold" items of the Wechsler- Bellevue, the deterioration quotient also from this 119 instrument, and the Form Level Rating furnished by the Prog nostic Rating Scale. Statistical evaluation was carried out in two ways: (l) Comparison of the distributions of differ ence or "improvement" scores of the experimental and control groups by means of the chi square test, and (2) comparison of pre- and post-treatment scores for the experimental group by means of t-tests. The chi square test revealed no sig nificant differences between groups except for the Form Level Rating in which very significant differential improve ment was demonstrated for the experimental group. The t- tests showed no significant negative differences between pre- and post-treatment scores for this group, with one variable very significantly improved. While only a few of the many possible measures of deterioration were used, it is felt that the general context of improvement in a wide variety of intellectual and personality functions in which these scores were obtained permits the inference that no appreciable de terioration in intellectual functioning occurred as a result of treatment. Limitations of the Study Size of Sample The first of several cogent criticisms that might be made of the present study, and one which can be applied to a regrettably large proportion of psychological research in the clinical area, relates to the smallness of the samples used 120 to study phenomena which are both exceedingly complex and difficult to assess. The N for both experimental and control groups was 25, near the minimum for practicable statistical evaluation. If larger N's had been possible, however, so high a degree of homogeneity in the groups would not have been so essential and many of the controls might have been relaxed. Other nosological groups might have been studied, for example, with the result that such findings as were then obtained would have been more broadly applicable. Of the first importance is the consideration that larg er N's would undoubtedly have provided more nearly normal distributions of scores so that, at least with the Wechsler- Bellevue scores and possibly with some of the other measures as well, more sensitive parametric statistics could have been employed. Many statistical penalties are incurred in the use of very small samples which effectively remove all chance of attaining desired significance levels, penalties which could obviously be avoided if the use of larger N's were possible. An example of this in the present study was the necessity of applying Yates's correction for insufficient cell frequencies in calculating chi square for several of the Prognostic Rating Scale variables. As a general prac tice, it is probably better to obtain large samples in which truly chance variations can cancel each other out than to attempt to eliminate such variation through the use of elab orate control devices. 121 Lack of Anonymity In this study the examiner was always aware of whether the individual being tested was an experimental or a control patient and whether the particular tests were the subject's first or second examination. This obviously constitutes a source of bias in the administration of the tests, but it poses problems that appear to be almost insoluble. In the normal testing session the great majority of patients make frequent personal references, allusions to their hospital experiences, and the like, which cannot fail to inform the examiner of their treatment status. If test protocols, par ticularly Rorschach records, are to be submitted to judges, they must, in fact, be carefully edited to keep such cues from appearing. The only possible corrective would appear to be to have patients designated for testing at a particu lar time by a third party who would then enter into a con spiracy of silence with them as regards the subject of their treatment status. However, this approach could not be ex pected to succeed with patients who were seriously disturbed or in poor contact, as many of them would be. It would im pose a difficult task even on patients in good contact and undoubtedly many slips and revealing chance remarks would occur with the most co-operative patients. It is ques tionable also whether this artificial restriction on his be havior in the testing situation would not have inhibiting or other adverse effects on the test responses themselves and 122 thus Introduce another source of bias. There are, in addi tion, many subtle cues, difficult or impossible for the pa tient to conceal however he tried, which would not escape the notice of the experienced examiner. The minimization (though not the elimination) of this possible source of er ror is discussed in the following section. Restricted Applicability of Findings The degree of homogeneity obtaining in the patient groups used in this study is thought to be one of its princi pal merits. The detailed controls established for these pa tients have eliminated a great number of the contaminating factors which are too often left completely uncontrolled in such studies, a fact which probably accounts in large part for the absence of any clearly indicated results in so many of them. Nevertheless, such controls, by their very nature, restrict the generality with which the findings may be said to apply to the universe which is being studied through (hopefully) representative samples. In the present study, for example, no females were included, the age range was se verely restricted, only first incidence schizophrenics were used, et cetera. While good and sufficient reasons are giv en for all these exclusions, it can fairly be said that the results presented herein are applicable to only a narrow sector of the total population of potential candidates for brief stimulus electroconvulsive therapy. The use of much larger samples, which would make possible the relaxation of 123 some of the controls, would appear to be the only acceptable solution, a solution which was not feasible in the present work. Further Analysis of Data The testing instruments used in this study supply a wealth of data which implies an almost endless variety of analytic approaches. The permutations and combinations of raw scores, composite scores, ratios, considerations of con tent, of approach, of sequence, and the like in the Ror schach alone, which find some precedent in the literature, would exceed the scope of many such studies as this. It may well be that the analysis of other groups of variables would have proven more fruitful, and it is almost certain that further analysis of other aspects of the data would add to our understanding of the effects of electroconvulsive thera py. However, some selection had to be made and the present one, limited though it undeniably is, was obtained not by diligent scrutiny of all possible sources of significant dif ferences, but by considerations of theory and of precedent which suggested that these measures specifically should be particularly appropriate for the testing of the specific hy potheses proposed. The suitability of the instruments thus chosen is attested to, in fair measure, by the results ob tained . 124 Suggested Improvements Refinements in the Present Design A number of suggestions for improving the present study are implicit in the preceding discussion of its limitations. The value of larger samples in assuring more representative groups, in reducing the number of essential controls, in permitting greater statistical freedom and more general ap plicability of results need not be further emphasized here. In fact, the emphasis may be needed in the other direction. Unless much larger samples are obtained than are generally employed in clinical research, no other means than strict controls can avoid the contaminating factors which appear to have clouded or canceled out conclusive findings in so many studies. Two relatively minor but nonetheless important technical improvements in the study can be suggested immediately. The first of these is the use of the Wechsler Adult Intelligence Scale (WAIS) in place of the older Form I. This scale was not available for use when work on the present study was be gun. It is said to possess several advantages over Form I, including greater reliability and higher correlation with Form II. Standardization of the test was more thorough and better sampling techniques provided more representative pop ulations on which to base the norms. Administrative proced ures are more carefully controlled. On the whole, it ap pears to be a more sophisticated instrument in whose design 125 can be seen the fruits of experience with its predecessors. The second of these technical improvements was dis cussed in some detail earlier (see pp. 75-76, 93~95) and will receive only brief mention here. It is concerned with the small number of scale intervals in the first three scales of the Prognostic Hating Scale. The effect of the arbitrary restriction of these scores to vary is to produce numerous zero difference scores when two performances are compared. Since this is not a logical expectation, these zero differ ence scores are believed, in the main, to be artifactual and not real differences. It would seem to be a simple matter to modify these scales so as to provide a larger number of scale units. Greater sensitivity might be achieved merely by using raw scores in place of the weighted scores proposed by the authors. Assuming a function of some importance for the weighted score in the first place, the best course would seem to be some modification of the process of weighting them. Ideally a study such as this one should be carried out by a team of workers, each of whom had certain specialized functions, and none of whom had complete knowledge of all phases of the experiment. The activity should be headed by a project director or co-ordinator who alone made the neces sary in-hospital contacts with patients and who retained knowledge of their physical management and treatment status. This individual would make all necessary arrangements with 126 the psychiatric staff. He would designate all patients for testing, mixing experimental and control subjects in random fashion, and he should make some attempt to enlist their co operation in concealing their treatment status. A number of examiners might be used but the same examiner should do both tests on a single patient. Examiners should be assigned ap proximately equal proportions of control and experimental patients. Examiners should have no association with the hospital or with the project except in fulfilling this test ing function. All scoring should be done by judges who have no other connection with the project. They would, of course, be un aware of whether a particular protocol was that of an exper imental or control subject, a pre- or a post-test. The project director would edit all records to remove any possi ble extraneous clues as to these facts. At least with the Rorschach and Mosaic records, scoring should be done by a panel of judges rather than a single one. Our experience in dealing with problems of the nature of those presented by the scoring or other evaluation of projective tests shows rather convincingly that pooled judgments in such matters are both more reliable and more nearly valid than individual judgments. Analysis and collation of the data could then be completed by the director. Methodological Criticisms Early in the planning of this study the decision was made to employ only the most conservative of statistical treatments in order to avoid the creation of spurious sig nificances or the inflation of existing ones so commonly seen when inappropriate measures are applied. Criticisms on this score are made of some of the most prominent workers in the field and are cited as the reasons for the failure of other workers to replicate their results. The course of conservatism carried to such a degree, however, leaves the research worker with only the crudest and most insensitive of devices with which to evaluate his data. Extensive in quiry was made concerning various other statistical proced ures, but, for one reason or another, none of these alterna tives could satisfy all the rigorous requirements estab lished by orthodox statistical theory. Pushed to a logical extreme, it would appear that none but the simplest, most molecular functions can meet all these criteria in unassail able fashion. The matter of equal scale intervals, for example, must be decided not only on the basis of consistency in the ex ternal operations of measurement, which is easy to achieve, but also in regard to the implications of unit values at different points along the continuum of measurement. This latter point cannot be determined on mathematical grounds, but must be decided by logical or, in the broad sense, phi losophical reasoning. A good case might be made for the in equality of scale units in such a universally accepted 128 measurement as that of blood pressure. The implications of an increment of, let us say, ten points from a systolic pressure of 80 to one of 90 are not at all the same as a like increment from 200 to 210. In fact, a phase change oc curs between these two segments of the scale so that the es sential meaning of one is quite the reverse of the other. Other differences in connotation could be cited, including the fact that the meaning of any given reading for this in tricately determined variable may differ widely among indi viduals. Such meanings can obviously be determined only by detailed study of the particular individual concerned and not by study of great numbers of other individuals. It has become increasingly evident that a great part of the basic substance of human behavior (including, of course, the internal, unseen, psychodynamic aspects of it), whose data are presented in the form of judgments, opinions, scor ing decisions, ratings or rankings, even as introspective reports, is to all intents and purposes excluded from con ventional statistical analysis. Many of the most clinically valuable procedures known to psychologists possess one or more of these characteristics. The information they supply is simply not considered to be acceptable data and it cannot be so regarded in terms of present statistical definitions. In an effort to meet these statistical stringencies, psy chologists appeared to react in two ways: (l) they at tempted to force their data into the established statistical 129 molds created for essentially different purposes, taking them out of context and losing much of their relevance in the process; and (2) they restricted their study to discrete bits of easily measurable behavior with apparently little concern for either theoretical or practical relevance. The results accruing from both of these courses of action speak for themselves. If the above considerations are correct in their essen tials, their implications are tantamount to the statement that the whole concept of the nomothetic approach as applied to basic psychological research is invalid in principle. This view, radical as it appears, is not without its active proponents at the present time. Proposing a whole new meth odology whose principal operations are based on the Q-tech- nique, Stephenson (1952, 1953) is one of the pioneer advo cates of an idiographic approach to the study of personality whose basic data are those opinions, judgments, ratings, and so forth, until now so lightly regarded. Since Stephenson's proposals represent not merely modifications or improvements of the present study but, rather, a complete revision of it, more detailed discussion of them is reserved for the follow ing section concerned with suggestions for further research in this area. Suggestions for Further Research 130 Follow-up Studies The need for further research on the electroconvulsive therapies generally, and in particular on the brief stimulus methods as one of the most promising of them, is great and urgent indeed. They are the most widely used of all treat ments for the psychotic disorders, and they are finding in creasing use with non-psychotic conditions both in and out of the hospital. Critical comments are not infrequently heard that they are being used indiscriminately and without proper justification, and such criticisms appear to be valid ones in many instances. The study of these therapies is a field rich in theory but poor in established or generally accepted fact. Little is truly known of the effects of shock treatment either at the physiological or at the psy chodynamic level. When we cast about for suggestions as to further research, we are faced with the literal truth that everything about these methods needs further investigation and analysis. In no particular is our knowledge complete or even adequate. Two recommendations are to be made here as extensions or modifications of the present work. The value of follow-up studies as a means of appraising the long-term implications of a treatment program is well recognized. Very few such studies have been done with the electroconvulsive methods. Little is known of the duration of the reduction in overt symptoms and probably less about 131 the underlying changes which, if they are of real and last ing significance, should be reflected in improvement in the general level of adjustment to a reasonable approximation of the pre-morbid condition. (This is applied, of course, to the patients in the present study in whom the onset of ill ness was quite recent.) Studies of this nature could tell us much more than the mere substance of their comments on adjustment levels or the duration of symptom relief. Fur ther test changes might be observed, of equal import, whether they occurred as extensions of gains already made or as re versals of trends established at the time of discharge. Possible potentiating effects of the treatment in areas oth er than those immediately affected could be noted. Further investigation might be made of the kind and amount of neuro logical or psychological deficit which might put in a de layed appearance. The information supplied by such studies might do much to answer the question of whether or not elec troshock provides mere symptomatic relief of short duration or whether it effects a more basic and permanent cure. An Alternative Approach Some of the major difficulties inherent in a study of this kind have been reviewed above with the implied sugges tion that Stephenson's Q-methodology might be particularly appropriate for use in this context. Only the briefest sketch of this writer's views can be attempted here, for a full exposition of which the reader is referred to his book 132 on the subject (Stephenson, 1953). Stephenson's basic postulate is that the proper subject for psychological investigation is the individual human per sonality as a functional entity. He feels that we can never get to know more about this personality by the study of large numbers of other personalities, but only by intensive study of the single case. He therefore rejects completely the nomothetic approach based on the study of individual differences and on the gathering of discrete bits of infor mation about large numbers of individuals. Such information is necessarily superficial and the composite personality de scribed by the central tendencies of distributions of such measures never existed in nature. The number of cases studied is one, but the number of variates is as large a number of opinions, judgments, or test scores as is believed necessary to provide the desired information. The Q-sort is the basic technique employed in handling these data. Essentially it forces the data on any particu lar variate into a prescribed number of ranks and frequen cies in an approximation of the normal distribution. The Q- sorts are then correlated and factored, but with a differ ence as compared to the usual procedure in factor analysis. The conventional factorial methods are forms of interdepen dency analysis in which no thought is given to the possible nature of the factors that may ultimately emerge. In fact, these factor theorists pride themselves on their 133 "objectivity” in reserving judgment in this manner. The at tempt is made to identify them and to describe them in psy chologically meaningful terms only after they have appeared. Stephenson proposes a dependency method in which, so to speak, the factors are built into the experimental design by means of the instructions to the judges for fulfilling the demands of the Q-sort. If these factors then appear as clearly identifiable clusters, conclusive evidence may be said to have been obtained about the proposition at hand. If these clusters of decisions pointing in the same general direction do not appear, the proposition may be rejected. Thus Stephenson terms his a postulatory-dependency methodol ogy rather than the hypothetico-deductive principle which served as the basis for most of the older systems. The advantages of Stephenson's system are clear. It accepts the kind of data that are normally given in the clinical situation. Its use avoids the endless strictures on experimental design and statistical analysis imposed by adherence to more conventional methods, strictures which often violate the basic nature of the data or the conditions under which they should be obtained. As a result negative findings are inevitable. His method provides for a sensi tive and discriminating analysis of the individual personal ity, ever the supreme object of clinical effort, and it per mits allowances to be made for idiosyncracies within that personality. It does not demand adherence to a set of norms 134 in a number of unitary traits as the sole measure of excel lence. It recognizes the principle of compensation in in tellectual and personality functioning, and the endless va riety of combinations of interrelated traits or qualities as the final determinant of capacity or potential. In the present study, tests were used as such, as more or less discrete indices of defined traits or abilities. This is true even though in two of them, the Rorschach and Mosaic tests, rating scales were interposed between test and statistic, allowing at least some elements of judgment and of context to be effective. This effectiveness could have been greatly enhanced, however, if the knowledge and skills of competent judges could have been used more directly and more extensively. The Q-methodology approach would permit much more di rect attack upon the fundamental questions we wanted to ask about this form of treatment. Q-sorts could be devised to inquire into whatever basic aspects of the anticipated per sonality changes we wished to know about. A wide variety of opinion could be obtained from anyone thought to be compe tent to judge such matters. Thus the judgments of psychia trists, based on other methods of analysis and other profes sional training and experience, could be made an integral part of the design. Tests could be given as now but our fi nal results would depend in no way upon the mere counting of responses, however disguised. Judges would be asked to make 135 certain specific decisions, all bearing upon the central questions involved. The judge is not asked how he arrived at his decision. Often he may not know. He may use test indications, history, interview material, clinical observa tions, or any combination of these. He is, above all, never asked to assign a test response or an overt symptom to a normatively derived psychodynamic correlate. The final re sults are the only measure of his effectiveness and he has complete freedom to utilize whatever material he deems rele vant in whatever way his own particular knowledge or skill dictates. If the anticipated clusters do not appear to confirm the a priori predictions, there is still the same opportuni ty for obtaining others of a clearly defined nature and for identifying them in a psychologically meaningful way as is the case with the conventional factorial approach. The total worth of an experiment cannot be restricted to its function in confirming or denying assertions made before the investi gation even began. It may be of even greater value when it opens up new avenues of investigation or when it leads to totally unexpected conclusions. In a small way, the present study may be said to have done this. CHAPTER V Summary and Conclusions Problem The history of research on the electroconvulsive meth ods reveals such confusion as to results and such a lack of general agreement on even the most basic aspects of these therapies that the need for further investigation and clari fication of the nature of their effects on the patient is considered to be urgently needed. The contradictory results and the failure of much of the research to achieve definite conclusions appears to be inherent for the most part in de fects in design, in inappropriate statistical analysis, in improper use of testing devices, or other methodological er ror. The brief stimulus methods have aroused great interest in more recent times, principally because they seemed to ob viate many of the serious criticisms made of the conventional methods. The writer's own clinical observation of patients under this form of treatment tended to confirm the opinion of many specialists in this field that this newer treatment method was of potentially great promise. Based on these preliminary clinical observations and on the results of a pilot study, several major hypotheses were proposed and an experimental design was perfected to test them. The principal effort in this study was directed to ward the elimination of the methodological and procedural 136 137 errors so often encountered, and toward the formulation of hypotheses which were logically related to the apparent changes which occurred in patients in therapy on the one hand, and to the demonstrated capacities of specific testing instruments on the other. Three major hypotheses were tested: (A) that patients receiving brief stimulus electroconvulsive therapy would demonstrate a general improvement in the level of intellec tual functioning; (B) that treated patients would give evi dence of increased ego integrative capacity or basic ego strength; (C) that treated patients would demonstrate affec tive-cognitive integration as related to the mastery of con temporary reality situations. As a corollary, it was postulated that treated patients would reveal no decrement in intellectual functioning of the sort usually associated with organic deterioration. Method An experimental and a control group of 25 patients each were obtained from first admissions to the hospital who met the following criteria: white males between the ages of 20 and 40; first incidence of serious emotional disturbance; unanimous diagnosis by the psychiatric staff and by the writ er as having some form of schizophrenic disorder; minimum of eight years formal schooling; IQ of 85 or above; no compli cating physical illness; no previous treatment for any form 138 of mental illness; no markedly deviant cultural background; no marked discrepancy in admission or diagnostic procedures while at the hospital. The physical management program was exactly equivalent for all. Patients received a series of six brief stimulus treatments at the rate of two per week. All parameters of these treatments were held constant, including intensity and duration of current, electrode placement, and pre-treatment sedation. No other treatment of any kind was used during this interval and every effort was made to keep patients' general hospital experience the same throughout. All pa tients were voluntary and could have withdrawn at their own request. The management program for control patients was exactly the same as for experimental subjects except that they merely resided on the ward without treatment for a pe riod of time equal to that occupied by treatment in the lat ter group. Pre- and post-condition testing was administered to the two groups alike. All testing was done by the writer. Forms I and II of the Wechsler-Bellevue Intelligence Scale were used alternatively to test Hypothesis A, concerned with improvement in intellectual functioning. The Rorschach Prognostic Rating Scale, based on the standard Rorschach Test, was used to test Hypothesis B, which postulated gains in basic ego strength. The Mosaic Test, as interpreted by the Peak Mosaic Rating Scale, was used in the testing of 139 Hypothesis C which proposed that treated patients would show gains in affective-cognitive integration as reflected in the mastery of contemporary reality situations. The Wechsler- Bellevue tests were scored by the examiner. The Rorschachs were scored in collaboration with an independent authority on this subject. Mosaic productions were evaluated by the author of the rating scale. He was presented with the pre- and post-condition productions of each patient and was asked to choose the better of the two mosaics. A check three months later yielded a 97 per cent reliability for these judgments. Results Analysis of results obtained with the Wechsler-Bellevue tests by means of the chi square test revealed very signifi cant differential improvement for the experimental group on the VQ, IQ, Comprehension, and Picture Completion scales, and significant improvement in Vocabulary. The t-test was applied to the distribution of difference scores for the ex perimental group as an ancillary measure of significant trends toward improvement. This group showed very signifi cant gains on the VQ, PQ, IQ, Information, Comprehension, Vocabulary, Picture Completion, and Block Design tests, and significant gains on the Picture Arrangement and Object As sembly tests over their pre-treatment scores. As contrasted with these findings, the control group showed no significant 140 gains on any of the scales and one significant loss on the Comprehension test. Pre-test equivalence of the groups was established. Analysis of the distributions of difference scores for the Prognostic Rating Scale by means of the chi square test with the grand median of the two distributions taken as the point of dichotomy revealed very significant differential improvement in the experimental group in the Form Level Rat ing, and significant improvement in the shading variable. None of the other variables of this scale, including the composite weighted score, approached significance. The distribution of difference scores for the Mosaic Rating Scale showed very significant differential improve ment for the experimental group as indicated by the chi square test in the variables of Closure, Articulation, and in the final weighted score, and significant improvement in the variables of Complexity, Spoiling, and Variety. In only one variable of this scale, that of Color, were the results non-significant. The "Don't Hold" items and the deterioration quotient provided by the Wechsler-Bellevue, and the Form Level Rating obtained from the Prognostic Rating Scale were used to test the corollary hypothesis of absence of indications of organ ic deterioration. No significant negative differences were revealed by the chi square test on the combined distributions of difference scores, and a very significant differential 141 improvement was noted in the Form Level Rating. One of the most sensitive and discriminating of these scales, the Block Design test, showed a very significant gain in the experi mental group from pre- to post-treatment testing, while no such improvement was noted in the control group. Conclusions It is felt that Hypothesis A, postulating a general im provement in intellectual functioning, is confirmed by the evidence of extensive, statistically significant improvement in the Wechsler-Bellevue scores for the experimental group. These gains are, of course, in excess of any such improve ment made by the control group. Comparison of the pre- and post-test scores for this group, in fact, shows no signifi cant improvement and one significant decrement. The claim can be made that the tests used herein do not sample all possible Intellectual functions, a claim that cannot be dis avowed. However, in view of the generally accepted concept of the global nature of intelligence and of the high inter correlations among the subtests of the Wechsler scales, it is quite improbable that such gains as were registered re flect improvement in only a few discrete, unitary functions. The more likely proposition is that some appreciable gain was made even in those areas in which no significant differ ences occurred. No assertion is made that this form of elec troconvulsive therapy adds to the basic intellectual capacity 142 of the Individual but, rather, that the treatment has a fa cilitating or potentiating effect which enables him to real ize such potential as he possesses more fully in a practical application. The evidence in support of Hypothesis B, which postu lates improvement in basic ego strength, is limited to sig nificant gains in only two areas which, though important in their own right as indices of improvement in several funda mental aspects of personality functioning, are believed to be insufficient to confirm the general hypothesis. The Prognostic Rating Scale was chosen for this use because it purports to be a measure of basic potential, whether or not such potential is available for use. In the light of the strikingly dissimilar results obtained with the Mosaic Test, which emphasizes contemporary levels of functioning in a practical situation, it must be concluded that this form of electroconvulsive treatment does not add to the store of ba sic ego resources. In contrast to the contraindications of any improvement in basic ego integrative capacity as reflected by the re sults of the Rorschach analysis, the Mosaic productions ap pear to offer substantial confirmation of the treated pa tients' improved ability to function in a concrete, reality- oriented situation. We thus have two contrasting sets of results which appear to argue strongly that, while brief stimulus electroconvulsive therapy adds nothing to whatever 143 basic resources the individual had to begin with, it does render them more readily available for immediate use. It is felt that similar implication's can be seen in the intelli gence test results, another instance of practical function ing in a contemporary situation. It is suggested that this improvement in the availability of resources, in the ability to function in the immediate present, applies as well to the psychotherapeutic as to the test or practical life situa tions . While only a few measures purporting to be valid indi cators of intellectual deterioration were used, none of them showed any significant decrement as tested by either chi square or the more sensitive t-tests. A very significant differential improvement in Form Level was noted. Inter preted in the context of general and extensive improvement in a wide variety of personality and intellectual functions, it is felt that these results substantiate the corollary hy pothesis and permit the inference that no appreciable deteri oration in intellectual functioning occurred as a result of treatment. A number of limitations of the study were discussed, including the smallness of the samples employed and the sta tistical strictures this entails; the necessity of estab lishing strict controls which then limit the applicability of the findings; the examiner's unavoidable awareness of pa tients' treatment status; and the necessary limitations on 144 the number of test parameters which could be subjected to analysis. The principal suggestion for improvement of the study proposed a team approach in which the functions of testing, scoring, and general direction of the project could be kept separate. Attention was called to the often unrealistic strictures imposed upon both design and interpretation by statistical considerations, considerations which, in the opinion of many, are based upon a methodology which is far from ideally appropriate to the problems presented in clini cal research. Suggestions for further research in a field which ur gently needs it proposed follow-up studies as a means of ap praising the long-term implications of the treatment program, about which very little is known at present, and the appli cation of Stephenson's Q-methodology to the problem of eval uating the effects of this particular therapeutic agent. It is suggested that the idiographic approach characteristic of this method, the obtaining of data in full context and rele vance by means of the Q-sort, and the interpretation of such data by means of factorial studies would obviate most of the difficulties which now so seriously encumber research ef forts in this field. A P P E N D I X E S APPENDIX A DIFFERENCE SCORES EXPERIMENTAL GROUP APPENDIX A D i f f e r e n c e S c o r e s , E x p e r i m e n t a l G roup W e c h s l e r - B e l l e v u e VQ . . PS. . IS. I n f Comp DS A te Sim V oc PA PC BD OA DS DQC 7 -4 3 2 4 0 0 -1 1 1 1 -1 -1 -2 -12 -3 16 6 1 0 -2 -3 0 1 6 2 6 -2 -2 -6 0 11 7 0 2 0 -1 1 -2 2 4 1 2 0 -7 6 11 7 1 -1 -1 5 4 0 5 4 1 -1 0 2 7 6 8 0 2 0 3 -2 0 3 -2 1 3 0 5 17 6 12 0 4 3 5 2 2 -3 3 2 2 0 6 2 16 13 2 6 0 5 -5 1 1 1 4 5 2 10 4 13 9 -2 3 -2 6 1 -1 -1 5 2 2 1 7 16 8 12 4 2 4 2 2 1 0 -1 2 4 -1 0 3 6 4 1 0 1 -2 3 -1 -3 4 1 -1 2 -1 3 -4 -1 2 6 0 -6 1 1 -4 3 1 0 -3 -26 11 0 7 3 3 4 -2 4 -1 -1 3 0 -2 0 -1 7 9 9 0 2 3 -1 1 2 3 1 1 3 1 -6 12 12 14 1 3 -2 4 6 1 0 2 5 1 1 8 7 -1 4 2 1 3 0 2 0 -1 1 0 2 -3 -12 -3 5 2 2 -2 -3 -1 -4 2 2 -2 0 4 0 -l6 11 7 12 1 3 3 -3 4 2 4 4 0 0 -2 -19 8 -15 -3 1 2 2 2 2 1 0 3 0 -4 0 7 6 2 5 2 3 3 -2 0 1 1 2 0 3 -2 -31 5 6 5 1 3 0 2 2 0 4 3 -1 1 -2 -11 -10 3 -3 2 -1 -4 -4 -3 0 3 -2 0 0 1 -14 17 27 21 1 4 3 5 2 3 6 2 6 3 5 28 5 -2 2 -2 3 3 -1 -2 1 0 -2 1 1 -2 5 0 -2 0 2 1 -2 -1 -1 1 5 -5 2 -2 -2 2 7 10 9 -1 3 2 -3 . 1 2 1 -1 6 1 2 M 5.80 5.84 6.56 1.04 2.24 0.84 O.56 0.64 0.68 i.4o 1.40 1.32 1.16 -0.28 -3.20 147 APPENDIX A (continued) VQ, PQ IQ, ' Inf Comp DS A te S im V oc PA PC BD = " OA DS "DO5” CTDM 1.384 1.829 1.236 0.305 0.433 0.544 0.713 0.597 0.250 0.604 0.542 0.435 O.537 0.406-2.771 4.l91b3.193b5.307b3.4lOb5.173bl-544 0.785 1.072 2.720b2.3l8a2.583b3.035b2.l60a0.690 1.155 Significant at the 5 per cent level. bSignificant at the 1 per cent level. CDQ difference scores reflected to show increased deterioration in positive terms. 148 M -1 0 0 0 0 0 0 0 0 1 0 2 0 -1 0 0 2 0 -1 -1 0 2 -1 0 0 Difference Scores, Experimental Group Prognostic Rating Scale FM m Shading Color Form Level 0 0 0 0 0.2 0 0 0.3 3.0 0.2 0 0 0.6 0.9 0.5 0 0 0.5 -0.1 0.2 0 -1 0.6 0.4 0.2 0 0 0.2 -3-3 0.6 0 1 -0.5 -0.8 -0.2 0 1 -0.3 -0.2 -0.1 0 0 3-5 2.5 0.9 0 0 0.8 0.7 1.6 0 -1 1.5 0 0.1 1 0 0.5 0 0.3 0 0 1.0 2.0 ~o. 6 0 0 1.0 0 0.1 0 0 -1.0 0 -0.2 0 0 0 1.5 -0.1 1 0 -1.1 0.3 1.1 0 1 0.3 0 0.2 0 0 0.3 0.4 0 0 -1 1.0 1.2 0 0 0 0.2 0.5 0.4 0 0 0.4 -0.2 0.3 -1 0 -0.6 0-3 0.1 -1 0 -0.4 0.3 0 0 -1 -0.3 1.5 1.4 o in i —i 0 u o o C O a 3 S o 3 u C O o r H c t f - P U G 0 c d 0 ) r H > 6 C t i • H O U C O C D a to b l X c H W - H a ■ P C O • » c i i c q p s 0 u o O - H r H O 0 o C O 0 3 O O 0 s a £ 0 s u o 0 o a a • H n - p < o i H O r o t-- VO O nj- ID t— lO V O O O O O OOOO D --1- O PO D — [>• O V O COlOrHrOOCOlOlOlO rH OOiAlACO moD miO H l A C O H H ** *• •• *•• ••• •• •* ••• ••• ♦*• O O O O r H O O O O O O O O O O O O O O O O O O J r H r H I I I in n f < - i oiko in-=J-^i--=t rH oj no ro in oo O J rH on oj rH ro o\ O 1 1 1 1 —I l—1 i — I i — i O O i — I O OJ O i — i O i — 1 i — f O C V i O O O OJ O i I i I OJ OJ OJ OJ 1 I I OOOlMfMWHHOHrlrlrHOHrlrlOOHHOW OO r II II O J i H r H O O O O J O O r H O J O O r H O r H O O r H O O r H O J O O I I I OJ C M O O O O i— I i— 1 i— I i— 1 O O O r — I O i— I O i— ! O i I O O OJ O i I I I I HHOOHrHOHHr-lOOrHOJHOrHHOrHrHrHOJOJPn I I r l H O O O l O l O H H r l O H H H H H r l O O H H H f O O l f O I I I APPENDIX B DIFFERENCE SCORES CONTROL GROUP APPENDIX B D i f f e r e n c e S c o r e s , C o n t r o l Group W e c h s l e r - B e l l e v u e VO PQ. IQ Inf Comp DS Ate Sim Voc PA PC BD OA DS DQb 6 -10 0 1 -2 0 5 1 2 1 -3 -1 -1 -1 17 -6 0 -3 -1 0 -2 -2 1 -1 -3 0 4 -3 1 10 -5 0 -1 1 -1 -3 1 -1 -1 2 -3 0 2 -1 -13 1 6 3 2 -2 -2 2 3 0 -3 2 1 2 2 -6 -1 -4 -4 -1 -4 3 1 1 -1 0 -3 1 -3 1 38 -4 -7 -6 -1 -2 -4 2 0 0 -4 -1 -2 3 -1 -12 -3 -1 -3 4 -1 -2 0 -1 -1 0 -1 1 0 -1 -8 -4 0 -3 -1 1 0 -5 2 -1 1 0 1 -2 0 -6 -4 -1 -3 1 -2 0 -4 2 -1 -3 1 3 0 -4 -12 -7 -3 -5 -3 -1 0 3 -4 -1 -2 -1 1 0 1 18 -11 4 -4 -2 -5 -2 0 -2 1 5 1 -1 -1 0 -2 13 19 17 0 4 7 1 1 -1 3 8 3 -4 3 30 -2 -4 -3 0 -2 1 -3 -1 0 -2 1 -2 3 3 -10 2 13 8 1 -5 2 2 3 0 5 -2 1 6 0 2 4 5 6 -2 4 0 3 -2 1 0 -1 2 2 1 12 10 15 14 2 2 -1 2 4 1 3 1 1 0 0 -2 -7 8 0 -1 -4 0 1 0 -2 5 -2 2 3 -1 9 -9 -7 -8 -1 -7 0 -1 0 1 -7 -1 2 1 1 5 -10 9 -1 -2 -4 0 0 -1 -1 1 5 2 0 -2 -3 -3 -8 -8 2 -1 -2 -4 0 0 -3 -1 2 -4 -2 -10 1 0 1 -3 -2 2 2 2 1 2 0 -3 2 1 7 3 10 7 2 -2 0 1 1 0 -1 2 -4 9 0 -33 1 8 5 2 -3 1 0 -2 1 7 0 -2 -1 2 -2 -3 -4 -3 2 -2 1 1 -3 0 -3 0 3 -3 -2 10 5 12 8 2 2 0 0 1 0 1 0 1 4 1 -10 M -1.400 2.400 0.562 0.161-1.560-0.040 0.322 0.200-0.123 0.200 0.080 0.641 0.600 0.080 1.162 152 APPENDIX B ( c o n t i n u e d ) VQ PQ IQ I n f Comp DS A te S im V oc PA PC BD OA DS DQU ODM 1.314 1.703 1.410 0.407 0.584 0.478 0.522 0.427 0.212 0.734 0.530 0.444 0.68l O.36I 3-282 t a 1.066 1.409 O.397 0.393 2.671 0.084 0.613 0.468 0.566 0.273 0.151 1.441 0.881 0.222 0.353 S i g n i f i c a n t a t t h e 5 p e r c e n t l e v e l . ( I t w i l l b e n o t e d t h a t t h i s i n d i c a t e s a d e c r e a s e i n p e r f o r m a n c e . ) ■ j ^ DQ d i f f e r e n c e s c o r e s r e f l e c t e d t o show i n c r e a s e d d e t e r i o r a t i o n i n p o s i t i v e t e r m s . H V J I 00 O O O O M O O O O O O O O O O O M M O O O M O M O 154 D i f f e r e n c e S c o r e s, C o n t r o l G roup P r o g n o s t i c R a t i n g S c a l e FM m S h a d in g C o lo r Form L e v e l w.s. 0 0 -0.1 0 0.5 0.4 -1 0 0.9 -0.3 0 -1.4 0 0 -0.4 0 -0.2 -0.6 -1 1 -0.4 -0.6 0.1 -1.9 0 1 0.9 -0.1 -0.1 1.7 0 0 1.0 0.5 0 1.5 0 0 -0.1 0 -0.1 -1.0 0 0 -1.1 0 0.3 0.2 -2 0 -1.5 -4.2 0 -8.7 0 1 2.3 -1.8 0.2 1.7 0 0 -0.7 0.3 -0.1 -0.3 0 1 -0.6 1.5 -0.1 1.8 0 0 -1.0 2.0 -0.1 0.9 0 0 0.1 0.9 0 2.0 0 0 0 -0.6 -0.2 -0.8 -1 -1 0.3 2.3 0 0.6 0 0 1.5 1.0 -0.2 2.3 0 0 -0.8 -2.0 -0.1 -2.9 0 -1 0 0.5 -0.1 -0.6 0 0 -0.5 -0.3 -0.1 -0.9 0 0 -0.2 0.1 1.2 2.1 0 0 -0.4 0 0 -0.4 0 0 0 0.8 0 0.8 0 0 0.4 -0.6 -0.1 -0.3 0 -1 1.1 0 -0.7 -0.6 O r - f O r H r H O O H O O O O O r - t H H O O O O O O r H O r H 155 D i f f e r e n c e S c o r e s , C o n t r o l G roup M o s a ic R a t i n g S c a l e C lo A r t Com C o l S p l g V a r Sum S c o r e -1 -1 0 -1 0 -3 -0.50 -1 -1 0 -1 -1 -5 -0.83 0 0 0 0 -1 -1 -0.17 1 1 -2 0 0 1 0.16 -1 0 1 1 1 1 0.17 0 0 1 0 0 1 0.16 -1 -1 0 0 0 -2 -0.34 -1 -1 0 0 -1 -4 -0.67 0 0 -1 1 -1 -1 -0.16 0 0 -1 0 0 -1 -0.16 0 0 1 0 1 2 0.33 0 0 1 0 0 1 0.17 0 0 0 0 -1 -1 -0.17 -1 0 -1 0 0 -3 -0.50 0 -1 1 -1 0 -2 -0.33 -1 0 0 -1 0 -3 -0.50 0 0 0 -1 0 -1 -0.17 0 0 0 -1 -1 -2 -0.33 -1 0 1 1 0 1 0.17 0 0 1 0 0 1 0.17 0 0 2 1 0 3 0.50 0 1 0 1 1 3 0.50 0 0 -1 -1 0 -3 -0.50 0 0 0 0 1 1 0.17 1 0 -1 0 0 1 0.17 APPENDIX C RAW SCORES EXPERIMENTAL GROUP APPENDIX C Raw Scores, Experimental Group Weehsler-Bellevue Initial Test VQ, PQ , IQ Inf Comp DS A te Sim Voc PA PC BD OA DS DQ 107 106 107 11 11 10 9 11 12 10 10 12 12 11 7 98 93 96 10 13 9 6 8 9 6 10 8 14 7 29 104 100 103 10 9 11 11 8 10 10 10 7 12 6 12 134 123 131 14 16 14 13 13 17 11 11 13 13 13 -1 126 124 127 11 14 17 13 16 12 12 15 14 13 12 -11 108 110 110 12 9 10 11 12 11 11 9 13 13 10 1 107 110 110 12 11 7 4 16 12 11 14 10 9 7 35 109 105 108 10 13 9 9 11 11 9 10 11 9 9 0 104 93 99 9 12 9 8 10 12 8 9 8 8 9 8 101 102 102 9 11 10 9 8 11 10 11 10 13 7 17 116 129 125 12 11 14 12 10 12 13 12 14 14 13 -11 101 113 107 8 13 7 8 10 11 10 11 12 14 11 13 106 76 91 11 10 10 10 11 10 5 7 7 5 9 -9 110 109 " 110 13 10 9 9 10 14 11 12 7 11 10 25 90 115 102 6 10 7 6 8 9 14 10 12 11 10 0 130 124 129 13 15 14 13 16 15 11 15 15 12 9 2 103 96 99 10 11 7 15 8 8 6 9 10 14 9 0 117 117 119 9 13 14 13 12 11 11 11 14 16 9 -7 102 96 99 9 11 7 12 10 8 9 9 7 7 10 -19 122 125 126 14 10 14 13 10 14 10 12 13 13 15 -9 124 125 127 11 14 13 15 14 13 13 14 16 14 12 -8 93 98 96 10 10 3 6 9 9 7 9 9 11 5 36 119 118 120 15 13 7 10 17 14 12 15 12 12 11 28 109 122 117 10 10 13 10 10 12 6 15 14 16 12 5 101 96 99 10 11 9 10 10 8 11 9 10 6 9 -16 157 Raw Scoresj Experimental Group Weehsler-Bellevue Pinal Test VQ PQ IQ Inf Comp DS Ate Sim Voc PA PC BD OA DS DQ. 114 102 110 13 15 10 9 10 13 11 11 11 11 9 19 95 109 102 11 13 7 3 8 10 12 12 14 12 5 35 104 111 110 10 11 11 10 9 8 12 14 8 14 6 19 140 134 138 15 15 13 18 17 17 16 15 14 12 13 -3 133 130 135 11 16 17 16 14 12 15 13 15 16 12 -16 125 116 122 12 13 13 16 14 13 8 12 15 15 10 ~5 109 126 123 14 17 7 9 11 13 12 15 14 14 9 25 113 118 117 8 16 7 15 12 10 8 15 13 11 10 -7 120 101 111 13 14 13 10 12 13 8 8 10 12 8 8 104 108 106 10 11 11 7 11 10 7 15 11 12 9 18 119 125 124 14 17 14 6 11 13 9 15 15 14 10 15 112 113 114 11 16 11 6 14 10 9 14 12 12 n 14 113 85 100 11 12 13 9 12 12 8 8 8 8 10 -3 122 121 124 14 13 7 13 16 15 11 14 12 12 n 17 97 114 106 8 11 10 6 10 9 13 11 12 13 7 12 127 129 131 15 13 n 12 12 17 13 13 15 16 9 18 114 103 111 11 14 10 12 12 10 10 13 10 14 7 19 125 102 116 10 15 16 15 14 12 11 14 14 12 9 -14 108 98 104 11 14 10 10 10 9 10 11 7 10 8 12 127 131 131 15 14 13 15 12 14 14 15 12 14 13 2 114 128 124 13 13 9 11 11 13 16 12 16 14 13 6 110 125 117 11 14 6 11 11 12 13 11 15 14 10 8 124 116 122 13 16 10 9 15 15 12 13 13 13 9 23 109 120 117 12 11 n 9 9 13 11 10 16 14 10 3 108 106 108 9 14 14 12 7 9 13 10 9 12 10 -14 v_n co M. 1 1 0 2 1 3 1 0 2 0 3 l o 0 1 1 0 1 1 2 2 1 3 1 1 Raw Scores, Experimental Group Prognostic Rating Scale Initial Test FM m S h a d i n g C o l o r Form L e v e l 0 1 1.5 1.5 1-7 0 0 2.5 0 1.6 0 1 2.4 o . 6 0.4 1 1 1.7 2.3 1.6 0 2 1.3 2.4 1.8 1 1 1.6 1.8 1.0 1 1 1.5 2.1 1-5 0 0 2.0 1.9 1.5 0 0 -1.5 0 1.0 0 1 1.5 2.3 0.4 1 1 1.0 3-0 2.2 0 0 2.5 0 1.6 0 0 2.0 1.0 0.1 0 1 1.3 1.7 0.4 0 0 1.9 2.3 1.7 1 1 0.8 1.5 1.8 0 1 1.4 2.3 0.4 1 0 2.3 2.0 1-5 0 1 2.1 1-9 1.9 1 2 1.2 0.6 1-7 1 1 2.3 2.5 1.9 0 1 1.9 2.5 1.6 1 2 1.5 2.0 1.0 1 1 0.9 2.1 1.2 0 1 3.0 1-5 0.3 R aw S c o r e s , E x p e r im e n ta l G roup 0 o 1 — 1 fe 3 O CO f c O u G -P o •H G Q i — ! -P 0 o 3 E h o K i — 1 O 3 •H C -P * H bC CO pc, c O •H c T} f a O 3 o -C u CO 0 -1 CTvVQ -=t CVJ CTVVO CO COCO OMHinlN LT\-=J- COCO t~-00 n co h -3- incovo o c o v o v o inco o v H o o i n i r inooco o\r-o o h o c o v o c o rH rH rH rH ovco cr\co o vo co-3- c riO c o c m n in tn [> -in !> -o M > -c o c n rH C u r- * * * * • • • • • • • • • • • • • • • • • • • • • t —|r—| O i —ICVlt—It—It—1 i —1 CVJ OJ i —iOOrHi—it I* ! 1 I* iCVlr ! i Ir I i I I inoncviooincoiH-tnoo o n -c o o v o o toco o m m o • • • • • • • « • • • o • ♦ * • • • • • » • • • • H f O H O K V I H H H O J O n t O C O H CM C O CM C V J O J .H C O C V J O J O J C O I in c o o cvj o v co o t H - o c o i n o o c o ctvco covq -3- oj in co o w n t - i—I O J C O O J i—I t—I rH i—i O J O J O J C O C O O J CD CD O O J O J O J O J O J CD O O J H O H H H H C V I r l O H O O O r l O H H i H r l r l H r l W H O O O O H O H H O O O H H O O O H r H H O r l H O O O O O r — I O CVI rH CO r — 1 O CVJ r — 1 CO CO O r — I H r — t CVJ i — I O H CVJ CO CM rH r — 1 I l6l Raw S c o r e s , E x p e r i m e n t a l G roup M o s a ic R a t i n g S c a l e I n i t i a l T e s t !1o A r t Com C o l S p l g V ar Sum S c o r e 5 4 3 2 2 2 18 3-00 2 2 2 2 2 2 12 2.00 3 3 3 4 2 2 17 2.83 5 5 5 5 2 4 26 4.33 3 4 3 3 3 2 18 3.00 1 2 2 3 1 2 11 1.83 3 3 4 2 2 2 16 2.67 2 2 2 o 2 3 14 2.33 2 2 3 2 2 2 13 2.17 4 4 3 4 4 3 22 3.67 3 3 5 3 2 3 19 3.17 2 3 3 3 2 2 15 2.50 2 2 2 2 2 2 12 2.00 1 1 5 2 1 1 11 1.83 3 3 3 1 2 2 14 2.33 2 3 4 3 3 3 18 3-00 2 2 2 3 2 2 13 2.17 4 4 4 4 3 2 21 3.50 1 1 1 2 2 2 9 1.50 4 4 5 3 2 4 22 3.67 2 2 4 2 1 4 15 2.50 4 4 4 3 2 4 21 3.50 1 3 3 3 1 1 12 2.00 1 1 3 2 1 1 9 1.50 2 2 2 2 1 1 10 1.67 C lo 4 3 3 5 5 3 3 3 3 5 3 3 3 2 2 3 3 4 1 3 3 5 4 3 5 LnooroininrorororoiAromoorocJcooOLr\iHcoroir\incom Raw Scores, Experimental Group Mosaic Rating Scale Final Test A r t Com C o l S p l g V a r Sum 5 4 2 3 23 4 1 2 3 16 3 3 2 2 16 5 5 4 4 28 3 3 5 3 24 2 3 3 2 16 3 4 3 4 20 3 3 3 3 18 4 2 2 3 17 4 3 3 3 23 5 5 1 4 21 3 3 3 3 18 2 2 3 2 15 4 3 1 3 16 3 1 1 2 11 5 4 2 3 20 2 3 3 2 16 5 4 3 4 25 1 3 2 2 10 4 3 3 3 19 4 2 2 3 17 4 4 2 2 22 5 5 3 3 25 3 2 4 3 18 3 2 2 3 20 APPENDIX D RAW SCORES CONTROL GROUP PQ 104 102 115 118 108 121 122 121 92 107 101 96 112 104 118 90 94 109 105 89 88 73 121 94 94 APPENDIX D Raw Scores, Control Group Weehsler-Bellevue Initial Test IQ I n f Comp DS A te Sim Voc PA PC BD OA 94 6 11 7 1 9 9 9 13 9 12 102 7 13 11 5 10 13 9 15 5 14 115 10 12 10 12 13 12 9 14 13 12 117 11 11 16 8 10 11 14 13 15 12 109 9 14 11 9 9 11 8 12 11 12 115 11 12 11 7 11 12 14 14 14 12 128 11 12 16 15 13 14 13 15 15 12 117 11 9 11 11 9 13 11 15 12 14 99 12 9 7 11 9 13 11 11 6 12 105 11 11 7 6 11 11 8 14 8 13 105 11 13 9 9 12 10 7 12 8 13 92 8 8 3 8 10 9 8 7 11 14 123 15 15 13 15 13 14 11 10 11 8 103 12 13 7 6 8 10 4 14 12 10 111 13 11 6 6 11 11 14 14 8 ll 94 8 11 7 7 8 12 7 9 8 11 91 7 12 6 0 11 8 4 12 8 10 106 9 16 7 6 8 10 12 13 8 9 113 14 16 7 9 13 13 10 9 11 14 101 11 14 9 8 11 13 11 9 5 11 87 7 10 4 4 8 7 4 9 7 7 88 9 10 10 9 12 10 5 8 7 2 121 10 13 9 17 14 11 7 15 16 15 105 9 13 10 11 12 13 10 9 5 10 99 9 12 9 9 10 10 11 9 9 8 164 PQ 94 102 115 124 104 114 121 121 91 104 105 115 108 117 123 105 102 102 114 8 l 88 83 129 90 106 Raw Scores, Control Group Weehsler-Bellevue Final Test IQ Inf Comp DS A te Sim V oc PA PC BD OA 94 7 9 7 6 10 11 10 10 8 11 99 6 13 9 3 11 12 6 15 9 11 114 11 11 7 13 12 11 11 11 13 14 120 13 9 14 10 13 11 11 15 16 14 105 8 10 14 10 10 10 8 9 12 9 109 10 10 7 9 11 12 10 13 12 15 125 15 11 14 15 12 13 13 14 16 12 114 10 10 11 6 11 12 12 15 13 12 96 13 7 7 7 11 12 8 12 9 12 100 8 10 7 9 7 10 6 13 9 13 101 9 8 7 9 10 11 12 13 7 12 109 8 12 10 9 11 8 11 15 14 10 120 15 13 14 12 12 14 9 11 9 11 111 13 8 9 8 11 10 9 12 13 16 117 11 15 6 9 9 12 14 13 10 13 108 10 13 6 9 12 13 10 10 9 11 91 6 8 6 1 11 6 9 10 10 13 98 8 9 7 5 8 11 5 12 10 10 112 12 12 7 9 12 12 11 14 13 14 93 13 13 7 4 11 13 8 8 7 7 88 4 8 6 6 10 8 6 9 4 9 95 11 8 10 10 13 10 4 10 3 11 126 12 10 10 17 12 12 14 15 14 14 102 11 11 11 12 9 13 7 9 8 7 107 11 14 9 9 11 10 12 9 10 12 2 2 1 2 2 1 3 0 2 1 1 0 3 1 1 2 0 1 2 0 0 1 1 1 0 Raw Scores, Control Group Prognostic Rating Scale Initial Test PM m S h a d i n g C o l o r Form L e v e l 1 0 2.8 2.6 1.4 1 0 1.1 2.6 1.9 0 0 1.9 3.0 1.9 1 0 1.4 2.1 1.5 0 0 0.6 2.4 1.9 0 1 1.1 1.8 0.9 1 0 1.9 3.0 2.2 0 0 2.8 2.1 2.0 1 0 3.0 2.4 1.5 0 0 -1.0 1.8 0.9 0 1 2.3 2.3 1.8 0 0 1.1 0.8 1.7 1 0 1.5 1.0 1.7 0 1 0.7 0.6 1.0 0 1 1.3 1.7 2.1 1 2 2.0 2.7 2.3 0 0 1.5 2.0 1.6 0 1 2.8 2.5 1.4 1 1 3.0 1.0 1.8 0 0 0.1 1.8 1.2 0 1 2.5 2.0 0.5 0 0 0.9 0.5 0.5 0 0 1.0 1.5 1.8 0 1 1.7 2.1 1.6 0 1 0.4 1.5 2.0 t " - — « —1 CD U o o CO i —! CD > CD O s a u 3 <u O O rH Ph U cd CD O CO i — 1 O bO u u c -p o -P -H CO 1—I C -p 0 o O aJB o o PS i —l O Cd fflH C 0) -P H b£ Pi M fr, c O O •H o £ rd CO bO cd O x; *5 X CO cd a PS <-* c S Ph s CUOJOJrH'vDCOiHrHOJ-^'H^'HOO 00 VO ■=£ CQ Cvl OJ H LP\ H H CO o c— c~— m d c o p - o o - r H - ^ t c o L n o v o 'vDGJ!>-Lncr\OJcoaJvo i —I i —I »H 0 \(J\ b —V Q 00 C D i H P O I A H O W O V Q O 0 \C0 ^t O O 0 - rH t— LT\00 L A C O H H H r l H O W W H H r l H H H H W H H H H r l O H H H v£> m o i n m r o o h o q ^ o o o o i A H O O i n i n i r i H i n m i n i n « « • • • • • • • • • • • • • •« • •• •• ** OJ O J C O r-1 O J OJ OO cvj H O J O J O O O J H L n o n O H H W O O l H H I t - o i n o i n n o o i'-inco'o l t m a c o r o o o o o o ^ o o i n o H i n OJ OJi—ii—i i —IOJi —It —i<—I i —! i —l O O O f —iOJ OO OJ OO O OJ O i 1 OJ i I O O O i —I i — !i— l O O O i — I i — (t— I O i — Ii— I i — I O i — l O O i — I O O rl o H O O O O O H O r H O O O r H O O O O O H O O O O O O I O l H H H W H f O H H H H O f O H H O I O H O l O H H H H O 168 Raw S c o r e s , C o n t r o l G roup M o s a ic R a t i n g S c a l e I n i t i a l T e s t C lo A r t Com C o l S p l g V ar Sum S c o r e 2 3 3 2 2 2 14 2.33 3 3 4 2 3 3 18 3-00 3 3 3 2 2 2 15 2.50 2 2 2 3 2 2 13 2.17 4 4 3 2 2 3 18 3-00 5 5 6 3 3 3 25 4.17 3 4 4 3 2 3 19 3.17 4 4 4 3 3 3 21 3.50 2 2 2 2 1 2 11 1.83 3 3 5 3 3 3 20 3-33 2 2 2 2 2 2 12 2.00 2 2 2 2 2 2 12 2.00 3 4 4 3 2 3 19 3-17 3 3 3 3 2 2 16 2.67 3 3 4 1 3 3 17 2.83 3 3 3 3 3 2 17 2.83 3 3 3 3 3 3 18 3-00 2 2 l 3 2 2 12 2.00 3 3 2 1 1 1 11 1.83 2 2 2 2 2 2 12 2.00 2 2 2 1 1 2 10 1.67 2 2 2 2 1 1 10 1.67 3 3 3 3 3 2 17 2.83 1 1 1 1 1 1 6 1.00 2 2 3 3 2 2 14 2.33 C lo 2 2 3 3 3 5 3 3 2 3 2 2 3 2 2 2 3 2 3 2 2 2 2 1 3 Raw Scores, Control Group Mosaic Rating Scale Final Test A r t Com C o l 2 2 2 2 3 2 3 3 2 3 3 1 3 3 3 5 6 4 3 3 3 3 3 3 2 2 1 3 5 2 2 2 3 2 2 3 4 4 3 2 3 2 3 3 2 2 3 3 3 3 3 2 1 3 2 2 2 2 2 3 2 2 3 2 3 2 3 3 2 1 1 1 3 3 2 S p l g V a r Sum 1 2 11 2 2 13 2 1 14 2 2 14 3 4 19 3 3 26 2 3 17 3 2 17 2 1 10 3 3 19 2 3 14 2 2 13 2 2 18 2 2 13 2 3 15 2 2 14 2 3 17 1 1 10 2 1 12 2 2 13 2 2 13 2 2 13 2 2 14 1 2 7 2 2 15 APPENDIX E SAMPLE MOSAICS APPENDIX E Sample Mosaics Fig. 2. Pre-treatment mosaic of Patient A, an experimental subject. This design was given a rat ing of 2.67 (see illustration of the use of the rat ing scale, Appendix F). This subject's post-treat- ment production appears on the following page. 171 172 Pig. 3- Post-treatment mosaic of Patient A. This mosaic was given a final score of 3*33 which represents a clear improvement in quality over the pre-treatment production. 173 Fig. 4. Pre-treatment mosaic of Patient B, an experimental subject. This design was given a rat ing of 3-0* It may be compared with the same sub ject's post-treatment production on the following page. 174 Fig. 5- Post-treatment mosaic of Patient B. This design was given a rating of 3.87, which rep resents a significant gain over his first produc tion (see preceding page). 175 Pig. 6. Initial mosaic production of Patient C, a control subject. This mosaic was rated 3*50. For this patient's second production, see following page. 176 Pig. J. Second or retest production of Patient C, a control subject, after an interval of 25 days. This mosaic was given a rating of 2.83 which repre sents a significant decline in the quality of the design after this period of hospitalization without treatment. APPENDIX F ILLUSTRATION OF THE USE OF THE PEAK RATING SCALE IN EVALUATING MOSAIC DESIGNS APPENDIX F Illustration of the Use of the Peak Rating Scale in Evaluating Mosaic Designs Scoring Item Design Number ... 27 Design Number 28 Variable: Closure 3 3 Articulation 3 3 Complexity 4 3 Color 2 4 Spoiling 2 3 Variety 2 4 Summated Score 16 20 Average 2.67 3.33 Note: The above table shows the detailed scoring of the productions of Patient A, an ex perimental subject, before treatment (Design 27) and after treatment (Design 28). For photo graphs of these productions, see Appendix E. 178 APPENDIX G INSTRUCTIONS TO THE RATERS IN THE USE OF THE PEAK RATING SCALE APPENDIX G Instructions to the Raters in the Use of the Peak Rating Scale The following six variables (A to F) define the six- level rating scale at its upper and lower limits. The upper limit is represented by level number 6. The lower limit is represented by level number 1. Evaluate each of the designs by each of the six variables, assigning to the design a val ue in the range of 1 to 6. When you have made all six judg ments, sum your results and divide by six to obtain an aver age. This average is the rating scale value of the design. The limits for the scale levels are: 6-5-5-6.0; 5~4.6~5.4; 4-3-5-4.4; 3-2.5-3.4; 2-1.5-2.4; 1-1.0-1.4. Enter the num ber of each design and your judgments of it on the Judge's Rating Sheet provided for you. Average your judgments and enter the totals, averages, and final ratings in the columns provided for those purposes. Use the pictures of the sample mosaics as aids in helping you to arrive at your decisions. A. Closure. From a very high degree of closure in which the meaning or organization is immediately apparent to complete or nearly complete failure to achieve a recogniz able gestalt. B. Articulation or organization. From a high degree of relatedness or integration among the several pieces or units of the designs to complete or nearly complete lack of 180 integration or relatedness among the same. 181 C. Level of aspiration, or complexity. From the con cept of a very complex, highly differentiated, organized, and meaningful design to the concept of a very simple, unor ganized, or confused design. D. Color. From the appropriate use of color in repre sentational designs or the aesthetically effective use of color in abstract designs to the inappropriate use of color in representational designs or the ineffectual use of color in abstract designs. E. Spoiling. From a design in which only minor incon sequential flaws can be detected to designs in which the gestalt quality is badly spoiled by serious weakening or de structive features. P- Variety. From the use of all or most of the shapes and colors in highly integrated designs, with a high degree of closure to a complete or nearly complete failure to use a variety of shapes and colors in any effective way. G- Level. In regard to the above specifications, the design is, in general, on the level of one of the samples shown. APPENDIX H INSTRUCTIONS FOR THE LOWENFELD MOSAIC TEST APPENDIX H Instructions for the Lowenfeld Mosaic Test The following material is quoted verbatim from the printed instructions supplied by the Psychological Corpora tion to American users of the Mosaic Test: The instructions below have been especially developed for use with the Lowenfeld Mosaic Text in the United States by Dr. Margaret Lowenfeld and Dr. Thomas I. McCulloch of Letchworth Village. The Examiner should place the opened box on the table and say: "Here is a box of colored pieces of different shapes and sizes. I'll show them to you. There are five different shapes. (Pick up one of each shape, each of a different color except white and lay them in turn on the table, replacing each after demonstration.) Each of these shapes is in five colors in this box: blue, red, yellow, black, green and white also. (Point out the various colors in the box.) "Now I want you to do something with the pieces in this box on this tray. (Place the tray in which a closely fit ting piece of paper has been laid in front of the individual to be tested.) You may use as few or as many pieces as you like and whatever of the shapes and colors you want. You may make anything you want to and take whatever time you like. Tell me when you are finished. Would you like me to 183 184 say this again?" If the instructions are not understood they may be re peated in whole or in part. If specific questions regarding procedure are raised which are not covered in the main in structions, the answer should be given that there is com plete freedom in the matter. When the design is finished, the examiner should discuss with the maker the significance the production had to him, determining for every design whether it is representational, and if so, the object or idea represented and if this idea was present or not in his mind when he began. This should be done without the use of leading questions, in a manner appropriate to the individual being tested. A time record should be kept from the moment that instructions are completed until the maker indicates that his design is finished. Careful note should also be taken of the order in which the pieces are used. Arrangement of pieces in the box. For each shape the colors are arranged from left to right or from bottom to top as follows: all whites, all greens, all blacks, all yellows, all reds, all blues. There should be eight of each color of diamonds and isosceles triangles, four of squares, six of equilateral triangles, and twelve of scalene. In the stand ard sized box of 456 pieces, the arrangement of each half of the box should be identical. The small box should be pre sented with the direction of the pieces as in the standard box. The former is opened with the lid to the examiner's left; the latter with the lid away from the examinee. The scalene triangles should be at the top and the squares at the lower left corner as seen by the examinee. APPENDIX I ILLUSTRATION OF THE USE OF THE RORSCHACH PROGNOSTIC RATING SCALE APPENDIX I Illustration of the Use of the Rorschach Prognostic Rating Scale No. 27 Response M FM m Shading C Form Level I 1. 2. 1 0 1.5 1.5 II • 5 • 5 1.5 III 1 2 1 2.0 IV • 5 1 2.5 V 1 1-5 VII • 7 2.0 VIII .8 1 2.5 IX • 5 0.0 X • 5 0.0 Totals 1-7 3.3 2.0 1-5 3-5 15.0 W.S. 1.0 1.0 1.0 Sum - 8.1 1-5 - Group II 2.1 1*5 Note: The above table shows the detailed scoring of the pre-treatment Rorschach of Patient A, an experimental sub ject. It may be compared with the post-treatment scale giv en on the following page. 187 188 No. 28 Response M FM m Shading C Form Level I 1 1 2.5 II • 7 1 • 5 • 5 1.5 III 1 . 2. 3- 1 1 1 • 5 1 1.5 2.0 1.5 IV 1 1.0 V 1 1.5 VI 1 • 5 -.5 2.0 VII 1 . 2. .8 1 • 5 1 2.5 2.5 VIII 1 • 5 1-5 IX 1 . 2. 3- • 5 -.5 0 0 2.5 2.0 2.0 (-.5) X • 5 1.0 Totals 2.8 3-7 4.0 2.0 3.0 27.5 w.s. 1 1 Sum 2 - 6.6 1.0 - Group II 1.3 1.8 - .5 = 1.3 Note: The above table shows the detailed scoring of the post-treatment Rorschach of Patient A. Comparison with the scale given on the preceding page reveals a decline of 1.5 points in the final rated score under treatment. This drop in Prognostic Rating Scale score as contrasted with a significant improvement in mosaic quality is a characterist ic finding in this study. (For examples of this patient's mosaics and rating scales on them see Appendixes E and F.) REFERENCES REFERENCES Adrian, E. D. The spread of activity in the cerebral cortex. J- Physiol., 1936-1937, 88, 127-161. Alexander, L. Non-convulsive electric stimulation therapy. Amer. J. Psychiat., 1950, 10£, 241-250. Alexander, L. Psychic and neural dynamics in the treatment of mental disease. Conf. Neurol., 1952, 12, 292-297- Alexander, L. Treatment of mental disorders. Philadelphia: W. B. Saunders Co., 1953- Alexander, L. The therapeutic process in electroshock and the newer drug therapies. J. Amer. Med. Ass., 1956, 162, 966-968. Alexander, W. P. Intelligence, concrete and abstract. British J. Psychol., 1935- Arai, K., and Obonai, T. A hypothesis of Memory Stratum IV; a study of amnesia resulting from electro-convulsive shock. Jap. J. Psychol., 1954, 25, 155~l64. Arieff, A. J. Threshold studies in electrical convulsions using a square wave stimulator. Quart. Bull. Northwest. Univ. M. School, 1948, 22, 10-17 Arieti, S. Interpretation of schizophrenia. New York: Robert Brunner, 1955- Arndt, R. Die Elektrizit&t in der Psychiatrie. Archiv. f. Psychiatrie und Nervenkronlcheiten, 1869, 2, 259_337- Augustin, Friedrich Ludwig. Vom Galvonimus. Berlin, 1801. Baruk, H. Psychiatric treatment involving central nervous system alteration: shock and psychosurgery. J. Psy chol . norm, pathol. , 1955, 52., 356-374. Baughman, E. E. Rorschach scores as a function of examiner differences. J. Pro.j . Tech. , 1951, 15, 243-249. Bayles, S., ejt al. Square waves (BST) versus sine waves in electroconvulsive therapy. Amer. J. Psychiat., 1950, 107, 34-49. Beck, S. J. The Rorschach test as applied to a feebleminded group. Arch. Psychol., 1932, 136, 1-84. 190 191 Benedikt, M. Nervenpathologie und Elektro-Therapie. Leip zig, 1874. Beran, M., et aJ. Psychological studies on patients under going non-convulsive electric stimulation treatment. Amer. J. Psychiat., 1952, 109. 5- Bowen, Barbara. An extension of the Mosaic Test designed to increase the prognostic value. J. Pro,1 . Tech. , 1954, 18, 5-10. Boynton, E. P., and Hines, M. On the question of threshold stimulation of the motor cortex. Am. J. Physiol., 1933, 106, 175-182. Brill, N. Q., Crumpton, E., Eiderson, S., Grayson, H. M., Heilman, L. I., Richard, R. A., Strassman, H. D., and Unger, A. A. Investigation of the therapeutic compon ents and various factors associated with improvement with electroconvulsive therapy: A preliminary report. Amer. J. Psychiat., 1957, 112, 997-1006. Brower, D., and Oppenheim, Sadi. The effects of electro shock therapy on mental functions as revealed by psycho logical tests. J. Gen. Psychol., 1951, 4^, 171-188. Cerletti, U., and Bini, L. L 1ellectroshock. Arch, gen. di Neurol. Psichiat. Psicoanal., 1938, 1£, 266^268. Cohen, J. The efficacy of diagnostic pattern analysis with the Wechsler-Bellevue. J. Consult. Psychol., 1955, 19, 303-306. Colm, Hanna. The Mosaic Test in conjunction with the Ror schach and Binet tests. Rorschach Res. Exch., 1948, 4, 216-233. Cronbach, L. J. Statistical methods applied to Rorschach scores, a review. Psych. Bull., 1949, 46, 393-429- Delmas-Marsalet, P. L'electro-choc par courant continu. C. R. de la Soc. de Biol. de Bordeaux, 11 Feb., 1942. Diamond, B. L., and Schmale, H. T. The Mosaic Test. Amer. J. Orthopsychiat., 1944, 14, 237-250. Duchenne, G. B. C. De L1electrisation localize et de son application a la pathologie et a la therapeutique. Paris, 1855• Dusser de Barenne, J. G., and McCulloch, W. S. Amer. J. Physiol., 1936, 114, 692. 192 F a b i n g , H. D. The n e u r o l o g i c a l b a s i s o f memory d e f e c t s f o l l o w i n g e l e c t r o c o n v u l s i v e t h e r a p y . J . n e r v . m e n t . P i s ., 1955, 121, 19-25- F e r g u s , A. A r e p o r t on t h e e l e c t r o c o n v u l s i v e t r e a t m e n t o f 102 l o n g - t e r m s c h i z o p h r e n i c p a t i e n t s . A m er. J . P s y c h i a t . , 1952, 102, 439-443. F l y n n , J . P. R o r s c h a c h and W e c h s l e r - B e l l e v u e c h a n g e s f o l l o w - EST i n t h e a g e d . U n p u b l i s h e d d o c t o r a l d i s s e r t a t i o n , L o y o l a , 1954. F o r d , M. The a p p l i c a t i o n o f t h e R o r s c h a c h t e s t t o y o u n g children. Univ. of Minn. Inst. Child Welfare. Mono graph, 1946, 23. F r e id m a n , C . , and W i l c o x , P. H. E l e c t r o s t i m u l a t e d c o n v u l s i v e d o s e s i n i n t a c t humans b y m eans o f u n i d i r e c t i o n a l c u r r e n t s . J . n e r v . m e n t . P i s ., 1942, £6, 56-63. F u n k e n s t e i n , P . , G r e e n b l a t t , M ., and S o lo m o n , H. P s y c h o - p h y s i o l o g i c s t u d y o f m e n t a l l y i l l p a t i e n t s . I . The s t a t u s o f t h e p e r i p h e r a l a u to n o m ic n e r v o u s s y s t e m a s d e t e r m i n e d b y r e a c t i o n t o e p i n e p h e r i n and m e c h o l y l . A m er. J* P s y c h i a t . , 1949, 106, 16-28. Gellhorn, E. Analysis of autonomic hypothalamic functions in the intact organism. Neurology, 1956, 6, 325_343« G e l l h o r n , E. A u to n o m ic i m b a l a n c e and t h e h y p o t h a l a m u s . Univ. of Minn., 1957* G e r v i t z , M. S. Some d e f e c t s o f t h e W e c h s l e r - B e l l e v u e . J . C o n s u l t . P s y c h o l . , 1952, 16, 124-126. G r e e n e , W. F . Ego f u n c t i o n i n g i n R o r s c h a c h r e s p o n s e s . Un p u b l i s h e d d o c t o r a l d i s s e r t a t i o n , U.C.L .A ., 1954. G r o s s , F . A s u r v e y o f t h e f i r s t tw o y e a r s i n a l a r g e p r i v a t e h o s p i t a l . A m er. J . P s y c h i a t . , 1952, 109, 32-34. G u n d e r s o n , M a x in e . An e x p l o r a t o r y i n v e s t i g a t i o n o f t h e u s e f u l n e s s o f p s y c h o l o g i c a l t e s t s i n p r e d i c t i n g t h e o u t come o f e l e c t r o s h o c k t h e r a p y w i t h s c h i z o p h r e n i c p a t i e n t s . U n p u b l i s h e d d o c t o r a l d i s s e r t a t i o n , U.C.L.A., 1953. H e a th , R. S . , and Norman, E. E. E l e c t r o s h o c k t h e r a p y by s t i m u l a t i o n o f d i s c r e t e c o r t i c a l s i t e s w i t h s m a l l e l e c t r o d e s . P r o c . S o c . E x p e r . B i o l . and Med. , 1946, 63» 496. 193 Heath, R. G., Munroe, R. R., and Michel, W. A. Stimulation of the amygdaloid nucleus in a schizophrenic patient. Amer. J. Psychiat., 1955* 111> 862-863. Heatherington, R. Psychological changes due to electrocon vulsive therapy as measured by a battery of tests given repeatedly. J. ment. Sci., 1952, 28* 334-341. Hemphill, R. C., and Walter, W. G. The treatment of mental disorders by electrically produced convulsions. J. ment. Sci.., 1941, 82* 256. Himmelweit, H. T., and Eysenck, H. J. An experimental analy- sys of the Mosaic Projection Test. Brit♦ J. med. Psy chol . , 1945, 20, 283-294. Hirschfeld, G., and Bell, J. Differential cerebral stimula tion. Pis. Nerv. Syst., 1951* 2* 3~7- Histoire de L'Academie Royale des Sciences. Paris, 1744. Holzberg, J. D., and Cohen, Eleanor. The relationship be tween psychiatric improvement and certain pathologic changes in the Rorschach during electroconvulsive ther apy- J- Clin. Exp. Psychopathol., 1952, 12, 237~246. Howard, L. R. C. Intelligence and ECT prognosis. Shiksha, 1956, 2(1), 52-58. Jackson, C. V. Estimating impairment on Wechsler-Bellevue subtests. J. Clin. Psychol., 1955* 11* 137~l43- Jacobs, J. S. L. Personal communication. (1954) Janis, A. L. Psychologic effects of electric convulsive treatments. J. nerv. ment. Pis., 1950, 111, 359-469- Janis, A. L., and Astrachan, M. The effects of electrocon vulsive treatments on memory efficiency. J. Abn. Soc. Psychol., 1951* 40* 501. Jastak, J. Ranking Bellevue subtest scores for diagnostic purposes. J. Consult■ Psychol., 1953* 1Z* 403, 410. Jeans, R. F., and Toman, J. E. P. Anxiety and cerebral ex citability; prolongation of seizure latency by anxiety and other factors in patients undergoing electroconvul sive treatment. A.M.A. Arch. Neurol. Psychiat., 1956, 12* 534-547. J o h n s o n , E. Z. K l o p f e r ' s P r o g n o s t i c S c a l e u s e d w i t h R a v e n ' s P r o g r e s s i v e M a t r i c e s i n p l a y t h e r a p y p r o g n o s i s . J n l . 194 Pro.i. Tech. , 1953, 11, 320-327- Kalinowsky, L. B., and Hoch, P. H. Shock treatments and other somatic procedures in psychiatry. New York: Grune and Stratton, 1946. Kalinowsky, L. B. Clinical evaluation of electric convul sive therapy. Res. Pub. Ass. nerv. ment. Pis., 1951, 21, 232-238. Kendall, Barbara, Mills, W. B., and Thale, T. Comparison of two methods of electroshock in their effects on cogni tive functions. J. Consult. Psychol., 1956, 20, 423“ 429- Kerr, M. The validity of the Mosaic Test. Amer. J. Ortho- psychiat., 1939, 2.> 232-236. Kirkner, F. J., Wisham, W. W., and Geidt, F. H. A report on the validity of the Rorschach Prognostic Rating Scale. J. Pro.i . Tech., 1953, 11, 465-471. Klopfer, B., and Kelley, D. M. The Rorschach technique: a manual for a projective method of personality diagnosis. Yonkers: World Book Co ., 1942 . Klopfer, B., Kirkner, F. J., Wlsham, W. W., and Baker, Ger trude. Introduction: The development of a prognostic rating scale. J. Pro.i . Tech., 1951, !5>, 421. Klopfer, B., Ainsworth, Mary, Klopfer, W. G., and Holt, R. R. Developments in the Rorschach technique. New York: World Book Co., 1954. Klopfer, B., Crumpton, E., and Grayson, H. M. Rating scales for ego functioning applicable to diagnostic testing. J. Pro.i . Tech., 1958, 22, 70-81. Korngold, M. An investigation of some psychological effects of electric shock treatment. Unpublished doctoral dis sertation, U.C.L.A., 1953- Krasner, J. D. The psychological effects of electroconvul sive therapy. Dissertation Abstr., 1952, 12, 591- Levit, H. I. A study of the effects of electroconvulsive therapy on certain psychological and physiological func tions in paranoid schizophrenia. Dissertation Abstr., 1955, 12, 1440-1441. L e w in , M. L. V a l i d a t i o n o f t h e M o sa ic T e s t . J . C o n s u l t . P s y c h o l., 1956, 20, 239-249. 195 Liberson, W. T. New possibilities in electric convulsive therapy: "brief stimuli" technique. Digest Neurol. & Psychiat. , 1944, 12, 368-369. Liberson, W. T. Time factors in electric convulsive therapy. Yale J. Biol. & Med., 1948, IX, 571-578 (a). Liberson, W. T. BST: Physiological and clinical observa tions. Amer. J. Psychiat., 1948, 105, 28-29 (b). Liberson, W. T. Current evaluation of electric convulsive therapy. In Psychiatric treatment. Res. Pub. Assoc. Res. Nerv. & Ment. Pis. Baltimore: The Williams and Wilkins Co., 1953- Liberson, W. T., and Wilcox, P. H. Electric convulsive ther apy; comparison of "brief stimuli" technique with FWR technique. Digest Neurol. & Psychiat., 1945, 8, 292- 302. Lopez-Ibor, J. J. Respective indications of shock methods. Congres international de psychiatrie, Paris, Rapports, 195071, 85-134. Lowenfeld, Margaret. The Mosaic Test. Amer. J. Orthopsy- chiat., 1949, 1£, 537-550. Lowenfeld, Margaret. The Lowenfeld Mosaic Test. J. Pro.i . Tech., 1952, 16, 200-202. Lundin, W. H., and Schpoont, S. Application of the Rorschach Prognostic Rating Scale to one intensively followed case. J. Pro.i . Tech. , 1953, IX, 295~300. Maher, B., and Martin, A. Mosaic productions in cerebroar- teriosclerosis. J. Consult. Psychol., 1954, 18, 40-42. Malish, H. B., Hanlon, T. E., and Kurland, A. A. A prognos tic indicator of treatment failure in schizophrenia. Arch. Neurol. Psychiat., 1957, 78, 177-193- Malzberg, B. Rates of discharge and rates of mortality among first admissions to the New York Civil State Hos pitals. Ment. Hyg., New York, 1952, 36, 618-638. Marilyn, A. C. Post-ECT confusion. J. nerv. ment. Pis., 1952, 116, 352-354. Marks, M. R. A criticism of the use of the Wechsler-Belle- vue Scale as a diagnostic instrument. J. Gen. Psychol., 1953, 42, 143-152. 196 Medlicott, R. W. Brief stimuli electroconvulsive therapy. New Zealand Med. J ., 1948, 4l, 29-37* Meduna, L. V. Versuche liber die biologische Beeinflussung des Ablaufes der Schizophrenie; Kompfer und Metrazol- krclmpf e. Arch. Psychiat., 1935, 105 , 92-135* Michael, S. T. Impairment of mental function during electro- convulsive therapy. Arch. Neurol. Psychiat., 1954, 71, 362-366. Miles, W. G. Brief stimulus convulsive therapy. J. Florida Med. Ass., 1949, 36, 25-26. Mindess, H. Predicting patient's responses to psychotherapy: A preliminary study designed to investigate the validi ty of the Prognostic Rating Scale. J. Pro.i . Tech., 1953, II, 327-335. Mintz, M., Schmeidler, G. R., and Bristol, M. Rorschach changes during psychoanalysis. J. Pro.i . Tech., 1956, 20, 414-418. Monroe, Ruth. Objective methods and the Rorschach blots. Ror. Res. Exch., 1945, 2, 59“73- Morrow, J. L. The advantages and disadvantages of e.s.t. from the standpoint of psychodynamic psychotherapy. Q.uart. Review Psychiat. Neurol. , 1951, 6, 153-157• Morrow, R. S., and Mark, J. C. The correlation of intelli gence and neurological findings on 22 patients autop- sied for brain damage. J. Consult. Psychol., 1955, 19, 283-289. Motoaki, H., Tomita, M., and Yumato, Y. A study of form lev el rating in the Rorschach test. Jap. J. Psychol., 1957, 28, 86-93. Offner, F. Stimulation with minimum power. J. Neurophysiol. , 1946, 2, 387-390. Oppenheim, Sadi, and Brower, D. Effects of electroshock therapy as revealed by the Rorschach technique. Amer. J. Psychiat. , 19*12, 98, 828-833. Parsons, E. H., Bonnet, S. R., Keating, J. U., and Lawrence, J. H. Modern methods of electroshock therapy. Amer. J. Psychiat., 1955, 112, 18-22. Peak. H. M. A quantitative approach to the Lowenfeld Mosaic Test. Unpublished doctoral dissertation, U.S.C., 1953* 197 Perlson, J. A case of schizophrenia treated with 248 elec- troconvulsive treatments. Arch. Neurol. Psychiat., 1945, 111-125. Piotrowski, Z. Prognostic possibilities of the Rorschach method in insulin treatment. Psychiat. Quart., 1938, 12, 679-689. Piotrowski, Z. The Rorschach method as a prognostic aid in insulin shock treatment of schizophrenia. Psychiat. Quart., 1941, 15, 807-822. Piotrowski, Z. The alpha formula for the differential diag nosis of schizophrenia versus neurosis. Psychiat. Quart., 1950, 24, 543-552. Piotrowski, Z. The practical application and confirmation of the alpha formula for the differential diagnosis of schizophrenia versus neurosis. Amer. J. Psychiat., 1955, 112, 443-460. Rabin, A. I. Effects of electric shock treatments upon some aspects of personality and intellect. The American Psychologist, 1947, 2, 284. Rapaport, D. Diagnostic psychological testing. Chicago: The Year Book Publishers, Inc., 19^9* Reiman, Gertrude. The Mosaic Test: Its applicability and validity. Amer. J. Orthopsychiat. , 1950, 20, 6OO-615. Rioch, Margaret. The Mosaic Test as a diagnostic instrument and as a technique for illustrating intellectual disor ganization. J. Pro.i . Tech., 1954, 18, 89~94. Rosenbleuth, A., and Cannon, W. B. Cortical responses to electric stimulation. Amer. J. Physiol., 1942, 135, 690-741. Salzman, L. An evaluation of shock therapy. Amer. J. Psy chiat . , 1947, 102, 669. Sargant, W., and Slater, E. An introduction to physical methods of treatment in psychiatry. Baltimore: Wil liams and Wilkins Co., 1954. Scherer, I. W. The effect of brief stimulus electroconvul- sive therapy upon psychological test performances. J. Consult. Psychol., 1951, i2, 430-434. Scherer, I. W. The effect of different types of electrode application (BST) upon retrograde amnesia, post- 198 convulsive recognition, and dynamometer readings. J. nerv. ment. Pis., 1955? 116, 226-232. Schiedemantel, P. C. G. Die Leidenschaften als Meilmittel betrachtet. Hildburghausen, 1797* Sheehan, J., Frederick, C., Rosewar, V/., and Spiegelman, M. Use of the Rorschach test in predicting the outcome of psychotherapy in a group of stutterers. J. Pro.i • Tech., 1954, 18, 233-239. Spearman, C. The proof and measurement of association be tween two things. Amer. J. Psychol., 1904, 1J5, 72-101. Spiegelman, M. Rorschach Form Level, intellectual function ing and ootentiality. J. Pro.1 . Tech., 1956, 20, 325~ 343. Smykol, A., and Wilson, M. 0. Wechsler-Bellevue subtest score changes resulting from electroconvulsive therapy. Proc. Okla. Acad. Sci., 1950, 31, 148-149. Steiper, D. W., Williams, M., and Duncan, C. P. Changes in personal and impersonal memory following e.c.t. J. Clin. Psychol., 1951? Z? 361-366. Stephenson, W. Q Technique and the Rorschach Test. Chapt. VIII in Beck, S. J., The six schizophrenias. Res. Mon- ogr., No. 6, Amer. Orthopsychiat. Ass. , 1952. Stephenson, W. The study of behavior: Q,-technique and its methodology. Chicago: Univ. of Chi. Press, 1953- Stewart, U., and Leland, L. American versus English Mosaics. J. Pro.i . Tech., 1952, 16, 246-249. Stewart, U., and Leland, L., and Streiter, E. Mosaic pat terns of eighth grade children. J. Pro.i . Tech., 1957? 21, 73-79. Stone, C. P. Losses and gains in cognitive functions as re lated to electroconvulsive shocks. J. Abn. Soc. Psy chol. , 1947? 24, 206. Summerskill, J., Seeman, W., and Meals, D. W. An evaluation of post-electroshock confusion resulting from the use of the Reiter EST apparatus. Abstr. The Amer. Psychol ogist, 1951? 5? 344. Sweel, A. Combined electroconvulsive-electrostimulatory therapy. J. Clin. exp. psychopathol., 1958, 2, 118-127. 199 Sym on ds, P. M. A c o n t r i b u t i o n t o o u r k n o w le d g e o f t h e v a l i d i t y o f t h e R o r s c h a c h . J . Pro.i . Te ch ., 1955; 1£; 152- 163- Taulber, E. S. The use of the Rorschach test in evaluating intellectual levels of functioning in schizophrenics. J. Pro.i. Tech., 1955; !2, 163-170. Toman, J . E . , and Goodman, L. S . C o n d i t i o n s m o d i f y i n g c o n v u l s i o n s i n a n i m a l s . In E p i l e p s y . R e s . P u b . A s s o c . Res. Nerv. Ment. Pis. Baltimore: The Williams and Wil- kind Co., 1947. Pp. 141-163. Ullersperger, J. B. Die Anwendung der Elektrizitclt. Munich, 1875. Walker, H. M. Degrees of freedom. J. Educ. Psychol., 1940, 21, 253-260. Wechsler, D. The measurement of adult intelligence. Balti more: The Williams and Wilkins Co.^ 1944. W ertham, P . , and G o ld e n , L. A. A d i f f e r e n t i a l d i a g n o s t i c method of interpreting mosaic and colored block designs. Amer. J. Psychiat., 1941, 98, 124-131. Wideman, H. R. The application of quantitative procedures to the scoring and validation of the Lowenfeld Mosaic Test. Amer. Psychologist, 1953; 8, 455-470. Wideman, H. R. Development and validation of an objective scoring method for the Lowenfeld Mosaic Test. J. Pro.i . Tech., 1955; 12, 177-192. Wilcox, Katharine. Intellectual functioning as related to electroconvulsive therapy. Dissertation Abstr., Univ. of Michigan, 1954. Wilcox, P. H. Electroshock therapy: A review of over 23;000 treatments using unidirectional currents. Amer. J. Psychiat., 1947; 104, 100-125* Wilcox, P. H., Funderburk, W. H., and Wilcox, Katharine. In Spiegel, E. A. (Ed.), Progress in neurology and psychi atry . Vol. 10. New York: Grune and Stratton, 1955* Williams, Marya. Memory studies in electroconvulsive thera- py. J. Neurol. Neurosurg. Psychiat. , 1950, 12, 30-35- Wolff, G. E. Electric shock treatment. Amer. J. Psychiat♦, 1955, 111. 748-750. 200 Worchel, A., and Gentry, G. Electro-convulsive shock and memory. Comp. Psychol. Monogr., 1950, 20, 95- Wyss, 0. A. M., and Obrador, A. Adequate shape and rate of stimuli in electro-stimulation of the cerebral motor cortex. Amer. J. Physiol., 1937> 120, 42-51. Y a t e s , F. I n S n e d e c o r , G. W., S t a t i s t i c a l m e t h o d s . Ames, Io w a : C o l l e g i a t e , 1937- P. l 6 l . Zirkle, G. A. Learning in the immediate post-electroshock period. J. Consult. Psychol., 1956, 20, 399-402. Zubin, J. Failures of the Rorschach technique. J. Pro.i . Tech., 1954, 18, 303-315. Zubin, J., Eron, L. D., and Sultan, F. Current status of the Rorschach test. Symposium, 1955* 1* A psychomet ric evaluation of the Rorschach experiment. Amer. J. Orthopsychlatr., 1956, 26, 773-782. Z u c k e r , L. The c l i n i c a l s i g n i f i c a n c e o f t h e M o s a ic and R o r s c h a c h m e t h o d s . A m er. J. P s y c h o t h e r■, 1950, 4, 473-487.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Relationship Of Dependency To Verbal Learning Without Awareness

PDF

Prediction Of Therapeutic And Intellectual Potential In Mentally Retardedchildren

PDF

Measuring Thought Process As An Ego Function In Schizophrenic, Mentally Retarded And Normal Adolescents By Means Of The Rorschach

PDF

Ego Diffusion In Women With Behavioral Disorders And The Integrating Effects Of Psychodrama In Identity Consolidation

PDF

First Steps In An Attempt To Construct An Objective Test Of Character Structure

PDF

Psychological Variables In The Mother Related To Infant Feeding Patterns

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

Perceptual Defense And Somatization: A Comparison Of The Perceptual Thresholds Of Obese And Peptic Ulcer Patients

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

Effects Of Different Treatment Procedures On Reading Ability And Anxiety Level In Children With Learning Difficulties

PDF

Self-Worth, Future Goals, Mood And Time Perception

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

On The Relationship Between Anxiety And Aggression In Nine-Year-Old Boys

PDF

A Semantic Differential Investigation Of Critical Factors Related To Achievement And Underachievement Of High School Students

PDF

An Investigation Of The Response To Stress Of Patients Hospitalized For Anxiety State And Peptic Ulcer Patients

PDF

An Investigation Into The Communication Style Of Suicidal Individuals

PDF

The Effects Of Group Experiences On The Aged

PDF

Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

PDF

Use Of College Students In A Social Therapy Program With Chronic Schizophrenics To Produce Changes In Mood And Social Responsiveness

PDF

The Visual Fusion Threshold (Vft) Test As A Measure Of Perceptual Efficiency In Kindergarten And First Grade, And As A Possible Predictor Of Later Reading Retardation

Asset Metadata

Creator Sharp, James Henderson (author)

Core Title Psychological Test Changes In Schizophrenic Patients Under Brief Stimuluselectroconvulsive Therapy

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 Seward, Georgene H. (committee chair), Carnes, Earl F. (committee member), Meyer, Mortimer M. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-49279

Unique identifier UC11357930

Identifier 5905025.pdf (filename),usctheses-c18-49279 (legacy record id)

Legacy Identifier 5905025.pdf

Dmrecord 49279

Document Type Dissertation

Rights Sharp, James Henderson

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