University of Southern California Dissertations and Theses

Susceptibility To Associative Inhibition Development Of Construct Validity Through A Paper And Pencil Encoding Task Given To Children, Adolescents And Adults

(USC Thesis Other)

Susceptibility To Associative Inhibition Development Of Construct Validity Through A Paper And Pencil Encoding Task Given To Children, Adolescents And Adults

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 71- 27,941 MARGOLIS, Harold Jules, 1933- SUSCEPTIBILITY TO ASSOCIATIVE INHIBITION DEVELOPMENT OF CONSTRUCT VALIDITY THROUGH A PAPER AND PENCIL ENCODING TASK GIVEN TO CHILDREN, ADOLESCENTS AND ADULTS. University of Southern California, Ph.D., 1971 Education, psychology University Microfilms, A X E R O X Company, Ann Arbor, Michigan © Copyright by HAROLD JULES MARGOLIS 1971 THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED SUSCEPTIBILITY TO ASSOCIATIVE INHIBITION DEVELOPMENT OP CONSTRUCT VALIDITY THROUGH A PAPER AND PENCIL ENCODING TASK GIVEN TO CHILDREN, ADOLESCENTS AND ADULTS by Harold Jules Margolls A Dissertation Presented to the FACULTY OP THE GRADUATE SCHOOL UNIVERSITY OP SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Education) June 1971 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA S 0 0 0 7 This dissertation, written by under the direction of /li.?. Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments of the degree of D O C T O R OF P H IL O S O P H Y Dean D a te .../A . . DISSERTATION COMMITTEE hrmart ACKNOWLEDGMENTS The major source of data for this study came from students in the Sioux City and the Los Angeles jCity School Districts. I am grateful for the cooperation lof the participating pupils and staff members. I thank my graduate professors who have been i iinvolved in this doctoral dissertation. I am especially jgrateful to Dr. James Magary, Dissertation Committee I Chairman, for his guidance during the pursuit of my doctoral studies and Dr. Frank Fox, for his encouragement land critique pertaining to the theoretical basis of jthis investigation. To my wife Hannah and to my sons, David and Jonathan, I offer my deepest appreciation for offering - a limitless source of encouragement, patience, and most i of all endurance over the eight years of graduate i i studies and for the six years of research culminating in this dissertation. ii TABLE OP CONTENTS Page ACKNOWLEDGMENTS..... ii LIST OP TABLES .................................. V LIST OP FIGURES ................................. vii Chapter I. THE PROBLEM............................... & Purpose of the Study Importance of the Problem Statement of the Problem Hypotheses Definition of Terms Organization of the Remainder of the Dissertation II. REVIEW OP THE LITERATURE .............. 7 Associative Inhibition Difficulties in Evaluating Presence of Associative Inhibition During Paired- Associate Learning Summary III. METHOD.................................. 20 The Learning Task Experiment #1 Experiment #2 Experiment #3 Experiment #4 IV. RESULTS................................. 44 Experiment #1 Experiment #2 Experiment #3 Experiment #4 Discussion of Results V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS FOR FURTHER INVESTIGATION...... 68 APPENDIXES iii Page APPENDIX A. THE LEARNING TASKS.............. 73 APPENDIX B. DIRECTIONS FOR GROUP ADMINISTRATION OF THE CODING TASK ............ 77 | BIBLIOGRAPHY ................................... 79 iv LIST OP TABLES Table Page 1. Scaled Values (M) of the Paired-Associate Members Selected for the Two Forms of the Experimental Encoding Tasks.....................23 2. Number, Mean and Range of Chronological Age, IQ and Reading Grade of Subjects Assigned to Sequences XY or YX for Experiment #1............25 3. Number, Mean and Range of Chronological Age, and Sequence Assigned by Grade/Age Levels for Experiment #2..............................29 4. Number, Mean and Range of Chronological Age and Sequence Assigned by Grade/Age Levels for Experiment #3.............................. 36 5. Number, Mean and Range of Chronological Age and Sequence Assigned by Grade/Age Levels for Experiment #4..............................42 6. Number, Mean and Standard Deviations of Subjects Performance on Forms X and Y and Parallel Form Reliability According to Sequence of Task Administration for Experiment #1............... 46 7. The Relationship Between Performance on the Form Administered First, Chronological Age and Standardized Test Scores for Subjects in Experiment #1...............................49 8. Relative Level of Difficulty of Parallel Forms as Evaluated by Mean Number of Correct Responses Produced According to Grade/Age Level of Subjects for Experiment #2.................. 50 9* Relative Level of Difficulty of Parallel Forms as Evaluated by Mean Number of Correct Responses Produced According to Grade/Age Level of Subjects for Experiment #3............ 50 10. Number, Mean and Standard Deviations of Subjects Performance on Forms X and Y, and Parallel Form Reliabilities According to Sequence of Task, by Grade/Age Levels for Experiment #2............................... ...52 v Table Page 11. Number, Means and Standard Deviations of Subjects Performance on Forms X and Y, and Parallel Form Reliabilities According to Sequence of Task, by Grade Levels for Experiment #3............................... 53 12. Number, Means and Standard Deviations of Performance on Forms X and Y, and Reliabilities of Forms and Difference Scores of 3rd Grade Subjects When Tested 7 Months Later for Experiment #4...............56 13. Number, Means and Standard Deviations of Performance on Forms X and Y, and Reliabilities of Forms and Difference Scores of Junior College Adults When Tested 4 Months Later for Experiment #4......................57 LIST OF FIGURES I Figure Page I. The Four Sequences of Form Administration Studied for This Investigation..., 2. Experimental Design for Experiment #1---- 3. Experimental Design for Experiment #2----... 34 4. Experimental Design for Experiment #3___ 5. Experimental Design for Experiment #4____ ! I i vii CHAPTER I THE PROBLEM Purpose of the Study The purpose of this investigation was to identify and isolate a variable yielding a measure of one's susceptibility to associative inhibition occurring during a learning task. An attempt was made to develop a con struct type of validity of such susceptibility through constructing alternate forms of the task and developing i the validity construct through administration of the forms to children, adolescents and adults. Importance of the Study Complex behaviors such as learning disabilities, i I mental retardation, manifestations of anxiety, and funct ional language difficulties could be more fully under stood if their underlying learning dynamics could be ! described. For in all these behaviors there appears to j be prior learned behavioral responses that seem to inter fere with the acquisition of new behavior that is more adaptive. The lack of standardization of task materials has 1 been a major obstacle in the development of empirically based theories to explain such complex behaviors. It is apparent to investigators that a severe limitation to theory building is the lack of standardization of task and procedural-related variables. If such procedures were standardized, it would measurably speed the investi gation process. This investigation attempts to provide a proto type of a paired-associate learning task in such form so as to be presented to groups of subjects in paper and pencil form. Such a task may substitute, in special cases, for the classical and tedious paired-associate method requiring individual administration and a great deal of consideration on the part of the experimenter as to the lists construction, administration and scoring Statement of the Problem A major aspect of this investigation was to develop two alternate forms of the paired-associate learning task with the following characteristics. 1. The forms should have characteristics similar to that depicted in classical paired-associate verbal learning tasks. 2. The forms should be easily administered to groups as well as to individuals. 3. The forms should yield a criterion measure 3 that can be scored objectively. 4. The forms should be easily administered to subjects of all school ages. 5. The task should yield reliable scores. 6. The task should yield a sensitive measure of one's susceptibility to associative inhibition. 7. The forms should be efficient in terms of time required to administer and score considering the kinds of information the task yields. Pour experiments were conducted for this investigation. The first experiment consisted of individual administration. The purpose was, in part, to determine whether the forms could be reliably given to children of all school ages, whether the forms would yield comparable scores, and to determine if the forms could yield a criterion measure that could be related to one's susceptibility to associative inhibition. The second experiment consisted of a develop mental cross sectional study of children, adolescents and adults who were administered the tasks in their classroom as a group. The third series of experiments was a cross sectional developmental study of the effects of practice and fatigue. The third experiment also consisted in a further reliability study of the two forms. The fourth series of experiments consisted of a 4 special reliability study of task performance on the part of children and adults over an interval of several months. Hypotheses The major hypotheses are presented below. The designation, H:l, refers to the experimental hypothesis. The designation, H:0, represents the hypothesis in null form. Difficulty Level of Forms H:1— Forms X and Y are of equal difficulty level. H:0— Whenever forms X and Y are administered first, in order, to subjects of any age/grade level, the mean number of correct responses produced by subjects on Form X will be equal to the number of correct responses produced by subjects on Form Y. Effect of Fatigue H:2— No evidence will be observed for a performance decrement as a consequence of previous experience with the same form. H:0— Whenever Form X or Y are administered twice, in rapid succession, to subjects of any age/grade level, the mean number of correct responses produced during the first task will be greater than the mean number of correct responses produced during the second task. Associative Inhibition 5 H:3— Whenever two different forms are administered to subjects of any age/grade level in rapid succession, whichever form is administered second, in order, will result in greater difficulty. H:0— Whenever alternate forms are administered to subjects of any age/grade level in rapid succession, the mean number of correct responses produced during the 2nd task will be greater than, and not equal to the mean performance on the task taken first, in order. Definition of Terms The following terms were used in this study with these implied meanings: 1. Associative Inhibition. Simply stated— when any two items, as a and b, have been associated, it is more difficult to form an association between a and third item m. 2. Learning Task. The learning task for this investigation consisted of two "parallel forms" designed so that whichever form is given second, in order, will result in more difficulty. The learning task shall be described in Chapter III and can be found in Appendix A. 3. Parallel Forms. The parallel forms, X and Y, of the learning task each consist of a "list" of four paired-associate members to be learned and produced. 4. Sequence. The sequence refers to the order 6 in which the forms, X and Y, were presented to the subjects. Sequence XY for example, indicates that Form X was given first, in order, and was followed by administration of Form Y. 5. Age/Grade Level. These terms describe the sample selection. Cross sections of grade levels were selected so that the age ranges do not overlap. Organization of the Remainder of the Dissertation The remainder of this study reviews literature relevant to decremental factors occurring during learning, following practice and upon recall. Emphasis will be on the procedural, task and subject variables that complicate research of associative inhibition. Following is a description of the learning task, sample procedures and methods of analysing the data. Next, the presentation of the data, accompanied by an interpretation and discussion are given. Lastly a summary is presented with conclusions and recommendations for further study. 7 CHAPTER II REVIEW OF THE LITERATURE The purposes of the bibliographic search for this chapter are: (1) to determine the background and history of developments in verbal learning theory through out the fields of education and psychology pertaining to decremental factors occurring during learning, follow ing practice, and upon recall; (2) to determine the historical background leading to present concepts of associative inhibition; and (3) to review material that might identify the procedural, task and subject variables influencing performance on paired-associate learning tasks where associative inhibition is expected. Associative Inhibition Although aneddotal evidence was published earlier by Munsterberg (1889) , the first systematic account of the phenomena of what we refer to today as "associative interference” or "associative inhibition" was reported by MQller and SchOmann (1894). This lengthy monograph appearing in a German publication was concerned with differences among subjects learning various verbal and musical tasks. During their investigation of task 8 variables, they had discovered a means of demonstrating that which, Munsterberg earlier alluded to, but only in anecdotal form. Mttller and Schttmann noted the marked difficulty their subjects encountered with the list of nonsense syllables. The subjects were aksed to learn the series of syllables followed by the administration of the same list, but with the modification of the list so that the pairs of associates to be learned were re paired with one another. Mttller and Schttmann referred to this reorganization of the paired syllables, in the case of the second successive list to learned, as an umstellungsreihen, or transposition series. Six years after the Mttller and Schttmann study of 1894, G.E. Mttller, in collaboration with A. Pelzecker reported a lengthy study, also appearing in a German journal. In this study appearing in 1900 Mttller and Pelzecker developed a psychological theory of forgetting, perhaps considered an outgrowth of the earlier study by Mttller and Schttmann in 1894. According to this theory a period of fixation is required following learning to allow the newly learned "traces" to become consolidated. Mttller and Pelzecker posited that interpolation of other activity disrupted this process of fixation. Thus, the underlying theory was provided for a theory of retro active inhibition. Although Mttller and Schttmann, in this publication, 9 gave little attention to the systematic alteration of learning paradigms, they did arrive upon a very interest ing conclusion or generalization, that over time has taken the stature of a law. They were able to state the following generalizations ....Ferner ist mit der Moglichkeit zu rechnen, dass die Erlernung der Umstellungsreihen in merkbarem Grade durch associative Hemmung beeinfluss worden sei. Ist n&mlich eine Silbe a mit einer anderen, bisher auf a gefolgten Silbe & associiert, und kommt nun in einer neuen Silbenreihe die Silbe a so vor, dass nicht b, sondern eine andere Silbe m auf dieselbe folgt, so kann die Tendenz zur Reproduktion von b, welche sich an die Silbe a anschliesst, sei es durch wirklich Reproduktion von b (associative Hemmung durch aktuell ReproduktionX, sei es durch einen nur unbewusst verlaufenden Vorgang (associative Hemmung durch nur virtuelle Reproduktion) die neu zu stiftende Association zwischen a und m erschweren*- The generalization, simply stated— when any two items, as a and b, have been associated, it is more diffi cult to form an association between a and a third item m— appears to the author to be a prototype of what is some times referred to as the Mttller-Schttmann Law of associative inhibition. XI As there has been semantic problems in regard to the terminology originally used by Mttller and Schttmann, the author has presented the original German. Following is the author's translation into English: ....Furthermore the possibility must be considered that the learning of transposition series, to a considerable degree, has been under the influence of associative inhibi tion. When a syllable a becomes associated with another syllable, b (up to now has followed a), and then appearing in a new series of syllables in whicfT not b, but another syllable, m, follows the same, thus in this manner the in clination Tor the reproduction of b which has become connected with the syllable a, be I t through associative 10 The early investigations of associative inter ference and negative transfer appear to have been under taken so as to provide support for the "retroactive inhibitory theory of forgetting" promoted by Mttller and Pelzecker (1900). From 1900 to the present day the retroactive paradigm has been widely used for developing the retroactive theory of forgetting. The retroactive paradigm can be best described by the operations outlined below: Task 1 T&sk 2 Recall Prior Interpolated Original Learning Learning Learning Experimental Group X X X Control Group X - X If in the recall the experimental group has better performance, the effect can be termed "retroactive facilitation*" If the experimental group has less recall, the effect can be termed "retroactive inhibition." Retroactive effects during learning has been studied in ways other than supporting the theory of for getting described above. Through efforts by Ausubel (1957), inhibition, through a real reproduction of b (associative " inhibition through immediate reproduction), or be it I through an involuntary development (associative inhibition ithrough skillful reproduction)— that renders the association to be established between a and m more difficult. 11 in his attempts to develop a theory of "Meaningful Learning," and studies by Odom, Shapira and Attwell (1966), Baumeister, Hawkins and Holland (1965) and McMannis (1967), which involve individual differences on the subject's part, attention has been diverted from a theory of forgetting per se, to the identification of individual differences. Empirical studies involving proactive effects during transfer did not appear until Atwater (1953) presented the notion that proactive inhibition also plays a vital role in the process of forgetting. The proactive paradigm can be best described by the operations outlined below: Task 1 Task 2 Recall Prior Interpolated Task Learning Learning 2 Experimental Group X X X Control Group X X If the experimental group performs better on recall (Task 2) than the control group, the effect can be termed "proactive facilitation*" In the event that the experi mental group performs less well then the control group bn recall (Task 2), then the effect may be termed ["proactive inhibition*" Studies by Underwood (1949), Atwater (1953), 12.. Mandler and Heinemann (1956), Iwahara and Sugimura (1960), Postman (1962), and Ausubel (1969) were all con cerned with the investigations of inhibitory processes as a function of the degree of prior learning. These studies were primarily concerned with proactive effects. Underwood (1949) and Atwater (1953) discussed the importance of proactive effects as they relate to for getting and interference theory. Difficulties in Evaluating Presence of Associative Inhibition During Paired-Associate Learning The procedures employed, method of design, and selection of learning materials will have a marked effect on the outcome of any learning experiment. The balance of the chapter consists in a brief summary of the kinds of problems encountered in the use of paired-associate methods accounting for difficulties in theory building. Task-Related Variables Warm-Up Phenomena McGeoch and Irion (1952) suggest that benefit from warm-up prior to recall is a function of the amount of warm-up activity, the similarities of behaviors involved in the warm-up and the learning activity, and the time interval between the warm-up and the recall. Vividness Van Buskirk (1932) demonstrated the phenomena 13 when he. had subjects learn a list of nine syllables in which ordinary black-on-white paper was used. In the experimental list, one syllable was printed in red on a green background. This syllable was retained better than the corresponding syllable in the control list. von Restorff Effect von Restorff (1933) demonstrated that the "isolated" item in a list will be better recalled. Gibson (1940) explained this effect in terms of generalization and differentiation. Length of List Rock (1957) pointed out that the longer the length of a list, the greater is the degree of reinforcement of pairs were associated early in practice trials. Criterion of Learning Generally, the higher the degree of learning the greater is the retention. However, diminishing returns are reported by Rubin and Rabsin (1941). Stroud and Schoer (1959) discussed the relationship between various criteria of learning and individual differences among subjects in their learning ability. Affective Character According to Hovland (1962) , research on this topic stems largely from Freud's conception of forgetting as an active process that protects the individual from | disturbing memories. On the basis of more frequent recall j i ! of pleasant experiences, many studies have concluded j that pleasant material is better retained than unpleasant, j j Meaningfulness Since the time of Ebbinghaus's work in 1885 the relevance, significance, and construction of the verbal material has made a great difference as to how readily the material could be learned and be reproduced | later. Schulz and Underwood (1960) have compled an excellent summary of the implications of meaningfulness ! | for verbal learning. The position was taken that the i j frequency with which verbal units have been experienced | is the fundamental variable responsible for the character- I istics which have been used to define meaningfulness. 1 ! As a test of this notion, various types of learning I material, differing in meaningfulness, were translated j | into frequency terms. Then a hypothesis was developed ; i to explain how frequency differences mediate the facts j relating meaningfulness and learning. This generalization, referred to by the authors as the spew hypothesis, states that the order of availability of verbal units is directly related to the frequency with which the units have been experienced. Other things being equal, there fore, the more frequently a verbal unit has been i experienced, the more quickly will this become a response in a new associative connection. 15 Procedural Variables Distribution of Practice Investigations regarding massed vs. distributed practice has yielded results that distributed practice leads to better retention. However, the interaction with meaningfulness, length of list, and other task and subject variables must be considered. Criterion of Learning The criterion of learning that an experimenter selects will often influence conclusions drawn from a learning experiment. Recall scores as described by Vergason (1965) were found to be inadequate as a criterion measure in conjunction with mentally retarded youngsters. Some studies incorporate the recall score, some incorp orate the recognition memory (matching type), other experimenters will utilize the reconstruction method (rearrangement) and some studies report the method of retained members (typical classroom method). Often number correct is used and sometimes errors. Differing criteria of "learning" is often observed even in the same experiment without explanation as to what the conclusions might have been if another criterion had been used. The savings method was deployed by Ebbinghaus (1885). The subject relearns the material he has already mastered. The relearning is compared with the original learning. The savings score can be presented as 16 as a percentage of material learned, number of trials to reach criterion, or number of errors produced. Subject Variables Learning Ability Mflller and Schttmann (1894) obtained a positive relationship between the time taken for their subjects to learn lists of syllables and the length of time necessary for them to relearn them to the same criterion 24 hours later. Since the turn of the century a great deal of work has taken placed in this essential area. Notable studies were reported by Pyle (1911) who studied the relationship of words, objects, pictures, syllables and connected, thought passages. Pyle found that faster learners are superior in immediate memory for these kinds of materials. However, when these materials have been learned to a more difficult criteria, the retention differences are not so great. Other classics are by Lyon (1914), Gillette (1936), McGooch and Irion (1952), Underwood (1954) , Stroud and Schoer (1959) and Stroud (1962). These investigators have all been confronted with the problems discussdd above regarding task, procedural and subjects variables. Effects of Motivating Aspects of Drive (D) Spence and associates (1953) claimed they were able to demonstrate a relationship occurring between rate of learning, "drive states," and presence of "competing 17. responses" occurring during PAL. These findings were derived from a Hullian-Spence conception of drive (D) as being a function of the strength of the emotional response (Spence, 1958). Janet (Taylor) Spence (1956) proposed the hypothesis that the level of this internal drive state would be reflected by the responses in an inventory referred to as the Taylor Manifest Anxiety Scale (MAS) . It has often been subsequently demonstrated that those individuals scoring high in the scaled direction of the MAS tend to require more trials to reach a learn ing criterion in a "complex task." The Spence-Hullian model would predict that when competing responses are operative, of which only one response is scored as correct by the experimenter, the effect in D is considered to depend upon the number and comparative strengths of the competing tendencies. Many investigators have attempted to relate various indices of "anxiety" with performance on various kinds of paired-associate tasks. Because of the various kinds of procedures, selection of materials and various subject variables, it has been difficult for investigators to make simple generalizations regarding the behavior of low and high anxious people during paired-associate learning tasks. Bernstein and Dorfman (1959) and Margolis (1966) have reported reduction of associative interference with 18 administration of meprobamate. Margolis (1966) found reduction of associative inhibition on a A -B,C transfer task when subjects were given a mild tranquilizer clinically reported to reduce "anxiety" Introversion-Extroversion Eysenck (1957) reported a theory of personality supported, in part, by findings that performance on a PAL task (where interference is expected) that subjects scores are differentially related to their scores on a test for introversion and extroversion. Fatigue Laslett (1928), Weiskotten and Ferguson (1930), Warren and Clark (1937) and Williams, Lubin and Goodnow (1959) reported relationships between fatigue and perform ance on verbal tasks. Mental Ability Cassel (1957) , Pryor (1960), Hawkins and Baumeister (1965), McMannis (1968) and Gilliland (1969) reported studies of the relationship between ability levels and various indices of learning performance. Such studies are difficult to perform because of the lack of standardization of materials, task-related, procedural- related and subject-related variables. Summary Since the development of the perseverative-trace theory of forgetting, attempts have been made throughout the years to "explain" forgetting from the standpoint of inhibitory phenomena occurring during and following practice. This tendency among investigators to explain forgetting from the standpoint of inhibition seemingly acted to take attention away from the study of associative inhibition per se. The attention given to the Mftller and Pelzecker study of 1900 impeded the refinement of methods and techniques of measuring effects of associative inhibition. Because of the wide differences in the kinds of materials, difficulty of materials, intratrial intervals, schedules of reinforcement and practice, criteria of learning, and little emphasis upon individual differences among subjects, it is often impossible to make any meaningful comparisons among studies or generalizations regarding individual differences among children in their susceptibility to associative inhibition. That associative inhibition occurs, is not disputed the problem is that nothing has been reported as to what individual differences among people account for the greater or lesser degree of its presence in their per formance . CHAPTER III METHOD The purpose of this chapter is to present a description of the learning task, sampling selection, procedures, data analysis and the experimental design. The first section of the hhapter describes the two forms of the experimental encoding task in terms of their development, meaningfulness (M) and performance expectations of subjects when administered in varying sequences. The second section of the chapter describes the series of four experiments. Each experiment will be described in terms of the sample of subjects, sequence of administration of forms, procedures, methods of data analysis and the experimental design. The Learning Task For this investigation, two lists of paired- associate letters were developed. The learning task was of the same transfer paradigm type referred to by Porter and Duncan (1953) as the A - Br (repaired) type. This paradigm was reported by Porter and Duncan to produce the greatest amount of negative transfer when compared to other transfer paradigms, including the A - B,C type. 20 f i i I jRefer to Appendix A for a copy of the learning task. Specific letter combinations were selected on the I basis of the M value of the two letter pairs scaled by Underwood and Schulz (1960) . The scaling procedures and jtable of values can be found in their book. Meaningfulness and Verbal Learning.1 j | The combination of letters selected for this invest igation were so chosen so that the total sum value of leach of the two forms would approximate one another as I i [closely as possible in terms of the sum value. Refer to j Table 1 for a description of the scaled and M values. The combination of letters were so selected so i jthat greater skill in reading ability might reflect i I itself in terms of better performance on each of the two : l forms. I I | The experimental encoding tasks were also developed j [so as to maximize the likelihood for the occurrence of associative inhibition of the Mttller-Schtimann type such that when alternate forms are to be given in rapid succession, whichever form would be given second, in order, would result in more difficulty. 1 A total of 273 subjects were presented with all possible two letter combinations and were asked to write down for leach the two first letters that came to mind. Tables were |presented by the authors giving the number of subjects responding with each letter to each stimulus member. For example, to the two letter stimulus SA, 67 subjects respond ed with T, 21 subjects with P, 8 subjects with W, etc. 22 Four sequences of form administration are studied throughout the four experiments. The four sequences of form administration studied for this investigation are described in Figure 1. Experiment #1 The purpose of this pilot study was to determine whether the task could be readily administered to individuals of all school ages, yield a criterion measure that could be scored objectively, and yet be efficient in terms of time required to administer and score considering the kinds of information it yields. The Sample The sample of this study consisted of 63 children and adolescents attending the Sioux City Schools, Sioux City, Iowa* These subjects, 7 through 17 years of age, were referred by their school principals for psychological services. These subjects are a sample of the total population of about 550 pupils identified during the school years, 1964-1966 by the school faculty because of learning and/or behavior problems observed in the classroom. Refer to Table 2 for a description of the number, chronological age, IQ and reading ability of the samples. The 63 subjects do not represent a random selection of all the students referred from the schools with learn- 23 Table 1 Scaled Values (M) of the Paired-Associate Members Selected for the Two Forms of the Experimental Encoding Tasks Form X Stimulus Stimulus Response! 21 12 91 31 Sum M 122 127 Sum M Figure 1 The Four Sequences of Form Administration Studied For This Investigation Sequence Form Taken First Form Taken Second XY X Y YX Y X XX X X YY Y Y 24 ing and/or adjustment difficulties; they did, however, offer an opportunity for a pilot study to investigate some preliminary aspects of the construct validity of the task. Procedures The subjects were administered the two experi mental forms midway through their psychological testing (upon completion of their individual intelligence testing. The forms were administered in the manner prescribed in Appendix B, with the exception that wording was appropriate for individual administration. Half of these subjects were given Sequence XY and the other half were given Sequence YX. Every other student was given one or the other sequence in order of their appear ance for psychological testing. All subjects were given the training task beforehand so as to acquaint themselves with the method and to provide a more common learning set. Children that could not follow directions for this training task were not given the experimental lists and were deleted from all data. These children who were not able to complete the training task, about 3 or 4 in number, consisted of children about 6 or 7 years of age who were later found to be classified as qualifying for special education under provision for the mentally retarded. Table 2 Number, Mean and Range of Chronological Age, IQ and Reading Grade of Subjects Assigned Sequence XY or YX For Experiment #1 Sequence N Chronological Age IQ Reading Grade"" Mean Range Mean Range Mean Range XY 33 10.8 7.3— 16.3 92.8 65— 121 3.8 1.5— 10.1 YX 30 11.5 7.3— 17.5 92.8 64— 125 4.6 1.2— 14.2 26 Data Analysis A study was made of the difficulty level of the two forms, as measured by the number of correct responses produced within 120 seconds. A t-test for independent samples was utilized. A study of the difference scores, i.e., the sequencing effect yielding the criterion measure of susceptibility to associative inhibition for each subject, was made for the two sequences. A t-test for paired measures was utilized and a directional one tailed test was made of the difference scores. A study was also made of the parallel form reliability of Form X and Form Y through use of the Spearman product-moment correlation. Experimental Design Figure 2 describes the experimental design for Experiment #1. The designation, "Sequence XY" indicates for this group of subjects Form X was given first and immediately followed by administration of Form Y. The designation, "Sequence YX,? indicates for this group of subjects Form Y was given first and immediately followed by Form X. The correlation coefficients each give an estimate of the parallel-form reliability of the forms, one regarding Sequence XY and the other regarding Sequence YX. Figure 2 Experimental Design for Experiment #1 kge/Grade Level Sequence Form Taken First Form Taken Second Reliability 1-12 XY X Y rxy YX Y X ryx 28 Experiment #2 The purpose of this experiment was to replicate the pilot study to determine if the predictions would hold for normal children of various age levels, normal adolescents and normal adults when the experimental forms are administered to groups. The Sample For this experiment 170 subjects were drawn from populations of children, adolescents and adults. The three populations differ from one another in their level of socio-economic status, racial and cultural background. The three populations, however, appear adequate for approaching the developmental problems discussed in this investigation. Refer to Table 3 for a description of the number of subjects assigned to sequences for the various age/grade levels. The sample of 108 children were selected from the 1st, 3rd, and 6th grades of the Yorkdale Elementary School of the Los Angeles Independent School District. The age range of the samples do not overlap one another. This sample of subjects represents about a third of the total population of this relatively small school in the Highland Park District of Los Angeles. The sample represents about half of the entire student body of thes three grades that were present that day in school. The Table 3 Number, Mean Chronological Age and Range of Subjects and Sequence Assigned by Grade/Age Levels For Experiment #2 Age/Grade Level Sequence XY Sequence YX N ' Mean Range N Mean Range 1st 22 6.9 6.3— 7.6 21 6.8 6.2— 7.6 3rd 12 8.9 0 0 t u > 1 1 vo • o 9 0 0 * 0 0 8.1— 9.2 6 th 12 11.5 10.9— 13.0 9 11.9 10.9— 3.0 11th 15 16.9 15.8— 18.1 13 16.9 15.8— 18.1 Jr. Coll. 45 27.0 18.0— 53.0 30 particular class sections were selected on the basis of discussion with principal and teacher as to which class section best represent that particular grade according to ability and skill levels. The school was selected as it seemed, according to the principal and to the experimenter, that it represented fairly well the mainstream of socio-economic classes throughout the Los Angeles City School District and a general cross section of minority group children. The performance of the school on group IQ tests fell well within the average range and achievement test scores of pupils appeared typical of schools in the Los Angeles City School District for such mean IQ's. The 28 adolescents were selected from an 11th grade class of U.S. History at Grant High School of the Los Angeles City School District. Upon discussion with the teacher, it was agreed that the socio-economic status of the attendance area, and the class as a whole appeared largely middle class, largely Caucasian and college bound. The class represents a random selection of 11th graders from that high school with an academic major. The 34 adults were selected from two sections of an evening division course of General Psychology at Los Angeles Trade-Technical College, a junior college of the Los Angeles Community College District. The 31 classes from which these students were drawn are required for an Associate of Arts Degree and are transferable to California colleges and universities. Most of the students are working full-time during the day in various semi skilled positions, are Negro and live in small homes or apartments in southwestern Los Angeles. Procedures Permission and discussion ensued among the teachers and principal of Yorkdale Elementary School as to the appropriateness of the sample, availability and timing of test administration. A cross section of the 1st, 3rd and 6th grades were selected at the end of the year, 1970 that represented regular classes of pupils for that grade and age level. The teachers were oriented as to the purpose of the testing and were asked for about 15 minutes of classroom time. Upon introduction by the teacher, the experimenter proceeded upon the standard method of task administration described in Appendix B. Difficulties were encountered in the group testing of the 1st grade, largely because not enough care was taken on the part of the experimenter to pre sort the different forms, so as to insure that the subjects were administered the forms in proper sequence. The alternate section of the 1st grade was tested and subsequently all groups of subjects had "kits" of the forms prestapled so as to insure against such a mix-up on the part of the subjects or experimenter. The methods for the group administration followed quite closely that described in the individual administration. Subjects that could not demonstrate an understanding of the pilot task were not given the experimental forms and were deleted from all data. About 5 subjects from the 1st grade class were not able to give indication of mastery of the training task, even when given help by the teacher and experimenter, in the five or ten minutes allotted for instructions. The total amount of classroom time required for instructions, administration of the training task, administration of the experimental forms and collection of the materials was somewhat less than 20 minutes. The same procedures were utilized in the sample of 11th graders. The experimenter was introduced to the class, the training task was quickly mastered and the experimental list sequences were given and collected. The time allotted for performance was limited to 90 seconds for each form. The total classroom time re quired was less than 15 minutes. The same procedures were utilized in the sample of college students as that given to the children and adolescent groups. The training task was quickly mastered and the experimental sequences were assigned to subjects 33 in random order. The time allotted for performance was limited to 90 seconds. The total classroom time required was somewhat less than 15 minutes. Refer to Table 3 for a description of the number of subjects for each grade/age level assigned to various sequences. Data Analysis The analysis was similar to that of the pilot study, including a study of the difficulty level of the two forms, a study of the difference scores, and a study of the reliability of parallel forms. These studies were made for each age/grade level. Experimental Design Figure 3 describes the experimental design for Experiment #2. All grade levels were given both Sequence XY and YX with the exception of the junior college adults who vere given only Sequence XY. Experiment #3 After completion of Experiment #2, an attempt was made to estimate the immediate-test-retest reliability of the two forms, as well as to study the effect of practice and fatigue that might result from a subsequent administration of a form. The experiment also offered an opportunity to compare the relative difficulty level of the two forms for different age and grade levels of Figure 3 Experimental Design for Experiment #2 Age/Grade Level Sequence Time/Seconds Form Taken First Form Taken Second Reliability 1st XY YX 120 X Y Y X rxy ryx 3rd XY X Y rxy YX 120 Y X. ryx 6th XY X Y rxy YX 120 Y X ryx 11th XY YX 90 X Y Y X rxy ryx Junior Coll Adults XY 90 X Y rxy 35 subjects. The Sample The sample for this series of experiments con sisted in 171 subjects, all of whom were naive regarding the experimental forms. Refer to Table 4 for a description of the mean age and ranges and number of subjects assigned to sequences for various age/grade levels. A sample of 110 subjects were selected from the 1st, 3rd and 6th grade classrooms at Yorkdale Elementary School during the middle of the year {January, 1971). This sample of children from Yorkdale Elementary School represented about half of all children attend ing these grades at the sbhool. The classes were selected so as to represent a typical cross section of children from that school. The school population was described earlier during the discussion of Experiment #2. A sample of 31 11th grade students in a required section of U.S. History were selected to represent the adolescent population. These 11th grade students, as in the sample for Experiment #3 were all in an academic program, college bound program attending Grant High School. The classroom teacher described the class as largely middle class, from middle siaed homes, Caucasian with fathers in small businesses, professions or sales. The junior college adult sample was similar to Table 4 Number, Mean Chronological Age and Range of Subjects and Sequence Assigned by Grade/Age Levels for Experiment #3 Age/Grade Sequence XX Sequence YY Level N Mean Range N Mean Range 1st 22 6.6 6.3— 7.6 21 6.7 6.2— 7.6 3rd 10 8.7 8.3— 9.2 12 8.7 8.1— 9.2 6th 30 11.2 10.4— 12.6 27 11.2 11.1— 12.6 11th 15 16.5 15.5— 16.8 16 16.4 15.3— 17.1 Jr. Coll. Adults 13 29.2 19 — 49 17 28.6 19 — 46 37 that described in Experiment #2, but from another section of General Psychology, the following semester. A sample of 45 adults were enrolled and appeared very similar in background to that described in Experiment #2. Procedures Procedures were similar to those employed for Experiment #2. Permission and discussion again ensued among the teachers and principal of Yorkdale Elementary School as to the appropriateness of the sample, avail ability and timing of test administration. The sample for this series of experiments were tested during the middle of the year (January, 1971) and represented regular classes of pupils for that grade and age level. The teachers were oriented as to the purpose of the testing and were asked for about 15 minutes of classroom time. Upon introduction by theteacher, the experimenter proceeded upon the standard method of administration described in Appendix B. The experimental forms were stapled beforehand forming kits. Every other pupil was given a different sequence. About 3 subjects from the 1st grade were not able to complete the training task and were not given the experimental task. The same procedures were utilized in the case of the sample of 11th graders. The experimenter was introduced to the class. Performance time for each form iwas limited to 90 seconds. The random assignment of task sequences, instruction administration of the forms and its collection required less than 15 minutes of classroom time. The same procedures were utilized for the sample of college students as that given to the children and adolescent group. The training task was quickly mastered and the experimental sequences were assigned to the subjects in random manner. Performance time for each form was limited to 60 seconds. The total classroom time required was less than 15 minutes. Data Analysis A study was made of the immediate test-retest reliability of the two forms. The Pearson product-moment correlation yielded test-retest reliability coefficients for each of the two sequences. Paired measures provided the data for evaluation of the effects of fatigue or practice and tested by a t-test for paired measures on a one-tailed test. The relative difficulty level of the two forms were tested by a t-test for independent samples by mean comparison of number of correct responses for whatever list was taken first, in order. These statistical studies were made for each grade level. Experimental Design Figure 4 describes the experimental design for Experiment #3. All grade levels were given Sequences XX Figure 4 Experimental Design for Experiment #3 Age/Grade Level Sequence Time/Seconds Form Taken First Form Taken, J Second Reliability 1st XX X X rxx YY 120 Y Y ry y 3rd XX X X rxx YY 120 Y Y ryy 6th XX X X rxx YY 120 Y Y ry y 11th XX X X rxx YY 60 Y Y ryy Junior Coll, XX X X rxx Adults YY 60 Y Y ryy 40 Experiment #4 Experiment #4 consisted of a study of the relia bility of the forms over a lapse of several months' duration. It also consisted in a reliability study of the difference scores occurring over a lapse of several months' duration. The Sample For this experiment, 12 pupils from the 3rd grade who had been given either Sequence XY or YX during June of 1970 (included in the data for Experi ment #2) were followed up and given Sequence XY or YX seven months later when in the 4th grade. A special study was undertaken of these 4th graders to determine the reliability of the lists and the reliability of the difference scores with a long lapse of time occurring between testing sessions. Refer to Table 5 for a description of the subjects, age/grade level and sequence assigned. The second sample consisted of 14 adult students attending the fall semester of the Evening Division at Trade-Technical College who had been given Sequence XY during September of 1970 (included in the data for Experiment #2) and given Sequence XY again four months later in January of 1971. Refer to Table 5 for a 41 description of the subjects, level and the sequence assigned. Procedures The administration procedures and time allotted for performance were identical with their previous testing. The 4th grade pupils were assigned to either sequences XY or YX on the basis of random assignment. The college students were given Sequence XY, the same sequence they had been given four months earlier. Experimental Design Figure 5 describes the experimental design for Experiment #4, The 3rd/4th grade sample represent a reliability study of parallel forms and a reliability of the difference scores over a period of seven months. The difference scores are based upon parallel forms. The junior college adult sample represents a reliability study of test-retest reliability and the reliability of the differences over a period of four months. The difference scores for the junior college adult sample are based upon similar forms (test-retest). Table 5 Number, Mean and Range of Chronological Age of Subjects Assigned by Grade/Age Levels for Experiment #4 Age Srade Level Prior Testing Subsequent Testing Chronological Age Chrono]Logical Age Sequence N Mean Range Sequence N Meani~ Range 3rd XY 7 8.9 8.6 — 9.2 XY 4 9.4 9.2— 9.5 YX 3 9.5 9.4— 9.7 YX 6 9.0 8.8 — 9.2 XY 3 9.5 9.5— 9.6 YX 3 9.6 9.5— 9.8 Junior College XY 14 27.9 22 — 39 XY 14 28.1 22 — 39 Figure 5 Experimental Design for Experiment #4 Age/Grade Prior Testing Subsequent Testing Level Sequence Lst Form 2nd Form 1st Form 2nd Form 3rd/4th XY X Y X Y YX Y X Y X Junior Coll. XY X Y X Y CHAPTER IV , i RESULTS I ! i This chapter presents the results of this invest!- : gation and evaluates the null hypotheses relative to these jreaftlts. The four series of experiments are presented in order with the null hypotheses relative for each experi- i ment. Levels of statistical significance fibr all testing of i null hypotheses was set at the .05 level; consequently this significance level was set as the maximum probability j I at which the null hypotheses could be rejected. The I chapter will conclude with a discussion of the results in terms of the development of the construct validity of the task and relevant aspects of past and recent research literature* Experiment >1 i | Difficulty Level of the Forms This null hypothesis stated that whenever Forms X and Y are administered first, in order, to subjects i of any age/grade level, the mean number of correct responses produced by subjects tin Form X will be equal to the number of correct responsespproduced by subjects on 45 Form Y. The difference between mean performances of Form X and Form of 5.4 was not significant as tested by a t-test for independent samples (P=.05). It was concluded that the two forms appear to be of equal difficulty level for the sample of subjects selected for Experiment #1. Refer to Table 6 for a description of the data resulting from Experiment #1. Associative Inhibition This null hypothesis stated that whenever two different forms are administered to subjects of any age/grade level in rapid succession, whichever form is administered second, in order, will result in greater difficulty as measured by the mean number of correct responses produced. One-tailed t-tests for paired measures controlling for individual differences indicated for both sequences XY and YX that the null hypothesis must be rejected (P=.05). It was concluded that when alternate forms were administered to the subjects selected for Experiment #1, whichever form was given second, in order, resulted in more difficulty. Refer to Table 6 for a description of the results of Experiment #1. Reliability of Parallel Forms The obtained coefficients of correlation of .94 for Sequence XY (*xy) and .90 for Sequence YX (ryx) both resulted in correlation coefficients of such magnitude that the probability of such an occurrence by chance to Table 6 Number, Mean and Standard Deviations of Subjects Performance on Forms X and Y and Parallel Form Reliability According to Sequence of Task Administration For Experiment #1 Sequence N List Takeij First List Taken Second Difference t-ratio Parallel Form Reliability Form Mean S.D. Form Mean S.D. Mean D S.D. XY 33 X 45.0 24.6 Y 30.8 16.0 14.2 17.3 4.70* rxy=s939* YX 30 Y 50.4 25.8 X 31.4 21.2 9.0 11.5 429* ryx=903* * P < .05 47 be beyond the .05 level. Refer to Table 6 for a description of the results. Correlational Analysis The Pearson product-moment correlation coefficients between the subjects peformance on the form given first and chronological age, mental age derived from WISC IQ scores, reading grade as derived from the Wide Range Achievement Test and raw scores to the WISC Encoding Subtest were all of such magnitude that the probability of such an occurrence by chance was beyond the .05 level. Refer to Table 7 for a description of the magnitude of the coefficients of correlation. Experiment #2 Difficulty Level of the Forms This null hypothesis stated that whenever Forms X and Y are administered first, in order, to subjects of any age/grade level, the mean number of correct responses produced by subjects on Form X will be equal to the number of correct responses produced by subjects on Form Y. Such mean differences between forms as that obtained from the 1st grade, 3rd grade and 6th grade samples could have occurred by chance, the data forcing us to accept the null hypothesis of no significant difference for these grades (P=.05). The t-test applied to the 48 mean difference of the performance of the 11th grade indicated the null hypothesis of no difference between forms must be rejected. The data suggests in the case of the 11th grade sample that Form Y was more difficult, at least in terms of number of correct responses subjects were able to produce. Refer to Table 8 for a description of the data. Associative Inhibition This null hypothesis stated that whenever two diff erent forms are administered to subjects of any age/grade level in rapid succession, whichever form is administered second, in order, will result in greater difficulty as measured by the mean number of correct responses produced. The null hypothesis of no difference must be accepted for the performance of subjects of the 1st grade (P=.05). The two sequences, XY and YX, failed to produce the ex pected differences. These results appear to force the conclusion that the generalizations of susceptibility to inhibitory effects as measured by this task for the 1st grade is not warranted. The t-tests for paired measures controlling for individual differences indicated that the mean differences among the performance of subjects selected from the 3rd grade, 6th grade, 11th grade and junior college adults approach levels of significance when their respective data for sequences XY and YX are combined (P=.05). The null hypotheses for these grade 49 Table 7 Relationships Between Performance on the Form Administered First, Chronological Age and Standardized Test Scores For Subjects in Experiment #1 Measures Product-Moment Coefficient of Correlation N Chronological Age .71* 63 Cental Age .70* 63 Reading Grade .77* 59 Nechsler Encoding Subtest .93* 53 *P <T.05 Table 8 50 Relative Level of Difficulty of Parallel Forms as Evaluated By Mean Number of Correct Responses Produced According to Age/Grade Level of Subjects for Experiment #2 Age/Grade Level Form X Form Y r . Mean Difference t Ratio N Mean S.D. N Mean S.D. 1st 22 13.0 6.9 21 13.3 6.2 0.3 0.15 3rd 1 2 si. V 23.5 9 51.2 18.2 0.5 0.06 6 th 8 67.3 19.4 9 73.2 19.8 5.9 0.58 11th* 15 63.8 13.3 13 ~9'4TT 29.6 io.3 2.2o** Jr. Coll. * 45 89.9 14.7 > * 90 second trial ** P .05 Table 9 Relative Level of Difficulty of Parallel Forms as Evaluated By Mean Number of Correct Responses Produced According to Age/Grade Level of Subjects for Experiment #3 Age/Grade Form X Form Y Mean t Level N Mean S.D. N 1Mean S.D. Difference Ratio 1st 15 19.1 7.6 6 20.8 8.5 0.5 0.05 3rd 10 48.5 17.3 12 34.4 15.3 14.1 1.95 6th 30 60.7 26.9 2? 67.4 20.2 13.3 2.41*** 11th* 15 727T" 6.5 16 73.1 6.5 11.6 0.36 Jr. Coll. ** 13 58.4 l6.4 17 51.9 17.3' 6.5 0.79 * 90 second trial ** 60 second trial *** P -05: 51 levels must be rejected. Refer to Table 10 for a description of the results. Parallel Form Reliability When the data for each age/grade level was combined (Sequences XY and YX) , all coefficients were significantly different from zero (P=.05). Refer to Table 10 for a description of the reliability coefficients for Sequence XY (rxy) and Sequence YX (ryX) for all age/ grade levels tested. Experiment #3 Difficulty Level of the Forms This null hypothesis stated that whenever Forms X and Y are administered first, in order, to subjects of any age/grade level, the mean number of correct responses produced by subjects on Form X will be equal to the number of correct responses produced by subjects on Form Y. Mean differences between forms as obtained from the 1st grade, 3rd grade, 11th grade and junior college adult samples could have occurred by chance, i.e., the data forcing us to accept the null hypothesis of no significant differences occurring for these age/grade levels (P*».05). In the case of the 6th grade sample it appeared that Form Y resulted in significantly less number of correct responses produced, suggesting in this sample of the 6th grade that Form X was easier to perform. Refer to Table Table 10 Number, Means and Standard Deviations of Subjects Performance on Forms X and Y, and Parallel Form Reliabilities According to Sequence of Task, by Age/Grade Levels For Experiment #2 \ge/Grade Level Sequence N List Paken First List Taken Second Difference t ratic Parallel Form Reliability Form Mean S.D. Form Mean S.D. Mean D S.D. 1st XY 22 X 13.0 6.9 Y 13.8 7.7 0.76 0.5 0.1- rxy=969* YX 21 Y 13.3 6.2 X 117 9" 6.2 0.6C 0.5 0.1C ryx=903* 3rd XY 12 X "5177"23.5 Y 33.1 16.0 16.6 t t: i 2.35 *5 rxy=824* YX 9 Y 51.2 18.2 X 35.0 11.3 16.2 20.4 1.3C * ryx=071 6th JtY 8 X 67.3 19.4 Y 54.4 .T3.T 12.9 12.4 1.6 j * rxy=.771* YX 9 Y 73.2 19.8 X 59.9 18.7 1'3 73' 18.6 1.59 t ryx=.753* 11th XY 15 X 83.9 13.3 Y 58.3 16.7 25.5 13.6 3.57 rxy=626* YX 13 Y T47T 29.6 X 70.5 12.6 23.5 14.0 3.68 ** ryx=»216 Jr. Coll. Adults XY 45 X 89.9 14.1 Y 58.9 12.6 31.0 12.0 6.61 * rxy=731 * P <1.05 + P -<.05 when data combined in t o Table 11 Number, Means and Standard Deviations of Subjects Performance on Forms X and Y, and Parallel Form Reliabilities According to Sequence of Task, by Age/Grade Levels For Experiment #3 Age/Grade Sequence N List Taken First List TaRen Second Diff<srence t Test-Retest Level Form Mean S.D. Form Mean S.D. Mean D S.D. ratio Reliability 1st XX 15 X 19.1 7.6 X 23.7 10.1 4.7 5.5 1.59 .846* YY 6 Y 20 . 8 8.5 Y 27.0 '3.9 6.2 5.7 1.55 .919* 3rd XX 10 X 48.5 17.3 X 49.6 10.7 1.1 9.0 0.21 .913* YY 12 Y 34.4 15.3 Y 41.4 13.5 7.0 8.1 1.56 .853* 6th XX 30 X 80.7 20.9 X 86.0 17.0 5.3 9.7 1.25 .902* YY 27 Y 67.4 20.2 Y 75.0 21.1 7.6 9.2 0.59 .891* 11th XX L5 X 72.1 8.5 X 76.1 7.7 3.9 6.0 1.27 .430* YY L6 7 73.1 6.5 Y 79.1 5.6 6.1 6.9 1.68 .338 Jr. Coll. XX L3 X 58.4 16.4 X 65.5 14.0 7.2 6.3 2.1* .922* Adults YY L7 Y 51. 9 17.1 Y $1.0 18.9 9.2 "570"' 3.6* .886* * P<^ .05 U 1 u > 54 9 for a description of the data for Experiment #3. Effect of Fatigue This null hypothesis stated that whenever Form X or Y are administered twice, in rapid succesion, to bubjects of any age/grade level, the mean number of correct responses produced during the first task will be greater than the mean number of correct responses produced during the second task. The 1st grade, 3rd grade, 6th grade, 11th grade and adult junior college sampling group all indicated that performance, as measured by mean number of correct responses produced, was greater for that task which was administered second, in order. No evidence was observed for a performance decrement as a consequence of previous experience with the same form. Refer to Table 11 for a description of the results of Experiment #3. Test-Retest Reliability The Pearson product-moment coefficients of correlat ion between subjects performance on a form and their re taking of the same form immediately thereafter were significantly different from zero. Refer to Table 11 for a description of the magnitude of the reliability co efficients for each of the two samples of sequences administered to each of the age/grade levels. ! Experiment #4 I Reliability of the Tasks With a Lapse of Time Pearson product-moment coefficients of correlation were obtained between subjects performance on a form and their retaking of the same form (and parallel forms) over |a lapse of several months. A t-test applied to the data obtained from the 3rd graders indicated that the reliabil- j ity coefficient for the task given first (combination of test-retest and parallel forms) was significantly differ- |ent from zero (P=.05). A t-test for the significance of the reliability coefficient (combination of test-retest i iand parallel forms) for the second task indicated that |such a correlation coefficient could have arisen by chance |(P=.05). Refer to Table 12 for a description of the data obtained from retesting of the 3rd grade. The t-tests applied to the data obtained from the I junior college adult sample indicated that the test-retest (reliability coefficient over a four month lapse of time i i (appeared significantly different from zero for form X and Form Y (P=.05). Refer to Table 13 for a description of the data obtained upon retesting the adult sample. Reliability of the Difference Scores I The obtained coefficient of reliability of the difference scores of .29 of the 3rd graders retested Table 12 Number, Means and Standard Deviations of Performance on Forms X and Y, and Reliability of Forms and Difference Scores of 3rd Grade Subjects When Tested 7 Months Later for Experiment #4 Sequence List Mean S.D. | rxy t “T Dilirerence Scores N 1 ratio Mean D S.D. rxy t XY and YX 1st 12 58.9 77.3 24.1 16.8 .52 2.46* 18.4 17.4 .28 1.32 2nd 12 40.2 23.0 .37 i 1.57 16.4 16.9 56.6 13.2 1 . 1 *t K .05 in Table 13 Number, Means and Standard Deviations of Performance on Forms X and Y, and Reliabilities of Forms and Differences Scores of Junior College Adults When Tested 4 Months Later for Experiment #4 S lence List Mean S.D. 1 rxx Difference Scores N ratio Mean’ Tf S.D. r XX t 1ST 16 90.8 12.1 1.84* XY 95.0 9.0 .38 32.8 10.9 .36 1.72 2nd 16 88.0 13.6 .77 6.23* 32.9 16.1 62.1 16.6 * P -C.05 U 1 -vj I I !seven months later, as indicated by t-test was not of suf- j i I ; ficient magnitude to be considered significant (P=.05). i | The method of correction for attenuation yielded a cor rected coefficient of correlation of .12 indicating the maximum correlation one might expect between the two lists if no random errors of measurement were present in either set of scores. The coefficient of reliability of .36 on the part of the junior college adults retested with the same sequence four months later, as indicated by t-test, was not of sufficient magnitude to be considered significant i (P=.05) . Correction for attenuation yielded a corrected j I correlation coefficient of .67 indicating the maximum i correlation one might expect to obtain between the two difference scores if no random errors of measurement were I j ’ present in either of the two forms. The reliability of j I the difference scores appear to reflect some variance i existing over a four month period on the part of the | I junior college adult students. Refer to Table 13 for a i j description of the data. | I | Discussion of Results j | This section contains a disucssion of the results of this investigation relative to the major hypotheses and to relevant aspects of past and current research. 53 Difficulty Level of the Two Forms There was no consistent pattern that might suggest that one or the other form was easier or more difficult. Moreover, whatever difference might exist in terms of their level of difficulty, had no measurable influence in the reliability or the construct validity of the task. That Forms X and Y are similar in difficult level supports the generality of the Spew Hypothesis as dis cussed by Underwood and Schulz (1960). The Spew Hypothesis states that all definitions of Meaningfulness (M) can be translated into frequency terms, leading to the principle that the higher the M of a verbal unit, the more frequently has that unit been experienced. That Forms X and Y appear similar in difficulty level suggests there is a value in relating encoding methods as introduced in this investigation for study of the kinds of problems typically reserved for classical paired-associate verbal kinds of materials. Effect of Fatigue No evidence was observed for a performance decrement as a consequence of previous experience with the same form. A tendency was noted for a facilitating effect when such a sequence as XX or YY was administered. All groups of children, adolescents and adults performed better, as measured by the mean number of correct responses, when afforded a second opportunity to perform 60 the same task as they had already taken. This finding of no measurable fatigue effect is important considering that the construct validity introduced in this dissertation proposed that when alternate forms are administered, that whichever form is given second, in order, results in more difficulty. Had the results indicated that there was evidence of fatigue, i.e., a decrement of a performance as a conse quence of previous experience with the same form, the data might have otherwise forced the position that it might not have been the associative inhibitory process per se that led to difficulties during the second administration of the form, but that moreover, it might have been attributed to the fatigue factor. Associative Inhibition The term, associative inhibition, is not presented in this study as a construct inferred to be actually existing. Neither is it conceived as having a one-to-one relationship between verbal or motor responses and inner mental events. Associative inhibition, as presented in this study is a theoretical construct accounting for a decrement of performance because of, or caused by, the presence of an identifiable relationship occurring between stimuli and their responses. Mtiller and Schttmann (1895) suggested that the relationship occurring among items, when associative inhibition is expected, can be 61 ' related to conscious or unconscious processes that renders the association with the new response more difficult. In any event, the position taken by the experimenter is, wittingly or not, the individual is impeded in his attempt to block the older response from occurring, and that, in part, it is the failure to block the older response that led to feelings of discomfort so often displayed on the part of the subjects partaking in this learning experience. The simplified design incorporated in this investi gation was expected to maximize the likelihood for conditions that might lead to the development of associ ative inhibition when the forms were given in sequences XY and YX. An A - Br arrangement was constructed by repairing the first list stimulus-response members on the transfer task as reported by Porter and Duncan (1953) to produce marked negative transfer. Because of the ruling out of the fatigue effect (H:2), it is possible to present a strong argument for the presence of associative inhibitory processes underlying the negative transfer occurring throughout the sequences XY and YX. Quite likely a similar process occurs throughout the learning of A - Br paradigm. The results of this investigation indicated that associative inhibition was present throughout the sample of subjects selected from age/grade levels from the 3rd, 4th, 6th, 11th grades as well as during the learning of the college adult samples. Evidence was not seen for the presence of associative inhibition throughout the learning tasks by the 1st grade pupils. It was pointed out earlier in the Results section that when another 1st grade sample was given Sequence XX and Sequence YY, the difference of the "difference scores" between groups XX and YY and groups XY and YX were significant beyond the .05 level. This latter finding suggests the presence of either associative inhibition in the XY and YX groups, or positive practice effects in the XX and YY groups and/or both conditions being operative during the per formance of the 1st grade samples. The results are suggestive that the employment of the simplified design, through the use of the added control groups ruling out the fatigue effect, effectively isolated a measure yielding an index of the extent to which subjects were susceptible to the detremental effects of associative inhibition occurring during an encoding learning task. It was outside the scope of this investigation to relate such measures to external criteria such as IQ scores, tests for manifest or task-related anxiety, personality or measures sensitive to learning disabilities. Limitations imposed by the reliability of the forms are presented in the following sections. I Reliability Studies i Immediate Test-Retest Reliability | I The range of test-retest reliability coefficients i i t varied from a low of .85 to a high of .92 for children ! | of elementary school age. The high coefficients for I | I immediate retesting on this timed test indicates the ! ; j i importance of the familiarization process and the | ; memorization of the stimulus-response members. The effect i observed during sequences XX and YY indicates the | ! importance of considering practice effects for children j of this age. Adolescents drawn from the 11th grade were found j to have coefficients of .43 for Sequence XX and .34 for i I : Sequence YY. The amount of time allotted for their I performance was limited to 60 seconds for production I of correct responses. The suggestion arises that 60 i I ' | seconds was not sufficient time for this age/grade group. The junior college adults were found to have a test- immediate-retest reliability of .92 for Sequence XX and a coefficient of .89 for Sequence YY. Performance time for this group of subjects was 90 seconds. Perhaps 90 seconds of allotted time for adolescents and adults provide for greater reliability than does 60 seconds for performance on the experimental forms. 64 Reliability of Parallel Forms The sampling pool of children and adolescents tested individually for the Sioux City pilot study were found to have a reliability coefficient of .94 for Sequence XY and a reliability coefficient of .90 for Sequence YX. The sampling groups of 1st graders, 3rd graders, 6th graders, and 11th graders all indicated that the reliability of Sequence XY (rxy) was greater than for Sequence YX (ryx). There is a clue as to what accounts for the relatively lower reliability coefficients of parallel forms as compared to that obtained from the test-retest method. Most likely the inhibition present accounted for the poorer reliability of alternate forms. Note that the effect occurred regardless of which form was administered first. The time allotted for performance appeared to effect the magnitude of the reliability coefficients. The parallel form reliability coefficients for the 11th grade sample tended to be higher than the test-retest reliability of a different sample of 11th graders who were allotted 60 seconds for performance. This latter group of students complained of hot having enough time, giving the experimenter the impression of frustration after completion of the first form. The amount of time has a marked effect on a task such as these timed encoding tasks. The optimal amount of time to be allotted for performance for differing age groups has not been established. However, 120 seconds appeared sufficient for dependable performance for elementary school children. Reliability of Forms with a Lapse of Several Months The reliability of forms over a lapse of several months appears undependable. Although correlation coefficients vary from .38 to .77, they are considerably lower than most of the test-immediate-retest or "immediate parallel form reliability coefficients that were obtained. Both correlation coefficients were significantly different from zero. These results suggest that performance on these tasks vary with time lapse between testing and, therefore, could be expected to be limited as predictors of performance over a lapse of several months, especially when subjects were given alternate forms. Reliability of Difference Scores with Several Months Lapsing" The difference score, when alternate forms are administered to the same group of subjects, is the measure which is being investigated and studied in terms of its value in contributing to the construct validity of the task in measuring one's susceptibility to associ ative inhibition. A sample of 3rd graders yielded a reliability co efficient of the difference scores of .28, which was not significantly different from zero (P=.05) when retested over a 7 month lapse of time. Such reliability of the difference scores are limited in their intensity by the magnitude of the reliability of each of the tasks. The list that had been given first, in order, had a parallel-test-retest reliability over a 7 month lapse of .52, whereasethe test that had been given second, in order, had a parallel-test-retest reliability of .37. Correction for attenuation yielded a correlation co efficient of .65. When retested over a period of 4 months, a sample of junior college adults yielded a reliability coefficient of the difference scores of .36, not significantly greater than zero (P=.05). A 90 second period was allotted for performance. The same sequence was ad ministered on both testing occasions. Correction for attenuation yielded a correlation coefficient of .66. These results pertaining to the study of the reliability of the difference scores over lapses of time indicate that it is possible to refine the reliability of the measure over time so that some prediction becomes feasible. However, it seems that the nature of the construct, "susceptibility to associative inhibition," as measured by this instrument 67 can be more readily studied as a transient state rather than as a trait that is consistent over lapses of such length. As the construct greatly depends upon the reli ability of each of the two forms that constitute the "difference” socre, the prediction of such diffeeences over time is hazardous. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS FOR FURTHER INVESTIGATION Summary The purpose of these investigations was to identify and to isolate a variable yielding a measure of one's susceptibility to associative inhibition occurring during a learning task. Construct validity of such a susceptibility was attempted through the con struction of a learning task. The learning task consisted of an encoding task developed on the paired- associate Earning model. An important aspect of the study was to provide a paper and pencil test substituting in special cases for the classical paired-associate method which requires individual administration. Two forms of the experimental encoding task were developed for this investigation. The learning task was of the same transfer paradigm type referred ho by Porter and Duncan (1953) as the A - Br (repaired) type. Specific letter combinations were selected on the basis of the meaningfulness value (M) of the two letter pairs scaled by Underwood and Schulz (1960) . The combination of letters 68 69 selected for this investigation were so chosen so that the total sum value of each list would approximate one another as closely as possible in terms of their scaled value. The expectation, then was that the lists would be similar to one another in terms of their difficulty level. The forms were developed so as to maximize the likelihood for the occurrence of associative inhibition of the Mtlller-Sbhilmann type such that when alternate forms are administered to subjects in rapid succession, whichever form is given second, in order, would result in more difficulty to perform. Fatigue and practice effects were studied by successive administration of the same form. Immediate test-retest and immediate parallel form reliability studies were taken for samples of children, adolescents and adults. A reliability of test- retest and parallel forms over a period of several months' duration was accomplished with a sample of children and adults. A reliability of the difference scores was also attempted for a period ensuing over several months. Results indicated that the forms were of comparable difficulty level for children, adolescents and adults. No evidence was observed for effects of fatigue when a second task was given in rapid succession. Immediate test-retest and parallel form reliability appeared adequate for use with children when allotted 120 seconds for production of correct responses. Reli ability appeared less than optimal for use with adoles cents and adults, most likely because of the lack of adequate time and/or allotted number of pairs to complete the task. When alternate forms were administered to groups of children, adolescents and adults, whichever of the two forms were given second, in order, resulted in greater difficulty in producing correct responses. This diffi culty on the part of the subjects was attributed to their susceptibility to the associative inhibition and inter fering response tendencies that interfere with the adjust ment to the new task. This measure— the difference of the number of correct responses between alternate forms when given in rapid succession— was of major concern in this study. Conclusions The following conclusions relative to the problems under investigation are suggested by the evidence pro vided within the limitations of these investigations, and in relation to previous research findings pertinent to the same problems. 1. When alternate forms of the experimental tasks are given to a group of subjects, one after another in rapid succession, whichever form is given second, in order 71 can be expected to result in more difficulty. This difference score yields an estimate of the extent that any given subject is susceptible to the associative interference produced by the prior learning of the alter nate form. 2. The associative interference produced by the learning tasks can be attributed to the interfering competing tendencies established during the prior learning experience with the alternate form. 3. Performance on both forms of the experimental task appear to be highly related to reading ability, chronological age and mental age. 4. Performance on both forms of the experi mental task appear to be moderately related to perform ance on the Wechsler Intelligence Scale for Children's encoding subtest. 5. The two forms of the experimental task appear to be approximating equal levels of difficulty for different age levels of subjects. 6. The reliability of the experimental forms varies somewhat with age of the subjects and time allotted for production of correct responses. However, the reliability of the instrument appears adequate in its present form for application for children of elementary school age. Applications of its use should be limited to studies concerned with the development of construct 72 validity, learning theory and further study of individual differences in learning ability and motivational states. Recommendations For Further Investigation Relevant to the results of this investigation it is recommended that further study be undertaken to: 1. Determine the optimal number of trial blanks to be utilized for experiments determining reliability of adolescents and adults on the two forms. 2. Determine the optimal time to be allotted for adolescents and adults to produce the correct responses. 3. Determine effects of having subjects complete the first task before being administered a second task. 4. Relate performance on this encoding task to measures relating to anxiety. Scales of manifest anxiety, test-related anxiety and various indices available to measure the presence of anxiety in children may lead to further developments of the construct, susceptibility to associative inhibition. APPENDIX A THE LEARNING TASKS R JJ N A N D S E E V 0 U s E R U A N Y 0 o 5 E A N Y 0 S ' E A N R U Y 0 R U A N S E Y 0 S b Y 0 S E A N R U Y 0 A N S E A N ■ j s - 75 s H P A 0 E L IV G T i H 0 S A C U P E S A P E H 0 C U P E H 0 S f \ C u P E S A c o H i 0 1 c u s A Q-Lu s A P E 0 C U P E H 0 C U S A H 0 S A P E C U P I E ( H 0 S A P E H 0 C u 5 A C U H 0 S A P E S A C V J P E H 0 S A P E H 0 C U H 0 P E s A C 0 S A W 0 P E S A C U W 0 P E C U P E H . 0 i • f 75 ! I J X s H P C A 0 E U IV & T P P E S A P E H 0 C U P E H 0 S A C U P E S A C U H 0 C u S A H 0 P E H 0 C U S A P E H 0 S A P E R 0 C u p E H 0 C U S A M 0 S A P E C U P E H 0 S A C U H 0 C u S A H 0 C u p E R 0 c u 5 A C u H 0 S A P E S A C U P E H 0 S A P E H O C U P E c U H 0 S A P E C U S A P E S A C 0 S A H 0 P E S A C U H 0 P E C U P E H 0 S A H 0 S A c u p E S A P L H 0 C U 76 P E G C U T S A P w 0 s A C U P E 5 A P E H 0 C U P E H 0 S A C U P E S A C U ( C U S A P E S A P E H 0 C u p E H 0 c u s A H 0 S A P E C U 1 H 0 S A P E H 0 C u 5 A C U H 0 s A P E S A C U P E H 0 S A H 0 c u H 0 P E S A c U 5 A W 0 P E S A C U H 0 P E C U P E y P E G C u T S A P H 0 W ■ % J p E 5 A P E H 0 C U P E H 0 S A C U P E S A C u 1 4 0 C U S A H 0 P E H 0 C U S A P E H 0 3 S A P E H 0 C 0 p E H 0 C U S A H 0 S A P E C U P E H 0 S A C U H 0 c u s A H 0 c 0 p E 3 E H 0 C u 5 A C U H 0 S A P E S A C U P E H 0 S A P E H 0 C U P E c U H 0 S A P E C u 5 A H 0 P E S A C U 5 A H 0 P E S A C U W 0 P E C U P E H 0 S A 1 4 0 s A C U P E S A P E H 0 C I) APPENDIX B DIRECTIONS FOR ADMINISTRATION OF THE LEARNING TASK 78 DIRECTIONS FOR GROUP ADMINISTRATION OF THE CODING TASK The experimenter passes out three stapled sheets, face down, to all the students in the classroom. These three sheets include the sample sheet. Form X and Form Y. Half of the class shall be assigned Sequence XY (Form X first, and Form Y second) and half of the class shall be given Sequence YX (Form Y first, and Form X second). The sequence of the packets are pre-arranged in terms of order Packets are given randomly. The method is taught through the warm-up with the sample sheet. The students are asked to write their names on the sample sheet. Once the experimenter is assured all students grasp the method, they are told: O.K....turn over the first sheet. See, there are divided boxes. Here we have SA with a letter, PE with a letter, HO with a letter and CU with a letter. You see...they make words. When I say begin (not yet!) start at the beginning and fill in as many boxes as you can until I tell you to stop (not yet!) . ...Are there any questions?...When I tell you to start do as many as you can. Ready? Begin! After the allotted time the students are told: Stop! Turn the page you have just completed over like this. The experimenter demonstrates how the page is placed. With as little delay as possible the experimenter goes through the same instructions as given with the first experimental form. After the allotted time the students are told: Stop! Note: As soon as the students begin work in completing the task, the experimenter asks them to do one at a time, to work quickly, and not to skip any spaces. In the event that a student finishes his 100 paired associates before time expires, he is told to wait. Students in the 7th grade or higher may be given 90 seconds for completion until told to stop. BIBLIOGRAPHY Atwater, S.K., "Proactive Inhibition and Associative Facilitation as Affected by Degree of Prior | Learning," J. Exp. Psychol., 1953, 46, 400-404. Ausubel, D.P., Robbins, L.C., & Blake, E., "Retro active Inhibition and Facilitation in the Learning 5 of School Materials," J. Educ. Psychol., 1957, 48 | 334-343. | i Ausubel, D.P., Stager, M., & Gaite, A.J., "Proactive Effects in Meaningful Verbal Learning and Retention," J. Educ. Psychol., 1969, 60, 59-64. Baumeister, A.A., Hawkins, W.F., & Holland, J.M., "Retroactive Inhibition in Short-term Recall in Normals and Retardates," Am. J. Ment. Def., 1965. Bernstein, E. & Dorfman, D., "Some Effects of Mepro bamate on Human Learning," J. Psychol., 1959, 47, 81-86. I Ebbinghaus, H. Uber das Gedachtnis: Untersuchung zur j Experimentelen Psycholo^ie. 1885 (Trans.' as Memory; A Contribution to Experimental Psychology by H.A. Ruger & C.C. Bussenius. Columbia Univ. Coll. Educ. Reprints, No. 3, New York: Teachers College, Columbia University, 1913). Eysenck, H.J., "Classification and the Problem of Abnormal Behavior," Handbook of Abnormal Psychology Eysenck, H.J., Ed., New York: Basic Books, Inc., 1961. Gibson, E.J., "Retroactive Inhibition as a Function of Degree of Generalization Between Tasks," J. Exp. Psychol., 1941, 28, 93-115. Gilliland, H.G., "Presentation and Response Modes Affecting Paired-AssociatevLe&rning in Retarded Children," Doctoral Dissertation, University Southern California, 1969. Hovland, C.I., "Human Learning and Retention," Handbook of Experimental Psychology. Ed. by S.S. Stevens, New York: JohnWiley & Sons, 4th Printing, 1962. 80 11. Jost, A., "Die assoziationsfestigkeit in ihrer Abhangigkeit von der Verteilung der Wiedholungen," Z. Psychol., 1897, 14, 436-472 12. Lair, C.V., Moon, W.H., & Kausler, D.H., "Associative Interference in the Paired-Associate Learning of Middle Age and Old Subjects," Devel. Psychol.* 1969, 1, 548-552. 13. Laslett, H.R., "Experiment on the Effects of Loss of Sleep," J. Exp. Psychol., 1928, 11, 370-396. 14. Lyon, D.O., "The Relation of Length of Material to Time Taken for Learning, and the Optimum > ■ Distribution of Time," J. Educ. Psychol., 1914, 5, 155-163. 15. Margolis, H.J., "Associative Interference: Effects of Meprobamate on Normal Adult's Performance on a Mtlller-SchOmann Type Learning Task," Psychopharmacologia (Berl.), 1966, 8, 379-388. 16. McGeoch, J.A. & Irion, A.L. The Psychology of Human Bbhavior. New York? David McKay Company, Inc., 1961. 17. McMannis, D.L., "Retroactive Inhibition in Paired- Associate Learning by Normals and Retardates," Amer. J. Ment. Def., 1967,71, 931-936. 18. Mtiller, G.E. & Pelzecker, A., "Experimentelle Beitrtige zur Lehre vom Gedtichtniss," Psychol., I960, BrgBflT 1., 1-288 19. Mtiller, G.E. & Schtimann, F., "Experimentelle Beitrtige zur Untersuchung des Gedachtniss," 5 ^ . Psychol., 1894, 6, 81-190. 20. Mtlnsterberg, H., "Beitrtige zur experimentellen Psychologie," Freiburg: Mohr, 1889, Heft 4. 21. Odom, R.R., Shapira, A. & Attwell, A.A., "Effect of Anxiety on Retroactive Inhibition," Calif. J. Educ. Res., 1966, 17, 33-40. 22. Porter, L.A. & Duncan, C.P., "Negative Transfer in Verbal Learning," J. Exp. Psychol., 1953, 46, 61-64. 23. Pyle, W.H., "Retention as Related to Repetition", J. Bduc. Psychol., 1911, 2, 311-321. dL 24. Rock, I., "The Role of Repetition in Associative Learning," Amer. J. Psychol., 1957, 70, 100-107 186-193. “ 25. Spence, K.W. et. al., "Anxiety (Drive) Level and Degree of Competition in Paired-Associate Learning," J. Exp. Psychol., 1953, 45, 265-272. 26. Spence, K.W., "A Theory of Emotionally Based Drive (D) and its Relation to Performance in Simple Learning Situations," Amer. J. Psychol., 1958, 13, 131-141. 27. Stroud, J.B. & Schoer, L.A., "Individual Differences in Memory," J. Educ. Psychol., 1959, 50, 282-285 28. Stroud, J.B. & Carter, L.J., "Inhibition Phenomena in Fast and Slow Learners," J. Educ. Psychol., 1961, 52, 30-34. 29. Taylor, J.A., "Drive Theory and Manifest Anxiety," Psychol. Bull., 1956, 53, 303-320. 30. Underwood, B.J., "Speed of Learning and Amount Retained: A Consideration of Methodology," Psychol. Bull., 1954, 51, 276-282. 31. Underwood, B.J. & Schulz, R.W., Meaningfulness and Verbal Learning. Chicago* Lippincott, i960. 32. Van Buskirk, W.L., "An Experimental Study on Vivid ness in Learning and Retention." J. Exp. Psychol., 1932 15, 563-573. 33. Vergason, G.A., "Retention in Retarded and Normal Subjects as a Function of Amount of Original Training," Amer. J. Ment. Def., 1964, 68, 623-629 34. Weiskotten, T.F. & Ferguson, J.E., "A Further Study of the Effects of Loss of Sleep," J. Exp. Psychol., 1930, 13. 35. Williams, H.L. & Lubin, A. & Goodnow, J., "Impaired Performance with Acute Sleep Loss," Psychol. Monog., 1959, 73, #14.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Effects Of Early Educational Stimulation On Selected Characteristics Of Kindergarten Children

PDF

An exploratory study of the psychological sequelae of surgery and hospitalization in male children

PDF

Modifying Appropriate And Interfering Behaviors In Autistic Children Using A System Of Behavior Modification, Therapeutic Teaching

PDF

The Development And Evaluation Of A Personal Management Program And Its Effects Upon The Self-Concept And Perceived Locus-Of-Control Of Blind Adolescent Females

PDF

Simple Auditory Reaction Time Of Young Adult And Elderly Subjects In Relation To Perceptual Deprivation And Signal-On Versus Signal-Off Conditions

PDF

A Study Of Relationships Between Grades And Measures Of Scholastic Aptitude, Creativity, And Attitudes In Junior College Students

PDF

The Systematic Isolation And Validation Of Personality Determiners In The Handwriting Of School Children

PDF

Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

PDF

An Experimental Study Of Intelligibility Thresholds Of Hypacusic Adults Using Forced-Choice And Yes-No Techniques

PDF

A Normative Study Of Oral Sensation And Perception: Two-Point Discrimination, Form Identification, Tactile Pattern Recognition, And Mandibular Kinesthesia

PDF

The Relation Of Sense Of Humor To Creativity, Intelligence, And Achievement

PDF

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

PDF

An Exploratory Study Of Pupillary Dilation As A Measure Of Difference In Attitude Of Three Groups Of Educators Toward Orthopedically Handicapped Children

PDF

The relationship of scores on a measure of test-wiseness to performance on teacher-made objective achievement examinations and on standardized ability and achievement tests, to grade-point average, an

PDF

Effects Of An Audience On Serial Association

PDF

A Measure Of Cultural Deprivation

PDF

Reading Rates And Comprehension As Affected By Single And Multiple-Ratio Schedules Of Reinforcement Within A Token Economy As Measured By Precision Teaching Techniques

PDF

A Study Of Relationships Between Group Test Of Creativity (Gtoc) Scores And Achievement Test Scores Of Students With Spanish And Non-Spanish Surnames

PDF

The Critical Behaviors Of Certificated Employee Council Representatives In The Meet And Confer Process

PDF

An Exploration Of Interpersonal Behavioral Possibilities And Probabilities

Asset Metadata

Creator Margolis, Harold Jules (author)

Core Title Susceptibility To Associative Inhibition Development Of Construct Validity Through A Paper And Pencil Encoding Task Given To Children, Adolescents And Adults

Degree Doctor of Philosophy

Degree Program Education

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

Tag education, educational psychology,OAI-PMH Harvest

Format dissertations (aat)

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Magary, James F. (committee chair), Fox, Frank H. (committee member), Garwood, Victor P. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-522977

Unique identifier UC11362115

Identifier 7127941.pdf (filename),usctheses-c18-522977 (legacy record id)

Legacy Identifier 7127941.pdf

Dmrecord 522977

Document Type Dissertation

Format dissertations (aat)

Rights Margolis, Harold Jules

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