, V ngfit WWWOFWDWMOUGH MWOFWNEIMVWL W Thai: for tho 0m 0f Pk. D. MW” Sfi‘ATE UNIVERSITY: Ronald Irwin Rib!” 1957 Igicst? This is to certify that the thesis entitled The Detection of Brain Damage through Measurement of Deficit in Behavioral Function presented bg Ronald I. Ribler has been accepted towards fulfillment of the requirements for Ph .D . Psychology degree in Z%W Major professor Date %¢z/1/é/;f7 0-169 THE DETECTION OF BRAIN DAMAGE THROUGH MEASUREMENT OF DEFICIT IN BEHAVIORAL FUNCTIONS By RONALD IRWIN RIBLER AN ABSTRACT Submitted to the School of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology Year 1957 Approved AN ABSTRACT BY Ronald Irwin Ribler- This study was designed in an attempt to gain a more complete understanding of behavioral deficit in the brain damaged individual. It was hypothesized that a battery of psychological tests, which measure the functions of abstract thinking, perceptual-motor coordination and memory, would be more valuable than any of these tests individually. This battery of tests was administered to 25 brain damaged subjects, 25 schizophrenic subjects and 25 "normal", non-psychiatric subjects. This battery was composed of the Grassi Block Substitution Test (22), the Hunt- Minnesota Test for Organic Brain Damage (3A), and the Bender Visual Motor Gestalt Test (8). Scoring criteria were derived from the first seven subjects in each group. By converting the test scores to standard scores, a sin- gle battery score was developed by addition of the indi- vidual standard test scores. It was found that positive battery scores differentiated the non-brain damaged from -2- Ronald I. Ribler the brain damaged at a very significant level. The second hypothesis of this study was that there would be more deficit revealed by the brain damaged group than by the schizophrenic group. This was found to be true, since the performance of these two groups revealed statistically significant differences, with the brain damaged group revealing more deficit. The final hypothesis was that we could expect re- duction in deficit as a function of time after brain damage. Theeight most recently injured and the seven least recently injured subjects in the brain damaged group were compared, using the battery score. It was found, on these If; subjects, that there was perfect discrimination of the recent and old injuries; the most recent revealing more deficit. Therefore, we may conclude from this study that: 1. There are very likely multiple factors of brain func- tion which are susceptible to injury and that some of these factors may vary with the individual. Thus, a bat- tery of tests which is designed to measure these several aspects of behavior is more effective than each individu- al test. 2. Schizophrenics are not as likely to reveal behavioral -3- Ronald I. Ribler deficit in the same areas as the brain damaged indivi- duals. Although the schizophrenic may function at a lower level of personality integration than the brain damaged, there are discreet functions which are retain- ed. This brings the problem of brain dysfunction in schizophrenia into somewhat sharper focus. 3. It seems reasonable to conclude that there is recov- ery from behavioral deficit in the brain damaged indivi- dual even though there is no neural regeneration. THE DETECTION OF BRAIN DAMAGE THROUGH MEASUREMENT OF DEFICIT IN BEHAVIORAL FUNCTIONS By RONALD IRWIN RIBLER A.THESIS Submitted to the School of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1957 TABLE OF CONTENTS ACKNOWLEDGEMENTS................................... LIST OF TABEES..................................... I. INTRODUCTION................................... A. Qualitative Tests......................... B. Quantitative Tests........................ 6. Perceptual and Mbtor Tests................ II. PILOT STUDY.................................... A. Procedure of Pilot Study.................. 3. Purpose................................... 0. Mhin Hypothesis of Pilot Study............ D. Results of Pilot Study.................... E. Discussion of Pilot Study................. III. EXPERIMENTAL STUDY............................. A. Hypotheses of Experimental Study.......... B. Procedure for Experimental Study.......... 0. Results and Discussion of Experimental Study. e e e D. Theoretical Implications.................. E0 Clinical Implicationseeeeeeeeeeeeeeeeeeeee Fe Summary and Conclusionl....o.............. BIBLIOGRAPHIOQOOOOOOOQOQOeeeeeoeeo00000000000000.0000 ii iv 17 23 29 38 #2 #5 47 #9 #9 A9 52 68 3 O. O APPENDICES Al - Grassi Block Substitution Test Record Form Scoring Criteria for the Bender Visual- thor Gestalt Test. 12 B1 - 5 Score Conversion of Grassi EST raw scores. BZ - 3 Score Conversion of Hunt-Minnesota Test for Organic Brain Damage. s Score Conversion of Bender Visual-Motor Gestalt Test. 33 ACKNOWLEDGEMENTS It is impossible to convey the depth of feeling to those who have helped with the undertaking and comple- tion of this study. Her is it possible to give adequate mention to all those people, without whose cooperation this work would have been impossible. I should like to convey the deepest gratitude to Dr. G. M; Gilbert, committee chairman, whose guidance and bolstering were so instrumental in the completion of this work. Thanks are also due to Drs. D. M. Johnson, A. G. Dietse, M. R. Denny, H. C. Smith and A. I. Rabin who offered their guidance, criticisms and assistance whenever called upon throughout this undertaking. I should also like to take this Opportunity to extend ,Iy gratitude to the many persons who made the collection of the data possible. These include the staffs of Dr. S. G. Armitage at the Battle Creek VA H03pital, Dr. J. J. Brownfain of the VA Hospital, Dearborn, Dr. A. Kahn, neurosurgeon of the University of'MHchigan Hospital, Dr. G. Hover of the Ann Arbor VA Hospital, and the Commander of the 1st USAF Hospital at Selfridge Air Force Base. Perhaps the one person offering the most inspiration and moral support throughout the years of work which have gone into this study is my wife, Nancy. LIST OF TABLES Table l Matched Variables for the Three Groups....... 2 2 Score Conversion of Test Scores............ 3 Chi Square Analysis of the Three Tests on the Three Groups Of Subjects................. 4 Raw Scores on Three Tests with Three Groups.. Ranges of Raw Scores on Three Tests with Three Groups.....OOOOOO....0...00.000.000.00. 6 mean Vocational Score, A e & Education, for Three Diagnostic Groups N-75)............... 7 Total Number of Misclassifications on Each Test and on Battery Scores (Derived Criteria) 8 Errors Using Original and Derived Cut-Offs on Individual Tests.......................... 9 Chi Square Analysis, Comparing the Number of Persons Misclassified by the Individual Test Scores and by the Battery Scores............. 10 Chi Square Analysis Comparing the Number of Persons Misclassified by the Original Author's Criteria and by the Derived Criteria on the Grassi Block Substitution Test........ 11 Chi Square Analysis Comparing the Number of Persons Misclassified by the Original Author's Criteria and by the Derived Criteria on the Hunt-Minnesota Test for Organic Brain DamagQOOOOOOOOOOO0.0.0....OOOOOOOOOOOOOOOOOOO 12 Chi Square Analysis Comparing the Number of Persons Misclassified by the Original Author's Criteria and the Derived Criteria on the Bender Visual-motor Gestalt Test...... ~4v- Page #1 Ah #5 46 46 SO 52 53 53 5k 5k 55 Table Page 13 Comparison of Derived Cut-Off Scores and Original Author's Cut-Off Scores.............. 55 1h Ranges of Raw Scores of Three Tests with Three Diagnostic Groups.....C...........00.... 56 15 Results of Analysis of Variance by Ranks (Kruskal-Wallis H-Test) Comparing Battery Scores and Scores of Individual Tests on Three Groups.................................. 57 16 Chi Square Analysis Comparing Brain Damaged and Non-Brain Damaged Subjects on the Battery score (smned Z).COCOOOOCOOCCCOCOOOOOOOOO0.... S7 17 Chi Square Analysis Comparing Brain Damaged and SchiZOphrenic Subjects on the Battery ScoreOOOOOIOOOOOOOOO00.0000...000.00.000.00... 58 18 Chi Square Analysis Comparing SchiZOphrenic and Normal Subjects on the Battery Score...... 58 19 Fisher's Exact Test Relating Score to Time Since injuryOOO0.00000000000000000000.0.0.0... 6]. -v- I. INTRODUCTION Ever since they were first seriously noted around the turn of the last century, extensive investigations have been made into the problem of cerebral lesions with respect to their psychological and physiological effects. Descriptive data dealing with the effects of brain damage have been forthcoming since the time of Hippocra- tes. Harlowis description of Phineas Gage, the man who accidentally had an iron rod driven through both his frontal lobes, in now classical (26). Behavior changes and observable personality anomalies have been subsequent- ly reported in a manner not unlike Harlow's report. Since these early observations, numerous scientific experiments and observations have been reported with both humans and animals. The most significant accomplishments having appeared in the last twenty to thirty years. This work has been done in the areas of physiology, medicine, and in psychology by learning theorists, perception theor- ists, physiological psychologists and clinical psychol- ogists. There are many sources of confusion in studying changes of behavior in patients suffering brain lesions. 1" -2- One is that we may never know exactly where the lesion is. Another is that we may not have seen the patient before he suffered his lesion to know just what changes in behavior to ascribe to the lesion. These are rather obvious difficulties. There are also some other import- ant factors, which have only recently been receiving prop- er emphasis. One of these is individual differences in individual brain structure. Just as people have differ- ent shapes of faces and bodies, so their brains vary in both gross and microscopic structure (A0). The positions of the primary fissures and sulci certainly vary from one person to another. So, too, do the shapes and sizes of some of the major gyri and convolutions. Such individual differences naturally enter the clinical picture, In two different individuals, lesions that sometimes appear to involve exactly the same regions of the brain may actually involve somewhat different functional areas. If the le- sions do not produce the same deficits in behavior, it may be due to individual differences in the size and shape of the parts of the brain. There is the fact, too, that almost any behavioral deficit manifested after a brain lesion has been incurred occur in the so-called functional states (A8). Stage fright may make a person just as speechless as he sometime is -3— after certain brain lesions which produce aphasias. In some of the postepileptic states we see people unable to understand written or spoken language, just as though they had severe lesions of the brain. In hysteria and other neurotic disorders we sometimes see cases of in- ability to recall names of familiar persons or objects; apparently the same kind of inability that occurs in some brain injuries. It is, in fact, possible by sug- gestion and hypnosis to duplicate most of the syndromes of impairment that occur in brain damage (66). If, when we may have purely functional disorders, we obviously have much cause for confusion and the prob- ability of interaction between the two. People seldom suffer brain damage without being rather seriously af- fected in their lives, their personal relationships, finances, employment and so on. Thus, almost certainly, there are functional symptoms of illness that are not directly due to the brain lesion. There are many cases mistaken for brain lesions because they revealed behavior symptoms classically ascribed to brain disorder. We have good reason to believe that individual dif- ferences in psychological make-up have a great deal to do with psychological behavior accompanying brain le- sions. Our evidence on this point, however, is rather -4- indirect and not very specific. According to Klebanoff, et al., "The correlation of psychological test perform- ance with specific areas of brain damage has been found to be limited by vast differences in brain pathology caused by serious limitations in the techniques of ana- tomical localization. In general, psychological instru- ments have proven incapable of differentiating patients with presumptive injury to specific cortical areas." (39). \ Morgan and Stellar (A6) make the statement, "As psychologists we are not interested in brain lesions nor in the anatomical areas of the brain that may suffer disease. we are interested in the physiological mechan- isms of behavior. To get at these mechanisms, however, we must 'catch as catch can' -- we must take whatever brain injuries nature provides and see as best we can what kinds of disturbances of behavior go along with these injuries. That is about all we can do, ..." 4 Before entering into a discussion of intracranial pathologies, a definition of brain damage is required. Organic brain damage, as it shall be referred to in this dissertation, is "an injury, structural change (of the cortex) of pathological type, or wound to tissues" (27). -5- The need for tools to assist in the understand- ing of organic brain damage is widely recognized. Brain lesions of many kinds are difficult to recog- nize because of their size or location. Medical instru- ments and methodology, such as neurological examination, electroencephalograms, arteriorams, pneumoencephalograms have been developed for this purpose, but there is still considerable need for further work in this area. This is expecially true insofar as differential diagnosis is concerned. The clinician is frequently faced with the problem of differentiating between hysteria and epil- epsy, schizophrenia and brain damage, etc. Psychologic- al tests of many types have been used and are still be- ing used to screen and assist in the diagnosis of patients presenting these and similar problems. It is both incon- venient and expensive routinely to employ the complex medical diagnostic tools on all suspect cases. In addi- tion, these medical tests are not infallible. Thus one of the major roles of the clinical psychologist in the diagnosis of brain damage is that of screening agent. Since none of the tests designed to identify this organ- 'ic pathology have been successful in all cases, they fre- quently yield only a suggestion for further investiga- tion of the suspected pathology. -6- The fact that current periodical literature re- veals studies on many and varied devices and tests de- signed to diagnose the brain damaged individual suggests that there still is no reliable method for the clinical paychologist to use in his routine diagnostic work. Goldstein studied the effects of gunshot wounds of the head during World war I while he was working with the German Armies. He observed many gross changes, both phymclogical and psychological. These included diffuse vasomotor disturbance, feeling of dizziness, palpita- tion, headache, feeling cold, shivering, sensitivity to change of weather, etc, The psychological changes re- ported by Goldstein include impairment of abstract func- tion which he considered basic for the following potentia- lities: “Va. Assuming a mental set voluntarily; b. shifting voluntarily from one aspect of a situation to anoth- er, making a choice; 0. keeping in mind simultaneoiu- sly various aspects of a situation; d. grasping the essential of a given whole, breaking up a given whole into parts and isolating them voluntarily; e. abstract- ing common properties, planning ahead ideationally, assuming an attitude toward the 'merely possible', and thinking or performing symbolicall ; f. detach- ing the ego from the outer world.” (20 . These changes are most commonly found in lesions of the preforntal areas of the brain. Many of the instr- uments used for the detection of brain damage have these -7- clinical observations of Goldstein as an integral part of the test. It shall be one of the aims of this paper to examine the major psychological tools used and developed for the purpose of studying the brain damaged individual and, fur- ther, to determine the possibility of combining some of the existing tests to develop a more adequate battery for the purpose of gaining a finer understanding of the psychologi- cal mechanisms and greater accuracy of differential diagno- sis. With this information,‘further investigation of theories of brain function may be compared with the perfor- mance of the brain damaged patient. These will be discuss- ed more fully after the following survey of the literature concerning tests for organicity. Yates (72) has set down conditions which he feels must be fulfilled before the validity of any test can be discussed. These are: "a. Adequate samples including brain damaged, functional and normal groups. b. The degree of possible error should be given. The optimum cut-off point may be given, the point beyond which no normals or functionals fall should be given. If -3- the distribution is normal it should be so stated so that deviations may be considered by the clinician using the instrument. 0. Reliability of the data should be verified by using the test on another, independent sample or samples. The findings should be tested by another worker on a differ- ent sample in another locale. d. As many variables as possible should be controlled." Most of the problems encountered clinically in the diagnosis of brain damage are primarily those which require the utmost sensitivity to the disorder. It is not infrequent that the clinical psychologist receives referrals for differ- ential diagnosis between hysteria and brain damage or epi- lepsy. This is true of many varied types of questions ask- ed of the clinician. Most frequently there is a behavioral symptom which initially raises the question of cerebral involvement. This may vary from headaches to convulsive seizures. In addition, this referral is frequently made after a neurological examination has been performed. Thus, it is quite clear that the major contribution of tests of brain damage must be efficacious in this twilight of bor- derline zone, where there may be the possibility of seri- ous function disorder or CNS involvement. As will be seen from the review of the literature, there are very few tests -9- which even claim to be sensitive to all cases of known cerebral involvement. The primary concern of this study, however, is not to validate nor to standardize a test or tests for the diagnosis of brain damage, but to determine measurable deficits of psychological behavior manifested by the brain dameged patient. Therefore, it seems reason- able that a combination of testspurported to measure brain function along different parameters would be the ideal tool. There are of course many difficulties in this area. In the first place, localization of function of the various areas of the brain is a disputable fact. On the other hand, there may be obvious neurogenic signs of brain damage such as paralysis or hemianopsias which can only be caused by some anomaly in thenfunctioning of cellular structure of the brain or parts of the brain. It is common to find no deficit in intellectual functioning or other psychological changes except in specific areas. The problem then, seems to be one of utilizing a test or battery of tests which will identify brain damage whatever the locus, on the basis of changes affecting perception, memory and flexibility in thinking. We do not need psychological tests to re- veal a lesion in the motor cortex when we find a hemi- plegia. It is difficult to make a theoretical found- ation for the rationale of tests'such as these, for in re- ality we know very little about that which we are measuring. -10- The neruophysiology of the brain is still very much a mystery. Investigation of the literature reveals that there are many complicating aspects to the study of the brain injured individual, and in fact, to the concept of brain damage, or "organicity" itself. Investigations have revealed that one part of the brain can take over the function of another part and that parts of areas can manage, after a time, to do the job that the whole area normally does (A0). We know also that the brain is symmetrical. The left side is like the right. That raises the question of the equivalence of the two sides. Do we need both sides or will one do? Do we normally use both sides or only one? If we use one and lose it, can the other take over the job? These have been important questions for a long time, and they enter into much of the behavior we see after brain lesions. For each of the things we do most of us have a major side and a minor side. We write, throw balls, and carry objects with our right hands. A few, of course, are left- handed. Some are "switch hitters" -- they can use either hand. We see right-handedness and left-handed ness not only -11- in people but also in animals. So there can be little doubt that laterality, as we may call it, is a general biological characteristic of the higher animals (17,68). Even though there may be a general tendency for one side or the other to be the major one, there is still variety in specific habits as to which side is major and which is minor. We know that there is frequently definite re- covery from paralysis that follows lesions of the motor cortex so long as these lesions are not too severe. We run into the same kind of phenomena with disorders of memory. After a brain lesion there may be great losses of memory, but memory may gradually come back in the course of time. Sometimes even after severe lesions, recovery is so great that one would hardly know that the patient had had a brain injury. Such recovery has long been a problem. In many cases the recovery of memory after a lesion is quite slow and there is plainly an opportunity for the patient to relearn. On the other hand, there are recorded instances of rather sudden reappearance of memories weeks or months after a brain injury knocked them out. Such sudden recoveries certainly cannot be a matter of release from physiological shock or the return of nerve cells to -12.. normal functioning. For many years now they have been the subject of much debate and theory. Lashley (A8) has examined the records of cases that seem to illustrate these delayed sudden recoveries. He believes that they all have had some opportunity for re- training and that this is the explanation. Sometimes in- tense excitement of the patients helps them recover lost memories. In any event, the patients always seem to have some opportunity to practice the functions that recover. The sudden return of memory, therefore, is not a spontane- ous return but rather one that follows some retraining -- so Lashley holds. In some cases the retraining may be of some function that is central to a number of habits, so that practice in the function may bring back memories of many habits. Whether memories come back slowly or rapidly or whether they return singly or all together, one conclusion is sure: there is a change in the function of the brain in the course of recovery of memories. If a brain lesion des- troys a memory, it means that the memory must somehow depend upon the area destroyed. If after the lesion the memory comes back, it means that the memory now depends on the parts of the brain left uninjured. The question is how this change of function from one part to another comes about. -13- Neurologists have struggled with the question for a long time. Their answers fall into two general categories. One is vicarious function. This is the answer that almost any part of the brain can take over a function if necessary and can function in place of an injured area. Because an entire system, such as the motor cortex or the visual cen- ters, may seem to be destroyed without preventing some re- covery of function, it has often looked as though the var- ious parts of the brain could take over almost any function, i.e., function vicariously. Modern experimental evidence, however, points more and more in the direction of a second possible answer, viz., equipotentiality of a system. This term means that the parts, and only the parts, remaining in a system may take over functions of that system. Lashley (A0) and others who have carefully examined the clinical evidence believe that this principle also explains recovery of memory. They conclude that there can be recovery of memories only so long as parts of the system normally used or available for use in a memory are left intact. Other parts of the brain out- side the particular system cannot function vicariously. Adding support to the conclusion and providing another important point about recovery is that fact that recovery of function depends upon the mass of the regions of the -lA- cortex that are involved in a memory and that are left un- injured. A good illustration of this point is supplied by Lashley. In reviewing the literature on aphasia, he brought together 18 cases of indiviuals who had lesions in the left third frontal convolution, which has been regarded as the "speech area“ in man. Lashley ranked these patients for the amount of this area that was damaged by the lesion. He also ranked them for the degree to which they Later recover- ed from the aphasias produced by their lesions. The corre- lation between these two rankings was .90. Interpreted, this correlation indicates that the more of the "speech area" a patient had left, the more he recovered his ability to I use speech. The mass of the cortical area remaining was therefore a crucial factor in recovery (A0). It turns out that we must distinguish between simple and more complex functions. In the simple sensory-motor capacities and learning capacities, a brain lesion in in- fancy does less harm than it does in an adult. This is seen in comparing the motor capacities and the motor skills of persons injured in infancy and in adulthood. Those with the lesion in infancy do better. This point has been test- ed in animals with the same result. Hebb (29) ran a number of tests upon patients with adult and infant injuries. He tested one group of individuals between ten and nineteen -15- years of age who had suffered brain injuries in infancy. They showed impairment of both verbal and non verbal funct- ions. In general, however, their verbal capacity was aff- ected most. On the other hand, it was remarkable that in- dividuals who got their brain lesions as adults had no im- pairment on the verbal items of the intelligence test but were rather impaired on nonverbal capacities. Hebb consid- ered carefully where the brain injuries were in his two groups and has given us convincing evidence that the places of injury do not account for the differences between them. we need, then, a concept for understanding why in- fant injury causes impairment in both verbal and nonverbal functions while adult injury impairs nonverbal functions the most. The hypothesis that comes to mind is that the hemispheres and different areas of the cortex are more in- terdependent in learning habits than they are in the memory of the habits once they have been acquired. More specific- ally, nonverbal habits are instrumental in the development of verbal memories, but once acquiered, the verbal functions are not affected by injuries that affect the nonverbal fun- ctions. That is what Hebb says in a somewhat different way: Two factors are at work in psychological performance, one being "present intellectual power, of the kind essential to normal intellectual organization and behavior induced -16- by the first factor during the period of development" (29). By testing the hypotheses which follow (see section II-C) an attempt will be made to ascertain to a level of relative certainty, the critical time lapse after injury to the brain when identification of the injury is optimum, and that the recoverability of function extends beyond the motor, speech, and memory parameters. In addition, residual deficit of old injuries will be measured in an attempt to determine how certain we may be of diagnosing cerebral in- jury, and thereby gain further understanding of the concept of psychological deficit following injury to the brain. Another consideration before attempting to use tests of this type is to look at the work of others in the area. There are many reports of studies which did not find re- sults comparable to the original work. Perhaps in some of these cases an adjustment in cut-off score of change in the method of administration may have been a critical factor. A test may be perfectly valid using one criterion score in a given part of the country or type of institution, but this may vary widely from one place to another. A case in point is an incomplete study by Ziegler at the Neuropsychiatric Institute at the University of Michigan. Ziegler has found that the Grassi Block Substitution Test is not adequate when Grassi's criteria are used, but if the cut-off score is -17- raised to 18 it becomes quite discriminative (personal ver- bal communication). The existing tests for organic brain damage will be discussed in some detail, keeping in mind the critical as- pects as outlined by Yates, and the demands of the clinical situations where these tests are to be used. A. Qualitative Tests: The Goldstein-Scheerer Tests (21): The Goldstein-Scheerer Tests include the Goldstein- Scheerer Cube Test, The Gelb-Goldstein Color Sorting Test, The Gelb-GoldsteinAWeigl-Scheerer Object Sorting Test, and The Goldstein-Scheerer Stick Test. All five of these tests are qualitative in nature and are designed to test the ability of patients to use abstract thinking, as described earlier on page 6. To elaborate further on Goldstein's concept of the deterioration of the abstract ability in brain damage: "What the deterioration affects and modifies is the way of manipulating and operating with ideas and thoughts. Thoughts do, however, arise but can become effective only in a con- crete way" (21). A detailed description of these tests will not be given here because of the amount of space required to accomplish this. Briefly however, these tests may be described as qualitative tests which require the use of manipulative -18- ability as well as conceptual abilities. All of them in- clude the necessity of changing set, of being able to shift in a sorting task from one approach to another, or of re- producing abstract printed designs using tangible materials, such as blocks, sticks, etc. Among the main criticisms of the Goldstein Tests is that there are no quantitative data. Goldstein defends the lack of quantification by stating, "The reason (for this lack of quantification) rather is amethodological one which is intimately connected with the nature of the case material to be examined with these tests; we are dealing with sick individuals, with defective human beings" (21). Are we not in most clinical situations dealing with sick individuals, many of whom are defective in some way? In addition to the lack of quantitative data, the au- thors of the Goldstein Tests have offered no validity studies, and little work has been done by them to utilize these tests to differentiate schizophrenics from persons suffering from brain lesions. Although it is very likely true that Gold- stein and his co-workers do obtain diagnostic differentiation with these tests, others have not been able to utilize them consistently because of the subjective considerations inher— ent in the evaluation of the patients' performance. The absence of objective criteria for diagnosis has proven a definite handicap to the practicing clinician. -19- Hutton (35) believes that the failure on the Cube Test is due to overabstraction rather than failure of abstractions. If this is true, even the theoretical foundations of the Goldstein Tests may be threatened. Another shortcoming of the Cube test was found by Boyd (9). Using 5A normal hospitalized subjects, he found perfect performance if the IQ of the subject was over 100. Thus, the factor of intelligence enters into the problem of clinical use of this test, creating further complica- tions and difficulty in differential diagnosis, particu- larly when the differentialdiagnosis may be between mental deficiency and brain damage. Armitage (5) had little success with the Kohs blocks (used in the Goldstein Test) in attempting differential diagnosis of normal, neurotic and brain damaged patients. Tooth (67) used the Kohs and Weigl Color Form Sorting Test on a group of Naval officers and found insufficient discrimination. An important factor was discovered by Halstead (25) when he found that Goldstein's Test was valid in cases of frontal lobe damage but was not as definitive when the damage was present in other areas of the brain. Halstead used a rather small sample of 11 brain damaged subjects and 11 normals. Further study of this type seems to be indicated, since localization of lesions in the brain is -20- probably the most difficult of all diagnoses to make with the present state of our knowledge and tools. Recently, Maslow (A3) found that schizophrenic patients did not perform significantly poorer on the tests of ab- stract ability after lobotomy than they did before the sur- gery. This datum suggests that whatever is being tested by tests of concept formation may also be found in schizo- phrenia as well as in brain damage. Maslow did not use Goldstein's tests although he was attempting to test Goldstein's hypothesis regarding the loss of abstract ability. This study did not include a normal control group and the sample was not large. The concept of the "abstract attitude" is, however, an invaluable aid to the clinical psychologist in this area. Other authors (72) have made use of this concept, with the addition of ob- jective scoring systems, based on statistical analysis of a fairly large sampling of subjects. One of the major shortcomings of the Goldstein Tests is that schizophrenic patients frequently exhibit deficits similar to the brain damaged patient in their performance. The Grassi Block Substitution Test (22): This test also employs the Kohs blocks, as does Goldstein's Cube Test. Grassi made several innovations in -21.. the design of his test, however, which lend themselves to easy quantification and administration. Instead of using pictures for the subject to reproduce with the blocks as Goldstein does, Grassi has made a series of five three- dimensional designs consisting of four blocks glued to- gether which the subject must reproduce in four different ways. The first step in the reproduction of the design model is to make a design exactly like the tOp of the de- sign model, using the same colors and the same design. Grassi calls this the "simple concrete" step. The second task is to reproduce a design like the top of the model but with different colors. This is referred to as the "simple abstract" step. The third reproduction requires the subject to make his blocks conform to the three dimensions of the model, both in design and color. This is called the "complex concrete" step. Finally the second step is repeated on all three dimensions. This is the "complex abstract" step. The scoring procedure is as simple as the administration. The total score is the number of designs reproduced correctly plus or minus easily calcu- lated time credits. The maximum attainable score is thirty (see Appendix Al). Grassi reports better than 90 percent accuracy with almost no overlap using a cut-off score of 16 as indicative of brain damage. In his original study, he used a sample consisting of 86 schiz0phrenics, -22- half of whom were "deteriorated" and half of whom were not, 72 organics, 30 post lobotomy cases, and 86 normals. This is quite a large sample for this type of work. Even more unusual are the extraordinary results reported. Grassi found a test-retest reliability of .85 with no overlap of the groups. In spite of these rather remarkable results, Grassi found it necessary to qualify his study by stating that the quantitative score was not sufficient for diagno- sis and that qualitative evaluations of behavior must be made in many cases. The reason for inclusion of the qual- itative evaluation is not too clear since the obtained statistical results were so significant. In spite of these reservations, this test seems to offer promise. Another claim made by the author is that there is no correlation with intelligence in the performance on this test. He claimed that subjects with an IQ of 70 were able to ac- complish all of the designs perfectly. If this is true, further work should be considered for validation of this test on a mentally defective group. Recently Harris (28) did a study of the validity of this test and obtained results far short of Grassi's. She found that A2% of the brain damaged subjects scored above the cut-off point of 16, and that 85% of the psychotic and non-NP subjects scored above this point. Her results also -23- suggested that intelligence may be a factor, although her results were inconclusive. Further research with this test may offer clarification of several of these important fact- ors of correlation with intelligence and the validity of the scoring criteria. The rank correlations obtained from the present study are not significant. Ptacek and Young (52) compared the Grassi Test with the Wechsler-Bellevue on a group of brain damaged and non- brain damaged subjects and found that the W-B misclassified 50% as false negatives while the Grassi produced 25% false negatives; a difference significant at .05. The Grassi Block Substitution Test yielded no false positives while the W-B revealed 10% false positives. The Grassi was correct 8A% of the time and the W-B was accurate 7A%; not a stat- istically significant difference. B. Quantitative Tests: The following tests are based on the principle that brain damage leads to deterioration of an irreversible na- ture as contrasted with the functional disorders. The Hunt-Minnesota Test for Organic Brain Damage (3A): This test consists of three major divisions; a voca- bulary test, which has been found to be relatively insensi- -2A- tive to brain damage; a group of recall tests found to be sensitive to organic deterioration; and a group of inter- polated tests. The subject's age and vocabulary score determine his expected score. The amount of discrepancy between the expected score and the obtained score on the word and design associations (when corrected for age) is the basis for the diagnosis of brain damage. The test was deve10ped using a group of 33 patients suffering from brain damage, including many paretics and A1 controls con- sisting of 15 neurotics, ll normals, 6 psychotics, and 8 non-NP patients between the ages of 16 and 70. When Hunt used a cut-off score of 68 only one of the 50 standardization subjects was misclassified. His groups were not differentiated by the interpolated tests. The value of using persons suffering from diffuse brain damage (paretics) as his criterion group may be seriously challenged unless the test is going to be used for the diagnosis of advanced general paresis. Armitage (5) found that the criteria do not hold for traumatic brain injuries. There appears to be some question of the stability of vocabulary score in brain damaged subjects. Yacorzynski (71) and Capps (15), in separate studies question the validity of the statement that there is this stability in mental illness. Rabin, et al (53) found that vocabulary -25- scores in short-term schizophrenics are significantly lower than a normal group, but that long-term schizo- phrenics revealed a signifiCant lowering of vocabulary scores 0 Using the Hunt-Minnesota Test, Aita, et a1 (1) found that Al.3% of their brain damaged subjects obtained scores in the normal range. Only the mean of the severe- ly brain damaged group was significantly different from .that of the normal controls. Canter (1A) found that a group of A7 arteriosclerotic patients fell within normal limits. Malamud (A2) found that six out of ten members of the psychology department at her hospital scored in the pathological range on this test. She later tested 6A normal subjects and found that 5A.7% of these normal sub- jects obtained scores above Hunt's criterion score for brain damage. Other workers (36,AA) have reported similar results. There appear to be too many false positives among persons of high vocabulary level. As a consequence, if a normal score were to be obtained, the vocabulary score would have to be set at a maximum of 21 (36). The Shipley-Hartford Retreat Scale (62): This test is similar in design to the Hunt-Minnesota. -25- This test consists of two parts; a vocabulary test and a set of abstract reasoning problems. A Conceptual Quotient (CQ) is derived. The norms were established on 10A6 intel- ligent young normals. It was then validated on 171 state hospital cases and 203 private hospital cases. Shipley and Burlingame claimed discrimination between normals, schizophrenics and brain damaged subjects. Yates (72) criticizes this test on the basis of un- representativesampling. There was no control for age, sex or intelligence. Aita (l) and co-workers, Canter (1A), Nar- garet and Simpson (Al) were unable to obtain results Similar to Shipley. Garfield and Fey (19) found that CQ,declined steeply as a function of age. The use of the vocabulary score in this test is subject to the same criticism as in the Hunt-Minnesota (15,71). Findings of Nansen and Grayson, Kohler, Fleming, and Ross and McNaughton (72) in separate studies, further reveal the evidence of low validity. wechsler DI (70) (From the Wechsler Tests of Intel- ligence): The use of this test assumes that organic deterior - ation is similar to the deterioration accompanying age, differing only in its early onset. 'Wechsler found that the index discriminated between young normals and young brain damaged patients with a high percentage of overlap. /\ -27- The DI consists of comparison of tests which supposedly hold up with age: Information, Comprehension, Object Assembly, Picture Completion and Vocabulary; and tests which do not hold up with age; Digit Span, Arithmetic, Digit Symbol, Block Design, Similarities, and Picture Arrangement. The former group Wechsler calls the "Hold" tests and the latter the "Don't Hold" tests. To obtaina measure of deterioration, the sum weighted scores of the "Hold" tests are compared with the "Don't Hold". The deterioration quotient is deter- mined by the formula: .DI=DON'T HOLD -f— HOLD X 100. Wechsler defines deterioration: "an individual may be said to show signs of possible deterioration if he shows a greater than 10% loss, and of definite deterioration if a loss greater than 20% than that allowed for by the normal decline with age." (70). Modifications have been made by Reynell and Hewson (56,32). These are based on various combinations of the subtests. Neither of these are designed to diagnose brain damage per se, but rather to determine the degree of deteri- oration in cases of known brain damage. Gutman (23), Allen (2), Rogers (57), Andersen (A), Kass (38), Diers and Brown (16) and others have found that the DI was inadequate as a -28.- ldiagnostic tool using several types of brain damaged Sub- jects. Yates (72) concludes that the indices of deteriora- tion are of little clinical use in their present form. Marrow and Mark (A7) performed autopsies on brain damaged and non- organic psychiatric patients (males) and found a "typical" WB II test pattern of significantly lower scores 6n DigitT Symbol, Block Design, Digit Span, Arithmetic and Similarities. The full scale scores and the Performance Scale scores were also found to be lower in the brain damaged group. Vocabulary, Information, and Comprehension showed no differences. They also found that the frontal lobe was not dominant for intel- lectual functioning in this sample. The present knowledge of the DI makes its use doubtful for the purpose of this study. Halstead (25) has developed a battery of tests which discriminated between normals and patients having frontal lobe lesions of 3-5% of the cortex, with occipital lesions being most difficult to detect. Another difficulty in using this test battery is the amount of expensive equipment and the time (about 3 hours) required for the administration. In addition, considerable skill is required for the adminis- tration of this battery. The primary objection to the test, however, is that it was developed using only normal and brain injured subjects. No functional psychoses were in- -29- cluded in the development of the norms. The quantitative tests comprise a relatively small percentage of the total number of tests designed for the study of the brain damaged. The greatest number of tests is to be found in what might be termed perceptual and motor tests, some of which were designed for purposes other than testing for brain damage. Among the most prominent of these is the Rorschach Test. C. Perceptual and Motor Tests: Rorschach's Test: Piotrowski (50) found ten signs in protocols of 33 persons: 18 brain damaged, 10 with non-cerebral CNS dis- turbances, and 5 cases of conversion hysteria. The brain damaged group produced a mean of 6.2 signs and the other (groups produced a mean of 1.5 signs.t There was no over- lap between the groups. Thus 5 or more signs were con- sidered as indicative of brain damage. Piotrowski later found that personality changes were responsible for a number of the signs, and that the number of signs tended to increase with age. He also found that schizophrenics and neurotics also produced some of the signs. -Ross (59) has attempted to replicate Piotrowski's work, however, he was not as successful in differentiating -30- pathology as was Piotrowski. He found that 55% of the brain damaged patients showed 5 or more signs, but so did 30% of those with non-cortical lesions of the CNS, and 20% of the psychotics, as well as 1A% of the neurotics. Ross later divided the signs into A groups and developed what he called "instability" and "disability" ratings. The stand- ardization was carried out using normals, neurotics, and brain damaged subjects; thus he obtained ratings along both neurotic and organic dimensions. Hughes (33) derived lA signs from a factor analysis of 22. He assigned weights to each of these signs and standard- ized this using 218 subjects, including 50 brain damaged, 68 schizophrenics, 7A neurotics, A manic-depressives, and 22 normals and found validity of .79. When he used a cut- off score of 7 he correctly identified 82% of the organics, whereas success was only 20% using Piotrowski's signs. In both cases, although the signs could identify organics, the amount of overlap was still large. Diers and Brown (16) found that Hughes' signs were valid only with subjects of high intelligence since there was an inverse relationship between W-B IQ and the Hughes score, independent of intracranial pathology. Dorken and Kral (72) developed an innovation. They -31- investigated what signs the organic would not show. Seven signs were developed. By weighting these signs, a maximum score of 10 could be obtained. Scores from 3 to 10 exclud- ed a diagnosis of brain damage. They report that 92.9% of the organics were identified, and 83.3% of the non-organics were correctly identified. Comparison revealed that Pio- a trowski's signs identified only 50p of the organics, while Ross' "disability" ratio identified 75%. Buhler, Buhler and Lefever (12) using 30 normals, 70 neurotics, 50 psychopaths, 27 schizophrenics and 30 organics, developed a Basic Rorschach Score. They reported that this score was capable of separating clinical groups in a stati- stically reliable manner, but admitted that this Basic Ror- schach Score was not sufficient for individual diagnosis. Thus, this technique is more useful in giving an estimate of adjustment or ego-integration. A replication by Buhler et a1 (13) confirmed the results of the original study. More recently, Hertz and Loehrke (31) made use of Piotrowski's signs on a group suffering post-traumatic encephalopathy. They found that 5 or more signs reflect organic pathology but there was, again, considerable over- lap in the direction of finding false negatives. Fisher et a1 (18) developed A Rorschach systems for determining the presence or absence of brain pathology and found that three of the four systems distinguished at -32- better than chance. Schreiber et a1 (63) report that they were able to distinguish brain pathology using the Rorschach. "The major criticisms of the Rorschach as a test of brain damage seem to be that in all the scoring systems, except that of Dorken and Kral, the greatest weighting is given to those factors that are the most difficult to score and that depend, therefore, to the highest extent on the subjective evaluation of the examiner." (72). Further criticism of these Rorschach studies may be directed toward their lack of validation by other workers, and also the fact that most of the studies have not con- sidered the variables of age and intelligence. Other Perceptual and Motor Tests: Armitage (5) attempted to use the Patch Test and the Trail Making Test for the purpose of identification of brain damaged subjects. He used AA patients with known brain damage, A5 normals, and 16 mild neurotics. The groups were roughly matched for age, level of education and premorbid occupation. He reported that the Trail Making Test "A" discriminated best; he was able to cor- rectly identify 32 of the AA organics. Test "B" identi- fied 39 of the AA, but misclassified 15 of 51 non-organics. The Patch Test correctly identified 26 of A3 organics and -33- misclassified 7 of the 51 controls. No psychotic patients were in the study. Reitan (55) used the Trail Making Test and found significance at .001 in differential diagnosis. His sample consisted of 27 matched pairs. No other work on this test has been reported. Shapiro (61) modified Goldstein's Block Design Test to use the amount of rotation of the designs as the criteria .for differentiation. He hypothesized that rotation would be maximized in brain damaged subjects. He compiled a set of A0 designs which could be reproduced using four Kohs blocks. Two groups were used; 19 organics and 19 non-orga- nic, psychiatric patients. The groups were carefully matched for age and sex and were tested under identical conditions. It was found that the organics rotated an ave- rage of 8 degrees per card, whil the non-organics rotated an average of only 2 degrees per card. Using a cutting score of 6 degrees rotation per card, the test correctly identi- fied 1A of the 19 brain damaged patients and misclassified only one functional patient. Replication has yielded similar results. Using the original two groups, Shapiro found that the Manual Dexterity Test of the USES battery of tests discriminated very significantly between function- als and organics. Both of Shapiro's studies have revealed some overlap, misclassifying both functionals and organics. -3 4.- The Bender Visual-Motor Gestalt Test (8); This test has been among the most widely used tests for the screening of patients suspected of being brain damaged. The test was originally designed using a series of patterns adapted from Wertheimer. As the name of the test suggests, it is designed to measure visual-motor perception, which is theoretically and physiologically a very vulnerable (to brain damage) area of function. Bender set up only rough qualitative standards for the diagnostic use of this test which have been found to be generally not discriminative in clinical practice. A quantitative scoring system was developed by Pascal and Suttell (A9) which has been reported to be very discrimina- ting. Their method is somewhat complex and requires con— siderable practice in order to gain skill and consistency in scoring. (See Appendix A2 for scoring criteria). Recently, there seems to have been an increase in the amount of research using sensory=perceptual techniques. Semmes and her co-workers have used visual and tactual tasks for the diagnosis of brain damage (6A); Ax and Colley (6) have used CFF type procedures along the visual parameters, plus audition, touch and electrical stimulation and found that when they used touch, audition and vision in a battery they could correctly identify 81% of the cases with relatively -35- little overlap. Price (51) has attempted to use spiral after-effect as a means of diagnosis. Using a large sample of brain damaged, psychotic and normal subjects, he found signifi- cance at .001. He also reports very little overlap. These last mentioned types of studies seem to have yielded rather good results but still remain to be validated by other workers. Further research in this area appears to be nec- essary. This concludes the descriptions of the major tools utilized for the determination of brain damage in adults. There have been several tests designed specifically for the diagnosis of brain damage in children. One of the best known of these is the Strauss-Lehtinen Battery (65). The authors claim that their battery has the capacity to discriminate normals, endogenous mental defectives and exogenous mental defectives but offer no quantitative evi- dence. The battery consists of the Marble-Board, Figure- Background, and Tactual-Motor Tests. The primary purpose of this paper is to review the tests used for the diagnosis of brain damage in adults.For this reason ani extensive re- port will not be included on these tests. Sarason (60) offers several criticisms of the Strauss—Lehtinen Tests -36- based on his long clinical experience and experience with defective children. Rafi (5A), using the Strauss-Lehtinen Battery on adult mental patients reports that the battery is cor- related with intelligence and that it will not discri- minate between adult mental defectives and brain damaged subjects. He performed a cross validation study and ob- tained the same results. Halpern and Patterson (2A) found statistically signi- ficant difierences between brain damaged and non-organic children, using the Goldstein-Scheerer Tests, but considered the overlap to be too great for clinical use. Beck and Lam (7) used the WISC in predicting brain damage in children (aged 6-16). They reported that brain injured children attained lower full scale scores than non- brain injured. They found that 31 of A2 were confirmed by neurological examination. The IQ of all the children was less than 80. Other Tests: All of the tests used for the detection of brain damage have not been listed or described, primarily because many of the tests are not currently in wide use, either clinically -37- or for research purposes. Most of the major tests for brain damage have been described. For further listings of other tests, Anastasi (3) is among the most recent and most com- plete texts on the general subject. Other references which may be considered of interest are Ross' (58) work with tactual tests, and Heilbrun's (30) study on the localization of cerebral lesions by the use of psychological tests, as well as Wahler (69), who worked on a memory test. Kahn (37) has devised a test of symbol arrangement. The list could be quite long, but the amount to be gained from an extension of the present long list would probably add little to our information, nor would it likely reveal more fruitful areas for research. Problem: The survey of the literature has revealed that there are essentially three major areas of function that are most susceptible to brain damage; abstract thinking, sensorimotor coordination, and memory. It was therefore decided to use tests that purport to measure these functions. Additional criteria for the selection of the tests were considered to be relative ease of administration and quantifiability. The Bender Visual-Motor Gestalt Test, which measures visual-motor functioning, the Hunt-Minnesota Test for Organic Brain Damage, which measures memory, and the Grassi Block Substitution Test, which measures sensorimotor coordi- nation at both concrete and abstract levels seem to meet these qualifications. One purpose of this pilot study is to determine whether it may be feasible to combine a number of tests into a battery that will function more effectively in distinguishing between brain damaged and non-brain dam— aged subjects than would the tests taken individually. Because of the widely variant reports of the efficacy of these kinds of tests, criteria will have to be develOped in -39.. the pilot study by which to evaluate the performance of the selected subjects on these tests. This will be nec— essary if further work is to be done with this method. A. Subjects and Procedure of Pilot Study: The subjects for this pilot study consisted of 21 white males, matched on the variables of age, intelligence, and education. Of these 21 subjects, seven were known to have definitely established brain damage of varying etiol- Ogy and in different areas of the cortex, seven schiz0phre- nics, and seven non-psychiatric patients who were military patients on Medical and Surgical wards of a U.S. Air Force HOSpital. The criteria for the selection of the brain damaged group were a definite history of a lesion, either as a re- sult of a penetrating wound, surgery, or a known disease entity which invariably produces cerebral involvement, with the exclusion of alcoholics, general paretics, and loboto- mized subjects because of other aspects of these syndromes which may occlude the results, and epileptics because of their frequently unknown etiology. The criterion for the selection of the schiZOphrenic group was the diagnosis of the psychiatric staff of the Veterans' Administration Hospital, Battle Creek, Michigan. ”HO‘ The criteria for the selection of the normal group was the absence of neurOpsychiatric diagnosis or treat- ment, as well as the absence of clinical symptoms of psychOpathology. In order to maximize the objectivity of the scoring systems, the Pascal-Suttell (AQ) scoring system was used in the evaluation of the performance on the Bender Visual- Motor Gestalt Test. The scoring systems developed by Grassi and Hunt were used on their respective tests. The 21 subjects comprising the pilot group were administered the three tests of the battery and the raw scores were recorded. When the entire sample was tested, the means and standard deviations of the scores for each test were determined and standard (2) scores were derived. These are recorded in Appendix B. The hypo- thesis of the pilot study was then tested statistically in order to determine the value of proceeding further with analysis of behavioral deficit, using these three tests imithis manner. It is expected that each of these tests makes a contribution to the differential diagnosis of the three groups studied. If this is the case, weightings may be determined and the tests may be used as a unified battery. The battery requires about one hour to administer. (The -41- matching data are given in Table 1.1 Table 1 Hatched Variables for the Three Groups .32. _§_ 10.12- Subjeet Age VS* Ed. Age VS* Ed. Age VS* Ed. 1. #1 25 12 A1 25 12 AA 27 12 2. 25 18 8 25 20 8 20 17 8 3. 26 21 12 25 29 12 2A 27 12 A. 25 1A 8 25 19 8 28 17 5 5. 38 20 12 32 23 8 38 20 12 6. 32 26 12 33 28 10 27 27 17 7. 23 20 9 27 20 8 22 20 13 *Binet Vocabulary score. The schizophrenic group was included because of the frequent difficulty of differentiating brain damaged sub- jects from schizophrenics due to the similarities of many 1The following abbreviations are used in the Tables: Grassi, Grassi Block Substitution Test; Hunt, Hunt- ‘Minnesota Test for Or anic Brain.Damage; Bender, Bender Visual-Mbtor Gestalt est; RD, brain damaged subjects; 8, Schizophrenic subjects; and Horn, "normal" subjects. ‘\ ‘\ -42- cf the symptoms. The normal group was included to pro- vide a standard by which to measure the discrepant func- tioning of the other two groups. The organic group was tested first, and each of the other two groups were match- ed to it. B. Eggpose: To determine the accuracy with which organic brain damage may be detected and distinguished from functional psychosis (schisOphrenia) and normality, by applying three different criteria of behavioral deficit simultan- eously: (a) sensory-motor coordination, (b) abstraction in perception, (c) memory. C. Main Hypothesis of Pilot Study: Combination of the scores on the Grassi Block Sub- stitution Test, the Bender‘Visual-Mbtor Gestalt Test, and the Hunt-Minnesota Test for Organic Brain Damage may be devised to measure the behavioral deficit related to dwa- ‘ age of the cerebral cortex better than any of them indi- vidually. Rationale of test selection: 1) The Grassi Block Substitution Test was selected as a test of both-censori- motor coordination with a high component of abstract -43 ee perception; 2) the Bender Visual-Motor Gestalt Test was selected as a test of visual-motor coordination and per- ception, with a minimal component of abstraction; 3) the Hunthinnesota Test for Organic Brain Damage was select- ed as a test of memory. The materials used were the Kohs blocks as modified by Grassi (22) (which had to be specially fabricated, _since none were available commercially when the study was begun), the Bender Visual-Motor Gestalt Test cards (8), and the Hunt-Minnesota Test for Organic Brain Damage (3A), Short— form. The scoring systems for the Hunt and the Grassi were the same as those set down by their respect- ive authors, and the Pascal-Suttell scoring (#9) was used to obtain an objective score on the BendereGestalt. A stop watch was employed to time the performances on the various tasks. The statistical procedure consisted of first deter- mining whether or not the three tests differentiated the three groups. This was accomplished using the chi square for significance of difference between the groups. An- alysis of variance yielded an "H" which was significant beyond .01. The second statistical step was to deter- mine whether there was any significance of differences between the groups when the tests were treated as -44- battery. This was accomplished by converting all of the raw scores to standard (2) scores so the several scores could be conveniently added. The scores for each sub- ject were summed and the result was a battery score, from which a cut-off point was empirically determined. (See Table 2.) Sub}. GrassI Hunt Bender Sum s BD I - .35 -l.l9 .68 - .86 2 O "' e 8 '1e19 -1e87 3 -1016 '1e19 -2e66 -5001 k "' e12 e33 "' e58 - e37 5 -2.91 - .BA ~2-12 -5.37 6 - e58 e93 ‘ e58 "' e23 7 -2.09 -1.19 .03 -3.25 S T .23 - .17 .03 .09 2 e70 leOl "’ e15 1e56 3 .A0 “-1.11 1000 035 A 1.05 2.11 .75 3.91 5 .58 .16 .93 1.67 6 e23 e25 ‘ 015 033 7 .70 -1.19 1.07 .58 Donn I .70 " 026 046 090 2 .23 .50 - .15 .58 3 1.05 " e85 lell 1e31 A .12 .8A .78 1.7A 5 .35 - .17 .25 .A3 6 .81 .16 .10 1.50 7 0 2.03 .39 2.A2 2 Score Conversion of Test Scores Table 2 -45- D. Results of Pilot Studlz The chi square tests of significance of the sever- al tests may be seen in Table 3. Table 3 Chi Square Analysis of the Three Tests on the Three Groups of Subjects 'Grassi Hunt Bender BD-S 10.28* 1.00 b.00 BDhNonm 10.28* 1.00 10.00* Norm-8 0.00 0.00 0.00 *Significant beyond .01. It can be easily seen that the Grassi and the Bender were more differential and sensitive than the Hunt- Mdnnesota Test, and also that, although there is a signi- ficant difference between groups on these two tests, there is still some overlap. (See Tables h and 5.) Raw’Scores on Three Tests with Three Groups -45- Table h Grassi Org. Sch. 16.5 19 18 21 13 20 17.5 22.5 5.5 20.5 15.5 19 13.6 20.5 Horn 21 19 22.5 18.5 19.5 21.5 18 20 Org. 8h 90 72 65 90 Hunt Sch. 78 6h 89 51 7h 73 90 ‘Nban Norm 79 66 70 118 86 159 66 101 78 ILA 7A 101 52 3# Org e Bender Sch. 8. 89 ‘57 6h 59 89 55 Horn 72 89 Sh 63 78 82 7‘ 82.57 7h.lh 72.1A 100.42 66.42 73.1k Table 5 Ranges of Raw Scores on Three Tests with Three Groups Grassi Hunt Bender 5.5-18.0 65-90 66-159 8 19.0-22.5 51-90 55-89 More 18.0-22.5 52-86 56-89 -47- The combination of the 3 scores produced perfect differentiation on the basis of positive and negative summed s scores, the organic group having all negative scores and the other two groups, not being significant- ly different from each other, having positive summed s scores. Careful inspection of Table 3 reveals that the HuntéMinnesota contributed only in the case of subject #1 to the extent that, without this score there would not have been perfect separation of the groups. B. Discussion of Pilot Study: The results seem to indicate that something close to perfect discrimination is possible if these three cri- teria of deficit are applied. The original hypothesis regarding the combination of tests into a battery is sup- ported by the data. The direct combination ef test scores in standard score fonm without the necessity of using a multiple regression equation suggests still fur- ther that there is really an advantage of the combination over the individual items. The fact that none of the tests, when applied singly discriminated the normals from the schizophrenics with any significance supports the assumption of a multi- criteria test of behavioral deficit. This suggest that -43- the battery is sensitive enough not only to differen- tiate nonmals from organics, but also to differentiate organics from schizophrenics. This may well have imp plications for theory development with regard to organ- ic involvement in schizophrenia; a subject only tangen- tial to this study. A possible reason for the lack of differentiation found between normals and schizophrenics consisted of those in "good contact” and in only 2 cases were ”regressed" or "deteriorated” subjects used. a 4‘ This battery appears to be effective in differen- tiation and it is also easy to administer, is not time consuming. The materials required are readily available in most clinics and (the Grassi is scheduled to be on the market soon) are neither expensive nor elaborate. It is objectively scored with relatively little difficul- tYe III. EIPERIMENTLL STUDY 1. Hypotheses of Egperimental Study: Hypothesis I: Combination of the scores on the Grassi Block Substitution Test, the Bender Visual-Mbtor Gestalt Test, and the HuntéMinnesota Test for Organic Brain.Damage may be devised to measure the behavioral deficit related to damage of the cerebral cortex better than any of them individually. Corollary hypothesis: Each of these tests is a measure of behavior correlated with brain damage and will therefore discriminate be- tween normals, schizophrenics and organics. Hypothesis II: The behavior (performance on these tests) of schizophrenics will reveal lessimpairment than the organics' performance. Hypothesis III: Recent lesions in the cortex will yield more behavioral deficit as measured by the test battery than will those of long standing. 8. Procedure for erimenta1 Stud : A total of 75 subjects were administered the Grassi Block Substitution Test (22), the Bender-Gestalt Test (8), and the Huntélinnesota Test fer Organic Brain Damage -50- (3b). This group consisted of 25 subjects who fulfilled the criteria for brain damage, as stated earlier, 25 sub- jects diagnosed by psychiatric staff decision as schizo- phrenic, and 25 subjects who carried no neuropsychiatric diagnosis and were considered ”normals". All subjects 'were matched on the variable of age, education, and vo- cabulary score, the last of which was used as a rough estimate of intelligence. Table 6 reveals the matching datae Table 6 Mean 78*, Age at Education, for 3 Diagnostic Groups (N-75) BD _§__ Norm Grand Mean Mean vse 22.28 25.1.8 23.88 23.88 [Mean Age 32.76 33.16 31.0L 32.65 Ilean.Education 11.36 10.80 10.88 11.00 a"Vocabulary score. All subjects were white males, thus insuring con- trol of the variables of sex and race. The scoring criteria were developed from the first 21 subjects, discussed in the pilot study results. The -51- scoring criteria thus derived were applied to the second group of 5b subjects in the form of the standard (z) scores. The derived standard scores with their corres- ponding raw scores for each of the three tests are tabu- lated in Appendix B. It may be recalled that the initial part of this study revealed that the sum of the standard scores for each test yielded a battery score. It was feund that this battery score was a negative number for the brain damaged group, and positive for the schizophrenic and normal groups. This applied to all cases in the pilot group, thus establishing the criteria that a negative battery score indicated brain damage, and a positive battery score was indicative of the lack of cortical damage. When this scoring system was applied to the re- maining 5h subjects, it was found that only h subjects 'were misclassified; 3 organics, 1 schizophrenic, and no normals. The result of this group differentiation may be seen in Table 7. -52- Table 7 Total Number of Misclassifications on Each Test and on Battery Scores (Derived Criteria) 132 a Long .8382 Grassi S 2 1 8 H unt S 13 5 23 Bender 6 h l 11 Battery 3 l O h Mean error on Individual Tests: 18.6% Error on Battery : 5.2% C. Results and Discussion of Experimental Study: The experimental group, consisting of Sh subjects, was administered the battery consisting of the three tests used in the pilot study. In all, there were 172 test administrations on this part of the sample. When the original authors' criteria were applied to the test re- sults of this sample, 57 of the 172 scores misclassified the subjects, while the criteria derived from the pilot study misclassified only 28 of the 172 cases. (See Table 8.) The percentage of error for the original authors' criteria was found to be 33, while the percent- -523- age of error using the derived criteria was found to be 16. When the scores of the pilot and experimental groups were treated together, the percentage of error for the original authors' criteria was found to be 37, While the percentage of error using the derived criteria was found to be 19. Thus, the results obtained in the experimental study were in line with the predictions made on the basis of the pilot study. -53- Table 8 Misclassifications Using Original and Derived Cut-Offs On Individual Tests (N=Sh) E Original Derived B2 S N orm Sum B2. q§ Norm Sum Grassi 7 1h 3 2h 5 2 1 8 Bender 8 1 O 9 3 2 O 5 H unt 2 1h 8 gg 3 10 2 12 Sums S7 ' 28 Percentage of error 33 16 See Tables 9, 10, 11, and 12 for chi square analysis of difference of groups and of criteria. Table 9 Chi Square Analysis Comparing the Number of Persons Misclassified by the Individual Test Scores and by the Battery Scores 2:! Individual Tests Battery Total Correctly Identified 183 71 25h Misclassified #2 - h M6 Total 225 75 300 Chi Square: 7.7, 1df, p.: beyond .01. -5h- Table 10 Chi Square Analysis Comparing the Number of Persons Misclassified by the Original AuthorFS Criteria And by the Derived Criteria on the Grassi Block Substitution Test Correctly Identified Nfisclassified Total Original Derived Scores Scores Total 39 67 106 36 i 8 44 75 75 150 Chi Square: 23.44, 1 df, p.: beyond .001. Chi Square Analysis Comparing the Number Table 11 of Persons Misclassified by the Original Author's Criteria And by the Derived Criteria on the Hunt- Nfinnesota Test for Organic Brain Damage Correctly Identified Misclassified Total Original Derived Scores Scores Total 39 52 91 36 23 59 75 75 150 Chi Square: 1.372, I age, pa: 005.0025. -55- Table 12 Chi Square Analysis Comparing the Number of Persons Misclassified by the Original Author's Criteria And the Derived Criteria on the Bender ‘Visual-Mbtor Gestalt Test Original Derived Scores Scores Total Correctly 6k 6k 128 Identified Misclassified 11 11 22 Total 75 75 150 Chi Square is not significant. The derived battery standard scores discriminated in 9h.8%»of the cases, while the original author's cri- teria (see Table 13) yielded accuracy of 89.3% on the Grassi, 69.3% on the Hunt, and 85.9% on the Bender (see, also, Table 7). Table 13 Comparison of'Derived Cut-Off Scores And ‘ Original Author's Cut-Off Scores Original Derived Norm Grassi 16 16-20 Mbre than 20 18 Hunt 68 Less than 76 Bender 100 85 -55- The range of scores achieved on all tests by the three groups may be seen in Table 1b. Table 1h Ranges of Raw Scores of 3 Test H73 Diagnostic Groups _§Q_ _§_ Norm Grassi k.5-22.5 15.5-2A.5 l7-25.5 Hunt 59-90 51-90 50-86 Bender 66-170 35-89 A5-89 In order to ascertain the level of statistically significant differences between the groups, the analysis of variance procedure for ranked data (Kruskal-Wallis ”H" Test) was used for each test in the battery, and for all the tests combined. This yielded results which are significant well beyond the .01 level. See Table 15. A further check was made by analyzing the data by the use of the chi square analysis between groups and be- tween tests. The results of this analysis are to be found in Tables 16, 17, and 18. -57- Table 15 Results of Analysis of Variance by Ranks (Kruskal- wallis H-Test) Comparing Battery Scores and Scores Of Individual Tests on Three Groups _§_’ df. 2. Battery Score 30.77* 2 beyond .001 Grassi Raw Score 16.13* 2 beyond .001 Hunt Raw Score 11.94* 2 beyond .005 Bender Raw Score 16.36* 2 beyond .001 *Beyond .01. Table 16 Chi Square Analysis Comparing Brain Damaged And Non-Brain Damaged Subjects on the ‘ Battery Score (Summed s) Sum z Sum z Less than 0 Nbre than 0 Total Brain Damaged ' ' 22-; 3 25 Non-Brain Damaged 1 #9 50 Total 23 ' 52 75 CHI Square: 26.87,‘1 d?., p.: beyond.OOI. -53- Table 17 Chi Square Analysis Comparing Brain Damaged and Schiz0phrenic Subjects on the Battery Score Sum 8 Sum z Less than 0 Mbre than 0 Total BD 22 3 25 S 1 2h 25 Total 23 27 50 Chi Square: 32.68, 1 df., p.:lbeyond .001. Table 18 Chi Square Analysis Comparing Schizo hrenic and Normal Subjects on the Battery core Sum z Sum z Less than 0 Mbrc than.O Totals S . 1 Zh 25 Norm 0 25 25 Totals 1 #9 50 Chi Square is not significant. The most consistently accurate test in the battery is the Grassi Block Substitution Test and the least con- tributory is the Hunt-Minnesota. This is seen in Table 7. Although the Hunt-Minnesota seems to be, itself, of little more than chance value when used alone, it does -59- add to the accuracy of the battery in identifying and separating the three groups. Thus the data support hypothesis 1: that a combin- ation of tests which are designed to tap functions of various parts of the cortex will be more sensitive in separating the brain damaged from the non-brain damaged. Having established this at a very statistically sig- nificant level, we may be reasonably certain that hypoth- eses 2 and 3 can be validly checked, since we are now in possession of data which is representative of the behavi- or of the groups. Hypothesis 2: that schizophrenics reveal less behavi- oral deficit than the brain damaged, is supported by the data, utilizing the analysis which revealed significant- ly lower scores among the brain damaged than among the schizophrenics. It is being assumed that the tests are valid, and are measuring behavior which requires the proper functioning of the cerebral cortex, as stated by the test authors and described in the introductory part of this study. The fact that the mean standard battery score for the schizophrenic group is .463, and for the brain dam- aged group is minus .867, a difference of 1.33, suggests rt -60- that the measurable behavior deficit is greater in the brain damaged group than in the schizophrenic. Further evidence for the markedly greater loss of facility in these behaviors is the finding that the mean standard battery score for the normal group was found to be .7h3, or 1.61 points at variance from the mean standard bat- tery score of the brain damaged group. The difference between the schizophrenic and normal groups is only .280, suggesting that behavioral deficit in schizophrenia is distinct from the deficit found in the brain damaged. All the tests of significance revealed that the differ- ences in performance of these two groups are significant beyond the .01 level of confidence. The clear-cut differentiation of these groups in quantitatively measurable behavior offers some insight into the qualitative as well as quantitative aspects of 2 clinical entities which have been so long confused, both theoretically and clinically. Insignificant corre- lations with intelligence were found in the performance of all three groups, further suggesting that the primary requisite for proper performance on this battery of tests is an intact cerebral cortex. This has further implica- tions, which will be elaborated upon in a later section. In order to test hypothesis 3, the brain damaged -61- group was divided into three parts; those having sus- tained cortical injury within one year prior to test- ing, those having injuries older than one year, and those who were suffering from active lesions. Since there were ten subjects suffering from active lesions, the scores of the remaining 15 subjects were used to determine the relationship between test scores and time since cortical injury. Table 19 Fisher's Exact Test Relating Score to Time Since Injury Most Recent Oldest Above Median (greatest deficit) 6 Median & Below 2 6 (least deficit) __ P. :18 .0211. Here, though only 15 cases were included, the results are significant. Table 19 reveals that there was signifi- cant separation of the old and recent cases; the most recent cases revealing the greatest deficit and the poorest scores, while the oldest cases achieved the least deficit. This finding is not inconsistent with Lashley's theories of equipotentiality and mass action. It also lends support to observations made in the Columbia-Greystone lobotomy -62- projects (h5), and it also helps to explain why it is frequently so difficult to detect brain injury by the use of existing psychological procedures when the dam- age is more than one year old. Unfortunately, it was impossible to localize or de- termine the exact extent of brain damage in this sample. For this reason, further inferences in this area cannot be made with any degree of confidence.' However, it seems safe to assume that there was diversified sampling of dam- aged cortical areas, since various manifestations of cor- tical damage were evident throughout the sample. Further work along these lines, utilizing autopsy reports, which can reveal the exact locus and extent of damage, could help lead to the answer of the old question of specifi- city of function of cortical areas. This, however, is beyond the scope of this study. Thus it appears that the data support all three hypotheses at very significant levels of confidence. The problems of definition of syndromes and of lev- el of psychological function have been continually in a state of relative confusion. The authors of tests de- signed to distinguish the brain damaged individual have met with frustration; for while their criteria seem to -63- hold for a standardization group, they frequently fail to yield similar results on different pepulations. This may have been true largely because these authors frequent- ly utilised one type of patient in a single institution. An attempt was made to overcome this problem in this study by restricting the sample of brain damaged subjects to those who were not paretic, lobotomized, or only epilep- tic, and also by limiting the age factor in order to slim- inate the senile disorders. ‘All other cases of known brain damage were included in the sample as they became available during the more than two years required to col- lect the data. In addition to using as wide a variety of subjects as possible, several institutions were used. These included four Veterans' Administration installa- tions of various types and the University of Michigan Hospital. It is true that all of these subjects were in the State of Michigan, but it was impracticable to ob- tain a larger, regional or national sampling. Many of the subjects used in this study, however, originated from other sections of the nation. If geographical fac- tors are important, further research in other parts of the country will have to be done before we will know to what extent these differences are significant. This may well be an important variable since we are interested in behavioral deficit resulting from brain damage, which is -64- based implicitly on the premise that the intact individu- al is capable of adequate performance on these tasks. It is possible that there may well be subcultural diff- erences in a national sample which would make the results of this study appear less valid if applied on.a larger scale. Therefore, if this study or similar studies are to be repeated in other sections of the country, a pilot study similar to the one in this study should be first accomplished so that there will be a basis for general- ization. At this point, it does not appear safe to ex- tend the generalization of behavioral deficit, as deter- mined by this study, beyond the population from which the sample was drawn. By illustration, a sample in a less industrialized section of the nation may have less basic facility in dealing with the construction of block designs. A case in point is that Grassi collected his standardization data in a Southern hospital and derived a cut-off score of 16 for his brain damaged group, while in the more industrialized State of Michigan, a cut-off score of 18 was derived. This is an hypothesis which will have to await the reports of other studies before it can be verified or refuted. The data.revea1ed, quite conclusively, that very likely there is some specific functions of various brain -65- areas, since the selected battery of tests proved to be far superior to any of them individually. The widely discrepant scores arrived at by these two approaches appears to be due to the fact that the several tests are tapping several behaviors (with some overlap, to be sure) which are sensitive to brain damage. It does not appear that the battery is more sensitive because of its increased length, but that we gain finer distinc- tion because we are investigating more aspects of beha- vior which have proven previously to be sensitive to brain damage. By the same token, the data tend to support the theory that there is generalized function of the cortex. The finding that deficit performance on one test of the battery by a brain damaged subject frequently was fol- lowed by similar performancc on the other tests in the battery is suggestive that mass action is operating. This, however, was not invariably the case, since in several instances there was adequate performance on one or two tests, and complete failure on the remainder. It appears on the surface that the Hunt-Minnesota Test for Organic Brain Damage did not add to the sensi- tivity of the battery, but a finer analysis revealed that it does add to the overall predictability of the -66- battery. This appears incongruous with the fact that, alone, it misclassified approximately 50% of the sub- jects. The fact that, in spite of its relative ineffi- ciency, it did add to the total value of the battery sug- gests that there may well be a degree of relative sensi- tivity to the functions and behaviors tapped by this test, and also that the errors in this test nullify po- tential errors in the other tests of the battery. In V any case, there appears to be value in maintaining this test in the battery. Ample evidence was gained that the deficit revealed by the brain damaged subjects is significantly greater than that revealed by the schizophrenic group. This find- ing may well clarify much of the clinical difficulty en- countered in the past by psychologists in efforts at diag- nostic differentiation. One complication which was not foreseen at the beginning of this study was the possibility that there may have been ”process" as well as ”reactive" schizophrenics in the sample. Since thisgmudy was under- taken, Brackbill and Fine (11) found significant differ- ences between these two diagnostic groups in performance on the Rorschach Test, using Piotrowski's signs, finding that the process group revealed more deficit. Theoreti- cally, process schizophrenia has a component of brain damage. It is possible, although not verified, that the -67- two schizophrenic subjects which were misclassified by this battery were of the process type. If this is true, 'we incidentally may have developed a tool which is easily capable of differentiating these two schizophrenic cate- gories. This is an avenue for further investigation and is only tangential to this study. The finding that the performance of the brain damaged subjects revealed sig- nificantly more behavioral deficit than the schizophrenic group is‘ most germane. From this point, finer discriminations may be attempted in_erdcr to gain a clearer understanding of areas sensi- tive to this type of behavioral deficit in schizophrenia. The final aspect of this study revealed that there is a significant tendency for initial deficit, seen shortly after incurring a cortical brain injury, to show marked recovery. This finding offers an explanation for the clinical phenomenon so frequently seen, i.e., that persons having known brain damage who do not reveal the usual signs of brain damage on psychological tests. This is an additional source of error in validation studies of tests for the detection of brain damage. For if there is recov- ery of behavioral functions and no residual disturbance of thought processes, does it not seem reasonable to clas- sify these individuals as ”recovered,” and no longer suffering -68- from brain damage? Thus, an individual may, at one time have received an injury to the brain, causing disturbances in his usual level of behavioral functioning, and then recovered within a period of several months. This is of- ten seen in the case of cerebral ventricular accidents, or "strokes." An individual may suffer a stroke and be completely incapacitated, yet recover fully and reveal none of the deficit which was initially evident. Even though cortical brain.tissue does not regenerate, neurolo- gists are of the opinion that the number of cortical cells is far in excess of the number needed to maintain the func- tions of the brain. Thus, if a relatively small number of cells are injured or destroyed in a cortical injury, perhaps the recovery processes are able to completely overcome the initial deficits. For all practical purposes, then, this individual should no longer be considered to be brain damaged, even.though there may be some cells which had been destroyed. D. Theoretical Implications: The findings of this study tend to offer some clari- fication of theories of behavioral deficit, not only in the brain damaged, but in the schizophrenic as well. The test battery concept utilized in this study led to the conclusion that there are several discrete as well as over- lapping functions which are susceptible to damage to the -69- human cortex. The results provided by the data revealed quite clearly that a battery of psychological tests de- signed to measure deficit in several aspects of behavior is superior to any of them individually. The fact that this proved to be true brings the theoretical problems of specificity versus commonality of brain function into sharper focus. This problem of brain function was con- sidered by Flourens in the nineteenth century. He con- ceptualized these aspects of function as action propre and action commune, a forerunner of Lashley's conceptions of equipotentiality and mass action. These concepts have held.thr0ughout this study. The findings that combination of measures is more sensitive than individual measures; and the clear-cut difference in deficit in recent and older injuries supports these concepts. In all but one case, the data differentiated the schizophrenic group from the brain damaged group, sug- gesting that if there is an organic component in schizo- phrenia, it does not create a behavioral deficit in anyway analogous to that of the brain damaged subject who is not suffering from a schizophrenic process. This might well have far-reaching theoretical implications in the under- standing of the etiology of schizophrenia. This study certainly cannot answer this important theoretical ques- tion, but it does offer a tool for distinction of the two -70- diagnostic classifications. In conjunction with this, is the problem of the pro- cess and reactive schizophrenias, which seem to be eti- ologically'differcnt. If there is truly a difference between these two classifications, then, perhaps they are dynamically as well as functionally different and require separate approach- es for treatment and understanding. A.recent article by Brackbill (10) offers a wide variety of theoretical and clinical approaches to this problem of cerebral involve- ment in schizophrenia. As revealed by his survey, this problem is far from solved and considerably more work is to be done before we gain a complete understanding of it. E. Clinical Implications: The implications for clinical practice are quite clear. Although this study is by no means a standardi- zation, it does suggest that perhaps it is time to take a closer look at our clinical criteria for the diagnosis of brain damage. The finding that a relatively small modifi- cation of cut-off scores significantly changed the categor- ization suggests that we may well have been rejecting the use of valid tests in the past, simply because we failed to make the necessary scoring adjustment for the population -71- with which we had been dealing. It almost goes without saying that, if this battery were to be standardized on a larger sample, we may truly have a tool which can easi- ly be used for the diagnostic differentiation of schizo- phrenia and brain damage. The results obtained in this study are quite significant and conclusive, and it is likely that the scoring criteria derived from this sample could be used clinically. The sample from which these scoring criteria were developed is not large, but the re- sults suggest that they might well be valid for clinical use. Further standardization is going to be undertakenz by this writer, since promise is dhown by these results. If the criteria hold under cross-validation on a. larger sample, then clinical psydhology will have added an extremely useful tool to its diagnostic procedures. E. Summary and Conclusions: This study was designed in an attempt to gain a more complete understanding of behavioral deficit in the brain damaged individual. It was hypothesized that a battery of psychological tests, which measure the functions of abstract thinking, perceptual-motor coordination and memory, would be more -72- valuable than any of these tests individually. This battery of tests was administered to 25 brain damaged subjects, 25 schizophrenic subjects and 25 "normal” non-psychiatric subjects. This battery was composed of the Grassi Block Substitution Test (22), the Hunt-Minnesota Test for Organic Brain Damage (3A), and the Bender Visual Motor Gestalt Test (8). Scoring criteria were derived from the first seven subjects in each group. By converting the test scores to standard scores, a single battery score was developed by addition of the individual standard test scores. It was found that positive battery scores differentiated the non-brain damaged from the brain damaged at a very significant level. The second hypothesis of this study was that there would be more deficit revealed by the brain damaged group than by the schizophrenic group. This was found to be true, since the performance of these two groups revealed statistically significant differ- ences, with the brain damaged group revealing more deficit. The final hypothesis was that we could expect reduction in deficit as a function of time after brain damage. The eight most recently injured and the seven -73- least recently injured subjects in the brain damaged group were compared, using the battery score. It was found, on these 15? subjects, that there was perfect discrimination of the recent and old injuries; the most recent revealing more deficit. Therefore, we may conclude from this study that: 1. There are very likely multiple factors of brain function which are susceptible to injury and that some of these factors may vary with the individual. Thus, a battery of tests which is designed to measure these several aspects of behavior is more effective than each individual test. 2. Schizophrenics are not as likely to reveal be- havioral deficit in the same areas as the brain damaged individuals. Although the schizophrenic may function at a lower level of personality integration than the brain damaged, there are discreet functions which are retained. This brings the problem of brain dysfunction in schizophrenia into somewhat sharper focus. 3. It seems reasonable to conclude that there is re- covery from behavioral deficit in the brain damaged individual even though there is no neural regenera- tion. -7h- The concept of behavioral deficit subsequent to brain injury has thus been made somewhat clearer. we can be relatively certain at this point that there is a qualitative difference between the deficit re- vealed by schizophrenics and by the brain damaged. By the same token, we have a clear demonstration of the recovery from behavioral deficit fellowing brain damage e Further work is indicated in the area of a more exact definition of the distinction between process and reactive schizophrenias. The findings of this study could be translated in terms of mental deficiency in terms of differentiating the endogenous from the exogenous. More work would be helpful in the determination of the correlation between extent and locus of a brain injury with the degree and dura- tion of behavioral deficit. Further investigation of behavioral deficit ap- pears necessary before we can consider it completely understood. However, the new approach utilized in this study, i.e., the use of a multiple factor battery of tests, may well lead to more complete understand- ing, since the results of this study indicate that there are several functions of the human cerebral cortex -75- which are interfered with when there is an injury to the brain. l. 2. 3. A. 5. 9. 10. ll. BIBLIOGRAPHY Aita, J.A., Armitage, S.G., Reitan, R.M;, Rabinowitz, A. The use of certain psychological tests in the evaluation of brain injury. ,1. ge . psychol., 1947, 3 3 225-2300 . Allen, R.M; Performance of the brain injured. i. clin. ps1chol., 19h7, 3, 225-230. Anastasi A. Psychological testing. lMacMillan Co. New York, 19 A. ’ Andersen, A.L. The effect of laterality of locali- zation of brain damage on wechsler-Bellevue indices igldeterioration. g, clin. psychol., 1950, 6, " 91h Armitage, S.G. An analysis of certain psychological tests used for the evaluation of brain injury, Psychol.lenog;., 19h6, 60, No. 1 (Whole No. 277). Ax, A.F., Colley,‘W;H. Temporal acuity and neuro- genifiépathology. g, consult. psychol., 1955, 19, 455- . Beck, H.S., Lam R.L. Use of the W130 in predicting organicity. 3. clin. psychol., 1955, 11, 154-158. Bender, Loretta. Instructions for the use of Visual Meter Gestalt Test. Amer. orthopsychiat. assoc., l9h6, 7pp. Boyd, H.F. A provisional quantitative scoring with preliminary norms for the Goldstein-Scheerer Cube Test. g, clin, psychol., 19A9, 5, 148-153. Brackbill, G.A. Studies of brain dysfunction in schizophrenia. Psychol. bull., 1956, 53, 210-226. Brackbill, G.A., Fine, H.J. Schizophrenia and central nervous system pathology. ,1. abnorm. soc. psychol., 1956, 52, 310-313. ' 12. 13. 1h. 15. 16. 17. 18. 19. 20. 21. 22. 23. -77.. Buhler, Charlotte, Buhler, K., and Lefever, D.W. Development 2; the Basic Rorschach Score with manual of directibns. Ios Angeles, Calif: Rorschach standardization studies, No. l, 1948. Buhler, Charlotte, Lefever ‘D.W., Kallstedt, Frances, E., and Peak, H.ML Qeyelopment 9; the Basic Ror- schach Score. Supplementary monograph, Ios Angeles, : Rorschach standardization studies, 1952. Canter, AtH. Direct and indirect measures of psycho- logical deficit in multiple sclerosis, Part II, :1. g_n_._npsychol., 1951, AA, 27-50. Capps, HdM. Vocabulary chan es in mental deteriora- tion. Arch. ps1chol;, 193 , AA, 27-50. Diers, W.C. and Brown, C.C. Rorschach "Organic Signs” and intelligence level. g, consult. psychol., 1951, 15, 3A3-3h5. - Downey, J.E. Laterality of function. Psychol. bull., 1933, 30, 109-142. Fisher, J., Gouda, T., and Little, K. The Rorschach and central nervous system pathology; A cross {glidation study. Amer. j, psychiat., 1955, 111, 7-1+92e ’ Garfield, S.L. and Fey, W.F. A comparison of the Wechsler-Bellevue and ShipleybHartford Retreat Scales as measures of mental impairment. ,g. consult. psychol., 1948, 12, 259-26A. Golbtein, K. After-effects of brain injury in war. New York: Grune and Stratton, 1942. Goldstein, K. Scheerer, M; Abstract and concrete behavior: An experimental study with special tests. Psychol. Monogg., 191.1, 53, No. 2 (Whole No. 239). Grassi, J.R. The Grassi Block Substitution Test for measurin or-Efiic brain atEoIo Sprinvfield "Ill. mgmafizsrla's‘j‘f ’ ° ’ ’ Gutman, Brigette. The application of the wechsler- Bellevue Scale in the diagnosis of organic brain disorders. g, clin. psychol, 1950, 6, 195-198. 2A. 25. 26. 27. 28. 29. 30. 31. 32. 33. 3a. 35. ‘78-'- Halpin, Virginia G., Patterson, Ruth M. The perfor- mance of brain-injured children on the Goldstein- gihggrer Tests. Amer. g, ment. Defic., 195A, 59, Halstead, W.C., Brain and intelligence: a Quantitative study 9; Egg frontal lobes. Chicago: Univ. of Chicago Press, 1947. Harlow, J.M5, In Coleman J.C. Abnormal Psychology and modern life. NeW'Tork: Scott, Foresman & Co., 1950. Harriman P.L. Ehgflggy dictionary of psychology. New York: Philosophica ibrary, 19L7. Harris, Pearl, Validity of the Grassi-Fairfield Block Substitution Test in differential diagnosis, g, consult. Psychol., 1955, 19, 330. Hebb, D.0. The effect of early and late brain injury upon test scores, and the nature of normal adult intelligence, Proc. Amer. philos. Sgg., 19A2, 85: 275‘292- Heilbrun, A. The localization of cerebral lesions by the use of sychological tests, Dissertation Abstr., 195A, 1A, 1 0A-1805, Abst. of PhD Thesis, 195A, State Univ. of Iowa. Hertz, Marguerite, Loehrke, Leah M. The application of the Piotrowski and Hughes signs of organic defect to a group of patients suffering from post-traumatic encephalopathy, g, proj. Tech., l95h, 18, 183-196. Hewson, Louise R. The Wechsler-Bellevue Scale and the Substitution Test as aids in neuropsychiatric diag- nosis, g, nerv. ment. Dis., 19A9, 109, 158-183, 2A6—266. Hughes, R.M§ A factor analysis of Rorschach diagnostic signs, i, gen. Psychol., 1950, A3, 85-103. Hunt, H.F. A practical clinical test for organic brain damage, 1, app . Psychol., 19A3, 27, 375-3 6. Hutton, E.L. The investigation of personality in patients treated by prefrontal leucotomy, g, ment. Sci., 19A2, 36. 370 38o 39. #0. Al. A2. 43- Ah. #5. A6. #7. -79- Juckem, Harriet, Weld, Jane A. A study of the Hunt- Minnesota Test for Organic Brain Damage at the upper level? of vocabulary, g, consult. Psychol., 1948, 2’ -5 e Kahn, T.C. Cross validation of the Organic Brain Pathology Scale for a test of symbol arrangement, 4;. consult. Psychol., 1955, 19, 130. Kass, W. wechsler's Mental Deterioration Index in the diagnosis of organic brain disease, Egans. Kans. Acad. Sci., l9h9, 52, 66-70. , Klebanoff, S.G., Singer, J.L., Wilensky, H. Psycho- logical consequences of brain lesions and ablations, Psychol. Bull., 1954, 51, 1-41. Lashley, K.S. Structural variation in the nervous system in relation to behavior, Psychol. Rev., 19A7, 5h, 325-33A. ,Magaret, Ann, Simpson, Mary M. A comparison of two measures of deterioration in psychotic patients, 1. consult. Psychol., 1948, 12, 265-269. Malamud, Rachel F. Validity of the Hunt-Minnesota Test for Organic Brain Damage, 1. Appl. Psychol., 19A6, 30. 271-275- Maslow, A.R. The effects of prefrontal lobotomy upon abstract behavior, g. clin. Psychol., 1955. Meehl, P.E., Jeffery,.Mary, The Hunt-Minnesota Test for Organic Brain Damage in cases of functional depression, g, appl. Psychol., 19A6, 30, 276-287. Mettler, F.A., Ed. Psychosurgical problems, by the Columbia Greystone Associates, second group, Phila.: Blakiston, 1952. .Morgan, C.T., Stellar, E. Physiological psychology, MbGraw-Hill, 1950, 2nd Ed. Morrow, R.S., mark, J.C. Intelligence of patients autopsied for brain damage, g, consult. Psychol., 1955 3 19 ’ 283-289 0 A8. A9. 50. 51. 52. 53. 55. 56. 57. 58. 59. -30- Nielsen, J.M. Aggosia, a raxia, a hasia. Their value ig cerebral loca zatign, ew Yor : Hoeber, 19A6, 2nd Ed. 1*— Pascal, J., Suttell, B.J. The Bender-Gestalt Test: uantification and validit for adults, New York: rune and Stratton, 1951. Piotrowski, Z. The Rorschach inkblot method in organic disturbance of the central nervous system, Q, nerv. ment. Qi§., 1937, 86, 525-537. Price, A.C., Deabler, H.L. Diagnosis of organicity by means of spiral aftereffect, i. consult. Psychol., Ptacek, J., Young, F.v Comparison of the Grassi Block Substitution Test with the wechsler-Bellevue in the diagnosis of organic brain-damage, g. clin. Psychol., 195A, 10, 375-378- Rabin, A.I., King, G.F., Ehrmann, J.C. Vocabulary Per- formance of short term and long term schizophrenics, g. abnorm. Psychol., 1955, 50, 255-258. Rafi, A. The discriminative power of the Strauss- - Lehtinen Battery in adult mental patients, g, consult. Psychol., 1955, 19, 135-138. Reitan, RgM. The relation of the Trail Making Test to organic brain damage,‘g. consult. Psychol., 1955, 19 3 393-39lfie Reynell, WkR. A psychometric method of determining intellectual loss following head injury, g, ment. §gi., 19AA, 90. 710-719. Rogers, L.S. A comparative evaluation of the wechsler- Bellevue mental Deterioration Index for various adult groups, g, clin. Psychol., 1950, 6, 199-202. Ross, A. Tactual perception of form by brain injured, g, abnorm. ggg, Psychol., 195A, A9, 566-572. Ross, W.D. The contribution of the Rorschach method to clinical diagnosis, g, ment. §gi., 19A1, 87, 331-3A8. Sarason S.B. ngchological roblems in mental defi- cienc’, New York: Harper, 18A9. A 61. 62. 63o 6A. 65. 67. 68. 69. 70. 71. 72. -31.. Shapiro, MAB. Experimental studies of perceptual anomaly, III, the testing of an explanatory theory, 51. ment. an” 1953, 99, 3944.09. Shipley, W.C., Burlingame, C.C. A self-administering scale for measuring intellectual impairment and deterioration, g, Psychol., 19A0, 9, 371-377. Schreiber, H., White, M.A. Diagnosing organicity on the Rorschach, Psychiat. Qpart. Suppl., 195A, 28, 255-277. Semmes, J., weinstein 3., Ghent, L., Teuber, H., Performance on complex tactual tasks after brain injury in man: analysis by locus of lesion, Amer. file PSYChOle, 1951+, 67, 220-211-09. Strauss, A., Lehtinen, L. Psychopathology and education f the brain-injured chil , New ork: Grune & Stratton, 1 A3?“ Teitelbaum, H.A. An analysis of the disturbances of the higher cortical functions, agnosia, apraxia, and apha- sia, g, nerv. ment. Dis., 19A3, 97, AA-6l. Tooth, G. On the use of mental tests for the measure- ment of disability after head injury. With a compa- rison between the results of these tests in patients after head injury and psychoneurotics,‘g. Neurol. Neurosurg. Psychiat., 1947, 10, 1-11. Tsai L.S., Maurer, 8. Right handedness in white rats, Science, 1930, 72, A36-A38. wahler, H. Analysis of the performance of brain damaged patients of a memory-for-designs test, Dissertation Abstr., 195A, 1A, 2A06-2A07, Abstr. of PEDIthesis, I95A, State Univ. of Iowa. Wéchsler, D. JMeasurement pf adult intelligence, Balti- more: Williams E WIIEins, 19AA. Yacorzynski, G.K. An evaluation of the postulates under- lying the Babcock Deterioration Test, Psychol. R2y., 19“, 1P8, 261-267e Yates, A.J. Validity of psychological tests of brain damage, Psychol. Bull., 195A, 51, 359-379. A1 A2 Bl B2 33 APPENDICES Grassi Block Substitution Test Record Form. Scoring Criteria for the Bender Visual- Motor Gestalt Test. 2 Score Conversion of Grassi BST raw scores. z Score Conversion of Hunt-Minnesota Test for Organic Brain Damage. 3 Score Conversion of Bender Visual-motor GBStalt TBBte (RIVIIID 10“! FORM 6-3A—RIV.-Bo40—BM-CARMICHAEL Appendix Al THE GRASS] BLOCK SUBSTITUTION TEST JOSEPH R. GRASSI RECORD FORM Name --- Age-----....... Date Education I. Q. Examiner TEST BEHAVIOR' (Describe the subject’s general behavior during the test, particularly in reference to coopera- tion, effort, tension, anxiety and other factors which may influence test performance.) G R A Y L Y N BOWMAN GRAY SCHOOL OF MEDICINE Winston-Salem, North Carolina Copyright 1949—J.R.G. TABULATION CHART DESIGN l H DESIGN II DESIGN Ill“ DESIGN IV DESIGN V T RHT R TJRHT R T R STEP 1 I STEP 4 I 6‘ pm i .l;c_;ll;_‘___l ACCURACY SCORE ........................ I _.—_. -_4 .—.. _.____ _._ >- t TIME CREDITS ................................ TIME DEDUCTIONS ................. . ....... TOTAL SCORE ................................ SCORING CRITERIA: I. Record under R in the Tabulation Chart, I credit for each correct response. No partial credit. 2. Record under T, above, the total time for completion of each step. 3. Credit an additional V2 point if the difference between the steps below is IO” or less. Step I and 2 Step I and 3 Step 3 and 4 Step 2 and 4 4. Deduct V2 point for each design completed in I20" or more. MAXIMUM SCORE — 30 points. TENTATIVE NORMS: I. 20 points or more — no evidence of intellectual impairment. 2. 16 to 20 points —- moderate intellectual impairment. 3. O to 16 points — severe intellectual impairment. Appendix A2 Scoring Criteria for the Bender Gestalt Test (From Pascal-Suttell (A9)) Design 1 Wavy Line (2) Dot, Dash Circle (3) Dashes (2) Circles (8) No. Dots (2) ea. 10-1A Double Row (3) workover (2) Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) Design 2 1. wavy Line (2) Dash or Dots (3) Shape Cir. (3) Cir. Miss. ext. (5) Cir. Touch (5 , Dev. Slant (3 No. Col. (2 ea.) 9-13 Fig. on 2 Lines (8) Guide Lines (2) Whrkover (2) . Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) Design 3 1. 2. 3. A. 5. 6. 7. 8. 9. 10. 11. 12. 13. Asymmetry (3) Dot, Dash Circle (3) Dashes (2) Circles (8) No. Dots (2) Extra Row'(8) Blunting (8) Distortion (8) Guide Lines (2) Wbrkover (2) Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) Design A Asymt Crv. (3 Break Crv. (A Crv. Not Center (1) Curls (A) Not Joined (8) Crv. Rotation (8) Touch-u (8) Tremor A) - Distortion (8) Guide Lines (2) Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) Design 5 l. 2. 3. A. 5 6. 7. 8. 9. 10. ll. 12. 13. De 1. 2. 3. A. 5. 6. 7. 8. 9. 10. ll. 12. 13. Des 1e 2. 3. A. 5. 6. 7. 9. 10. ll. . Ext. Asymmetry (3) Dot, Dash Dir. (3) Dashes (2) Circles (8) Join Dot 2) Ext. Rotation 3) No. Dots (2) Distortion (8) Guide Lines (2) Workover (2) Sec. Attem t (3 ea.) Rotation ( ) Des. Missing sign 6 Asymmetry (3) Angles (2) Pt. Crossing (2 ea.) Crv. Extra Dbl. Line Touch-up( Tremor (A) Distortion (8) Guide Lines (2) Workover (2) (£1 ea.) Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) ign 7 Ends No Join (8) Angles Ext. (3) Angles Miss 3) Ext. Scat. 3) Dbl. Line (1 ea.) Tremor (A) Distortion (8 ea.) Guide Lines (2) Sec. Attempt (3 ea.) Rotation (8) Des. Missing (8) Appendix A2 Design 8 l. 2. 3. 5. 6. 7. 8 9. 10. ll. 12. Ends no Join (8) Angles Ext. (3). Angles Miss. (3) Ext. Scat. (3) - Dbl. Line (1 ea.) Tremor (A) - Distortion (8 ea.) . Guide Lines (2) Workover (2) Sec. Attempt (3 ea.) Rotation (8) M Des. Missing (8) Confg. Design 1. 2. 3. A. 5. 6. 7. Place Des. A (2) Overlap (2 ea.)~ Compression (3) Lines Drawn (8) Order (2 No Order 58 I Rel Size Design Totals 1. 5. 20 6e 30 7. A. 3. Config. Total Raw Score..... Total Standard Score..... Appendix B1 2 Score Conversion of Grassi Raw Scores Raw Score §_§ggp§ 5.0 '3000 5e5 '2e91 6e0 -2Q80 6e5 “2069 700 “2057 7e5 ’2045 890 -2033 8.5 "2.21 9.0 '2e09 9.5 -1098 10.0 -1.87 10.5 “le75 lleo .1063 11.5 “1051 12e0 .1039 12.5 -1028 13.0 -1016 1305 ’1005 11.».0 - .93 11445 "’ 08]- 1500 "' O70 15e5 ‘ 058 16.0 -' Ohé 16e5 - 035 17.0 '" 021+ 17.5 - .12 18.0 0 18.5 .12 19.0 .23 19.5 035 2000 0’" 20.5 05 21.0 070 21.5 .81 22e0 ’93 22.5 1.05 23.0 1.16 23.5 1.28 2A.0 1.3 2A.5 1.5 25.0 1.63 25.5 1.75 26.0 1.98 27.0 2.21 27.5 2.33 Appendix B2 z Score Conversion of Hunt Raw Scores Raw Score z Score 90 -l.l9 88 "le03 87 ' e93 86 " e85 85 - .77 84 " .68 83 - .59 82 "’ o 50 81 ’ ehl 80 ’ 039 79 " .26 78 " e17 77 " 008 76 " .01 75 ~ .07 7A .16 73 .25 72 .33 71 .Al 70 .50 69 .59 68 .68 67 .77 66 .8A 65 .93 6A 1.01 63 1.10 62 1.19 61 1.28 60 1.36 59 1.AA 58 1.53 57 1.61 56 1.70 55 1.79 54 1.87 53 1.95 52 2.03 51 2.11 50 2.20 Appendix B3 z Score Conversion of.Bender Raw Scores z Score Raw Score z Score Raw Score 2 Score Raw Score 160 -2.70 119 -1.22 78 .25 159 -2.66 118 -l.19 77 .29 158 ~2.62 117 -l.15 76 .33 157 ~2.58 116 -1.11 75 .36 156 -2.5A 115 -l.07 7A .39 155 -2-50 111 -l.0A 73 .AZ 151 ~2.A7 113 -1.00 72 .A6 153 -2-h3 112 - .97 71 .A9 152 ~2-h0 111 - .93 70 .53 151 -2.36 110 - .89 69 .57 150 -2.33 109 - .85 68 .61 1A9' -2.29 108 - .82 67 .6 1A8 -2.26 107 - .78 66 .6 1A7 -2-23 106 - .75 65 .71 1A5 -2.15 10A - .68 63 .78 lhh “2.12 103 "eéh 62 .82 143 -2.08 102 - .61 61 .85 1A2 -2.0A 101 - .58 6O .89 1A1 -2.00 100 - .5A 59 .93 1L0 -1097 99 ‘ ehg 58 097 139 -l.9A 98 - .A6 57 1.00 138 -l.90 97 "eh3 56 1e0h 137 -1.87 96 - .A0 55 1.07 135 -l.80 94 - .3A 53 1.1A 13A -1.77 93 - .31 52 1.18 133 -l.73 92 - .27 51 1.22 132 ~l.70 91 - .23 50 1.26 131 -1.66 90 - .19 49 1.30 130 -1.63 89 - .15 48 1.3A 129 “1059 88 - e12 L7 le38 128 ”le56 87 - .08 #6 l.A2 127 -1.52 86 - .OA A5 1.A6 125 ~l-Aé at .03 A3 1.54 12“ “lehz 83 e06 #2 1.58 123 -1.38 32 .10 41 1.62 122 -1.3A 81 .13 A0 1.66 121 -l.30 80 .17 39 1.70 120 -1.26 79 .21 38 1.7 37 1.7 36 1.82 35 1.86 a“ (f) —-> n fif‘f‘i 1‘, 11.136.“ U c. ‘ ane- C (“f1 91? 11' S a “-4aqf‘