University of Southern California Dissertations and Theses

Immediate Memory And Its Correlates With School Achievement

(USC Thesis Other)

Immediate Memory And Its Correlates With School Achievement

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 6 7-5300 MEEKER, Mary Nacol, 1928- IMMEDIATE MEMORY AND ITS CORRELATES WITH SCHOOL ACHIEVEMENT. University of Southern California, Ed.D., 1966 Education, psychology University Microfilms, Inc., Ann Arbor, Michigan IMMEDIATE MEMORY AND ITS CORRELATES WITH SCHOOL ACHIEVEMENT by Mary Nacol Meeker A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF EDUCATION (Educational Psychology) August 1966 This thesis, written under the direction of the Chairman of the candidate’s Guidance C om m ittee and a p p ro ved by all members of the Com m ittee, has been presented to and accepted by the Faculty of the School of Education of The University of Southern California in partial fulfillment of the requirements for the degree of M aster of Science in Education. Date Dean Guidance Comm ittee Chairman ACKNOWLEDGMENTS The author would like to express her gratitude and appreciation to the members of her dissertation committee, professors C. E. Meyers, Robert McIntyre, and Elmer Wagner. A special debt of gratitude is acknowledged to J. P. Guilford for his help, encouragement, and generosity. This study was made possible through the coopera tion of the Centinela Valley High School District. Appre ciation is expressed for the use of their students and facilities, and especially to Marie Boryan, vice-principal and head counselor, whose aid cannot be estimated. The author is grateful to Robert Meeker, whose sympathy was unending, and to her children, Jessica, Valerie, and Heather, whose early independence training generalized to their mother's needs at the most propitious time; to her mother, Hattie Nacol, the ultimate in grati tude; and to her friend and typist, Marguerite Tolson, the deepest appreciation for her time, skill, and special knowledge. Computing facilities of Systems Development Corporation in Santa Monica enabled the author to do both factor analyses under the "on-tap" help of Miles Rogers and Harry Harman and to use for the first time in education TRACE (Time-Shared Routines for Analysis Classification and Evaluation), a program developed by Robert Meeker, Jerry Shure, and Bill Moore, which allowed data reduction and variable processing such that the most powerful statistical analyses could be used. TABLE OF CONTENTS Page ACKNOWLEDGMENTS ......................................... ii LIST OF T A B L E S......................................... v LIST OF ILLUSTRATIONS.....................................vii Chapter I. INTRODUCTION .................................. 1 Review of the Literature Implications of the Present Study II. PURPOSE AND ORGANIZATION OF THIS STUDY, WITH DETAIL FOR PART I .............. 4 5 Organization of Study III. PROCEDURES AND FINDINGS, PARTS ONE AND T W O ........................... 56 Part One Part Two IV. MEMORY FACTOR SCORES FOR CLINICAL GROUPS . . 90 Hypotheses Discussion of Hypotheses Summary V. SUMMARY, FINDINGS, DISCUSSIONS AND RECOMMENDATIONS............................107 REFERENCES................................................ 122 APPENDIX...................................................131 LIST OF TABLES Table Page 1. Distribution of Spans Made by the 90 Subjects on the Various Span Tests........................... 64 2. Communalities for 27 Variables and Computed Reliabilities for Memory Tests............................. 66 3. Final Correlations, Means, and Standard Deviations from Principal Factor Analysis for 25 Variables ........... 70 4. Final Eight-Factor Rotation for 25 Variables.................................. 71 5. Correlation Coefficients, Means, and Standard Deviations of Test Battery for Part Two Using Factors II, V, and VI as Predictor Variables ....................... 83 6. Reliability Coefficients for Tests Used in Part Two of the S t u d y ............... 84 7. Chi Squares or Fisher-Yates Significances on the Relationships between the Memory Factor Scores and Scores on Subject Tests Achieved by the Highest and Lowest Scorers on each Factor...................... 86 8. Means, Medians, Standard Deviations and Tests of Significance for the Memory Tests Given to Sixty Boys in Six Clinical Groups...................... 99 9. Mean Factor Score Approximations in z Scores for Six Clinical Groups .............. 100 v Table Page 10. Comparisons between Clinical Groups on the Three Memory Abilities (Mean Factor Score Approximations) ......... 102 11. Description of Tests Used in Part 1 and Part 2 and Score S h e e t ............132 12. Means, Standard Deviations, and Correlation Coefficients for 27 Variables.....................................139 13. Eight-Factor Rotation Matrix for 27 Variables.................................140 14. Four-Factor Rotation Matrix ................... 141 15. Six-Factor Rotation Matrix for 25 Variables.................................142 16. Six-Factor Correlations, Means, and Standard Deviations for 25 Variables .... 143 17. Factor Score Contributions in Standard Deviations ......................... 144 18. Individual Identifications of the Ten Highest and Ten Lowest Factor Scores on Member Memory Tests .............. 145 vi LIST OF ILLUSTRATIONS Figure Page 1. Percentage Frequency of Completely Correct Recall of 3-Consonant Trigrams....................................... 12 2. A Schematic Diagram Taken from Osgood, Showing Flow Lines from Stimuli to Responses............. 19 3. Theoretical Model for the Complete "Structure of Intellect" ......... 51 4. Schematic Showing of Dual Processes of Memory Storage............................ 53 5. Mean Factor Score Approximation for Six Clinical G ro u p s ..................... 98 vii CHAPTER I INTRODUCTION Memory as a special ability was not systematically investigated until the efforts of Ebbinghaus (McGeoch § Irion, 1952) to measure retention. His findings resulted in learning curves and forgetting curves which defined memory, historically, as retention of learning tested either with nonsense syllables or paired-associate words or letters. Events subsequently found by other investiga tors to be associated with Verbal Paired Associate Learn ing (VPAL), such as distribution of practice, the length- difficulty relationship, position effects, interference and its effects, reminiscence, forgetting, motivation, and incidental memory, among others, have come to be regarded as foundations for learning principles which may be ap plied when teaching subject matter. McGeoch and Irion (1952), in their review of the body of literature on VPAL, concluded that these are derived principles which do ex press relationships between practice and performance, and as rational equations are really quantitative hypotheses concerning the nature of the learning process, since they 1 specify precise relationships between the independent and dependent variables of the learning process. However, principles derived from the learning of serial lists and paired-associate lists have been based on the following implicit assumptions: (1) that memory as a process is not differentiated, (2) that a measure of retention is equally a measure of learning ability, and (3) that sensory mode responses are the same for all sub jects. It is the purpose of this study to investigate the assumption that immediate memory is molar and that certain tasks which are assumed to test memory not only demand but are predicated upon different kinds of memory abilities. This study seeks to differentiate memory abilities in ninth-grade boys of average intellectual ability and to determine whether specific memory abilities can be related to specific curricular tasks, dependent upon modal input. Auditory and visual immediate memory will be examined as the more critical for mastery of school learning. Review of the Literature It is likely that no other subject in psychology has been more surveyed than has memory. The following re view of the literature defines the problem leading to the writer's investigations and explorations. In order to make a cogent whole of this body of knowledge, the 3 literature was organized into four sections so that each would, by reason of lack of information to be reviewed, show the necessity for the present study. The four sec tions are: 1. The need for a theory of memory and those studies concerned with psychological issues which have pertinence to the development of a theory. 2. The factor-analytic approach to defining memory and its contributions to education. 3. The need to relate neuro-physiological data and chemical findings in studies of memory. 4. The place of psycho-physiological data in defining memory. Section 1. Issues Involved in the Development of a theory of Memory As a result of the 1963 meetings of the American Association for the Advancement of Science, theoretical psychologists and psychological studies of human short term memory sharpened critical issues involved in the development of a theory of memory. One issue with which learning theorists are concerned is in regard to appropri ate assumptions to be made about characteristics of memory traces that are products of experiences and repetitions of experiences. Hull (1943) , Spence (1955), Jones (1962) , Keppel 4 and Underwood (1962), among others, have made the assump tion that habit strength grows incrementally over repeti tions (memory traces studied with engrams, associations, bonds, habit strength, sHr's); as an alternative they pro posed an all-or-none conception; Estes (1960) has ques tioned the validity of this assumption. As a result of Underwood's (1957) discovery that proactive inhibition had been underestimated as a source of interference in forgetting, a second area which came under surveillance was the role of interference in a theory of forgetting. Barnes and Underwood's (1959) work was given more importance with their finding that the unlearn ing factor was a component of retroactive inhibition. Consequently this development attracted more attention to the interactions of memory traces during learning as well as their interactions at the time of attempted retrieval, whether tested by recognition or by recall. Since 1958, however, a third force has emphasized the need for more vigorous theorizing and research on im mediate and short-term memory. The nature of memory, the characteristics of the memory trace, and especially the relations between short-term memory and that memory which results from multiple repetitions, have been and are still areas for clarification. Prior to that time, research on short-term memory was concerned with the memory span as a capacity variable and very little else, and, as mentioned 5 in the introduction, assumptions for these studies were similar in that memory was considered an undifferentiated ability. When Broadbent, in Perception and Communication (1958), emphasized short-term memory as a major factor in human information processing, the renewal of investiga tions was predictable. There is, nevertheless, much con troversy among learning theorists concerning the question of the necessary and sufficient conditions for association (or memory trace) formation (Melton, 1963). What, then, are the necessary and sufficient conditions? Since investigators of memory have followed Ebbinghaus in employing rote learning (to avoid problems of selective learning and to ensure the contiguous occur rence of stimulus and response under conditions that demonstrably result in formation of an association), their emphasis has been on the storage and retrieval of that association, i.e., of the residual trace of it in the cen tral nervous system and on the ways in which frequency of repetition and other learning affect such storage and re trieval (McGeoch § Irion, 1952; Melton, 1963). The storage of traces, however, necessarily must be investigated separately from the retrieval of traces before one can determine which conditions are necessary and which are sufficient. Osgood and Miron (1963) and McCarthy and Kirk (1961) have demonstrated that perception (decoding), 6 association (storage), and retrieval (encoding) can be mea sured as independent processes. There are four issues in storage. First, should memory traces be given the charac teristic of autonomous decay over time (Thorndike's Law of Disuse recently championed by Brown [1958])? The alterna tive is the notion formulated by McGeoch (1932), Guthrie (1935), and others, that associations, once established, are permanent. The second storage issue is also hypothetical and concerns the autonomous process of enhancement or consoli dation of the memory trace rather than its decay. The hy pothesis of perseveration as strengthening a memory trace if it is allowed to continue without interruption was ad vanced by Muller and Pilzecker (Melton, 1963). Their im plication was that an automatic "inner repetition" occurs if repetition and duration are given equal opportunities to function. More recently, Deutsch (1962) and Hebb (1945), among others, have hypothesized that memory trace estab lished through an experience requires fixation through au tonomous reverberation or perseveration if it is to become a stable structural memory trace in the central nervous system. The alternate view would then be that every ex perience establishes a structural memory trace without the necessity of consolidation through reverberation. Melton (1963) qualifies this assumptive alternative, however, by stating that "this would not deny that perseveration, if 7 permitted, may strengthen the trace" (p. 7). A third issue about storage is actually reflective of the first issue above. It concerns the issue of asso ciation formation. Is the second incremental trial (or trial +n trials) the same trace system as the first? Is its activation a reactivation which is strengthened? Or is the reactivation a new (or enlarged) trace? The fourth issue about trace storage seems to be the most critical to development of a theory about memory. Theorists are concerned about, and taking positions on, the question of whether there are two kinds of memory storage or only one. Advocates of the dual processural storage, Waugh and Norman (1965) , Hebb (1949) , and BToadbent (1958) postulate that, on the basis of many findings, one type of storage mechanism is involved in remembering of an event recently experienced and that a different type is involved in the recall or other utilization of traces established by repetitive learning experiences, i.e., long-term memory or habit. The differences of theorists may be definitional. James in 1890 (Waugh, 1965) defined primary memory as an event which has never left consciousness and is a part of the psychological present, while an event recalled from secondary memory has been absent from consciousness and belongs to the psychological past. Perhaps definitional differences may be temporal, yet there are other consider ations. For example, the demonstration which evaluates 8 the subject's memory ability and which may be said to be part of the psychological present may indeed be so, but it may not have been internalized or perceptually integrated. For that reason, individual differences must then play a part in the mode of perception as well as in the individual temporal tolerance for the psychological present. Other definitional differences between short-term memory (STM) and long-term memory (LTM) are (1) that STM involves "activity'' traces, whereas LTM involves "struc tural" traces (Hebb, 1949, 1961); (2) STM involves autono mous decay, whereas LTM involves irreversible, nondecaying traces (Hebb, 1961); and (3) STM has a fixed capacity that is subject to overload and consequent loss of elements stored in it, for nonassociative reasons, whereas LTM is, in effect, infinitely expansible, with failure of retrieval attributable mainly to incompleteness of the cue to re trieve through interference from previously learned associ ations (Broadbent, 1958). Melton (1961) and Keppel and Underwood (1962) pos tulate therefore that a monistic view with respect to trace storage in general would accept the characteristics of LTM storage as the same characteristics of STM storage so that both types of traces of events that occur only once would have the same structural properties, same irreversibility, and same susceptibility to associational factors in retrieval. 9 Since the main distinction made between STM and LTM pertains to the obvious finding called "overload," an examination is necessary to see whether the overload is in the person's capacity to retrieve specifics as given or whether it is in his general performance. For example, the subject is given (or exposed to on a memory drum) some nonsense syllables or numerals and is asked to indicate recall. On the other hand, the same subject is accustomed to going to a movie where a story enfolds without a break from beginning to end and he is able to recall the plot perhaps for years accurately without having had any pos sible chance for rehearsal while observing. The sugges tion here is that perhaps the temporal aspect of over loading is also affected by the material to be recalled. It may be that resultant curves from paired-associate- learning experiments are actually curves which demonstrate examples of Miller’s plus or minus 7 (1956). That is, they may be norms of curves which have to do with sub jects' remembering bits of information rather than assess ments of their ability to retain. At least Melton (1963), after examining the as sumed dichotomy of the phenomena of short-term memory and long-term memory, after restating a number of current issues in learning theory as issues about the formation, storage, and retrieval of memory traces, and after relating the slope of the short-term forgetting curve to 10 the number of elements or recoded "chunks/* confirmed and extended Hebb's finding that there is a specific accumula tive strengthening effect of repetitions in the "immediate" memory situation beyond the span of immediate memory of human subjects. He concluded, however, that a continuum, rather than a dichotomy, is involved in short-term and long-term memory. Thus he rejected the postulate of auton omous decay of traces in the case of short-term memory and accepted the postulate of permanence of traces, once formed, throughout all varieties of memory. His rationale for clarifying postulates for a theory of memory was based on his belief that a theory of learning must encompass the processes of trace formation, trace storage, and trace utilization, because he believes that a general theory of memory is considered to be only a portion of the domain of a theory of learning, and he sees no need to differentiate between the dual and single mechanism theories of memory with respect to the factor of stimulus change in retrieval, although he does indicate that an implicit and undetected one may exist. Since Postman (1964) and McGeoch (1952) hypothe sized that irreversibility of traces and all failures of retrieval or utilization are interpreted as instances of stimulus change or interference (and thereby implying a one-mechanism theory of memory), it seemed necessary then that measurement of only those aspects of memory retrieval 11 where the temporal element is so short as to eliminate the possibility of rehearsal needed to be undertaken. There fore, not only would it be important to differentiate between mode of presentation, but it would also be neces sary to check for differentiation in retrieval of types of presented materials. One of the more critical issues related to retrie val has been the relationship between repetition and re trieval probability; to avoid this issue as contamination (that is, whether repetition multiplies the number of all- or-none traces or whether it produces incremental changes in the strength of a trace) the present study will be con cerned only with characteristics of traces as retrieved after a single repetition, with rehearsal controlled. Another issue seemed to be critical in retrieval and that is the stimulus itself. For example, when reten tion intervals are compared in Murdock’s replications of Peterson and Peterson's experiments (Fig. 1), the fact that there were two very different curves was explained in terms of Miller's (1956) "chunks." Melton (1963) indi cated that the implication here was that, other things being equal, the rate of forgetting of a unit presented once is a function of the amount of intra-unit interfer ence, and that this intra-unit interference is a function of the number of encoded chunks within the item rather than the number of physical elements such as letters or 12 < # > 5 s H U I o s o o 100 90 80 1 word 70 Murdock (1961) 60 50 3 words 3 consonants 30 20 10 Peterson £ Peterson (1959) _ 3 consonants 0 3 6 0 9 12 15 18 RETENTION INTERVAL (seconds) Figure 1. Percentage frequency of completely correct recall of 3-consonant trigrams (Peterson and Peterson, 1959, Murdock 1961, and 1-word and 3-word units, Murdock 1961). information units. On those replicated experiments where three conso nants were used, the curves were similar; when Murdock (1961) instead used one word, the visual curve is much higher. The difference in retention may easily, the writer submits, be accounted for in terms of factorial differ ences as found by Guilford, where the words had meaning attached to them and the consonants, of course, did not. The writer suggests that there also may be another 13 explanation for the different slopes, and that is that they may be due to the differences in modal stimuli. Section 2. Factor-Analytic Approach to Studying Memory Guilford (1959) has shown through factor analysis that symbols ”S" (numbers and letters) are factorially dif ferent from semantics "M" (words and ideas) when perfor mance is tested. His data, like those acquired in VPAL, were based on measured performances of young adults. He had hypothesized and demonstrated that memory as a compo nent ability of intelligence was factorially different from others. Orpet and Meyers (196$) and Carlson and Meyers (1966, unpublished), among others, have demonstrated that memory abilities factor differentially, confirming structure-of-intellect hypothesized abilities; however, in their studies the stimulus differences ("S" and "M”) did not separate in data on young children. They found that the digits-forward test, when administered with Binet pro cedure, tended to load with any auditory-forward content in a factor never saturating digits backward. The auditory-forward memory-span factor included meaningful as well as non-ordered sentences, digits, and letters, con firming the separateness from the reverse digits and from visual memory such as the cube test. At present other studies (Sitkei § Meyers, 1966, unpublished; Meyers, Sitkei, 14 and Watts, 1966, unpublished) on normal children of four and six, respectively, show the same: forward auditory forms one factor, regardless of whether letters, digits, fast digits as in ITPA, sentences, or words are used. Visual memory, if with pictures or other meaningful stimu li, stays separate. Digits backward and Knox cube tapping are separate also, going to other factors but without clear pattern except for being more likely to join a visual factor. The most analogous study to the above was per formed by Loeffler (1963) on retarded Ss having mental ages of about six to seven. Here a clear auditory-span factor emerged, while three visual-memory tests loaded on a figural reasoning factor. The famous McNemar (1942) analyses of the 1937 standardization data of the Binet were carried to three unrotated factors. The tests of digits forward and back ward, word memory, and sentence memory were among the tests tending to load beyond the first factor. Stott and Ball (1963) showed one or two memory factors in analyzing various sets of canned test data of infancy and pre-school years, based on Gesell, Binet, and Merrill-Palmer scales. Pioneer American work with children was done by Truman Kelley (1928), who showed a short-term auditory memory at kindergarten age. Historically, he anticipated for today’s investigators the possibility of separating memory 15 abilities. Some factor analyses have been performed on the Illinois Test of Psycholinguistic Abilities (McCarthy § Kirk, 1961). The digits-forward test of this battery tends to load as a singlet, or on a general factor, in the several analyses for different age levels in which only the nine subtests were entered into the matrix (McCarthy 6 Kirk, 1963). However, when other auditory span tests are used, there is no factorial doubt--the ITPA digits test loads with its kind (Meyers, 1966; Meyers, et al., 1966). French (1951) summarized the older factor-analytic studies published. Extracting from his numerous tables, the writer noted that Memory as reported by numerous in vestigators factored into a general memory factor with loadings reported from .26 to .80, a visual memory factor with loadings from .52 to .58, a paired-associate memory factor from .41 to .66 with one rotated as auditory (.56), and a musical memory factor with loadings from .33 to .53. Mary Tenopyr (1966), in a dissertation under Guilford, included in her factor analyses and correlation matrices tests sampling 50 other structure-of-intellect variables. Among these were consonant spans, visual and auditory, digit spans, auditory only, and nonsense word spans, visual only. The span tests loaded on one memory factor for forward spans regardless of whether the span consisted of words or digits, auditorially or visually 16 presented. Her sample included boys and girls who were ninth-grade students. The auditory-memory span tests unquestionably cor related higher with school achievement and grades. Duncanson (1964) administered a battery of ability tests to 102 sixth-grade children. The tasks related to rote memory included one verbal task, one numerical task, and one figural task. Factor analysis led to factor scores which in turn were entered in a factor analysis together with scores on ability measures; one factor specific to all abilities was interpreted as a speed factor. Verbal ability, rote memory, and reasoning were also obtained. It was concluded that although measured abilities were re lated to measured learning performance, learning perfor mance depended upon factors which were independent of ability measures; that although learning in one task may be related to learning in others, performance in concept- formation tasks was not related to the ability measures used nor to performance in rote-memory tasks. Duncanson was investigating the role of each of the above tasks in learning and so each task was repeated for 12 trials. In essence his work consisted of the separating of types of learning tasks rather than identifying contributions of various tasks to learning. The categorizing of learning tasks and validation of them upon sixth-grade students gives added support to the Structure-of-Intellect model (see Fig. 3 in Chap. II, p. 51). Brown (1966, unpublished) and Guilford and Hoepfner (1963) pretested 40 new memory tests, and of these, 27 were selected to be included in a factor- analytic battery for 175 high-school juniors. The factor MMU (Fig, 3) (memory for semantic units) is the ability to retain relatively isolated pieces of information; among other memory tests, it retained its separateness. Later, as part of an orientation program in a state college, 10 of the memory tests were administered to a group of over 500 entering freshmen. At the end of the academic year, grades were obtained for a random sample of 100. They were converted to a five-point scale and correlated with each of the ten memory tests. Nine of the ten tests corre lated significantly with the grade criterion, eight were significant beyond the .01 level. A multiple R of .48 was significant beyond the .01 level. It seemed apparent that tests measuring memory are significantly related to success in elementary psychology in a state college. In Guilford’s (1959) Structure of Intellect, memory is only one of five major operations which have been iden tified in factor-analytic approaches. The structure-of- intellect model does not assign weight, or value, to any one of the operations or cells. Nevertheless, without Cognition there could be no Memory. Without Memory there could be no Evaluation. It is possible that without 18 Evaluation there might be Convergent Production, and, as in the case of digit spans and word spans, it is possible that Convergent Production might occur without Memory as it is customarily defined wherein retention or intent to learn with storage is implied (Meeker, 1965). It is at this critical point in defining memory where theorists depart from each other and categorize memory either as a dual pro cess or as a monistic process. Therefore, a most important aspect to be explored in understanding educational poten tial is an intermediary process between Cognition and Con vergent Production, namely, memory. Osgood (1953, 1957; Osgood and Miron, 1963; Osgood and Sebeck, 1954) has termed his multi-stage psycholinguis- tic behavioral model the "Mediating Box" approach. Waugh and Norman (1965), on the other hand, have called the primary memory, or immediate memory as used in this paper, a func tion of the "Echo Box" capacity. Although both imply a lim ited capacity, Osgood introduces in his model the concept of "central" intermediaries which he assumes to mediate be tween observable or controllable stimuli and measurable re sponses. He postulated his concept in an effort to simu late an intervening variable which will account for mean ingful language behavior; his multi-stage theoretical model of behavior in which language is a special case (see Fig. 2) describes three levels of organization (projection, inte gration, representation), assumes three processes in the 19 THEORETICAL MODEL OF PSYCHOLINGUISTIC ABILITIES LEVEL PROCESS DECODING ASSOCIATION ENCODING Representational (conceptual) Predictive Motor Sensory / Integration Sequential Automatism r-r-r s-s-s Integrational (Perceptual- Imitational) Evocative Unit Automatism r-r-r s-a-s Reflex Projection (Sensory - Motor) r-r-r I Effectors s-s-s Receptors Signal Automatism Vocal Visual CHANNEL Auditory Motor Tactual FIG. 2. A schematic diagram taken from Osgood (1963) showing flow lines from stimuli to responses. 20 handling of information (decoding, association, encoding), and includes their communication channels (visual, auditory, tactile). This study will examine memory as an association process through the visual and auditory "S" channels for decoding and will use results from the tests as measures of encoding for both vocal and motor "R." A summary of the model may bring added clarifica tion of the theoretical issues which have so greatly con cerned Melton (1963) and others, as previously discussed. Description of the Osgood M o d e l -There are three levels of organization: (a) The projection level is conceptualized as a neurophysiological relay system between the peripheral receptors and effectors and the central nervous system; the sensory surface of the person is precisely manped upon the sensory cortex, and the voluntary muscle system is precisely mapped upon the motor cortex. To the projection level has been ascribed the two functions identified as isomorphism and unmodifiability (Osgood, 1957). In isomorphism, the receptors are "wired-in" pre cisely to the sensory cortex, whereas the motor cortex controls effectors in a point-to-point projection. It has been shown that for each point on the retina there is a corresponding one in the cortical area. For example, if electrodes are placed in the temporal lobe while 21 different tones are sounded, the electrical recordings show an orderly arrangement of places on the cortex correspond ing to those of the cochlea. Osgood described functioning at this level as not modifiable by experience or learning; he viewed the pro jection system as a "perceptual tabula rasa" (Osgood, 1963). Visual stimulation produces the same measurable activity in Area 17 at 20 years as it did at 20 months, regardless of experience. Thus, the projection level is regarded as a reflex level which receives, transmits, and expresses events in an orderly fashion such that it provides for organization of the peripheral, sensory, and motor events. The stability of these responses permits behavioral theo rists to regard stimulus-and-response observations as indices of sensory and motor signals. (b) The integration level of organization involves SS and RR associations rather than the S-R bond. It is concerned with overt receptive (decoding) and expressive (encoding) events characterized by regularities, pattern- ings, and orderings which have developed into automatisms and have come to be paralleled by equivalent covert organization within the sensory and motor nervous system. The integration mechanisms explain automatic phenomena such as closure, sequential memory, syllabication of speech, syntactical aspects of speech, and the reading and writing of words. The integration level is responsible for the 22 rapid retrieval of the proper selection and sequence of syllables and words in speech. Most messages received and sent involve a complex order of signs. Grammar of a lan guage forms the connective matrix within which messages may be communicated. The syntactical redundancies of language have grooved in certain expectancies in the speaker and listener (e.g., agreement between subject and predicate). Although this level of organization is not isomorphic or unmodifiable, it becomes, either through learning and/or practice, a relatively "driven" or auto matic system. The degree of automation is a function of such association variables as frequency, temporal contigu ity, and spatial propinquity. At this level the nervous system is assumed to reflect associations in experience. "The greater the frequency of co-occurrence of either stimu lus (SS) or response (RR) events in the experience of the organism, the greater will be the tendency for their cen tral neural correlates (ss or rr) to elicit each other within their own level." In this way Osgood postulates that we come to "mirror" and store in our nervous systems the redundancies of past experiences. The storage problem as Melton defines it becomes simplified within this opera tional definition, but the question of strength or fre quency of "S" is still not clarified. As shown in Figure 2, the integrational level is subdivided into predictive and evocative integration 23 levels of organization. Gestaltists have demonstrated that certain laws of perceptual organization exist: Law of closure, law of proximity, law of similarity, etc. (Hilgard, 1956). In the law of closure, for example, there is a tendency in perception for gaps to be filled in or completed. The dotted outline of an object may easily be recognized as continuous lines and not as so many disconnected lines. Our perception of objects is much more complete than the sensory stimulation we receive from the object. Osgood (1957) has extended these principles to include encoding and explains the evocative integration principle as fol lows: "With high frequency of stimulus or response pair ing, the occurrence of the central correlate of one will become a sufficient condition for the occurrence of the central correlate of the other." If there has been a close association between S-S-S...Sn, a part of a sensory signal (e.g., an incomplete figure) can trigger the pattern in the CNS to complete itself. Osgood believes that many perceptual activities on all channels of perception in volve such evocative integration from incomplete infor mation. Certain motor skills are also organized at the evocative integration level: for example, opening a door, typing, serving in tennis, syllabication in spoken lan guage, etc. More specifically, opening a door involves tightly integrated patterns of motor skill components: 24 arm-extension; hand opens, grasping; wrist twists; and arm flexes. The proper sequence of syllabication in speech represents a number of highly practiced motor patterns. In much of encoding, then, frequently repeated motor se quences are expressed automatically as tightly bound wholes once initiated. In decoding, this type of integration also underlies spelling and writing, the recognition of fami liar words when scanning or reading, and the rapid dis crimination among phonemes during speech. The greater the strength of relationship between S-S and R-R, the more one can reduce (chunk) the amount of information needed to complete the pattern. Evocative integration must be an all-or-none process or else percep tion and motor activity would be perpetual jumbles. The predictive integration level of organization, on the other hand, does not affect events on an all-or- nothing basis; instead, incomplete information facilitates and sets up an expectancy or readiness for decoding and encoding. Against a background of competing stimuli ("noise") certain cues increase the probability of a par ticular experience, and "with a lower frequency of stimu lus or response pairings, occurrence of the central corre late of one will merely 'tune up' or facilitate occurrence of the central correlate of the other" (Osgood, 1957). A predictive mechanism thus provides an explanation for the stability in much of decoding and encoding and establishes 25 a "set" for expected subsequent events. Evocative and predictive integrations have some instability in that they are still subject to disturbances and alteration from extraneous stimuli. There may be a syntax of behavior, just as in language, which provides a stability of perception and action relatively impervious to influence from "noise" and free from continuous volun tary decision making. The person is capable of integrating sensory signals into evocative and predictive automatism which parallel redundancies in experience; he is also able to integrate motor signals into evocative and predictive responses which reflect redundancies in his own past be havior. There is no sharp differential line between the evocative and predictive integration levels. The type of integration is a matter of degree and depends on the strength of association between SS for Decoding and RR for Encoding, the strength of association being a function of frequency, temporal contiguity, and spatial propinquity. Any stimuli and response, regardless of origin, can become operative at the integration level if sufficiently asso ciated in an automatic fashion. (c) The representational-mediation process is defined as "an implicit, stimulus-producing response assumed to mediate between observable S and observable R yielding a two-stage S-r-s-R process" (Osgood, 1963). The process is elicited by signs which represent original 26 significates for the person. This is an important mecha nism for associating sensory events with motor events because symbolic processes are involved. Decoding and Encoding divide the usual S-R para digm into two independent units. Decoding is the recep tive association of signs with representational mediators, while encoding is the expression of association of mediated self-stimulation with over-instrumental sequences. In other words, decoding is a process of interpretation or "reading” the stimulus while encoding is expression of that reception. Osgood points out that since each stage is itself an S-R process, all of the conceptual machinery and research of a single-stage S-R approach can be trans ferred to this two-stage model without new postulations. The dilemma which memory theorists have faced seems eased somewhat because of the contribution of the Osgood model. Hypothetically, the association processes involved in assessing memory are statistically differen tiated, but the acceptance of the model allows the placing of memory tasks along a continuum from the reflex, sensory- motor level through the representational or conceptual model. This is not the frame of reference Melton (1963) used to conclude that memory is a continuum, but the factor analytically based Osgood model schematizes a con tinuum in much the same manner. 27 Section 3, Chemical and Neuro- Physiological Effects on Memory Chemical.--Recent investigations have been made into the role played by ribonucleic acid (RNA), in reten tion of trained tasks. One of these, the work being done at the University of California, Los Angeles, by Jacobson, Jacobson, Babich, and Bubash (1965), has shown that "remem brance of things past" has now been found to be transferable not only from one animal to another of the same species, but to animals of different species. The investigators in jected ribonucleic acid taken from the brains of trained hamsters and rats and injected it into the brains of un trained rats. These animals then showed a tendency to perform a task that the trained rats had performed. The implication from this work is that, if it is shown to have significance for humans, it might assist in reaching an understanding of the role played by RNA in storing and retrieving memory. If the variables that influence RNA can be identified, further research might, it is hoped, facilitate increased learning capacity in humans. In their animal project, the investigators trained rats to approach a food cup at the click of a food dis penser. The rats learned to ignore the cup in the absence of a click. A group of untrained rats was then added as controls. Seven trained and eight untrained rats were sacrificed, and RNA was extracted from the brains of all 28 15. The material was injected into 15 new subjects. When the rats were then run through a series of trials, it de veloped that the animals that had been injected with the RNA of the trained rats displayed a significantly greater tendency to approach the cup area when the click was sounded. The California investigators cautioned that al though it appeared most reasonable that the effects observed were produced by the RNA material, it is con ceivable that other factors in the extract from the trained rats might have been involved. In subsequent experiments, Jacobson and his asso ciates achieved a similar "transfer of learning" from hamsters to rats. The researcher reasoned that if this effect could be shown to be dependent on associative learn ing rather than on a more general factor--for example, sheer amount of stimulation--a logical step in research would then be to find out what modifications in RNA are produced by training so that "memory transference" can occur. Jacobson would not speculate on the possibility that such research might lead to methods of improving re call and learning ability in humans. He did indicate, though, that better understanding of the human memory and learning processes could emerge. Toward this end, studies are in progress at McGill University in Montreal. At present, reports indicate that 29 the learning and recall ability of patients suffering mem ory loss has been dramatically improved after injection with yeast RNA. Neuro-Physiological.--Other animal experimentation of a neuro-physiological nature has concentrated on the loss of memory. Boycott (1965) has shown that in the octo pus's vertical lobe and the mammalian cerebral cortex, memory is both everywhere and nowhere in particular. He stated, however, that the difficulties such a conclusion presents to students of brain functioning may be due to a failure to distinguish experimentally between the two con stituents of a memory. He trained the octopi by condition ing with shock to force avoidance and non-shock to force acceptance. The purpose of the training was to elucidate the anatomy and connections of the animal's brain and relate them to its learning behavior. He hypothesized a complex associational interaction among all components in the brain such that removal of any one part has a slowing down effect so that the total carrying system (when devel oped to carry the expected neural transmissions) becomes congested, thus slowing down responses. The author concluded that parallels exist between results with invertebrates and epileptic patients who have been treated by surgical removal of the temporal lobes of the brain who score as well in IQ tests before 30 and after, who remember their past, their profession, and their relatives, but who cannot, however, retain new infor mation for more than short periods. Since this surgery involves the hippocampal system of the human brain, it seems that although man's cerebral cortex incorporates a long-term memory system, the hippocampal system is essen tial to the establishment of new long-term memories. Neurologically, the separation of memory into short-term and long-term systems is out of fashion, due to the fact that neither convulsive shocks nor cooling the brain to a temperature so low that all activity ceased seemed to abolish learned responses. It is now known that if such treatments are given during the early stages of learning before a memory is fully established, they do have an effect. Supposedly this is because interference occurred in the more active part of the process. As it happens, a long-term memory system is intact after removal of temporal lobes. A short-term memory sys tem must also remain, however, because the patients can remember new information for short periods. Boycott (1965) thinks, then, that the hippocampal system may have the role of linking the two memory mechanisms, just as the median inferior frontal lobe does in the octopus. The parallel is obvious. (The hippocampal gyrus in the human is located lower in the brain; however, it also lies as an undercoating of the cortex of the temporal lobe (Morgan $ 31 Stellar, 1950]). Conclusions Boycott has drawn from the findings support the theory that chemotactile and visual centers evolved out of primitive taste and bite reflex mechanism and became distance receptors. Environmental changes demanded responses which changed the animal so that learn ing had to become possible. The hypothesis that actual change represented by memory is produced by the small cells agrees with the facts found, and memory evolved to make inhibition last longer. Consequently, it seems that long-term memory evolved to perform a storage function, the limit of which depends upon the species, but storage functions certainly act variably in the human. This finding lends some cre dence to Waugh's model (1965) which shows Secondary Memory acting as a residual recipient of rehearsed material. The fact remains that at least in the octopus both visual and tactile memory systems embrace sets of brain lobes arranged in similar circuits, and it is Boycott's conviction that comparable circuits exist in the brains of mammals, includ ing man. The question may be raised here whether Boycott's results were attributable to the surgical removal of organs, or whether instead there was a total reduction of RNA which may have accounted for the findings. The Jacobson work needs to encompass the addition of further 32 research. Someone should discover methods of temporarily and successfully suppressing the action of RNA without actual withdrawal. Earlier Neuro-Physiological Studies.--Before 1930, Lashley (Garrett, 1930) was concerned with identifying the location in the brain of animals and in man which seemed to control learned behavior. He found that after excising the frontal lobe in cats, rats, monkeys, most recently learned tasks were not remembered. With humans who had suffered strokes the same phenomenon was observed. Franz (Garrett, 1930) controlled in later experiments for shock from the operative procedure and found the same results. In well-learned habits, however, there is persistence even when parts of the frontals are lost; he concluded that mechanized habits apparently are carried on through the agency of other parts of the brain. In cases where brains of humans had been injured by disease or wounds, retrain ing in colors, poetry, and prayers was possible. One man who died of pneumonia shortly after the experiments showed in post-mortem considerable destruction in the lower part of the cortex (temporal lobes). Franz points out that the slow but steady reeducation of this man suggests the pos sibility of the reestablishment of old brain connections or the opening of new ones. His feeling was that old path ways in the brain and nervous system, when still partially 33 intact, can be reached either through powerful emotion, motivation, or retraining. Lashley (Garrett, 1930), in later rat experimenta tion, substantiated that 1. The capacity to form motor habits is reduced by brain destruction, the reduction being independent of the place of the injury but roughly proportional to the amount of destruction. 2. The more complex the problem set, the greater the effect upon learning produced by any given injury. 3. Simple sensory habits such as that of bright ness discrimination are not greatly affected by cere bral lesions even when the whole sensory area is in volved, owing probably to the comparative simplicity of the habit itself. 4. Retention of simple motor habits (learning mazes) after forty days is significantly impaired by cerebral lesion, the degree of impairment depending chiefly upon the extent of the injury and the initial learning ability of the injured rats. (pp. 328-329) Garrett (1930), after review of the work of Lashley and Franz, concluded that the most far-reaching re .ult for human psychology of the above experimental studies is the finding that learning and retention of sensory and motor habits cannot be explained in terms of fixed nervous path ways, definite brain structures, or specific synaptic con nections. Complex learned activities must depend upon much more extensive brain patterns than was formerly thought probable, thus explaining why a single operation does not greatly disturb a given pattern, but the more ex tensive the injury the greater is the likelihood of some disturbance. The clinical evidence, as well as the experi mental evidence, indicates that this widespread substitution 34 of brain function in the case of rats, cats, and monkeys holds also to a high degree for man. To summarize, for years in physiological psycholo gy there has existed among theorists the dilemma which centered around attempts to explain how certain localized functions in the brain scientifically and logically can be made acceptable when the changes above were documented. RNA experimentation today may offer new solutions to sup plant the electrical conduit model proposed by cybernetics analogists (Morgan § Stellar, 1950). Chemical investigations, therefore, not only may confirm speculations resulting from early neural investi gations, but may also provide supportive evidence for the geneticists if RNA balances are found to be differential in animals and in humans. This latter speculation is interesting in light of French's (1951) speculation that certain factors such as reaction time and associative mem ory may be mainly genetic while learned factors may be mainly experiential in this culture. For the purposes of this study, then, the investi gator will measure non-practiced spans of unlearned imme diate memory, i.e., measurements of and differentiation of types of modal abilities in immediate memory will be attempted. Section 4. Psychophysical Aspects of Memory 35 Eidetic Memory (El) is one of the more intriguing types of psychophysical studies in memory. E. R. Jaensch investigated this phenomenon of memory and its physical manifestations. His work was interrupted by World War II and to date has not been pursued further. Johnson and Medinnus (1965) review it briefly, but Kluver (1931) pre sents a comprehensive review and description. He credits E. R. Jaensch in Marburg with having been the first one systematically to attack the problem of eidetic imagery in children. El, or, to use the German equivalent, Anschauungsbilder, are subjective visual, auditory, etc., phenomena which assume a perceptual character. An eidetic individual, after having been asked to look attentively at an object, is able, with eyes open or closed, to "see" this object again. This is possible either immediately or after a certain lapse of time, even after the passage of several years. These eidetic phenomena, as "images of hallucinatory clearness," are seen "im buchsta*blichen Sinne"; they are seen in the sense that after-images (AI) are seen. Although the stimulus object may be reproduced with almost photographic fidelity, El as a rule differ from the original stimulus object with respect to color, form, and the number of details. They differ from hallu cinations in that the eidetic individual does not believe in the objective reality of the phenomena but recognizes their subjective character; they differ from memory-images 36 in that the phenomena are really seen. These El, it is pointed out, are not of pathologi cal nature but must be considered as phenomena in "normal and healtly" persons. The investigation of the El is pre ceded by an examination of the A1 of the child. It was found generally that children with strong AI have weak El. There is no doubt that in general El's disappear at the time of puberty and that eidetic cases are rather rare among adults. When employing special methods for examining four- to-six-year-old children for AI, it was found that at this age negative AI exist. Kliiver (1931) reports that Kiesow, on the basis of a similar investigation in Italy, main tained that the absence of negative AI in four-to-six-year- old children is the exception rather than the rule; and that Roessler, who tried to clear up the whole problem of unitary type of El, found 401 of a group of 726 to be eidetic. Comparing different age levels, the highest per centages of eidetic cases and the highest "degrees" of eidetic imagery were found in the six-year-old group. In other words, the youngest group investigated represented the "acme" of the eidetic disposition. In fact, the greatest number of eidetic phenomena were observed in the first weeks of school life. Roessler interprets his re sults as lending support to Jaensch's view that the El is the primordial form of our perceptual and imaginal 37 responses. There seems to be general agreement that El range from "weak" to "strong," in other words, that there are "degrees" of eidetic imagery. Scales were constructed for indicating these differences. As emphasized by different investigators, visual memory and eidetic memory are largely independent of each other. Objects, forms, colors, etc., well remembered by the subject, are not necessarily seen in the El, and the intention to see them in the El may be entirely ineffec tive . Such reactions, commonly considered physiologically founded, are absent in other forms of El; still, this does not mean, as the experiments show, that El with a large number of physiologically founded phenomena are less easily influenced by the intention of the subject than those El which, in the sense of Jaensch, are visualized ideas. In eidetic vision we apparently have reactions on different levels occurring simultaneously in a fashion not yet entirely understood. In this connection it is worth not ing that the observation of a negative AI is often fol lowed by a train of El and interrupted sometimes by the reappearance of the AI. It seems that something similar occurs when AI are observed by adults under the influence of mescal; the negative AI was found either to become part of the visionary designs or in some way to "mix" with the mescal hallucinations. In any event it would be diffi cult for the patient to sort out the real from the unreal. Of the cited studies there was some indication that high El occurred among lower intelligence scorers on Binets, although some strong eidetics were found at all IQ levels, but when psychological functions required to pass certain IQ items were examined, there were types of corre lative abilities. Unfortunately, these correlative abili ties were not reviewed, other than to state that strong El was found in children of low intelligence, and that when intelligence-test scores were correlated with the results of tests for El, no correlation between El and general in telligence was found. A different picture presented itself if the various psychological functions involved in passing an intelligence test were considered separately in their relation to eidetic imagery or if different types of El were viewed in their relation to general intelligence. One investigator, in studying the formation of concepts in children, found that children of Jaensch*s B-type had great difficulties in replacing vivid El by abstract concepts. It was noted in an analysis of 109 answers given in a Binet test (items not elucidated) 25 per cent of the an swers were based on El and 30 per cent on the description of mental images. Another investigator found that children of one type of El, the B-type, were more suggestible than children of another type. Other experiments have brought 39 out the fact that El are utilized whenever tests demand observing visual objects or require certain forms of visual memory. It may be added that some interesting results were obtained from an example of El in feeble-minded chil dren. Results were not elucidated. Nor was there any reported comparisons between type of El and IQ level. What is the importance of Eidetic Imagery for edu cation? Some writers emphasized the great significance of eidetic imagery for classroom instruction, others warned against hasty generalizations in this respect. The analy sis of some adult eidetic cases showed the important role eidetic imagery may play in studying certain subjects. It appears that further investigations are neces sary before determining to what extent the results of research on eidetic imagery should be taken into account in the teaching of different school subjects. Of greater im portance than a possible change in didactic methods is un doubtedly the recognition, on the part of the teacher, that high or low scholarship in certain fields, ethical "defects," as well as the pursuit of certain "ideals," may have something to do with eidetic imagery. The forensic importance of certain kinds of El should be kept in mind, specifically as they may be visual mediators replacing or at times supplanting verbal mediators. With the developmental approach to perception typified by Fantz (1965) and others, one may speculate that 40 the early investigations of eidetic memory may be profit ably re-explored today within the frame of reference which would look upon El as an unlearned perceptual ability, and that this ability may be submerged during the process of visual training which is required for symbolic and semantic learning in the schools. Better still, perhaps El serve as associative mediators for finer visual discrimination. An interesting speculation is that the occurrence of learn ing at a critical time is undeniably important, but that learning then may be limited due either to genetic factors or to chemical factors such as RNA. Implications for the Present Study Each of the four preceding sections has shown the need for clarification, documentation, and application of knowledge about memory. In VPAL investigations, for exam ple, some of the current problems can be attributable to lack of recognition of a satisfactory model which can serve as an organization for developing learning principles or a theory of memory. Two experiments serve as examples of this difficul ty. Murdock's replications of Peterson and Peterson's studies (see Fig. 1) might have had explication for the different curves if the Guilford Structure of Intellect had served as the basis for selecting stimuli. A second example of findings which might have been 41 more easily explained if the structure of intellect had served as organization for the stimuli is the important replications of early experiments on serial and paired- associate (PA) learning wherein Jensen and Rohwer (1963) , Rohwer (1964), and Davidson (1965) used children for their subjects instead of college students. Rohwer, et al., emphasized a developmental approach to basic problems in the experimental psychology of learn ing. The reported analysis of variance table indicated that the ability to learn does not change with age since age interacted differently with serial and PA learning. That is, with increasing age there would be an expected increase in verbal experience, therefore PA learning would improve with increasing age. The implication here is that the apparent superiority of older children over younger children in many forms of learning would be interpreted as being due to greater transfer from prior learning and not due to any fundamental increase in the ability to learn, whereas the implication for serial learning would be that it is reflective of basic learning ability rather than reflective of transfer from prior verbal experience--that serial learning would not change with increasing age. It is possible that the conclusions of Rohwer, et al., though rigorously deduced, may not, however, be due so much to the lack of verbal mediation in serial learning as it may be due specifically to the 42 authors' lack of recognition that these two tasks actually are not comparable to begin with, therefore treating them with any facilitator could not have equal results. At least as these tasks, PA and serial learning, would fit into the logic of the structure of intellect, they would occupy different factor cells. Paired-associate learning demands that a relationship be imposed on "bits" by the subject while immediate storage, "chunking," is taking place for both stimuli (the forming of cell assemblies [Hebb, 1949J); whereas in serial learning the task con sists of tying a previously presented item (secondary mem ory [James, 1890]) to a primary memory (Waugh 8 Norman (1965) in immediate memory at a units level of presenta tion, thus a temporal element is introduced. Since it is the function of analysis of variance to break the variance of tasks and their measures into pieces, and since these were different tasks to begin with, the variance found was significant. The conclusions the authors arrived at may have completely different causative relationships. VPAL, if placed into the structure of in tellect, is testing memory and perhaps even cognition at a relations level; serial learning is testing memory at a units level; and both are dependent upon vocal recall of visually presented stimuli. It is also possible that some of the resulting differences may stem from another aspect of individual differences of the subjects, namely, that 43 some subjects may have been better at visual modes of learning while others may have been better at auditory presentation. One may argue that these differences may randomize themselves within the subjects, but it would be a cleaner investigation had they been controlled for separately. In fact, the discovery by the authors that this latter ability, serial learning, did not improve with age increments has relevance to the typical learning situation in public schools, even though the reasoning behind the conclusions differs in emphasis from that of Guilford (1959), Meyers, et al. (1966), Brown (1966), and Tenopyr (1966). Both the VPAL and structure of intellect are con ceptual approaches to understanding the construct, memory, in young children; they differ from each other in descrip tion in that in the first, the concern is the replication of traditional PA learning with children as subjects (Jensen § Rohwer, 1963), whereas in the second the interest lies in isolating an ability on which learning is predi cated. It was suggested that the issue of the dichotomy (or lack of it) in the theoretical problems of establishing a theory of memory as a sub-section in the establishment of a theory of learning (Melton, 1963), could be clarified within the Osgood model (Fig. 2). 44 The structure of intellect hypothesizes differen tial abilities as performed; the psycholinguistic model (Osgood, 1963) hypothesizes the loci of these differential abilities wherein the measured performance implies the loci through communications. In order to establish the usefulness of the above speculation, it will be necessary to develop and clarify the issue of whether differential inputs into storage do or do not establish differential memory systems. This study addresses itself to this problem and goes beyond factor-analytic findings to ask the question: If there are differential modal inputs, are the differential re trievals significant beyond the establishment of factor- analytic separations? If so, then do they have relevance or relationship to differential curricular tasks? The writer recognizes the importance of the neuro- physiological and psychophysical areas as reviewed, but relegates these internal attributes of memory to labora tory experimentation which may some day shed light on the reason why hypothesized measures are and can be differen tiated. CHAPTER II PURPOSE AND ORGANIZATION OF THIS STUDY, WITH DETAIL FOR PART I This study was concerned with some of the memory issues left by the present status of the literature. It sought to determine whether different modes of presentation of stimuli (visual and auditory) in ninth-grade pupils would result, as hypothesized, in differences in factor structure when analyses were derived from auditorially versus visually presented materials and in forward versus backward response tasks. The study further sought to determine whether any resulting differences would be reflec ted in special learning problems or aptitudes of ninth- grade pupils. Hypotheses were generated concerning the following questions: 1. Do the visual and auditory perceptual modes differ on factor-analyzed tests of memory for each for this age group? 2. Does order (forward or backward) required in reproduction of immediate-memory-span abilities make a difference when the test results are factor analyzed? 45 46 3. What specific relationships may be found between immediate-memory abilities and school achievement? 4. Can it be demonstrated that clinical groups found in any typical school population show critical differences in auditory and visual memory abilities? Organization of Study The study was organized into three parts, as follows: (1) The determination of memory factors as based on findings from a battery of tests, including memory tests, to ninth-grade children of average intelligence. (2) The study of school achievement correlates of different obtained span factors by identification of pupils having high and low factor scores. (3) The study of special sub-groups of ninth- grade pupils having specially identified learning problems or peculiarities of achievement, guided by hypothesized factor characteristics of these sub-groups. The study was limited to an investigation of ninth-grade pupils, the major factorial results based on those of average intelligence (IQ's 90 to 100 on the CTMM test), residing in the upper-lower and lower-middle class residential area of Hawthorne high-school district in 47 the Los Angeles area of Southern California. All of the subjects were Caucasian. The specially identified groups of Part Three of the study (as described in [3] above) were taken from the same grade in the same schools but were pupils other than the ones for which the factor analy sis had been made. They consisted of identified cases having particular academic peculiarities or special prob lems, including placement in special classes for the "educationally handicapped." The study did not inquire deeply into clinical features of these identified groups, but merely employed those items which were a matter of record. Limitations.--Rappaport (1945), Harris (1961), and Wechsler (1950), among others, view emotions as prime in hibitors of perception and memory, whereas Festinger (1957) and Fitzgerald and Ausubel (1963) hold that emotion does not play a significant role in the exercise of cogni tive power and that the reason cognition is sometimes impaired is that incoming information is at variance with information at hand and is therefore not selected for assimilation (or that there is no information at hand with which incoming information may be related and so it is not selected). The lack of definitive information on the role of emotion and affect on learning is such that this study will not encompass a review of aspects of emotional 48 maladjustment. Since both Wechsler and Rappaport based their hy potheses on the WAIS subtests, and Hopkins and Michael (1961) found the reliability of difference scores between subtests and scale score differences on the WISC not to be large enough to reject the null hypothesis of random vari ation at the .05 level, this writer concurs with his findings that: "before offering diagnostic interpretations of subtest scores, one should be reasonably sure that there is something other than chance variation to interpret" (pp. 116-117). This study was not concerned with the role of emotions or affect on immediate memory. To do so might have raised more questions than were within the scope of this paper. Fatigue, although not directly investigated, was carefully controlled for, just as obvious (if any) emotional reactions were; it was hoped that situational anxiety was kept at a minimum. The students were familiar with the counselling procedures in the high school in which this study was conducted; also, they had been alerted prior to the testing that at some time in the future they were to be called into the office for some short tests, the results of which would not affect their grades and would be given to them at a later date. Since it was the encoding aspect of immediate mem ory with which this study was primarily concerned, with 49 the exception of the DAT-Spelling Test only encoding re sponses were measured. Specifically designed into the experiment was the avoidance of time for rehearsal because in the real world of the child's life, when auditory instructions or lec tures are given, customarily the classroom procedure flows without deliberation time to see whether (or test whether) communications have been received. The items measuring visual short-term memory were also controlled for time allowed, so that no rehearsal took place. Each word or digit was presented for one second and removed. Visual immediate memory of necessity depends upon recognition as well as recall, and since the visual mode cannot be represented vocally, common three-letter words were presented, controlled for vowel sound, consonant placement, meaning, and time of exposure. This study delineated between short-term memory (immediate) and long-term memory, not because the writer supports a dichotomy in theorized memory but because fac tor analysis shows measurable differences. Only those correlates of the measurable differences which related to school achievement were investigated. Definitions.--Measurements of immediate memory based upon retrieval spans of auditory and visual stimuli so were investigated within the framework of factor analysis, particularly as memory is defined within the structure-of- intellect model hypothesized by Guilford (1966). The following definitions were used as operational parameters for the study: Factor-Analytic Model. Guilford (1966) has hypo thesized a model of intelligence in which Memory has been demonstrated to require different ability for performance than other operations involved in intellectual function ing. The figure below, Figure 3, shows the schematic. Following his definition of symbolic contents, this study employed digits only. Following his definition of seman tic contents, this study employed only three-letter words. Extracting from Figure 3, the definition for Memory is the ability to demonstrate how much you can remember, given a standard exposure to information (Guilford, 1966). Memory for symbols is defined and used herein as the recall of unrelated digits. Span for digits is the longest series recalled by a subject in correct order with a maximum of two trials. Memory for semantics is defined and used herein as the recall of a series of unrelated three-letter words where the series was controlled for vowel and consonant sounds so that no meaningfulness could be attached within the list and so that a minimum of interference would occur due to similarity of vowel or consonant sounds. Units Classes Re l a t i o n s Systems Transformations Implic atio ns c, . \ O in Fig. 3.--Theoretical model for the complete "Structure of Intellect." PRODU C TS 52 Span for semantics is defined and used herein as the longest series of three-letter words recalled by a subject in correct order with a maximum of two trials. Auditory input are those items which are exposed to the subject vocally. Visual input are those items which are exposed to the subject through visual presentation. Forward order refers to the encoding response where the subject repeats the digits or word series in the order presented. Backward order refers to the encoding response where the subject repeats the digits or word series in reverse order as presented. As explained in Chapter I, there is a theoretical argument about whether Memory as a construct is composed of two processes or whether Memory as a construct is one total process. For the purpose of this study that characteristic of memory which is schematicized as being related to the psychological present (PM, Fig. 4) (James, 1890 in Waugh 1965) was explored, namely, Immediate Memory. Immediate memory is that recall from the Primary Memory System as postulated by Waugh and Norman (1965) (see diagram), where memory depends only on the interval and remains invariant with changes which might take place in Secondary intervals. The model is based upon James' 53 definitions of primary memory as an event (PM), which has never left consciousness and is part of the psychological present, while an event recalled from secondary memory has been absent from consciousness and belongs to the psycho logical past (James, W., Principles of Psychology, Vol. I, 1890) . SECONDARY MEMORY PRIMARY MEMORY Rehearsal STIMULUS Forgotf ten Fig. 4.--Schematic showing of dual processes of memory storage (Waugh § Norman, 1965). However, it is necessary to point out that imme diate memory tests when factored tended to separate into forward immediate memory and backward immediate memory. This latter, backward immediate memory, was termed Pro cessing Memory since the subject evidently had to process backward the order given, although the recall was of the psychological present. Immediate Memory thus was defined specifically as that ability to recall digits or words presented single for one second exposure in correct order asked for forward presentation, in the psychological present. Processing Memory was defined as that ability to recall digits or words presented singly for one second exposure with correct backward order asked for after having 54 been presented in forward order. It was still considered to be part of the psychological present. Sample Description for the First and Second of Three Farts of this Study Ninety boys aged thirteen to fifteen who are Anglo-Saxon, of upper-lower and lower-middle socioeconomic status, were selected from the normal range of IQ scores on the CTMM (90-110) (1957 form, California Test Bureau). They were ninth-grade pupils who were not placed in any special classes. Thus, boys who scored either over 110 or under 90 and boys who were transferees for whom there was no CTMM data were also excluded. Because this high school serves a low socioeconomic area, the total number of boys out of approximately 200 who met prescribed qualifications was ninety; a large portion of the population scored below 90 and a few above 110. The issue of transiency (and consequent mortality for the study) was reviewed with the vice-principal. The total battery of tests demanded three different testings and the large number being individually tested meant a time interval of three weeks to a month including a week or more between testing for the first two parts of the study. The vice-principal and her counselors indicated that the majority of transients occurred among the lower IQ scorers. This proved to be substantially correct, for 55 of the 91 boys tested in Part One only one student left, leaving data for 90 rather than 91. CHAPTER III PROCEDURES AND FINDINGS, PARTS ONE AND TWO This initial phase of the study developed a com plex battery of auditory and visual immediate memory tests and factor analyzed the results. These results were then employed in Part Two as potential indicators of academic variability. Clinical groups were studied in Part Three. This portion covers Part One, the way by which the major Memory hypotheses were implemented by test battery and factor analysis, together with the results of the analysis. PART ONE Hypotheses for Part One A review of the major experimentation in memory has shown that there is a need for questioning some of the basic assumptions underlying studies on memory. It appears that people differ in having good memory in different functions and for different purposes. Both factor analysis and learning theory suggest variability among individuals. Therefore, the following hypotheses were tested by factor- analytic procedure. 56 57 Hypothesis 1.1. Immediate memory is not a single ability but yields different factors according to sensory input channels and also according to whether a simple reproduc tion is required as in forward spans or whether a processing is demanded as in backward spans. Four factors are hypothesized. Hypothesis 1.2. Auditory or visual input yields dif ferent factors. Hypothesis 1.3. Forward or backward order yields dif ferent factors. The Test Battery These plans and hypotheses thus called for suffi cient auditory and visual immediate-memory tests to deter mine whether separate functions were measured, and whether different processes of responses were involved for forward and backward tests, the former showing instant, non processed reflection of the immediate memory input, the latter showing an internal manipulation preceding overt response. The tests serving these hypothetical functions are indicated in Table 11 in the Appendix and are briefly described below. In order to determine whether pupil competence on memory tests was a separate function at all from the other 58 types of ability function, it was of course necessary to use from the records a larger battery of tests with which the memory tests were mixed for the factor matrix. Such would provide a basis for distinguishing separate memory factors, if indeed they would segregate. Thus the entire battery of eight memory tests was administered to the sub jects, and their scores used on twelve other tests as described below are overviewed here. Description of the Tests Revised for Memory Factors There were four tests of auditory memory and four tests of visual memory. Each visual test was as nearly identical to the auditory as possible. Forward Order.--For example, for each auditory test in which digits were used for forward responses in exact order, its identical visual test consisted of digits to be reproduced in exact order vocally by the subject. On auditory tests, digits were presented vocally by the examiner one per second; on visual tests, digits were pre sented one at a time on a card and exposed for one second. The digits were controlled for relationships such as 2, 4, 6, etc., so that no relationship could be imposed. The second auditory test in the series (matched with a visual test as described above) consisted of three- letter words as stimuli. Each three-letter word in a 59 series (series began with three words and went to nine words, two trials per series) was placed so that a minimum of interference would occur due to similarity of consonant or vowel sounds. The words were also controlled for mean ingfulness, in order to approximate as closely as possible the same lack of meaningful relationships as was designed for in the digits tests. It was necessary to control words for placement so that any words with plural meanings would not take on a meaning unsuspected by the examiner (exam ple: ewe) . Backward Order.--The last two tests in each series of the four modal tests were as much as possible identical to the first two tests, with the exception that a backward order was required. Two trials for each series, with one second exposure, was the same as that required for the forward tests. Thus there were two contents; digits and words, each used with forward and backward order and in visual and auditory modes. Many considerations were involved in the determi nation of the order in which the eight memory tests were administered. One was whether content practice effect would carry over from auditory to visual, and another was whether sets would form because similar tests were given in succession, thus overdetermining factors. Pretesting had determined an order which would eventuate in the least practice effects and would have the least effect in 60 producing artifacts. The tests are shown in the Appendix (Table 11). For one-quarter of the subjects the order was 1. auditory digits forward 2. auditory words forward 3. auditory digits backward 4. auditory words backward 5. visual digits forward 6. visual words forward 7. visual digits backward 8. visual words backward For another quarter, the four visual tests pre ceded the auditory tests. The auditory and visual forms were alternated; for example, if the preceding subject received the auditory tests first, the next subject re ceived the visual tests first. For the other half of the sample, one-quarter re ceived 1. auditory digits backward 2. auditory words backward 3. auditory digits forward 4. auditory words forward 5. visual digits backward 6. visual words backward 7. visual digits forward 8. visual words forward 61 and the other one-quarter received the visual forms first, alternately as described above. Use of Additional Tests.--In order to gain as much data as possible on the examinees, to use with the memory tests for their better location in factor space, the fol lowing test results were taken from their records: CTMM (California Test of Mental Maturity, 19S7, Junior High Form), administered in October, 1965; CAT (California A- chievement Tests, Reading and Arithmetic), administered in October, 1964, while they were in the eighth grade; Nelson Silent Reading Test, administered in January, 1966; DAT (The Differential Aptitudes Test), for verbal, numerical, abstract, spatial orientation, mechanical aptitude, speed and accuracy, spelling, and language-sentence usage, ad ministered in October of 1965. The employment of records secured at different times is not defended here; it was decided that it was better to use what information was available to increase the spectrum of abilities for the factor matrix in order to give the memory tests a broader basis for interpretation. Subjects For purposes of the factor structure, 90 subjects, all boys, were drawn from the ninth-grade classes at one of the larger high schools in the Centinela Valley Unified High School District. The school had a total population 62 of 2,200. Only boys were included, for the applications of the obtained factor scores were to be made almost exclu sively to those subjects giving rise to most of the clini cal interest of the study--top achievers, academic fail ures, disturbed and incorrigible boys, and students in special classes. Criteria for Selection.--The group intelligence score range of 90-110 was chosen for the purpose of ascer taining and identifying individual differences which exist within the so-called "normal" range. Community.--The school serves a community almost entirely composed of people of lower-middle and upper- lower socioeconomic levels; it is considered generally to have a very homogeneous school population. Its population is almost wholly Caucasian. Procedure of Testing.--A list was made of the 90 boys who met the prescribed qualifications described in the section entitled Criteria for Selection. This list was given to the secretary of the ninth-grade offices and she scheduled, by means of hall passes, those boys in the sam ple according to their availability. Since the high school actually ran on two sessions: early classes began at seven-thirty in the morning with an early lunch period and late classes began at eight-thirty in the morning with 63 a late lunch period, this meant that scheduling was a fac tor which, in essence, randomized the order of boys called. All 90 subjects were tested with the battery of eight memory tests. Fifteen to 20 minutes were allowed per sub ject and all tests were given in one session. Variables in the Factor Matrix.--Data for the anal ysis included the eight memory tests described above, and the DAT Verbal, Numerical, Abstract, Space Orientation, Mechanics, Speed and Accuracy, Language (Sentence-Usage), CTMM (1957 form J) total score, CAT Reading, CAT Arithme tic (1957, Junior High Level), Nelson Silent Reading Scores (form W) from the high-school reading laboratory, birth date information, modal order of testing, and date of testing were placed in score sheets (see Appendix). Also put into the matrix for the analysis was the following information taken from the records: favorite subject, English grade, mathematics grade, and teacher evaluation comment (on a rating scale from 0 to 9) for each. When all data were manuscripted, a total of 27 variables were submitted for the factor analysis. Comparisons of Obtained Spans Table 1 shows the spans obtained by subjects on each memory test. A t-test was computed for the identical auditory and visual forms of each test, and no matched modal tests differed significantly from each other. The TABLE 1 DISTRIBUTION OF SPANS MADE BY THE 90 SUBJECTS ON THE VARIOUS SPAN TESTS Tests 0 1 2 3 4 5 6 7 8 9 10 11 Mean S.D Auditory Digits Forward .. 6 22 15 30 13 2 2 .. 6.4 1.3 Auditory Words Forward • • • • * * 20 21 26 23 1.1 Auditory Digits Backward .. 5 19 41 11 11 3 .. 4.1 1.2 Auditory Words Backward .. 3 40 34 12 1 .8 Visual Digits Forward 1 1 10 27 31 13 7 .. .. 6.7 1.2 Visual Words Forward 1 .. 4 33 31 17 4 .9 Visual Digits Backward • • • « 4 17 23 28 14 4 1.2 Visual Words Backward 1 1 26 32 26 4 .. 4.1 .9 STANDARD ERRORS, DIFFERENCES BETWEEN MEANS, AND CRITICAL RATIOS Tests DM C.R. Auditory Digits Forward and Visual Digits Forward 1.8 .3 .16* Auditory Words Forward and Visual Words Forward 1.4 .7 .5 Auditory Digits Backward and Visual Digits Backward 1.7 1.4 .6 Auditory Words Backward and Visual Words Backward 1.2 .5 .4 *The obtained differences are not significant. 65 purpose here was to explore the success of matching for stimuli, and no further discussion is required. Reliabilities Table 2, which lists the variables employed in the factor analysis, also lists the column h? or communality, which can be taken as a lower-bound estimate of the relia bility since it is the sum of all the common factor vari ance of the test demonstrated in the present study, and thus is a value below which the reliability for the present study cannot be. The listed communalities are not those of the factor analysis employed (which for factor clarity was based, as discussed later, on an eight-factor solu tion) , but rather were based on the complete factor ex traction. The reliability information thus presented is not here discussed; it would not have been practical to in crease the values for those tests with low reliabilities. Only enough reliability was required to yield a factor structure of use, and this consequence did occur. K-R 21 estimates of reliability were calculated for the eight memory tests and are also listed. Factor Analyses First Eight-Factor Rotation The means, standard deviations, and 66 TABLE 2 COMMUNALITIES FOR 27 VARIABLES AND COMPUTED RELIABILITIES FOR MEMORY TESTS Variables h2 Reliabilities K-R 21 1. CTMM .59 • • 2. CAT Reading .79 • • 3. CAT Arithmetic .64 • • 4. Nelson Reading .74 • • 5. Auditory Digits Forward .74 .78 6. Auditory Words Forward .58 .63 7. Auditory Digits Backward .63 .64 8. Auditory Words Backward .46 .51 9. Visual Digits Forward .39 .45 10. Visual Words Forward .57 .62 11. Visual Digits Backward .39 .43 12. Visual Words Backward .50 .63 13. Birthdate .47 14. Examination Sequence .37 15. Best Subject (coded 1-9) .25 16. English Grade .70 17. English Evaluation .44 18. Mathematics Grade .76 19. Mathematics Evaluation .44 20. DAT Verbal .66 21. DAT Numerical .68 22. DAT Abstract .47 23. DAT Spatial Orientation .61 24. DAT Mechanical .47 25. DAT Speed and Accuracy .46 26. DAT Spelling .57 27. DAT Sentence-Language .58 67 intercorrelations of the 27 variables are presented in the Appendix in Table 12. This matrix was submitted to factor analysis, with employment of a Philco Q37 facility, by courtesy of System Development Corporation, Santa Monica, California. One product of the first runs was the commu- nality (h^) data presented in Table 2. The analyses were then carried on in the following manner: Factors were extracted by the principal-axis factor method and were rotated to the orthogonal varimax criterion. Examination of these factor loadings of the 27 variables suggested that Chronological Age and position in testing were resulting in an artifact in the factor analysis (see Table 13). Discussion of Findings from the First Eight-Factor Rotation.--The variables and their loadings are shown in Table 13. As is frequently found in factor analyses of school achievement and aptitude measures, a major factor emerged on which were clustered the following variables: CTMM, CAT-Arithmetic, CAT-Reading, Nelson Silent Reading Test, DAT Verbal, DAT Spelling, and DAT Language-Sentence Usage (7 of the 27 variables). This factor was tenta tively identified as a general, educational ability for taking verbal pencil-and-paper tests. Factor Two was cleanly composed of Auditory Backward tests. Factor Three was formed by the DAT Abstract, DAT Spatial Orientation tests, and the DAT Mechanical tests, seemingly identifying 68 an ability factor for visualizing figures oriented in space. Factor Four was identified as a Studiousness fac tor since English Grades and Arithmetic Grades with their evaluational comments from teachers loaded most heavily on it. Factor Five showed both Visual Backward and Visual Forward tests for words loading on it. Factor Six showed loadings from variables such as chronological age and position in the examination group; thus it was called an artificial factor. Factor Seven showed low loadings with the exception of a tentative spatial-orientation and cleri cal factor variable; and Factor Eight was composed of the Auditory Forward variables. Since two of the 27 variables formed an artificial factor, it was decided to remove these two variables (chro nological age and position in the examination group accord ing to order) and to resubmit the remaining 25 variables for intercorrelations, means, and standard deviations in the following manner: The factors were extracted by the principal-axis factor method and rotated to orthogonal varimax criterion for four factors and then for six factors. The purpose was expressly to gain information about the first splitting off of the memory tests. On the four-factor rotation the memory tests clustered together and split off to form Factor Two. Table 14 in the Appendix shows the four- factor rotation matrix. Tables 15 and 16 give the 69 resulting information for the six-factor rotations. In this rotation the auditory and visual memory tests split off saturating two factors. These interim rotations made it possible to ascertain the fruitfulness of the hypotheses so the variables were resubmitted for the final eight- factor rotation. The means, standard deviations, and reliabilities are shown in Table 3. The analyses of this matrix were then carried on as before: Factors were extracted by the principal-axis factor method and were rotated to the orthogonal varimax criterion. Discussion for Final Eight-Factor Rotation for 25 Variables The data, when resubmitted for eight-factor rota tion, loaded as shown in Table 4. The resulting informa tion was used for the rest of the study. On this eight- factor rotation, the auditory memory tests split into forward and backward components. The visual backward tests stayed together, claiming the visual words forward tests; it seemed likely that the use of words in the for ward visual memory tests, because there was an attachment of meaning, tended to define an ability which demanded a type of storage for retrieval similar to that for backward visual order. Factor One continued to be defined by paper-and- pencil group-test-taking ability verbal skills, as 70 M E 3 Variable 1 2 3 4 5 6 7 3 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1. Cffi 60 54 48 10 08 10 14 22 24 13 01 21 45 -07 45 -24 57 53 33 22 14 37 37 52 2, Oil Reading 62 79 19 14 19 22 29 28 23 13 23 49 05 39 -26 65 44 15 15 06 34 48 53 3. CAT Arithmetic 53 15 C5 20 13 22 19 27 15 29 49 -17 57 -35 50 66 27 20 08 47 45 53 A, Kelson Reading 21 17 19 25 30 27 26 14 18 40 OS 33 -13 56 35 13 16 -01 41 57 49 5, Auditory Digits Forward 63 44 49 23 34 22 37 26 11 19 03 06 33 07 10 29 08 -02 00 12 6, Auditory Words Forward 32 40 21 26 24 28 24 13 25 -03 15 30 01 03 10 04 00 05 11 7, Auditory Digits Backward 51 42 33 37 29 01 15 17 U -17 21 25 11 08 16 02 13 16 8, Auditory Words Backward 36 29 23 30 27 08 10 03 -06 19 07 08 05 08 07 15 25 3, Visual Digits Forward 30 16 25 10 19 07 20 00 17 21 -06 08 -06 13 22 25 10, Visual Digits Backward 43 52 15 27 14 31 -05 30 21 14 03 -06 07 30 36 11, Visual Digits Backward 41 13 if t . 0 10 19 -03 29 18 05 -08 -05 12 22 28 12. Visual Words Backward IE 27 03 12 -04 19 13 06 03 -12 07 25 15 13, Best aubject 24 05 ir' Oj -17 29 35 25 10 19 14 21 26 14, English Grade V i f - C;) -41 1C 57 22 22 13 36 51 36 1$, English Evaluation -29 37 CC -24 -04 •13 -02 04 -11 -13 16, Mathematics Grade J 47 67 43 41 24 40 38 52 17. Eathenatics Evaluation •19 -39 -10 -10 -13 -14 -27 -32 18, DAI Verbal 46 33 25 26 20 44 56 19, DAT Hunericai V, s"\ 35 30 37 33 45 20. DAI Abstract 42 37 21 20 34 21, DAI dpatiai Orientation 45 25 -06 12 22, DAT liechanical 04 -02 20 23, DAI Speed and accuracy 35 31 24. DAI spelling 43 25, DAI Bentence-ijanguaje Means 39.8 7,9 8.2 7,6 o,4 5,6 4,1 3,8 .7 -.9 .5 ,1 3,2 1.7 4.0 J 3.3 32 4 509 4r; 394 374 3i 254 Standard Deviations i,0 16,6 14.4 17.8 J 1,1 1 2 ,3 .2 ,971 ,2 .98 2 Q I ,32 2,7 ,99 2.5 22.1 29.5 25.1 26.8 24.8 25,1 25,6 20,5 Note,— Decinial points omitted. ;«refers to percentile, FINAL EIGHT-FACTOR ROTATION FOR 2 5 11 H H H > H O i O I / H C O O H O U ) l A N I 9 N C l ^ i f l t ' > N ® O O H l A t ' > O O H O O O N H N O H O M O O O H Q H O t O H O r lO H M > ®«0*MNIOHr'H*«OOCO'OiAf'>HOHt'>*NHlOOl O O M H O O O O H O O O O N O N O H M O N O iA H O H HNN<tmisrlHOO>HlAOtfNOH\ONNtO['>«»0«0 > OH O rirs^ N flH O rlN N H H H rlN O O N O O O O W w f t « < H a i > H ♦ ♦ ®Ol HHOOOOOlMONIflO>CO(00®OK)OWl/IHlfl O O O O N N N H H i O l f l O H N O N O f t r l H H H O I O N <fOOO^ONNIfllONHOO^IOO®OlflfONMOO® HO M O HH O OO OOO HI flO ifl WH ^O HOO NH H M H |'»Hr**lA<40Hf'N*O*NNlAONOrtNr<Ht'>«0 N O H O H O O O O O O O N H O ^ H t ' W O 'O ' O H O N H O H H H ( O H O U ) i / ) N H H H O H O H O H O O O O O H lA O H N ' O C O U W I O N O t f W N H O O H O O O C O O ' t O C O H \ O c O ' O M O O O H N N N O H < f O l O N O ^ H O C M l f l ' 0 d > d l Vd V - d ( 4 h - d skirt'd Hd Sd^it-ddti G* uSdf M ft o ta d N 2 U iS < f t dONU « i / i a, ca n Ubovnvn ft M < r i fiH fl.H 'd n n s 0 • H V fi d u 3 u U < d c 9 d d V odd s # o o H 'd H o a n v d 9 H i/i n «H > M H 3 0 ) 30ft H 0 9* vh ooG oov v nv n v d n dv UHMd n o u 'd 'H O 'H o , H 'd , H , duddnn uuHHdBd Cgd22Q2MfMVl)f>U0H‘ HdflBi.)'HC < H £ ( V H O 'H O riU ffl'H 'H 9 H G 'H d’ H H 0 uvBimNasai.0 vvftdvvjjMv a t< H ffff dftftdd^indUddC dffiO O O O iH H H H W nnSBdd^ftdH ftd H<ovvvvdddd * h * h o )«>>2<w 2u w w s n'H iH 'H H ddddV H H jd^ f h H H d d d d n n « «n oooovvHHHHHHHH H << 0 3 3 3 3 'H 'H H 'H dSfidd<«t<<<<«<< uuuz<<<o>>>««w22C3 3aaaaaQ d d v V H 6 0 n ¥ 6 « r l 0 a I I 0 4 ♦ 0 0 4 1(440 H H H H H H r l H H r t N N N N N N 72 exemplified by those necessary for the CTMM, CAT Reading, CAT Arithmetic, Nelson Silent Reading Test, DAT Verbal, DAT Spelling, and DAT Language-Sentence Usage. Factor Two was defined primarily by the auditory backward tests and carried with it the visual digits for ward tests. Since each card was exposed singly for one second, there is the probability that at this age the sub ject is more accustomed to visual spanning than he is to exact reproducing of auditory sounds, and thus may have been able to scan more than once so that a type of re hearsal may have occurred, causing this test to simulate a processing function and to cluster with processing abili ties found in backward auditory digits and words. The unusualness of the task itself may also have lent a quality to the visual test such that it clustered with the auditory processing tasks. Questions may be raised about the practice of naming a factor a backward factor when it loads on a for ward test as well, but the identification of it as such is defended because without question, even if the test for visual digits forward were removed, the factor would not be removed. Since this finding replicates previous findings reviewed in Chapter I, and since the findings also fit theory, there is justification for the title. Factor Three was defined by the Mathematics Grade .45, DAT Numerical .40, DAT Abstract .63, DAT Spatial 73 Orientation tests .62, and DAT Mechanical .62. The factor seemed to encompass an ability to comprehend figural com ponents in space. Factor Four included aspects of behavioral judg ment, or perhaps a personality trait such as studiousness or an attending type of behavior, since English Grades loaded .54 and English Evaluative comment about grades loaded .63; Mathematics Grades loaded .50, and Mathematics comment .59. Both measured performance in subjects and evaluations loaded together. Factor Five was defined by the Visual Backward tests plus the Visual Words Forward. As with Factor Two, if the forward test were removed, a loublet factor would remain. Interestingly, the second highest loading oc curred with DAT Spelling .31. Factor Six was defined by the auditory forward variables. This result indicated that forward digit spans required a type of storage which differed, as tested by retrieval, from that where backward order is demanded, i.e., the kind of processing necessary for immediate re production when forward order is demanded is different from the kind of processing necessary for a temporary holding in mind when a backward order is demanded. That this factor was named an auditory forward factor is justified on the basis of previous work cited in the review of the litera ture even though its identification was based on a doublet 74 factor. Factor Seven claimed DAT Clerical Speed and Accu racy, with Factor Eight indicating, as would have been expected, a ceasing of or limiting of factors. It looked as though the DAT might begin to pull apart on a factor rotation greater than eight, since it was originally a factor test. The fact that it loaded heavily on Factor One was, or may have been, due to influence from group paper-and-pencil test-taking ability. Hypothesized Factor Results.--The results of the final eight-factor rotation (Table 4), when examined, indicated that the auditory and visual modal inputs were not identical and evidently demanded separate abilities for their performance. Factors Two and Six, which carried the auditory tests, carried loadings from achievement tests which were different from those achievement tests loading on Factor Five (the visual factor). Specifically, Factor Two carried the arithmetic tests; Factor Six carried the read ing comprehension tests, and Factor Five carried achieve ment tests for spelling. Tentative support was given to the conclusion that immediate memory, whether shown as a deficiency or an aptitude, is variously and differentially related to school functioning. Summary 75 Part One of Chapter III has outlined the first part of the study where hypotheses concerning differentiated memory abilities were tested and through factor-analytic rotations were confirmed. These findings led to close examination of those variables which loaded on the three memory factors pertinent to the study: Factor Two (Audi tory Backward Memory), Factor Five (Visual Backward Mem ory) , and Factor Six (Auditory Forward Memory). Visual forward tests did not form a separate factor. The examination uncovered variables which loaded on one of the pertinent factors exclusive of each of the other two, respectively, and led to the generation of new hypotheses concerning the nature of immediate memory as it related to school achievement. The assumption underlying many earlier experiments in memory, that memory is a molar ability, was questioned and found to be untenable. Hypothesis 1.1. That immediate memory is not a single ability was demonstrated when three memory factors emerged through factor analysis. Hypothesis 1.2. The 4 auditory and 4 visual tests yielded three separate factors. Two were auditory and one was visual. Hypothesis 1.3. Order rather than content determined separation of the factors, i.e., the 76 backward tests formed two factors-- auditory and visual. The forward order formed only one factor; it was auditory. PART TWO Introduction Obtaining clear factors of memory functions, it was possible to pursue the next two steps of the study. The second step employed factor scores and is entitled Part Two. Here pupils who scored high and low in the separate memory factors were studied as to academic attributes pre dicted from their memory status. This is discussed in the remaining portion of this chapter. The other step (Part Three) is taken up in the subsequent chapter, in which a study was made of identified clinical groups of pupils not included in the above factor study who were given the mem ory tests for the purpose of verifying hypotheses about their nature as special pupils. When the results of the final eight-factor rota tion showed, as hypothesized, that abilities for perform ing immediate memory tasks were different depending both upon the modal presentation and on whether recall was measured as a simple span or demanded a processing before recall, the next step consisted of scrutiny of the loadings on certain factors to determine which were indicators of 77 specific academic aptitudes for each factor. From thisf hypotheses were generated for each of the pertinent factors. To test the hypotheses, factor scores for all sub jects were examined. In order to get at as clear-cut differences as the data would allow, it was decided to select the ten highest and ten lowest scorers on each fac tor to determine the significance of being high or low. Table 17 in the Appendix lists the factor scores in stan dard deviation form for each of the individuals chosen for testing the newly generated hypotheses.* Factor Two, Auditory Backward Procedure for Generating Hypotheses Since Factor Two had three common loadings in vari ables which were not common to either Factor Five or Six, it seemed fruitful to hypothesize that the general auditory backward ability might be contributing to the ability to perform or not to perform well on CAT-Arithmetic tests, DAT Numerical tests, and DAT Language-Sentence Usage tests. The question of why the latter test loaded on Factor Two was answered tentatively by speculating that the ability involved in knowing or not knowing the correctness of a system of letters or numerals may have an auditory memory ^Table 18 shows individual standings on each of the three factors used in this part of the study. 78 component. If this were the case it seemed advisable, then, to take the top ten and low ten factor scorers and use them as a group on which to test the hypothesis that auditory memory of a processing storage type seems to underlie the ability to perform well or not in mathematics and in lan guage. Hypothesis 2.1. There is an ability to process several symbolic elements at a time in the per formance of tasks involving a system of letters or numerals which is predicated upon an auditory component. Those boys who obtained one of the highest ten or lowest ten Factor Two scores are respec tively high or low in arithmetic. To test Hypothesis 2.1 for Factor Two, the Wide Range Achievement Test (Jastak, 1946) for arithmetic was selected on the basis of its validity, both concurrent and congruent, and its "virtual elimination of chance vari ance (recall format)'1 (Hopkins, Dobson, 5 Oldridge, 1962, p. 791). Because arithmetic is taught as a lock-step disci pline where auditory instructions are heavily relied upon for learning, the hypothesis seemed a reasonable one. The ten lowest and ten highest Factor Two scorers were given this test. Lesser differences or no differences were hy pothesized for other school tests. Hypothesis 2.2. Those pupils who scored high or low on 79 Factor Two are respectively high or low on an individually administered test of language. It seemed that this hypothesized relationship would be fruitful since children learn to listen to language be fore speaking and reading (Russell, 1956). The DAT Lan guage- Sentence Usage test was selected for individual ad ministration to the boys selected on the basis of their Factor Two low and high scores. Lesser differences or no differences were hypothesized for other school tests. Factor Five, Visual Backward Hypotheses Factor Five was composed of the two Visual Backward tests plus Visual Words Forward, and was identified as a visual backward ability. The only other variable loading on this factor independently of other factors was the DAT Spelling. It seemed fruitful to speculate that those indi viduals contributing the ten highest and ten lowest scores on the individual factor scores would score respectively high and low on an individually administered spelling test. Hypothesis 2.3. Factor Five had only one educational variable loading independently upon it--spelling; therefore the ten boys who scored highest and the ten boys who scored lowest on that factor are 80 respectively high and low in spelling. It was decided to test the relationship between visual memory components and spelling with the Jastak Wide Range Achievement Test (WRAT) for spelling. Since this test may be considered to be an encoding test, it also seemed interesting to explore whether there would be simi lar scoring ranges for individuals on a decoding type of spelling test. The DAT Spelling test, if administered individually, would fit the hypothesized specification because it required that the student decode spelling words for correctness. Lesser differences or no differ ences were hypothesized for other school tests. Factor Six Procedure for Generating Hypotheses Factor Six was identified as a forward auditory memory-span ability and seemed more related to an immediate retrieval ability. Since English Grades, the Verbal DAT, and DAT Space Orientation loaded independently on this factor, it was reasoned that perhaps this factor repre sented a behavioral attribute of immediate comprehension, or an ability to attend to details, which was necessary for achievement in verbal areas. Hypothesis 2.4. Those boys scoring highest and lowest on Factor Six are respectively high 81 and low on tests of immediate compre hension in reading. For Factor Six (auditory forward immediate memory) the Nelson Silent Reading Test for Comprehension of Para graphs was selected for individual administration (see Appendix for sample). Lesser differences or no differ ences were hypothesized for other school tests. A total of 42 boys composed the sample for testing factor hypotheses (see Table 18). Logistics All 42 boys were tested with the entire battery. As in the Part One testing, the list of names was given to the secretary for scheduling according to availability. Testing time was approximately one hour for each pupil; it was possible to test two boys at a time because the room was such a large one that tasks requiring individualized administration could be alternated between the subjects. Testing was completed within a week. Each boy was told he had been chosen from the first group where nearly all of the freshman boys had been tested because he had made scores which were exceptional. Each boy was also told that he was being given more tests this time to see how his memory tests related to his school achievement and that his scores would be available for his or his parents' perusal within three weeks. They were 82 assured that the tests were untimed and they could take as long as they wanted to finish the written tests. Control for guessing was carried out when each boy was instructed not to guess, to circle any item he was not sure of, so that guessed answers would not be included in his score, for if they were wrong they would lower his score. The data were submitted to several statistical analyses. Primary among these were chi squares and Fisher exact probabilities. The intercorrelations, means, and standard deviations of tests in this battery are shown in Table 5. Reliabilities of Tests Devised for the Battery for Part two Table 6 shows the computed and published relia bility coefficients for the tests used to determine the hypothesized relationships with Factors Two, Five, and Six. Since all tests were power tests, reliability coef ficients were computed in accordance with the harmony of the assumptions made for using the formula and the specific test. Guessing was controlled for as much as possible and the raw scores demonstrated that there were adequate floors and ceilings. A condition of the design prohibited the use of formulae for test-retest or alternate forms. It was hoped that the tests chosen were as homogeneous as was possible and that the fact that they were individually administered compensated for the lower number in the TABLE 5 CORRELATION COEFFICIENTS, MEANS, AND STANDARD DEVIATIONS OF TEST BATTERY FOR PART TWO USING FACTORS II, V, AND VI AS PREDICTOR VARIABLES (N=42) Variable 1 2 3 4 5 6 7 8 1. WRAT Spelling 72 62 28 78 62 [ 51 32 2. WRAT Arithmetic 50 29 60 76 23 04 3. DAT Sentence-Language 13 69 57 47 24 4. Nelson Reading Comprehension 45 19 00 27 5. DAT Spelling 65 54 31 6. Factor II 25 20 7. Factor V 35 8. Factor VI Means 54.1 40.7 23.4 17.1 29.6 2.0 2.0 2.0 Standard Deviations 20.3 16.0 12.1 2.6 10.2 .7 .7 .7 Note.--Decimal points omitted. TABLE 6 RELIABILITY COEFFICIENTS FOR TESTS USED IN PART TWO OF STUDY Test N Computed rtt Communality Published Jastek WRAT-Arithmetic 20 . 86a .84 Not Available Jastek WRAT-Spelling 20 . 95a .95 Not Available DAT-Language-Sentence Usage 20 . 78a .77 T3 o * 1 O 00 • DAT-Spelling 20 • 00 00 o .87 .80-.90c Nelson-Denny Reading Comprehension 20 . 86b .86 . 90b aKuder-Richardson Formula 20. ^Kuder-Richardson Formula 21. cSpearman-Brown, odd even split half. ^Buros (1959). 85 samples. The reliabilities were satisfactorily high. Discussion of Hypotheses for Factor Two It was hypothesized that the general backward memory ability, Factor Two, was a contributory factor to the ability to perform on individually administered DAT Sentence-Usage tests and the WRAT Arithmetic tests. To test the hypothesized relationships, a chi square or Fisher exact probability was employed. The value of .05 was selected for rejection of null hypothesis. Hypothesis 2.1. The data in Table 7 show support for the hypothesized relationship between auditory backward memory and the ability to perform on individual arithmetic tasks. All low Factor Two performers scored below the mean for the entire group and all high Factor Two performers scored above the mean. The chi square gave a significance at the .001 level. Hypothesis 2.2. The hypothesis that the ability to do a processing type of auditory memory is much the same GUI sjU A ilL d O n i ' l o l U i i J . I X - — t ' J i lO .i.lo Jj' > 01 : ill- ; It/G A rlO W sH IP L i l k , vk tiO iiY FiiC'JOii XL-it~>j <wLi X U i-L -is O h s U l e X i T iiG iG iiCIIXij’ / i j } hi ±LJu iI-Laa— nji .i-.L jO'-.Ij j . i jaLiL.iii'* ca Liallja 1'iiG xO )ii iL Jl sC O X LG 1 ' + k v j iV * i C ■ .'V- ‘ ^ L ' •- I J (.iUiiitory :;c; • r J -i.ii (-■ J 0/ uov: (.; i .) -H iA jlL i J . J i - S ijL /O iL iliitl-J (•/’.r.u.l - uelraard) ip;; I.".) hdv (1 10) I’ A O 'iX Jit lix (a u d ito ry 1 hiph 10) 2C0KXRJ ’ oruard) Lov; (II 10 /auiT -Xr It im et ic C-/’ ri i ■ 7 'i p 42-99 10 a 3 7 Level of sig n ific a n c e n.. n.a. Lki-Lanpuape 0-24 3 10 3 7 sentences 25-oC 7 I — 0 7 3 Level of d ip n ifican ce * l V J t .01 n.a. '.vlLiT-opellinr' C-59 7 L u 3 9 60-99 p * 7 1 uQvel of s ijn ific a n c e . PU 3 .005 LLVl-soellinp 0-29 3 ji' - 3 5 6 30-60 , \ \ .j 1 • 5 4 Level of a ip n if Icance . 1 ■ L ^ .001 n . a. Lelaon-Jenny 0-17 / p 3 6 iieadinp P a ra - 16-22 L ;i i r ^ * i * • 1 7 4 pruph.; Level of sig n ific a n c e \' ■ * ‘ • ' * n . e . n.n. oo O' 87 kind of ability as that which under lies correctness of syntactical ability is also supported. This is seen in the data of Table 7. The DAT Sentence-Language scores were cut at the population mean to determine high and low performance for comparing high and low factor scorers. The obtained relationship was significant at the .002 level. Discussion of Hypothesis for Factor Five Factor Five was identified as a visual memory abi lity and it was hypothesized that those individuals con tributing the ten highest and the ten lowest scores on the individual factor scores would score respectively high and low on the WRAT for spelling. This test was looked upon as an encoding spelling test; it was decided to see also what relationship might be obtained on a decoding test such as the DAT-Spelling. The WRAT-Spelling correlated .51 with Factor Five (see Table 5); and the DAT-Spelling correlated .54. Hypothesis 2.3. As Table 7 shows, a Fisher significance at the .003 level gives support to the hypothesis that visual memory plays an important role in the ability to spell as measured on the encoding WRAT- Spelling test. Of the ten top performers 88 on Factor Five, nine scored above the mean on the WRAT for Spelling, whereas eight of the lower ten scorers tested below the mean. To ascertain whether a decoding spelling test would be significantly related to visual memory, a chi square was computed between Factor Five low and high per formance scores and low and high OAT-Spelling scores. The relationship found between these two variables, as shown in Table 7, was even more significant (.001), indicating perhaps that simple spelling tasks are even more related to visual memory than are more complex spelling tasks. Discussion of Hypothesis for Factor Six Factor Six was composed of the two auditory for ward tests. The writer assumed from the other variables loading on Factor Six independently that the ability to make good grades in English depended a great deal on com prehension in reading which in turn implied an attending ability. The test chosen to test this assumption was the silent paragraphs portion of the Nelson-Denny Reading Test; it correlated .27 with Factor Six (see Table 5). Hypothesis 2.4. A Fisher exact probabilities computation for relationships between Factor Six (as shown in Table 7) and the Nelson- Denny Reading Test indicates that the 89 obtained relationships do not support the hypothesis that there is a signifi cant relationship between attending type of behavior as a component in individual tests of immediate compre hension and auditory forward spans of memory. Summary Part Two of the study showed the following predic ted relationships between backward auditory memory and arithmetic and backward visual memory and spelling perfor mance; it was demonstrated that (1) arithmetic and one language ability and (2) spelling ability, are highly de pendent upon auditory backward and visual backward immedi ate memory, respectively. Auditory backward memory ability was found to be more critically related to school achievement. The concern of-Part Two has been with the employ ing of factor scores in order to determine the meaning of these scores as they relate to school achievement. Chap ter IV will be concerned with clinical groups wherein factored memory abilities are dependent upon clinical group classifications. CHAPTER IV MEMORY FACTOR SCORES FOR CLINICAL GROUPS This chapter covers hypotheses for and identifica tion procedures of clinical groups to whom the eight mem ory tests were given. A section is included which dis cusses mean factor scores and their derivations for each of six clinical groups. Findings are reported both statis tically and in graph form. Procedure for Identifying Samples In the selection of the sample of 90 boys for the first part of the study, none was used who was either in special counseling or in special classes. Before the sam ple of boys for this third part was selected, the counsel ors had been asked to submit lists of ninth graders who fit into the following categories: boys with behavior prob lems, boys in educationally handicapped classes, boys making A's in arithmetic only, boys making F's in arith metic only, boys making A's in English only, and boys making F's in English only. In the Part One sample, grades had not been a screening criterion, so it was necessary to have each 90 91 counselor check his list to be sure none had participated as a member of the original sample of 90. One of the boys who had been categorized as a member of the F in arithme tic group did have to be excluded. Another criterion of selection for membership in one of the clinical groups was that the student must have made three consecutive A's or F's in his freshman year in arithmetic or English while making average grades in other academic subjects. All names were placed on one list and given to the secretary who was to handle scheduling of the boys for testing. Group membership was not described on the list so that no contamination or bias would occur in the test ing. The secretary was informed that the testing proced ure for each boy would take fifteen or twenty minutes, the same amount of time as in the first testing session for Part One. Description of Clinical Samples Group One consisted of 13 boys with behavior prob lems who were not failing any academic subject; they are hereafter referred to as the Behavior Problems group. Group Two consisted of 11 "educationally handi capped" boys whose placement in the Educationally Handi capped program was dependent upon their having average or better Binet IQ scores and being three to five years retarded in school achievement on group measurements for 92 verbal and arithmetic skills. They were considered to be potential dropouts who had learning disabilities which could perhaps be remediated in the school setting; here after they are referred to as EH boys. Group Three consisted of 11 boys who made three consecutive A's in English but not in other academic sub jects; their mathematics grades were average. They are hereafter referred to as the A-English group. Group Four consisted of six boys who had three consecutive F's in English but were not failing in other academic subjects; their mathematics grades were average. Hereafter they are referred to as the F-English group. Group Five consisted of 13 boys who made three consecutive A's in mathematics but not in other academic subjects, and are hereafter referred to as the A- mathematics group. Group Six consisted of six boys who had three consecutive F's in uathematics but not in other academic subjects^ They are hereafter referred to as the F- mathematics group. The purpose for specifics concerning grade criteria was to ensure clear typologies. A total of 60 boys from the same school, same grade, were selected as having met the various group qualifications. They served as the sam ple in this part of the study. None scored below 90 on group tests of intelligence. A criticism may be raised 93 here about the small N in each group even though non- parametric statistical tests are available for determining significant differences between groups. The criteria for membership in any one group were very stringent, particu larly concerning grades. Most students tended to make similar grades in their academic subjects, so students who made three consecutive A's or F's in one academic subject and who also made average grades in other academic subjects were not commonly found in this school, thus the restric tions imposed a strictness on selectivity. Similarly, it was difficult to compile a list of boys who do not fail academic classes and who also have such serious behavior problems that they are, for the most part, considered to be incorrigible by legal and/or social standards. The EH boys were all of the ninth-grade boys who had previously been tested on the Stanford-Binet, LM, and placed in the EH classes. So this sample was de fined ab initio. Procedure for Testing Problems posed in a testing situation are many: scheduling around assemblies, faculty meetings, early or late arrival and departure, early and late lunch periods, sports, tests, absences and tardies; therefore the secretary was asked to send any boy on the list according to his availability. The members of groups one through six were 94 given the same eight auditory and visual, backward and forward order tests used in the first battery on the Part One sample, and testing was accomplished within five days. The order of tests was alternated as described in Chapter III. It was hoped that their test scores would predict characteristics of their group membership. Hypotheses This section organizes the hypothesized relation ships between various memory abilities and the clinical groups by group membership. The hypotheses were suggested in large part by the findings in Parts One and Two. Group One, Behavior Problems Hypothesis 3.1. It is hypothesized that their mean factor score approximations will not show deficiencies. Boys of the lower socioeconomic level of society, possibly mimicking the significant male adults in their environment, portray "machismo" or pseudo-masculinity which has to do with recklessness, "getting away with things," general lack of taking moral responsibility, and a low level of regard toward education. The hypothe sis is made because this is a group comprised of boys identified not for 95 cognitive but for behavioral reasons. There is no implication of defect of cognitive abilities. Group Two, Educationally Handicapped Hypothesis 3.2. It is hypothesized that the boys in the EH group are low in both auditory memory factors (Two and Six). Group Three, Students Who Make A's inTnglisTi --------------- Hypothesis 3.3. It is hypothesized that A-English boys are high in visual backward mem ory (Factor Five). This hypothesis and the following one, 3.4, are sug gested by the fact that Factor Five was saturated by the spelling tests and the fact that the highest and lowest factor scorers on this factor were found to be respectively best and poorest in spelling. Group Four, Students Who Make P ' s in English Hypothesis 3.4. These boys have low factor score approximations for visual backward memory (Factor Five) and for auditory forward memory (Factor Six) abilities. 96 Group Fivef Boys Who Make A's in Mathematics Hypothesis 3.5. It is hypothesized that the A-mathematics students are high on auditory-backward ability (Factor Two). This hypothesis and the following one, 3.6, are suggested by the fact that Factor Two was saturated by arithmetic tests and the fact that the highest and lowest factor scorers on this factor were found to be respectively best and poorest in arithmetic. Group Six, Boys Who Make Fvs in Mathematics Hypothesis 3.6. These boys are low in auditory-backward memory (Factor Two). Obtaining Factor Score Approximations The factor analysis in Part One showed that for each memory factor there were two or three highly saturated memory tests. Since these clinical groups could not be given the whole set of 25 variables, it was decided to secure factor score approximations limited to the high- loading tests for each of the three factors. A boy's raw scores on each of the two or three memory tests for each factor were converted to z scores and the mean of these were used as a sufficient approximation of his factor score. 97 Mean Factor Scores of Clinical Groups After individual mean factor score approximations were computed, a mean score for each group was computed. Figure 5 shows these mean factor score approximations by group. Table 8 shows statistical derivations for each of the memory tests administered to the six groups, and Table 9 shows the mean factor score approximations in z score form for each of the six clinical groups. Discussion of Mean Factor Score Approximations Figure 5 demonstrates unquestionably that the ability to hold unrelated items in mind before processing (auditory backward immediate memory, Factor Two) was criti cal to performance in tasks of a mathematical nature. The F-mathematics students had a negative mean z score and were second only to the EH group whereas the A-mathematics students had the highest mean score. On the auditory forward ability (Factor Six) the A-mathematics students and Behavior Problems made the highest mean scores in contrast with the EH, F-mathematics, and F-English, all of whom had negative mean z scores. The A-English group, on the other hand, were high est on visual backward (Factor Five), second to the Behav ior Problems, while the F-English and EH groups as well as the F-mathematics group had negative mean z scores on this ability. The greatest spread occurred between mean factor MEAN FACTOR SCORE APPROXIMATIONS ( z scores) 98 Figure 5 MEAN FACTOR SCORE APPROXIMATIONS FOR SIX CLINICAL GROUPS AUDITORY FORWARD AUDITORY BACKWARD VISUAL BACKWARD (Factor Six) (Factor Two) (Factor Five) 1.00 .90 .80 .70 .60 .50 .40 .30 .20 .10 - 00 - .10 - .20 - .30 - .40 - .50 - .60 - .70 - .80 - .90 - 1.00 - 1.10 - 1.20 -1 .3 0 -1 .4 0 -1 .5 0 -1 .6 0 - 1 .7 0 Legend: A-M— A students in mathematics F-M— F students in mathematics BP — Behavior Problems EH — Educationally Handicapped A-E— A students in English F-E— F students in English TABLE 8 MEANS, MEDIANS, STANDARD DEVIATIONS AND TESTS OF SIGNIFICANCE FOR THE MEMORY TESTS GIVEN TO SIXTY BOYS IN SIX CLINICAL GROUPS t-value Chi Tests Mean Median SD (Means) Square P Auditory Forward Digits 6.28 6.00 1.35 Auditory Forward Words 5.61 6.00 .94 4.53 43.21 .01 Auditory Backward Digits 4.48 4.00 1.21 Auditory Backword Words 3.80 4.00 .86 5.21 41.49 .01 Visual Forward Digits 6.42 6.50 1.24 Visual Forward Words 4.53 4.00 .99 13.41 26.71 n.s. Visual Backward Digits 5.31 5.00 1.17 Visual Backward Words 4.10 4.00 .98 8.14 37.30 .10 TABLE 9 MEAN FACTOR SCORE APPROXIMATIONS IN z SCORES FOR SIX CLINICAL GROUPS Auditory Auditory Visual Group Forward Backward Backward (Factor Six) (Factor Two) (Factor Five) A-students in Mathematics N-13 F-students in Mathematics N«6 Behavior Problems N*13 Educationally Handicapped N*ll A-students in English N-ll F-students in English N-66 .39 - .21 .94 -1.3 .08 - .37 .87 1.2 .65 1.6 .33 .27 .18 - .19 .96 -1.5 .33 - .07 Note.--N * 60 boys. 101 scores for the EH and Behavior Problems with each group respectively lowest and highest where hypothesized. The Behavior Problems group shows no deficiency at all whereas the EH group had the lowest mean score on each memory abil ity. The A-mathematics group were superior to the A-English group on both auditory abilities. Comparisons of Groups The Mann-Whitney U Test was used to test for signi ficances of differences between groups. Table 10 shows the the comparisons between clinical groups on the hypothe sized relationships. A-math and F-math groups-When the 29 mathematics students' z scores were combined for factor approximations, ranked and compared on each of the memory ability factors, they were found to be significantly different only on Fac tor Two, auditory backward memory. Although their mean score approximations had a wide spread of factor six, the difference between groups was not significant. A-English and F-English Groups.--ltfhen the 17 Eng lish students' z scores were combined for factor score approximations and were ranked for the purpose of compari son between the two groups on each memory ability, the only memory ability which differentiated them was the hypothesized one, visual backward (Factor Two). The TABLE 10 COMPARISONS BETWEEN CLINICAL GROUPS ON THE THREE MEMORY ABILITIES (MEAN FACTOR SCORE APPROXIMATIONS) Level of Significance for: Groups Factor Six Factor Two Factor Five Auditory Forward Auditory Backward Visual Backward A-math with F-math n.s. * less than .02 n.s. A-English with F-English n.s. n.s. .002 Behavior Problem students with EH less than .02 less than .002 less than .002 *Mann-Whitney U Test of Significance. 103 probability that these boys were from the same population was less than .002. They were not differentiated on either auditory memory ability. Behavior Problems and EH Groups.--When the mean factor score approximations were ranked and compared for the 24 boys in these groups, their mean factor scores indicated that the groups were significantly different on each of the three memory factors: the probability that such differences would have occurred by chance was less than .02 on auditory forward (Factor Six), less than .002 on auditory backward (Factor Two), and less than .002 on visual backward (Factor Five). Discussion of Hypotheses Group One. Behavior Problems Group^ Hypothesis 3.1 This hypothesis was supported; the mean factor score approximations on each memory ability showed that the boys were not deficient in any ability; in fact, they were among the highest scorers, thus indicating that the prob lems they cause for teachers and staff in school are not related to memory deficits, but rather are behavioral and most probably arise out of environmental values which dif fer from those held by most of their peers and teachers. 104 Group Two, Educationally Handicapped. Hypothesis 3.2 This hypothesis stemmed from the writer's experience in testing many educationally handicapped students, boys and girls from all grade levels, and came about as a ques tion concerning results from their Stanford-Binet LM IQ's. The Binet is an auditory test; nearly all educationally handicapped students were unable to succeed on the immedi ate memory items; thus it was hypothesized that these 11 boys would not have high scores on auditory tests of im mediate memory. The hypothesis was supported, but it was also found that these boys were equally poor on visual backward memory. Since the backward memory tasks demand a kind of processing, and since these boys were suspected of being poor on the more simple immediate reproduction of stimuli, then perhaps it was inevitable that they would not be able to succeed at any task which required a longer or more complex ability such as processing and holding in mind those items regardless of modal input or order. These boys showed memory deficits which were sufficient to have affected retention of instructions within the classroom on a day-to-day basis. This may explain why they have in common school achievement which is at least three to five years retarded in verbal and numerical subjects. 105 Group Three, A-English Students f HypotheTis 3.5 The A>English group was found to be high in visual backward memory ability but not high in the auditory mem ory abilities, thus Hypothesis 3.3 was supported. The mean factor score approximations for the group on all mem ory abilities were positive and, when contrasted with the F-English students on Factor Five, visual backward memory, the Mann-Whitney U test indicated that the means of the two groups were significantly different. Group Four, F-English Students f Hypothesis 3.4 It was hypothesized that these boys would be low on visual backward memory as well as on auditory forward mem ory. The mean factor scores for the group had negative values on both factor abilities. Interestingly, their mean scores for auditory backward memory ability were similar to that of the A-English group, with both groups having scores with positive values. The hypothesis was confirmed. Group Five, A-Mathematics Studentsf Hypothesis 3.5 This hypothesis was concerned with the auditory backward memory ability (Factor Two) of the A-mathematics group. Their mean factor score on this ability was the 106 second highest of all six groups* thus supporting the hy pothesized relationship. Group Six, F-Mathematics Students y Hypothesis 3.6 It was hypothesized that this group would have low factor approximation scores on auditory backward memory (Factor Two). This hypothesis was confirmed; the mean fac tor score of this group was the secon lowest. The F- mathematics students, like the EH Ci.oup, obtained negative values as means for all three memory abilities. Summary This chapter covered hypotheses for and identifi cation of clinically unusual groups which are frequently found in the schools and are of great concern to educators. The purpose of Part Three was to attempt to make practical applications of the findings from Parts One and Two. The results show the value of using combinations of immediate memory tests for the purpose of prediction or diagnosis. A deficit in immediate memory is undoubtably critically related to assimilating subject matter and this chapter has demonstrated that the subject matter most critically affected is dependent upon the modal characteristic of the immediate memory ability. CHAPTER V SUMMARY, FINDINGS, DISCUSSIONS AND RECOMMENDATIONS Summary The nature of immediate memory and its relation ship to certain school subjects was investigated in a three-part study using ISO ninth-grade high-school boys who represented an Anglo-Saxon stratum of school popula tion in the Centinela Valley School District near Los Angeles. The district serves a lower socioeconomic level. The purpose of Part One of the study was to demonstrate, through factor analysis, that memory is not a molar abili ty, and that, in fact, immediate memory did, through analysis, separate into different memory factors. Toward this end, a battery of eight memory tests was devised for modal input differentiation, order differentiation, and content differentiation. These tests were given to 90 of the boys who were selected from the 90-110 IQ range and who also had not shown any special problems in school. The results from these auditory forward and backward order digits and words span tests, along with measures taken from the boys' records on 19 other variables, were 107 108 employed for the purpose of factor analysis. The analysis differentiated three immediate-memory abilities: auditory forward, auditory backward, and visual backward. In Part Two an additional test battery was devised to test for hypothesized relationships between memory abilities and school achievement. The complete battery was given to boys who obtained the highest or lowest ten factor scores on one or more of the three factors, although only selected relationships were hypothesized. Part Three of the study purported to give the first two parts practical application for school use. Hypotheses were developed about the nature of memory factors in parti cular clinical and academic groups. There were 60 more boys from the same grade and school who were involved in six groups. They were selected as typifying the most com monly found groups in education for which differential handling is usually required: Behavior Problems, Educa tionally Handicapped, A-students in English, F-students in English, A-students in mathematics, and F-students in mathematics. They were given the eight memory tests for the purpose of testing hypotheses relative to memory abilities of each group. FINDINGS Part One The study first attacked the assumption that memory 109 is a molar ability. The assumption was found to be unten able. Part One showed that immediate memory may be differ entiated according to modal input and also to order of reproduction of stimuli. The fact that order (forward or backward) and not content (words and digits) caused the formation of separate factors confirms the findings of Tenopyr (1966) and Meyers, et al. (1966), who obtained simi lar factorial separateness. However, if more memory tests had been used in the battery, it is possible that content would have caused further delineation of factors as found by Guilford (1959, 1966), Brown (1966), and Stott and Ball (1965). In the final eight factor rotation, a general edu cational factor carried most of the group tests of achieve ment; however, the independent loading of certain achieve ment tests on any one of the three individual memory factors led to hypotheses concerning relationships between certain academic performances and the identified memory abilities. Part Two Part Two was concerned with the correlated abili ties of the high and low factor scorers on each obtained memory factor. Relationships were explored between audi tory backward memory (Factor Two) and arithmetic as tested in the WRAT, visual backward memory (Factor Five) and 110 spelling, and auditory forward memory (Factor Six) and reading for immediate comprehension as tested in the Nelson-Denny Paragraphs Test. Factor Two (Auditory Backward Memory)-The hy pothesis was cdnfirmed that the ability to hold in mind an unrelated series of digits or words and to process them backward in order was unquestionably related to the abili ty to perform well or not in arithmetic computation in volving symbols only. The Fisher Yates test gave a significance at the .001 level for this relationship. Evi dently the same memory ability is also an integral compo nent in performance on tests such as the DAT-Language, Sentence Usage. This obtained relationship was significant at the .002 level. It was found that the low or high fac tor scorers on this factor were also respectively poor or good spellers as tested by the WRAT and DAT tests of spell ing. There are two facts, however, to bear in mind which are pertinent to the findings: one, the division between high and low scoring on the spelling tests is at a seventh- grade level and, in fact, the mean for the entire ninth grade on group tests of spelling was 7.4. This is due in part to the low socioeconomic level of the population. Two, the significance of the relationship between high and low Factor Two scorers and scores on the spelling tests may also be due to the unavoidable occurrence of dual or Ill occasional triple factorial membership of the ten highest or lowest factor scorers. There was no significant rela tionship between high and low Factor Two scorers and scores on the Nelson-Denny Comprehension of Reading Paragraphs. Factor Five (Visual Backward Memory)-The loading of spelling tests on Factor Five, visual backward memory, led to hypothesized relationships between this ability and the ability to perform on tests of spelling individually administered. Both encoding (WRAT) and visual decoding (DAT) spelling tests were used to test the hypothesized relationship. DAT-Spelling and Factor Five scorers showed relationships between their performances on both variables that were significant, (p».001), whereas the obtained relationship with the more complex task, WRAT-Spel1ing, was .003. The hypothesis was confirmed. There was no discrimination between the high and low factor scorers on performance on WRAT-Arithmetic or the Nelson-Denny. That the Factor Two scorers showed more relationships to be significant than the hypothesized, whereas the Factor Five scorers did not, may be explained partially by the order of the appearance of the factors with Factor Two being the second factor to form. The boys who made the ten high factor scores on Factor Five scored almost equally between high and low cut-offs for the DAT-Language Sentence Usage Test, but none 112 of the ten low scorers on that factor scored in the high range on the DAT-Language test. It may be surmised that a minimal level of the ability to do visual span processing in backward order had inherent in it a quality which is necessary for successful mastery of language tasks as tested by the DAT. Factor Six (Auditory Forward Memory).--In the final eight-factor rotation, auditory forward memory was one of the later factors to form; there were fewer loadings on it, but the reading comprehension group test, the Nelson- Denny Silent Reading Test, did load on factor six inde pendently and thus the hypothesis that an ability to do tasks involving immediate feedback of comprehension was thought to be related to auditory forward memory. The Nelson-Denny Paragraph Comprehension test was chosen to test that hypothesis. No significant relationship was found between the two variables; the high ten and low ten factor six scorers were distributed indiscriminately on the test. These 20 boys were given the entire battery, as were the others, and there was no significant relationship found between their respective abilities and the WRAT- Arithmetic, DAT-Sentence Usage, or DAT-Spelling. They did, however, distribute themselves differentially as high and low scorers respectively on the WRAT-Spelling. This find ing was interpreted as meaning that the spelling tasks for 1X3 the WRAT-Spelling demanded a similar ability to that in volved in successful performance or poor performance on the composite memory tests which saturated Factor Six. Part Three This part was concerned with the exploration into the possibilities of using combined memory tests as indica tors of clinical types frequently found in the schools. Its purpose was to give practical application to the first two parts. It was found that combining the tests as they had saturated each factor did differentiate groups. Of the 13 boys who made A's in mathematics and the 6 boys who failed mathematics, their mean factor scores on auditory backward memory ability did differentiate them; the hy pothesis was confirmed. A Mann-Whitney U test gave a probability of .02 that such a separation would occur by chance. Mann-Whitney U tests were computed for group dif ferences on auditory forward and on visual backward memory; the obtained levels were not significant. These boys did not differentiate themselves by group on the other memory factor abilities. However, the A-mathematics students had positive mean factor scores for all three factors, and the F-mathematics students had negative mean factor score ap proximations on all three memory abilities. In the comparisons of the 11 boys who made A's in English with the six who made F's, it was found that they 114 were significantly different groups on visual backward memory, (p«.002). Their mean factor score approximations did not differentiate them on factor six (auditory forward) or on factor two (auditory backward). Interestingly, the F-English group had mean factor profiles which were very similar in pattern to that of the A-mathematics group, with the difference being one of degree except in visual backward memory; the scores were close in value on visual backward memory, but the A-mathematics group mean was positive whereas the F-English group mean was negative. The largest difference between two groups, con sidering all three factor scores, was found between the Behavior Problems and the EH groups. The Behavior Prob lems group scored highest for auditory forward and visual backward memory and were second only to the A-mathematics group on auditory backward memory, while the EH students were the lowest on all three memory abilities. The EH students were undoubtedly critically deficient not only in auditory memory as hypothesized but in visual backward memory, too. Since all of the EH boys had been given Stanford-Binet IQ tests (LM) before placement in the special classes, and since all had achieved scores which placed them in the average or above average range, these findings indicate that general assessments of intellectual academic aptitude may give adequate appraisal of potential intellectual and academic aptitude, but do not contribute 115 information of specific deficits such as in immediate mem ory unless the responses are assessed in terms of a model of intelligence such as the structure of intellect. It is equally apparent that boys in this socioeconomic environ ment who are able to maintain passing grades in high school but who are incorrigible in their behavior are not suffer ing from memory deficiencies or perhaps even cognitive deficiencies, but rather that other variables such as personality or cultural values are involved. Mann-Whitney U tests showed that the above groups were not drawn from the same population, for all three factors discriminated between the two groups. For Factor Six, the probability was less than .02 that they were from the same population; for Factor Two, the probability was less than .002, and for Factor Five the probability was less than .002. Discussion The findings reported have demonstrated the use fulness of the structure-of-intellect model in studying individual differences among students. The study was in tended to show that memory is not a molar ability such that generalizations can be made about learning principles derived from the study of paired-associate learning and verbal learning experiments. Other investigators (Orpet § Meyers, 1966; Meyers, et al., 1966; Stott and Ball, 1965; 116 Brown, 1966; Duncanson, 1964; Tenopyr, 1966) have shown that memory is differentiated for younger students as well, and this study has also shown that the sensorimode used as stimuli for testing has differential results. It has also shown that the order required in processing for repro duction is equally important. The writer had hoped to show that backward order for a series is indeed a different kind of ability in immediate memory (because the task demands "holding" in mind and processing of information before reproduction) from that required when forward order for a series is demanded where reproduction is immediate without processing. Thus the basic assumption held by many VPAL investigators and by interpreters of traditional individual intelligence tests that memory is molar was found to be untenable. The finding that visual memory is an ability which is necessary for spelling and is unrelated to intelligence scores, given a minimum, (all of these boys came from a range of 90-110) is one which should in all probability give additional insight to those who hold spelling as an indication of ability to achieve and thus a prime indicator of ability to "reason" (Bloom, Engelhart, Hill, Furst, 6 Krathwohl, 1956). There is a need to' Took upon students as indivi duals with differential abilities, some of which are strong and some of which are not; there is a need to 117 acknowledge that differential abilities have relationship to success or failure in certain curricular areas. Few will deny the importance of memory in education but its specific importance needs to be clarified. Special con sideration needs to be given to the notion that memory may be separated factorially into different memory abilities each of which has specific correlates in subject matter. This study has shown that a combination of memory tests can have useful application toward that purpose in the schools. The fact, for instance, that auditory backward immediate memory is a necessary ability for the successful performance in mathematics and in correct language usage was established and confirmed when the memory tests were used as independent predictors and when they were used as dependent variables. If one then surveys the field of information avail able on the relative scarcity of students who not only prefer mathematics but also do well in the subject, espe cially at a time when mathematics is playing an ever increasing role in technology, one is led to question not whether there may be fewer people who are "auditory minded" so to speak, but whether there are fewer people who excel in mathematics because school tasks require visual memory ability from the earliest grades. Thus visual memory is perhaps over-trained early in school whereas tasks requir ing auditory memory are not only less frequent in the 118 curriculum but are often taken for granted and consequently seldom taught. Or, one may wonder whether children come to school for the first time with environmentally or geneti cally influenced predispositions for specific memory abili ties and that more are "visual-minded" than "auditory- minded" as a result. Pressey (1944) noted that students who did better at visual tasks in school were more numerous (about 8 to 2) than those who performed well in tasks pred icated on auditory instructions. The implication here is of serious import for educators, particularly in the face of contributions from S-R learning theory which stresses both conditioning and teaching to specific tasks. On the other hand, if one reads extensively in books on education, one finds a plethora of material about separate subjects and methods of teaching them. Rarely do allusions to individual ability to comprehend by the learner encompass characteristics other than IQ range or emotional adjustment. The relationships which have herein been shown to be significant, specifically visual memory of a processing nature and spelling, and auditory memory of a processing nature and arithmetic, are rich sources for further inves tigations. The importance of immediate memory cannot be underestimated in school learning. Perhaps the most startling result reported, how ever, has timely significance. It is the profile where 119 mean factor score approximations of two of the more im portant groups of high school students, the Behavior Problems and the Educationally Handicapped, are compared. Even though EH boys had normal or better Binet IQ's, the patterned deficit in immediate memory of all three kinds implies a causative relationship such that a critical deficit in any one of these areas can cause real inability to comprehend everyday subject matter. In contrast, excellent immediate memory ability does not necessarily ensure successful school achieve ment, for as the Behavior Problems group has demonstrated, personality variables have important influence. Neverthe less, the fact that the Behavior Problem boys were holding their own, gradewise, is indicative that perhaps memory abilities have some part in this success, since their IQ range was 90-110. One might speculate that a strength in any one of these memory abilities is sufficient for pass ing grades in related subjects when very little effort is expended. Much concern may be voiced that the current prac tice of many educational systems of placing students having serious deficit memory abilities such as EH students in the same classes with boys who have behavior problems places a grave burden on the teacher who is required to teach them. Recommendations This study was a second-step procedure in the sci entific application of the structure-of-intellect model of intellectual functioning. The findings carry two related implications for traditional modes of intellectual assess ment and academic planning. First, immediate memory is a necessary prerequisite for most instruments which assess intelligence; lack of immediate-memory ability may obscure or mask other facets of intellectual potential. Second, in the development of academic capability, the lack of imme- diate-memory ability could well serve as an inhibitor. This suggests that efforts be directed toward the training of memory abilities when there are deficiencies; it also suggests that since other investigators have shown differ entiated memory abilities among younger students, memory abilities need to be assessed at earlier levels if problems in academic achievement are to be avoided. It is recommended that, from the supportive quality of findings herein, an efficient psychometric battery of memory tests be developed for use with academically inter esting cases and that the results of these tests be pur sued in the course of curriculum planning for individuals. There is little doubt that educational psychologists and educators must take cognizance of the importance of rela tionships between separate memory abilities and curriculum correlates. 121 The general method in this study could be extended to the exploration of other dimensions of the structure of intellect among specifically interesting academic problems --one example, the patterning of divergent and convergent thinking which might be investigated using creative and gifted children. A second area which might be rewarding to educators is the comparison of evaluative abilities among non-problem students and problem students. A follow- up of this study would concern itself with differential memory abilities among gifted and average students in an upper-middle class community. REFERENCES 122 REFERENCES Barnes, J. M., § Underwood, B. J. "Fate” of first-list associations in transfer theory. Journal of Experimental Psychology, 1959, 58^ 97-105. Bloom, B. S., Engelhart, M. D., Hill, W. K., Furst, E. J., § Krathwohl, L). R. Taxonomy of educational ob jectives. Handbook I. Cognitive domain. New York: David McKay Company, Inc., 1956. Boycott, B. B. Learning in the octopus. Scientific American, 1965, 212, 3, 42-50. Broadbent, D. E. Perception and communication. New York: Pergamon, 1958. Brown, J. Some tests of the decay theory of immediate memory. Quarterly Journal of Experimental Psy- chology, 1958, 1£, 12-21. Brown, S. J. Semantic-memory and creative (divergent- production) abilities of senior-high-school stu dents. Unpublished doctoral dissertation, Univer sity of Southern California, 1966. Carlson, D. C., 6 Meyers, C. E. Two language, two memory, and two perceptual abilities in retardates of mental age four. Unpublished study, Socio- 123 124 Behavioral Laboratory, Pacific State Hospital, Pomona, California, 1966. Davidson, R. E. Mediation and ability in paired-associate learning. Journal of Educational Psychology, 1965. Deutsch, J. A. Higher nervous function: the physiological bases of memory. Annual Review of Psychology, 1962, 24, 259-286. Duncanson, J. P. Intelligence and the ability to learn. Research Bulletin (RB-64-29). Princeton, N.J.: Educational Testing Service, 1964. Estes, W. K. Learning theory and the new "mental chemi stry." Psychological Review, 1960, 6_7 , 207- 223. Fantz, R. L. Visual perception and experience in early infancy. Unpublished paper presented at the Social Science Research Council Conference, Univer sity of Minnesota, 1965. Festinger, L. Behavior and cognition. In Contemporary approaches to cognition: the Colorado symposium. Cambridge: Harvard University Press, 1957. Fitzgerald, D., § Ausubel, D. P. Cognitive versus affec tive factors in the learning and retention of controversial material. Journal of Educational Psychology, 1963, 54_, 73-84. French, J. W. The description of aptitude and achievement factors in terms of rotated factors. Psychometric Monographs, No. 5, 1951. Garrett, H. Great experiments in psychology. New York: Century Company, 1930. Guilford, J. P. Three faces of intellect. American Psy chologist , 1959, 1£, 46-79. Guilford, J. P. Intelligence--1965 model. American Psy chologist , 1966, 2_1, 1, 20-26. Guilford, J. P., 6 Hoepfner, R. Current summary of structure-of-intellect factors and suggested tests. Reports from the Psychological Laboratory, No. 30. Los Angeles: University of Southern California, 1963. Harris, I. D. Emotional blocks to learning. Glencoe: The Free Press, 1961. Hebb, D. 0. The organization of behavior. New York: Wiley, 1949. Hebb, D. 0. Distinctive features of learning in the higher animal. In J. F. Delafresnaye (ed.) Brain mechanisms and learning. New York: Oxford University Press, 1961. Hebb, D. 0., 5 Foord, E. N. Errors of visual recognition and the nature of the trace. Journal of Experi mental Psychology, 1945, 35, 335-348. Hilgard, E. R. Theories of learning (2n ed.). New York: Appleton-Century-Crofts, Inc., 1956. 126 Hopkins, K., 6 Michael, W. The diagnostic use of WISC subtest patterns. California Journal of Educa tional Research, 12, No. 3, 1961. Hopkins, K. D., Dobson, J. C., 6 Oldridge, 0. A. The con current and congruent validities of the Wide Range Achievement Test. Educational and Psycho logical Measurement, 1962, 22_, 4, 791-793. Hull, C. L. Principles of behavior. New York: Appleton- Century-Crofts, 1943. Jacobson, A. L., Jacobson, Anne, Babich, F., § Bubash. Memory transferred between animals via RNA. Psychiatric Progress, 1965, 1_, 2, 2. James, W. Principles of psychology. New York: Holt, 1890. Jastak, J., § Bijou, S. Wide Range Achievement Test. Copyright, 1946. Jensen, A. R., § Rohwer, W. D., Jr. Syntactical mediation of serial and paired-associate learning as a function of age. Child Development, 1963, 601-608. Johnson, R. C., 6 Medinnus, G. R. Child psychology: Behavior and development. Wiley and Sons, Inc., 1965. Jones, J. E. All-or-none versus incremental learning. Psychological Review, 1962, 69, 156-160. Kelley, T. L. Crossroads in the mind of men. Stanford: Stanford University Press, 1928. 127 Keppel, G., § Underwood, B. J. One-trial learning? Journal of Verbal Learning and Verbal Behavior, 1962, 1_, 1-13. Kluver, H. Eidetic imagery. In Handbook of Child Psycho logy (C. Murchison, Ed.). Worcester, Mass.: Clark University Press, 1931, 699-722. Loeffler, F. J. An extension and partial replication of Meyers, et al., primary abilities at mental age six. Paper read at Society for Research in Child Development, Berkeley, 1963. McCarthy, J. J., 5 Kirk, S. A. Illinois test of psycho- linguistic abilities. Examiner's Manual. Urbana, Illinois: University of Illinois Press, 1961. McCarthy, J. J., S Kirk, S. A. The construction, stan dardization, and statistical characteristics of the Illinois Test of Psycholinguistic Abilities. Urbana: Institute for Research on Exceptional Children, University of Illinois, 1963. McGeoch, J. a . Forgetting and the law of disuse. Psy chological Review, 1932, 39, 352-370. McGeoch, J. A., 5 Irion, A. L. The psychology of human learning (2nd ed.). New York: Longmans, Green, 1952. McNemar, Q. The revision of the Stanford-Binet scale. Boston: Houghton Mifflin, 1942. Meeker, Melton, Melton, Meyers, Meyers, Miller, Morgan, Murdock 128 Mary. A procedure for relating Stanford-Binet behavior samplings to Guilford's structure of the intellect. Journal of School Psychology, 1965, 3^, No. 3, 26 - 36 . A. W, Comments on Professor Postman’s paper. In C. N. Cofer (ed.) Verbal learning and behavoir: problems and processes. New York: McGraw-Hill, 1961. A. W. Implications of short-term memory for a general theory of memory. Journal of Verbal Learning and Verbal Behavior, 1963, _2, 1-21. C. E. A factorial view of the Illinois Test of Psycholinguistic Abilities. In progress, 1966. C. E., Sitkei, E. G., § Watts, C. A. Further ability factor hypotheses at six years. Unpub lished study, Socio-Behavioral Laboratory, Pacific State Hospital, Pomona, California, 1966. G. A. The magical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review, 1956, 63, 81-97. C., § Stellar, E. Physiological psychology (2nd ed.). New York: McGraw-Hill, 1950. , R. B. The retention of individual items. Journal of Experimental Psychology, 1961, 62^, 618-625. 129 Orpet, R. E., § Meyers, C. E, A study of memory, conver gent, divergent, and evaluative factors at six years. Journal of Educational Psychology, 1966, 57, (in press). Osgood. C. E. Method and theory in experimental psycholo gy. New York: Oxford University Press, 1953. Osgood, C. E. Motivational dynamics of language behavior. In M. Jones (ed.) Nebraska Symposium on Motivation. Lincoln: University of Nebraska Press, 1957. Osgood, C. E., 5 Miron, M. S. (eds.). Approaches to the study of aphasia. Urbana, 111.: University of Illinois Press, 1963. Osgood, C. E., 6 Sebeck, T. A. (eds.). Psycholinguistics: A survey of theory and research problems. Baltimore: Waverly Press, Inc., 1954. Peterson, L. R., S Peterson, M. J. Short-term retention of individual verbal items. Journal of Experi mental Psychology, 1959, 58_, 193-198. Postman, Leo. Short-term memory and incidental learning. In Arthur Melton (ed.) Categories of human learn ing. New York: Academic Press, 1964. Pressey, S, L., § Robinson, F. P. Psychology and the new education (revised). New York: Harper and Brothers, 1944. Rappaport, D. Diagnostic psychological testing, Volume I. Chicago: Year Book Publishers, Inc., 1945. 130 Rohwer, W. D. The verbal facilitation of paired-associate learning. Unpublished doctoral dissertation, University of California, 1964. Russell, D. H. Children's thinking. Boston: Ginn and Company, 1956. Sitkei, E. G., § Meyers, C. E. Comparison of four-year- olds of two ethnic groups on factored tests. Unpublished study, Socio-Behavioral Laboratory, Pacific State Hospital, Pomona, California, 1966. Spence, K. W. Behavior theory and conditioning. New Haven: Yale University Press, 1955. Stott, L., $ Ball, R. S. Evaluation of infant and pre school mental test. Detroit: The Merrill-Palmer Institute, 1963. Tenopyr, Mary. Symbolic memory tests as predictors of high-school grades. Unpublished doctoral disser tation, University of Southern California, 1966. Underwood, B. J. Interference and forgetting. Psycho logical Review, 1957, 6£, 49-60. Waugh, N. C., $ Norman, D. A. Primary memory. Psycho logical Review, 1965, 7_2, 89-104. Wechsler, D. Cognitive, conative, and non-intellective intelligence. The American Psychologist, 1950, 5, 78-83. A P P E N D I X 131 132 TABLE 11 DESCRIPTION OF TESTS USED IN PART 1 AND PART 2 AND SCORE SHEET Verbal directions for Auditory Forward digits and words tests: This is a test to see how well you can repeat in order some numbers (words) read to you. Listen care fully and repeat this sample: 5-1-4 Now, we will begin the test. You will have two trials for each length. They will get longer as we go along. Verbal directions for Auditory Backward digits and words tests: This time I will say some numbers (words) and you are to repeat them in backward order. Ready? Directions for Visual tests were identical except the opening read: You will be shown some cards on which individual numbers (words) will be typed. You are to watch each card carefully and when they are all face down, tell me the numbers (words) you saw. Give them in the same order as you saw them (or backward order as they were presented). 133 NAME_____________________ BD_____ SCORES: T-CTMM_ I.D. CIRCLE ORDER GIVEN FIRST: M______ Visual Auditory GROUP (Left half of Score Sheet A) AUDITORY TESTS: (Rate: one per second) Digits Forward RESPONSES 3-8-6 4-3-1-7 5-2-9-4-1 3-8-1-9-7-4 1-5-7-4-3-6-8 1-6-4-8-2-5-9-7 S-3-9-7-1-6-2-8-4 RESPONSES 6 - 1-2 6-8-S-7- 5-2-1-9-6 7-9-S-8-4-3 2-9-7-4-1-3-S 2-6-1-7-5-8-4-3 4-2-9-6-1-7-3-8-5 Words Forward (Circle number of first correct responses) act, tree, dog 3 fun, dad, top 3 cat, now, end, kid 4 eat, car, low, sat 4 fur, egg, too, put, ice 5 fat, art, see, toe, sky 5 won, set, ear, cup, oak, pig 6 has, buy, tub, who, for, bee 6 sun, cow, use, end, ate, man, why 7 hoe, wee, say, out, red, dig, rat 7 did, yes, hat, run, box, she, tag, try 8 can, saw, did, new, eel, how, pie, sew 8 boy, tin, sit, had, arm, met, ray, rug, bat 9 sir, not, are, get, day, all, old, ran, ask 9 134 AUDITORY TESTS (Continued) Digits Backward RESPONSES 5-7-4 7-2-9-6 1-4-2-5-8 1-6-5-2-9-3 5-5-9-2-6-4-7 6-9-1-4-7-3-8-5 4-7-2-S-9-1-4-8-6 Words Backward eye, leg, bow cap, toy, bed bid, pin, fly, yam mat, rid, ton, ewe off, pie, act, lit, bag fee, got, may, nut, let not, own, set, hot, art, tan bum, wow, sap, buy, hit, yet few, pat, hay, air, age, add, cry far, pin, sit, den, nor, til, war VISUAL TESTS: (Right half of Score Sheet A) Digits Forward RESPONSES First Second (Presented on 3 x 5 unlined cards, one at a time, in series of 3, 4, 5, 6, 7, 8, 9, with two trials. Second trial series was offered only if subject missed the first series.) Words Forward (Same procedure as above, using three-letter words instead of digits.) 3 3 4 4 5 5 6 6 7 7 RESPONSES 2-5-9 8-4-9-3 9-6-8-S-2 3-6-8-1-9-4 3-5-1-9-2-8-7 3-1-7-9-5-2-8-3 3-5-8-2-7-1-6-9-4 135 VISUAL TESTS (Continued) Digits Backward RESPONSES First Second (Same procedure as above with two trials controlled for patterning as in Digits Forward.) Words Backward (Same as above, using words instead of digits.) Extra words to add if needed: law, lie, act, ill Nelson-Denny Silent Reading Test The Nelson-Denny Silent Reading Test-Paragraphs (Form B) was used to test immediate comprehension. The paragraphs were ordered for difficulty. Each paragraph appeared separately on a 5" x 7U card which the subject read at his own rate. When he finished, a second card was presented on which appeared three questions. He gave the correct answers for them to the examiner, who recorded the answers. Example: 136 5. Jack had sat all morning on the bank of the lake with his pole and line, but had caught only two small bass. When he first came to the lake a week ago, he had caught three great big perch. Suddenly he felt a big tug on his line. Jack began to pull on the line excitedly. How hard he had to pull! It must be a large fish. What was Jack doing? 1 swimming, 2 fishing, 3 reading, 4 rowing a boat ................... What do you think happened next? 1 Jack pulled out a large fish, 2 Jack went home, 3 Jack fell out of the boat, 4 Jack cleaned his fish . . . r .............................. What kind of fish had Jack caught that day? 1 perch, 2 trout, 3 bass, 4 pickerel . . . . 137 Directions for DAT Language-Sentence Usage: On your paper are some sentences* This is a test of grammar and English. If the sentence doesn't sound or look right or correct, put the alphabetical letter which is under the phrase which seems incorrect in the left margin. An error may be in the grammar, spelling, or punctuation, or any combination of these. If the sentence seems correct, put a plus in the left margin. If you are not sure, put a circle around the number. Take all the time you like. Don't guess. Examples: 1. Where / did you / stop at / on your trip / to A B C D Chicago. E 2. Was it / him / who / got burned / when the A B C D boiler bursted? E 3. The dog laid / sleeping / after chasing John A B C and I / with hardly no / time out. D E Directions for DAT Spelling: On your paper are typed some words. If they seem to be spelled correctly to you, put a plus in the left margin; if they seem incorrectly spelled, or do not look right, put a minus in the left margin. If you do not know, do not guess; instead, circle the number of the word you are not sure about. Take as long as you like. Examples: 1. apointed 2. commission 3. limited 4. arival 5. comunity SCORE* SHEET, second battery 138 Test 1 Test 2 Test 3 Test 4 WRAT-Spelling WRAT-Arithmetic DAT-Sentence- Language Reading - Comprehension NAME + + 1. A. B. C. 2. A. B. C. • > • l B. c. 4. A. B. c. 5. A. B. c. 6. A. B. c. 7. A. B. c. Test 5. DAT-Spelling + TABLD 12 .iiiJ, j J U B a m * ,J COREMIOH CC2F7ICIEHK FOR 27 VARIABLES Variable j i i i .saurau D a ! j-;athi C rient& tios 24, DA! Aeclianicai i - i e a n s Standard Deviationc 3 - 3 3 7 ^ 9 1C 11 12 13 14 15 16 17 IS 19 20 21 22 23 24 25 26 27 1 , C M i 73 2, Oil Heading °2 3, OAT Arithmetic 4, ilelson leading 5, Auditory Digits Foward 6, Auditory 'lords Bernard 7, Auditory Digits Backward 8, Auditory .lords Backward 9, Tisuai digits Forward 10, Tisuai lords For-ard 11, Tisuai Jigias Backward 12, Tisuai lords oacbarc 13, Birth Date* ii, liss-Mion -eouenee* 15, less Object (oolei 0-,'j 16, anDlish Grade 17, jiglisi: -.valuation it, ijtotici 'Grade IT , U t o t i i i j a l u n s l o n 2 0 , DA. V erb al 2a, J.T! iir.cricai r -' 'f ' ' y 'y V - - m " f 24 13 01 21 45 -07 45 -24 57 53 33 22 14 37 37 52 12 1; 22 29 28 23 12 23 49 C5 39 -26 65 44 15 15 06 34 48 53 05 20 13 22 19 27 15 29 49 -17 57 -35 50 66 27 20 08 47 45 53 17 iv o 30 27 26 14 16 40 08 33 -13 56 35 13 16 -01 41 57 49 60 3; k 28 34 22 37 26 11 19 25 06 33 07 10 29 08 -02 00 12 jL vJ 2i 26 24 28 24 13 25 -d3 15 30 01 03 10 04 00 05 11 51 42 38 37 29 27 15 17 14 -17 21 25 11 03 12 02 13 16 ^ ! % 29 23 30 27 n < ? tC 10 03 -Cc 19 07 03 05 08 07 15 30 30' 16 25 10 1/ 07 20 00 17 21 -06 08 06 13 22 25 43 52 J - * ; 27 J , 31 -05 30 21 a 03 -06 07 30 36 Li ■ ’* ? -v r n SO 10 19 -0-3 29 18 05 -08 -05 12 22 28 i r 27 04 12 -04 10 13 06 02 -12 07 25 15 2 1 ?3 -17 pQ 35 24 10 19 14 21 26 -31 -0 -41 (5 57 22 22 13 36 51 36 -44 47 67 43 41 24 40 33 52 46 33 25 26 20 44 56 34 35 30 37 33 45 42 37 21 20 34 46 25 -06 12 04 20 20 35 31 2g, DAT speed and Accuracy ^ 2c, Uni spelling 27, D a ! sentence nan.u^e h . , ; i h , ? i,j u u i . 2 n so 1 .7 »&>%* «fi % ffi % a? . u s a ; a:. :a u , » u .* 2 . 1 . £ 2.7 . » « 22.2 2S.5 z;.i 26.1 M «.i s.4 li/ Ct/. /fC 1(2 ^ 2r(_ Ji) I'iObSt^'lJscLv^- poinw Cui.tw 6( * ( Correlations for Variable: 13 and U are oaitted as artificial factor corxorents. u, \ refer: to :iean percentile, EIGHT-FACTOR ROTATION MATRIX FOR 2 7 140 M H H > H H > n f t p H P U > P 8 <<« VJ > l/ l p 0 p u 8 > f t H H H H M H H l / l 8 f t A 8 f t P 8 > fttONNNO'OMOffl®*** O f t O f t f " O f t l O f t O Q N O f t f t f t O f t f t N O O N O r l O O OHmoONtO^OWl'OCs'OfOlO'O'OHHf'Ol'OOON'OHH H O H O H O O H H O O O O O O H O M O O N H i f l O M H H H H ^'Ort'OlflOlflNirtHO'Oi/lHNNOO^lflHOirtOOONOKl H O O f t f t O O O O f t f t O i f l i f l O t ' I O f t O O O O O O l ' W O N^MCOfllONCOflN^rtOlflNXIOOMOlONisOOMA OOrlONNNHH'O'fl'OHrlHrlONOHHHHHONN l / H O r l f l l O I M ' ) « O N H I / ) » ( H H O M # ® O O ^ H « O N i n O I H O I O O O H O O O O O O O O H l f l ' O ^ l f l r l i t O H O O r l H H N ftpmftp.NM'WOOftOlOWOftNNOl'lNlAiflNts N O H O H O H O H O O O O O l O H O l t H K l r t ^ l f l i O H O N NtOK>LflrOOHK15iHOOLfl^Ol^OvOOOtO^'P'^Hl/lHOOH OrlHHMNM/l^NHHrlHNOHOrlOHOOO OOH ooi'>H»r'0Niorta>fli/>Not')HoN0'0Hatfrt*H r v 0 0 ' 8 « O O O f t N N f t O f t f t N 8 , O'*N\0*NftO8'l/>'O 1 3 U PU P U 8 P 13 S P S d ^ PU jm3 n!^ pjUP<mM 0 P 8 U 1 8 , i O ftO«rtC>y^8 f t flOMOM i/l n H i 0 P Q < 4 < u U P DIP I/I ft A ftfiftUftUW l / l f t p f t OOP Mh4J 1/lP « 4 8 0 X 0 ft 0 f t P 8 P 8 P 8 8 P 3 8 0 8 3 8 U 3 0 ftUft 8 U 8 P 8 8 f t < i/l 8 m Pft 3 8 3 UW f t 0 8 t3 UH I M l j- H O 'H O 'H U 'H U U f i 8 8 l/l l/l U l H ' H 8 B U CBdCjSQiiWPOOPftOSPmUftftBSS-PS f t j j l ) f t O f t O W f t U U A f t f t 8 P P ' H S l ) f t # TjptfmmQfcQSOPB +JP43 0+JWX 4IH4J m,h m m f t 8paa88pgw8uoos Bh(jooooHrtHHodo«n»aBt «9flM S < M (84! OPPPP1 8 8 4 * fi'H 'H 'H 0 8>2<W2WWW S wftftftft3 3 3 3 0 gPftft,fi,e , , , l h H H i 3 l } ' 3 l ) i / l i / l i / l i / l M < /l A M P P H h H H H H H H UUU2<;<|<l<>|>>>UWBWW2j0Q00QQQ0 , , , , , I » I I # ♦ ♦ » * • • • • * * » * • 1 ' * * i-INH, )*lflOr*COaOHNfO«tlA'0NQOaOHNrt*l/>^ H H H H H H H H r t H N N N N N N N N P e f t o a g •H 0 8 0 1 8 P 0 2 HI > * r l t O O l O « ' r t f l r t K ) N l A H t ' K H l A l O O ' 0 ' 0 ' O l A « * H H O l O O I O n O H O H H H r t i n i A ' O l A O I A H O O H N N ^CONHOOOOlAM'lNO'tlflMHHOttOQNNint'' N O H O t O H H H O O H r l N H O r t H t O M l f l l ^ ' O H H H X H a b i z 0 H H < f t 0 A <& 0 f t U < f t i a ! p o u < H t ^OO' OHff lOO' OO^NHNHlOOOMrt COir t^HKIHOrt H Q H H N ^ l f l ' O ^ ^ ' f i ^ ' O M ^ H H O N H O O O O N N H A i M P 4 > \ O l f l M i W 0 l A H ^ H H ^ O I ' > H I A N I f l t ' ' H H M O l O O lOCO'OOOOOOHNNNOt'KOVN'O'tNHOV'O'O t) T j PO H c d f t 0 itPJlf'3 PO p j l U i t P jit 0P(«Ujif-SO ft OfflA P i U* ft rtOPHU i d u f t ort d y o c + J r t f t 1 i f l f t W .H fi-HTl'H'tf # 1 W C 0 •H P A It 0 0 3 •H'flH P It It i t P > V 3 U f t 0 X ' H 0 P I t I t P P 3 t U V PH 5 h o it'd p . H b O P M P P i n P i n ’O H IIP U fi M t ) e e v d ' H oh O'HTt'H'd it it m in u u h h i j b u d g (tOSQS MP Mp 0 p > u UH'H it It (J * H (J 'Hi# H O'H OtJOW'H'H It p PH rtTJ H t) ^ “^ ^ " ‘JwSSSjSS^HS BPdOOOOHHHH(jiflinS64)3^ P * 9*u c u o p p p p f l r t d f l .h h # « > z < i oswiom ? WtM'H'H 3 3 3 3PHH,C,fl . . . . . . S f t f t H ' O f l O f l i n i n i n i o i n M M P P f t f t f t f t f t f t f t f t Sjjl^53>55SSSSS«S5S2SS2 P P H 6 0 in p 3 H 0 p. E ■H u t) Q , t , , * « I ♦ t « I • • ‘ ♦ * • ' * * * ' * ' * HN I r f WUI Ot ' t OCl OHNKWI / t ' ONO Of l l OHNKWl f l 1 7 h h h h h h h h h h n n n n n n TABLE X 5 SIX - FACTOR ROTATI ONI MATRIX FOR 25 VARIABLES Variables Factors I I I XXI IV V VI X . CTMM 66 O 8 26 X 4 O 2 03 2 - CAT Reading 8 5 2 X O 2 04 03 02 3 . CAT Arithmetic 6 7 X 3 X 6 35 XO 04 4 . Nelson Reading 8 3 20 O 5 06 O 8 06 S . Auditory Digits Forward 04 7 X X 6 2 O 17 3 X 6 . Auditory Words Forward oe 5 7 O 8 28 2 O 2 S 7 - Auditory Digits Bac lew ard O 4 6 7 09 XO 23 X 7 8 _ Auditory Words Bac lew ard X X 6 S 04 04 X 3 09 9 . Visual Digits Forward 2 5 4 a X 2 09 X 3 O 3 XO _ Visual Words Forward 2 O 2 9 06 O 2 64 06 XX. VisuaX Digits Bac lew ard 20 24 O 2 0 0 2 5 2 O 8 X2. Visual Words Backward 04 3 X O 7 04 6 X X 3 X3. Best Grade X 9 30 2 8 X X X 3 O 7 X4. Eng 1 i sir Grade O 7 oa X 2 50 3 X 2 3 IS. Eng X i sir Eva I ua t ion O X X 7 O X 5 8 C 7 05 xa . Mathematics Grade 4 4 O 3 4 X 54 26 X 2 17 . Mathematics Evaluation 2 X oe X O 60 00 3 X 3 X8. DAT Verbal 64 2 X 36 OX 18 O O 1 X9. DAT Numer i c a X 4 6 X 2 38 5 X X O 06 20. DAT Abstract X 9 O 4 6 3 06 X 3 O X 2 1. DAT Spatial Orientation X X 09 5 7 X 2 X 2 4 5 2 2. DAT Mechanical O 2 09 63 08 X 6 02 2 3. DAT Speed and Accuracy 5 O O 7 X O X 3 O 7 X 9 24 . DAT Spelling 5 9 O X 06 20 3 2 X O 2 5. DAT Sentence Usage 5 8 X 3 29 18 2 2 2 2 Note. Decimal points omitted. 143 lABM 13 m m c o m o k , a i , m m m dmiioiib foe 25 m u l e s Variable 1 2 3 A 5 6 7 2 9 10 11 12 13 16 15 16 17 18 19 20 21 22 23 26 25 1. cm; bC 13 48 10 02 A 14 22 26 13 01 21 65 ■07 b "26 57 53 33 22 16 37 37 52 2. CAT Reading 62 79 19 14 19 22 29 28 23 12 23 69 05 39 -26 65 44 15 15 06 34 48 53 3. CAT Arithmetic 53 15 05 20 13 22 19 27 15 29 49 ■17 57 -35 50 66 27 20 03 47 45 53 4. Kelson Reading 21 17 19 25 30 27 26 16 18 40 08 33 -13 56 35 13 16 -01 41 57 49 5, auditory Digits Forward 68 id 29 28 36 22 37 26 11 19 C3 06 33 07 10 29 08 -02 00 12 6, Auditory Vords Forward 32 AO 21 26 26 28 26 13 25 -03 15 30 01 03 10 04 00 05 11 7, Auditory Digits Backward 51 62 38 37 29 27 15 17 16 -17 21 25 11 03 12 02 13 16 8, Auditory M s Backward 36 29 23 30 27 03 10 03 -06 19 07 08 05 08 07 15 25 9, Visual Digits Forward 30 16 25 10 19 07 20 00 17 21 -06 03 -06 13 22 25 10. Visual llords Forward 63 52 10 27 14 31 -05 30 21 14 03 -06 07 30 36 11, Visual Digits Backward 61 13 28 10 19 -03 29 18 05 -08 -05 12 22 28 12. Visual lords Backward 18 27 04 12 -04 19 13 06 26 -12 07 25 15 13, Best Grade 24 05 28 -17 29 35 25 1C 10 14 21 26 14. English Grade -21 65 -41 65 57 22 22 13 36 51 36 15, English Evaluation 29 37 Of, bu -24 -04 -10 -02 04 -11 -13 16, i'iatkiatics Grade -44 47 67 43 41 24 4 0 3 2 52 17, llatheaatics Evaluation JO -39 -10 -It -13 -14 -27 -32 IS, DAT Verbal cO 33 25 26 20 44 56 19, DAT numerical 34 35 30 3? 33 45 20, DaI abstract 42 37 21 20 34 21. DAT jpatiai Orientation 45 25 -06 19 22, DAI ilechanical 04 -02. 20 23. DAI speed and Accuracy 35 31 2A, Da T spelling 43 25, DAT Sentence language cleans 99.8 7.9 8.2. 7.6 6,1 5.6 4.1 3.6 6,7 ;.9 5.5 6,1 3,2 1,7 4.0 1,7 3,3 325* 375 5o;; 47? 4 .06 386 366 266 standard Deviations 6,016,616,4 17.9 1.3 1.1 1.2 .8 1.2 .9 1.2 ,9 2.9 .8 2.7 ,5 2,5 221,1219.5 25.1 26,8 26.7 25,1 25,6 20.5 Bote,— Decimal points omitted. *> refers to percentile. 144 TABLE 17 FACTOR SCORE CONTRIBUTIONS IN STANDARD DEVIATIONS 10 High Scorers 10 Low Scorers Subject Code Subject Code Number Score Number Score 2 1.20 15 -1.01 4 1.78 21 -1.28 26 1.49 22 - .80 Factor 34 2.22 29 -1.04 44 .87 30 - .86 Th j 51 .61 39 - .90 63 1.99 53 - ,95 65 1.47 54 - .80 67 2.09 56 -1.41 81 .95 60 -1.11 14 1.11 29 -1.09 19 .77 30 -1.13 20 1.11 32 -1.52 Factor 23 1.05 60 -1.02 34 1.56 61 -1.15 Five 44 .84 62 -1.15 78 .76 65 -1.07 80 2.06 83 - .83 81 1.12 85 -1.49 86 .96 89 -1.03 19 .68 10 -1.18 20 .87 15 -1.15 23 1.21 22 -1.65 Factor 33 1.05 27 -1.07 35 1.05 29 -1.13 Six 39 .94 32 -1.09 44 1.33 48 - .90 63 .66 69 -1.43 70 1.40 79 -1.07 82 1.03 83 -1.31 TABLE 18 INDIVIDUAL IDENTIFICATIONS OF THE TEN HIGHEST AND TEN LOWEST FACTOR SCORES ON MEMBER MEMORY TESTS Subject Code Factor II Aud i t ory- Baclcward Factor V Auditory Pot* wax'd Factor VI Vi sual Bac: lew ard Number Low High Low High Low High 2 4 - t - 1 0 ~ 14 1 5 - - 19 - t - 2 0 21 - Z Z - - 2 3 26 Z 7 - 29 — - — 30 - - 32 - - 3 3 34 35 39 — 44 48 - 5 1 - • * 53 - 54 - 56 - 69 — - 61 - 6 2 - 6 3 65 - 6 7 69 — 70 78 79 - 80 -1 - 81 82 83 - — 8 5 - 86 89 -

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

School Behavior And Attitudes Of Retarded And Average High School Boys Asconcomitants Of Intelligence And Socioeconomic Status

PDF

Effects Of Success And Failure On Impulsivity And Distractibility Of Three Types Of Educationally Handicapped Children

PDF

Auditory Perceptual Si Factors As Non-Predictors Of Reading Achievement In An Upper-Class And Upper-Middle-Class Population

PDF

Predicting Success In The Study Of Descriptive Linguistics

PDF

Maternal Child-Rearing Attitudes And Developmental Growth Of Rubella Deaf-Blind Children

PDF

Testing Attitudes Toward Physical Handicap: Stereotyping And Objectivityin A Partially Integrated Elementary Program

PDF

Academic And Non-Academic Achievement Of Freshmen And Seniors At Azusa Pacific College

PDF

Prescriptive Teaching As A Supplement To Behavior Modification In The Remediation Of Learning Disorders

PDF

Effects Of Associative And Category Cuing On Recall Of Items From Exhaustive And Nonexhaustive Lists

PDF

Value Orientations Of Parents Of Academically Successful And Unsuccessfulchildren

PDF

An Investigation Of The Use Of Focused Video Tape Feedback In High Schoolgroup Counseling

PDF

Anxiety, Physiologically And Psychologically Measured, And Its Consequences On Mental Test Performance

PDF

The Conditioning Of A Discriminative Stimulus Measured As An Orienting Reaction In Profoundly Retarded Blind Children

PDF

The effect of noise on intellectual performance as related to personality and social factors in upper division high school students

PDF

Confirmation Of Expectancy And Changes In Teachers' Evaluations Of Student Behaviors

PDF

Reading Achievement And In-Grade Retention Rate Differentials For Mexican-American And Black Students In Selected States Of The Southwest

PDF

The Effects Of Two Types Of Experimenter Intervention And Schedules Of Reinforcement On Verbal Operant Conditioning Of Affective Self-References

PDF

The Modification Of Maladaptive Behavior Of A Class Of Educationally Handicapped Children By Operant Conditioning Techniques

PDF

The Relative Efficiency Of Prompting And Confirmation Learning Paradigms

PDF

Conceptualization And Management Of Deviant Children In A Parochial School System

Asset Metadata

Creator Meeker, Mary Nacol (author)

Core Title Immediate Memory And Its Correlates With School Achievement

Degree Doctor of Education

Degree Program Educational Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education, educational psychology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Meyers, Charles Edward (committee chair), McIntrye, Robert B. (committee member), Wagner, Elmer E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-116298

Unique identifier UC11360189

Identifier 6705300.pdf (filename),usctheses-c18-116298 (legacy record id)

Legacy Identifier 6705300.pdf

Dmrecord 116298

Document Type Dissertation

Rights Meeker, Mary Nacol

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA