FIRST SERIES No. 110 JUNE 1, 1926 UNIVERSITY OF IOWA | STUDIES ee STUDIES IN CHILD WELFARE VOLUME Iil NUMBER 6 - THE LEARNING PROCESS IN YOUNG CHILDREN An Experimental Study in Association by Jutia A. Kirkwoop, Px.D. PUBLISHED BY THE UNIVERSITY, IOWA CITY Issued - Sebaecladpendl hea ugho ae gue ree Entered at the post office at Iowa City, Iowa, as 8 a élak aides the Act of October 3, 1917. ~LBIIOS 164 | L@r, UNIVERSITY OF IOWA STUDIES IN CHILD WELFARE Proressor Birp T. BALDWIN, PH.D., Editor FROM THE IOWA CHILD WELFARE RESEARCH STATION VOLUME III NUMBER 6 THE LEARNING PROCESS IN YOUNG CHILDREN An Experimental Study in Association by JuLIA A. Krexwoon, Px.D. PUBLISHED BY TIE UNIVERSITY, IOWA CITY, IOWA La oe ‘fine J ) 4 { \ }. Gos p ; ¥ ee oe : 1¢ or al a. 5 ~ . itera < asf . “2¢ ae gaan! nae - Fe } " AE peep Sala ia ae ~~ Ls te aida, I eS ‘ a ae: ~~ oy x s : * oe 7. # > y ies o 2 sal a a . a — x i Fy ade . * . ~ . % ‘ . ; * = + - Ven vol i f = ~“S . * he 1 — \ ~ + = , J a J ' ‘ a Sg f ~ ; ‘ wy et 2 > re e : , - ? ’ a - - Jka? , = , y ~ - x > e « Toa A P « = ay - “er : . - » ue FOREWORD The psychology of the learning process has been a fertile and interesting field for much experimentation in the learning of school children, in the analysis of the mental traits of adults, and in the behavior of animals, but a very limited amount of detailed work has been undertaken in this field with preschool children. This study of learning by association reaction by Dr. Kirkwood is one of the series of special studies undertaken in the Iowa Child Wel- fare Research Station as a part of a comprehensive program for the investigation of the nature and development of intelligence in young children. It is a portion of a thesis submitted in partial fulfillment of the requirements for the doctorate in child psychology. The author presents the results of experiments with 180 children between the ages of two and six years, grouped for study of various phases of the problem of learning and relearning. These include interesting data on the learning of the material according to various forms of presentation, on learning on successive and on alternate days, relearning after an interval of one year, an analysis of the associations as shown by the correct and incorrect responses of the children, types of individual learning curves, and the correlations between general intelligence and learning. The results will form the basis for a more general and comprehensive investigation into the learning situations involved in the daily activities of young children. Birp T. BALDWIN Office of the Director Iowa Child Welfare Research Station State University of Iowa May 1, 1926 ‘heel on 7 cade aa tga Hy one My orighe ; ay . ; va \ SIN - Pi ar eS a ; nih 1 ht a) : E a be rt Fy : ie, { yes phates MN, @ J 2 in 4 pe ; > f “14 b Be ey hae 4 conn ate mela i} Cth Orbe i-tel i a j payer S ue Le em) weit eatin re ta ; ais, ‘$f anaes aga eee! | ah ey 1tGh as 9 ne Cea CR AAS CPSs Sk Fn eo See ent % , catyhtel : . CONTENTS CHAPTER PAGE LEMETTEIA VERSIE EX 00. a2. ule GO een dag dt RRR aoe AE Sec SCL SS 3 I MareriaAu, ForMs or PRESENTATION, AND METHODS OF LEARNING IN AN ASSOCIATION REACTION LEARNING Ex- CRUG AS athe fsa dale a i aca re Ras fi VER UOT ID evi Re Ais kG nahn ot Bes 3 Ud eA 5 Lip na AST 9 Forms of Presentation of Material and Methods of ADCAIUIN Os ee niet Pee emt ee OM UST ee Mee Rn 8 be 9 LAT AT UM OCI erent se ere nee nom Met ne he Pt 10 Learning on Successive Days: Children Studied uC Tacl 11) sae eee Meee ce a ence Te acta 11 Learning on Successive Days versus Learning on Alternate Days: Children Studied by Pairs .......... LL VATIAT Gee OLIN Smee enn eee ny Oe) ays oie 12 pindard Form ntercnangved c.1 Ae oe pac Learning on Successive Days: Children Studied Gav eU yi Ce eT NN hig ak eMac mili el oe beet Al tel Jn SLM 13 Single Series: Series I and Series IT ...W.......0.... 13 Learning on Successive Days: Children Studied DV at rc eee ere cr arnt ee Mente ares is} 13 CCP ONETICALININIG oaee tee Une re ee een eee 14 Check on Learning of Standard Form ....................... 14 Check on Learning of Variant Forms .......................... 15 Relearning after an Interval of One Year ...................... 15 APOC SOL ALG t. paw cuenta ara me ern iS a el 15 BLOCOlU des GCie teen, ere me oat mM Shee me i 15 SLUT RES Aa tcl anda, SRO OER RRL eat ot yw Ry CaS re TL Oe 16 RSE VA LIONS 3 rsas oo tee ral ne ote Pear erie SDD) A 16 PeIMIUlatAverCCOrdUs@alt se oer Ae en ese Pa Le Pest Dilities Tors kh arthem StUCY 22s esear- tees se a 17 Piers ers: OF THE, XPERIMENT 22020 he i io Children in the Preschool Laboratories ............................ 19 POrey Iie eve MaNdergartens hs ene ee 19 banee: or ChronologicalwAces sere 26s cee 20 Hanes: of MentaltA ces ean ve ase in ee ee 20 Distribution according to Presentation of Material...... 22 PLise NESUUTS OF UXPERIMENTAL OTUDY seinen eS 23 MITITETION Ole LIGA LIEN Ovo crene tote ce eee noes ee Sy UE: 23 Absences as a Factor in Learning ...........0..0-..2.....---- 20 Quantitative and Qualitative Analysis of Data —_........ 27 Material ii: sin Seteee Basie ee Roentgen hs 27 Forms of Presentation of Material and Methods of AIGA PTS eee cree en eae team rN is ek 30 6 CONTENTS CHAPTER Standard (Form: 2.433 ee ee Learning on Successive Days: Children Studied Individually 23 a Learning on Successive Days versus Learning on Alternate Days: Children Studied by Pairs........ Variant. Forms 322 2 Sgt Ae ee eee Standard Form Interechanged: Children Studied Individually 22" 5 ee es ee ee yea i Single Series: Series I and Series II: Children Studied ‘by. Pairs 22 ee Cheek’ on earning). 2 ee ee Check on Learning of Standard Form ................-. Check on Learning of Variant Forms ....................- Relearning after an Interval of One Year _................ Learning Curves ick Soe Sa ean on ee SICOTOS seid Ls le alt ee ia ae an hea Analysis :0L Mrrors 422 hc ee eee Order of Difficulty in Forming the Required Asso- Cla tLONS | oc2.. 20 As ee ae Comments of the Children during the Experiment.... Statistical Analysis 4 c.4c2 ee eee ae Reliability: ofthe Experiment): (ee Correlations between the Number of Trials Necessary for Learning on the Standard Form of Presentation and Results of Stanford-Binet Examination .............. Mental Age (Stanford-Binet) by Groups ............... 2 Mental Age (Stanford-Binet) by Sexes ............... Correlations between the Number of Trials Necessary for Learning on the Standard Form of Presentation and Various Psychological Tests on Children of the Five and One-Half-Year Age Group ..................--.------- Detroit. Kindergarten: Test... 32 ieee Two Performance Tests (Pintner and Paterson) ..- Montessori Cylinders tac eget ee ee IV .STupy- OF SPECIAL CASES | hic eee ie ce el ee Analysis of Four Cases of Failure in Learning the Material: 2000 a ee Analysis of Four Cases of Success in Learning the Material @ 0.50005 90 BN eh ee UMMM ALY ee a ey ae V_ Reactions or A SELECTED Group or ADULTS TO THE LEARNING ‘MATERIA 20.0000. 2 i oe ee Reports of Introspection of Adults Order of Difficulty in Forming the Required Associa- [i (t)1! anos Sn MOM Memmi 2c To VI Summary, RESULTS, AND CONCLUSIONS ....-.-2-----c-c-0c0-----0------ REFERENCES 37 ot CHAPTER I MATERIAL, FORMS OF PRESENTATION, AND METHODS OF LEARNING IN AN ASSOCIATION REACTION LEARNING EXPERIMENT In looking over the field of experimental work on the learning process, one finds that the subjects used have been animals, pre- school children, school children, and adults. When thinking of learning, one is likely to think of it in connection with children, yet the literature shows that at least sixteen different learning tests have been used with adult subjects, four with school children, and two only with preschool children. Animals have been given, as a rule, either a maze or a puzzle box test in the study of their progress in learning. The first test of learning to be used with preschool children was the card sorting test ;'® later a specially de- vised test was used by one investigator to study the young child’s ability to learn to recognize words.’®? School children have been given learning tests based upon code translating, maze learning, addition, and substitution. The learning process of adults has been studied by means of the following tests: ball tossing, code trans- lating, substitution, maze learning, typewriting, telegraphy, addi- tion, multiplication, nonsense syllables, sensible material, marble distribution, card sorting, cube formation, mirror tracing, javelin throwing, and hand movements.1 As there have been so few investigations into the problem of how children learn, particularly how young children learn, this phase of the study of the learning process is one in which there is need for more experimentation in regard to many of the different factors involved in an analysis of how learning takes place and of the factors that influence the rate of learning. If children are to be taught efficiently, there must be thorough scientific knowledge of the process by which a child learns. This knowledge can be derived only from experimental results obtained by placing the 1. This study is Part II of a thesis on file in the library of the State Uni- versity of Iowa. Part I is a general review of the literature on the learning process, with references. The references are included in this abridged form of the thesis in order to indicate the scope of the material included in Part I. 7 8 IOWA STUDIES IN CHILD WELFARE child in a controlled learning situation and by analyzing quantita- tively and qualitatively the results obtained. It was in the hope of throwing light on some of the problems involved in the learning process of young children that this investigation was undertaken. Some of these problems, such as effect of distribution of practice and rest periods, the problem of interference and transfer, and relearning, as compared with original learning, have been studied by investigators of learning in animals and in adult subjects. One of the principal motives of the present investigation is a desire to discover whether or not the principles and laws derived from experimentation on learning in animals and adults are equally applicable to learning in children of kindergarten age and younger. One objection that may be raised against some of the materials used in studying the learning process is that the response by which the learning is studied is in itself too complicated and involves too many factors to be considered an adequate means of studying how learning takes place. For example, typewriting has been used quite extensively as a means of studying progress in learning; it seems advisable to choose a more simple problem than the ac- quisition of skill in typewriting in making a study of the psychol- ogy of learning, since typewriting is an exceedingly complex pro- cess involving both mental and physical elements. In the present investigation, the aim was to use as simple material as possible and to have the response of the child as simple as possible. There are various theories current as to how learning takes place. The theory that has received the widest recognition is the theory that learning goes on through a process of association, a connecting of stimulus and response. A stimulus is presented and a response given. Later presentations of this stimulus will eall out the response connected with it. The method used in the study of learning reported upon here is that of the association reaction type of response. The materials, forms of presentation, standard and variant forms, and methods of learning were planned as means of studying by association reaction the factors in the learning proc- ess of young children that have been studied with animals or with adult subjects and to discover experimentally whether there are similarities or differences in the results found in the present in- vestigation and those reported in the literature on learning. LEARNING PROCESS IN YOUNG CHILDREN 9 MATERIAL The material for this experimental study of learning through the process of association reaction consists of a series of twenty blocks and a series of twenty pictures. After the twenty blocks of geometrical shapes had been decided upon, simple outline pictures were drawn to represent objects that are in some way similar to the blocks. With one exception possibly, the pictures are of objects with which every child is familiar. Each picture suggests the block with which it is to be associated in the experiment, and this facilitates the formation of associations. The pictures were made with India ink on unruled index ecards, 3 by 5 inches. The plan of having a certain similarity between block and picture was de- cided upon in order that the learning involved would be not too difficult for preschool children. The approach to the study of the problem was made through blocks and pictures because of the universal appeal that these have for young children. The twenty blocks were devised in such a way that the complete series may be regarded as consisting of two parallel series ar- ranged in approximate order of complexity, from the simpler to the more complex. Each block of Series II, blocks 11 to 20, is a modification, or variant, of the block that parallels it in Series I, blocks 1 to 10. The material was designed in this way purposely in order to allow for variations from the standard form of pre- sentation and interchange of blocks associated with the pictures. Figure 1 shows the parallel arrangement of the blocks; the child, however, sees the blocks, placed before him, according to the ar- rangement shown in Figure 2. Figure 3 shows the pictures in the order corresponding to the parallel arrangement of the blocks; it is also the order in which they are presented to the child. The pic- tures are placed before the child, as shown in Figure 4, instead of the blocks, but in the order of the blocks originally placed before him, for a check on learning in order to test the permanency of the associative bonds formed during the learning. FORMS OF PRESENTATION OF MATERIAL AND METHODS oF LEARNING The material was presented according to three different forms and two methods of learning in order to investigate some of the im- portant problems involved in the study of the learning process. The forms of presentation are designated as the standard form and the variant forms, the first, standard form interchanged, in which 10 IOWA STUDIES IN CHILD WELFARE the associations between the blocks and pictures as given in the standard form are interchanged; and the second, the single series form, in which the child learns one half of the material only, either Series I or Series II. The methods of learning are by trials on successive days or by trials on alternate days. STANDARD FORM The form of presentation of blocks and pictures used with the largest number of children is referred to as the standard form. Figure 1 shows the order in which the child’s responses are given when the twenty correct associations have been formed according to this method of presentation. The material was presented to each child until he had attained three consecutive perfect scores of 20, or, 1f he had not made three consecutive perfect scores at the end of the twentieth trial, the test was discontinued. A large number of the children attained scores of 20 once or twice, and then their scores dropped 1 or 2 points. For this reason the criterion of three consecutive perfect scores was adopted as a measure of complete learning. The blocks are placed upon a low table before the child in two rows in the prescribed order (Figure 2); in the first row are blocks 5, 8, 138, 17, 20, 3, 16, 9, 2, and 14, and im the second row are blocks 11, 19, 4, 15, 1, 6, 12, 7, 18, and 10. According to this arrangement, no block is next to, above, or below the one that pre- cedes or follows it in the experiment. The child is seated at one side of the table in front of the middle blocks so that he can reach easily any one of the twenty. The examiner sits opposite the child (Figures 5, 6, and 7). When the child and examiner are seated, the latter says, ‘‘See these blocks,’’ and pauses slightly while she indicates the blocks upon the table, ‘‘ and see these pictures,’’ indicating the pack of twenty pictures that she holds in her left hand, arranged in nu- merical order as shown in Figure 3. ‘‘One of these blocks [The examiner points toward the blocks on the table and puts slight emphasis upon the words italicized here.] goes with each picture. I am going to tell you which block goes with each picture and then I want to see if you can give me the right block that goes with each picture.’’ The examiner holds up, in the left hand, the pack of pictures with the first picture, a clock face, on top, and says, LEARNING PROCESS IN YOUNG CHILDREN At ‘““‘When I show you this picture, you must give me this block. [The examiner picks up the round block, block 1, with the right hand and holds it beside the picture of the clock face.] Remember {slight pause] this is the block that goes with this picture.’’ The examiner then replaces the block in its former position upon the table. Using the same procedure, the examiner goes through the complete set of twenty pictures and twenty blocks, being careful to replace each block in its correct position before showing the next picture. When the last block has been replaced upon the table, the examiner says, ‘‘Now it’s your turn. Let’s see if you can give me the right block that goes with each picture. Give me the block that goes with this picture,’’ and the examiner holds up the first picture. The whole series of pictures is gone through in this manner, the examiner repeating, ‘‘Give me the block that goes with this picture,’’ as each picture is shown, and each time replacing the block the child has handed to the examiner so that in each case the child must choose from the total number. It was found advisable, after the children had learned many of the as- sociations and responded almost as quickly as a picture was shown, to discontinue saying ‘‘Give me the block that goes with this pic- ture’’ each time a picture was shown, and to repeat it only at fre- quent intervals throughout the whole series. This was due to the fact that often a response had been given before the examiner had repeated more than a word or two of the phrase. The method of learning by one group of children was by trials on suecessive days, by another group, on alternate days. Each child’s record was begun on, a Monday so that within the two groups there should be a comparable distribution of learning periods before Saturday and Sunday intervened. Learning on Successive Days: Children Studied Individually The majority of children who learned the material were given it according to the standard form of presentation on successive days. Their records were studied individually and as a group, since most of them fell within the five and one-half year age group and there- fore afforded sufficient material for establishing an age norm for this form of presentation. Learning on Successive Days versus Learning on Alternate Days: Children Studied by Pairs It was decided to present the material on successive days to one group of children and on alternate days to another group of chil- 12 IOWA STUDIES IN CHILD WELFARE dren, since the problem of the effect of distribution of work periods and rest periods upon learning is one of the most important phases of the learning process and one that has been the subject of much investigation. In order that the records might be com- parable, the children were paired on the basis of approximately the same mental age, height, and weight. One child of each pair learned the material according to the standard form, on successive days; the other child learned it also according to the standard form, but on alternate days. The records were compared in order to de- termine whether, for this type of learning, presentation of ma- terial on alternate days or on successive days resulted in greater economy of learning. VARIANT FORMS Two variant forms of presentation of the material were de- vised as means of approach to some of the factors involved in the psychology of the learning process that were not considered in the standard form of presentation. Standard Form Interchanged. For the. standard form of presentation interchanged the blocks are placed upon the table and the pictures presented as for the standard form (Figures 2 and 3). The object of interchanging the associations is to compare the learning by this method with the child’s first learning and to study the factor of imterference or transfer. When the first trial is to be given for learning the re- versed or interchanged associations, the examiner says, ‘‘ These are the same pictures and the same blocks, but now there is going to be a different block that goes with each picture.’’ [Special empha- sis is placed upon the word different.] ‘‘I am going to tell you which block goes with each picture, and then I want to see if you can give me the right block that goes with each picture.’’ The ex- aminer holds up, in the left hand, the pack of pictures, with the first picture, the clock face, on top and says, ‘‘When I show you this picture you must give me this block.’’ The examiner picks up block 11 with the right hand and holds it alongside the picture of the clock face. ‘‘Remember, now [slight pause] this is the block that goes with this picture.’? The examiner then replaces the block in its assigned position upon the table. Using the same pro- cedure, the examiner goes through the complete set of twenty pic- tures and twenty blocks, but this time the pictures of Series I are LEARNING PROCESS IN YOUNG CHILDREN 13 shown with the blocks of Series II, and vice versa, the pictures of Series II are shown with the blocks of Series I. Learning on Successive Days: Children Studied Individually.— The material was presented according to the standard form in- terchanged to a small group of children who had completed the learning according to the standard form of presentation, followed by a check on learning. The method of learning was by trials on successive days. Single Series: Serves I and Serves II When a child is given the total number of twenty pictures and twenty blocks to learn, that is, both Series I and Series II, the ef- fect that the learning of Series I may have upon the learning of Series IT is a factor that is not easily determined. Since Series I and Series II are so comparable, it may be that the learning of the first series facilitates the learning of the second series; on the other hand, the latter part of the material may be more confusing because of the number of blocks and pictures that precede, or a child, particularly one of the very young children, may become fatigued before the twentieth picture and block have been reached. In order to study experimentally the problem involved, a num- ber of children were paired according to comparable results upon a mental test, and to one child of each pair the blocks and pic- tures of Series I only were presented; to the other child, the blocks and pictures of Series II only were shown. Figure 8 shows the ar- rangement of blocks on the table when Series I only is being learned by the child and Figure 9 shows the arrangement of blocks on the table when Series II only is being learned. It is to be noted that the block of each series has relatively the same position as the block with which it is paralleled. For example, the first block of Series I and the first block of Series II are the second from the left on the lower rows of blocks, and the third block of each series is the middle block in the upper row in each grouping. The direc- tions are the same as for the standard form. Learning on Successive Days: Children Studied by Pairs.— Learning for the single series was on successive days only. The re- sults of each child’s trials were compared with the results of the child with whom he was paired in order to study any differences that might occur between learning on Series I and learning on Series II, that is, when results from comparable material are com- pared on comparable children. 14 IOWA STUDIES IN CHILD WELFARE CHECK ON LEARNING CHECK ON LEARNING OF STANDARD FORM When three consecutive perfect scores have been attained on the standard form, one additional trial of a different type is given as a check upon the learning. The learning situation is now reversed. The examiner places the pictures upon a low table in two rows, as shown in Figure 4, in the order corresponding to the prescribed arrangement of the blocks as shown in Figure 2. The blocks are then shown to the child, one at a time, in the order shown in Figure 1. When beginning this part of the experiment, the child stands at the table in front of the middle pictures. The reason for hav- ing the child stand is that the pictures extend sufficiently far to- ward his right and left to necessitate his walking up and down in order to locate the required picture. The smaller child would prob- ably be unable to reach the pictures at the extremes of the two rows if he were seated. When the child has taken his position be- fore the pictures arranged on the table, the examiner says, ‘‘ These are the same pictures and the same blocks, but now you are going to have the pictures and J am going to have the blocks. When I show you a block, you point to the picture that goes with the block.’’ The examiner has the twenty blocks arranged in order in four equal piles of five blocks each and picks up block 1 from the top of the first pile. As the block is held up before the child, the examiner says, ‘‘Show me the picture that goes with this block.’’ The whole series of blocks is gone through in this way, the ex- aminer repeating the last sentence as each block is presented to the child. One trial only is given in this manner. Since this reversal of the learning situation is used simply as a check on learning, it is not counted in the total number of trials required for complete learning. In order that this record may stand apart from the records of daily trials of the learning, it is marked ‘‘Picture-block’’ at the top of the score sheet used for this trial. The number of the trial on the record of the third successive score of 20 is counted as the number of trials necessary for complete learning and this number is used for comparing individual records, for obtaining averages by groups, mental ages, and by sexes, and in the various correlations. LEARNING PROCESS IN YOUNG CHILDREN 15 CHECK ON LEARNING OF VARIANT FORMS For the check on learning of the variant form referred to as the standard form interchanged, the arrangement of the pictures upon the table and the procedure are the same as for the check on learn- ing of the standard forms. Now, of course, when the examiner shows the blocks of Series I, the child responds by pointing to the pictures of Series II and, vice versa, when the blocks of Series IT are shown, the examiner responds with the pictures of Series I. For the check on learning the single series form, Series I only or Series II only, the pictures (Figures 10 and 11) are in the same position as the blocks with which they correspond and here, as in the arrangement of the blocks for learning Series I or Series IT only, the pictures of each series have relatively the same position. RELEARNING AFTER AN INTERVAL OF ONE YEAR Several investigations that deal with the learning process have in- eluded retention tests on material learned under experimental con- ditions from a few months to a few years previous to the giving of the retests. The majority of these studies have dealt with the -retention of learning in which the problems have involved the ac- quisition of motor skill, such as ball tossing, mirror tracing, or typewriting. In the present investigation, a small number of chil- dren who had completed the learning of the entire series of twenty pictures and twenty blocks according to the standard form were available one year later and were given the material again in order to compare each child’s results on learning with his results on re- learning after an interval of one year. RECORDS OF DATA RECORD SHEET Results of each trial are kept on a record sheet; the form in- cludes spaces for the name of the subject, date of test, chronological age, mental age, and the number of the trial; outlines of the twenty blocks arranged as for the standard presentation (Figure 1), and a space, the lower half of the sheet, headed ‘‘ Observations. ”’ When a child hands the correct block to the examiner, the latter places a plus sign upon the corresponding outline on the record sheet. When an incorrect block is given, the examiner draws it upon the score sheet beside the outline of the block that should have been given. In the analysis of the data, it is necessary to 16 IOWA STUDIES IN CHILD WELFARE have this exact record of the incorrect responses instead of simply the indication of an error by a minus sign, in order to determine not only how many errors are made but exactly what errors are made, which blocks are most frequently confused, and to determine the order of difficulty of the formation of the associations. SCORES The method of scoring the individual tests is 1 poimt for each correct response; thus the perfect score is 20, except for the variant form of learning one series only, for which form the score is 10. The results of the experiment are scored on the basis of the num- ber of trials required for complete learning and it is this number of trials that is used throughout the statistical analysis. The num- ber of trials necessary for complete learning is understood to be the number of the trial on which the third perfect score is attained. The check on learning is not included, since this is not an intrinsic part of the learning situation. In the tables in which the actual score on each trial is given, the numbers italicized denote the score on the check on learning and are not to be included in com- puting the number of trials required for complete learning. OBSERVATIONS The examiner records anything that may be of value in inter- preting the results. Usually a few minutes are necessary after each trial for the examiner to write up these observations. Some- times a few words in regard to many of the observations are jotted down while the child is making responses, but they are always en- larged upon immediately after the completion of each trial. Un- der ‘‘Observations’’ are noted the type of the child’s response, whether it is quick or slow, whether the child picks up just any block, a block near at hand, or seems to be searching for a particu- lar one, and whether one or both hands are used, if only one, which one. It is particularly important to record comments made by the child during the experiment. Very frequently a child men- tions and points out the similarity of block and picture. Many children remarked upon the ease or difficulty of finding certain blocks. Sometimes, at the end of a trial, a child made comments on the situation as a whole. A very few of the children localized the position of some of the blocks, and, when nearly all associa- tions had been learned, reached directly with the right or left hand “posuvyolopUL wWIOT JUSMISUBIIV OY, °Z “SUT pilepueys 94} 10F puw uorezyueseid JO wa40F Prlepuejs oY} LOF P[lyo oy} o10feq posed 8B S¥00]qG oy} Fo a ‘06 94 TT SYOTQ “TI Setmeg pure ‘oT 04 [ syaoiq ‘T Soldeg ‘sattes OMy 24} FO JUsMASUBIIG [oT[VIVd VY} MOYS 03 SB OS peoeld yuowtsodxe Surusree] UOTPOVOL UOTPVIOOSSB OY} UI SHOOT AZUSA OY, “T “SIT ‘Z OINSL] UL SYDOTG oy} FO yey} 04 Suripuodsat1o9 JUIUIESUBIIB OY} “SUIUIBET UO Yoyo oy} OF PII oy} o10fZoq peoyid se sainqord oy} JO JUOMESUBIIe OUT, “Ff "SLT “poSuBYpAOJUL ULLOF PABPUBIS OY} LOZ puUB UOTZEZUESoAd dy} 0} popuesatd sv sotngotrd Azuoaz oY} JO dopo oT, “E “OL G Fig. 5. Child being shown the series of twenty pictures and the blocks that are to be associated with each picture. Bigse 64 | Child= re- sponding with a block when the examiner shows a picture. Fig. 7. As a check on learning, the child responds with a _ pic- ture when a block is shown. Fig. 8. The arrangement of the blocks as placed before the child for the single series form of presentation when Series I only is to be learned. Fig. 9. The arrangement of the blocks as placed before the child for the single series form of presentation when Series II only is to be learned. Fig. 10. The arrangement of the pictures as placed before the child for the check on learning when Series I only is used, the arrangement corre- sponding to that of the blocks in Figure 8. Fig. 11. The arrangement of the pictures as placed before the child for the check on learning when Series II only is used, the arrangement corre- sponding to that of the blocks in Figure 9, LEARNING PROCESS IN YOUNG CHILDREN 17 for the correct block, according to its position on the table. The child’s interest and attention during the examiner’s presentation of the series of blocks and. pictures and during the child’s performance were recorded. An interesting point noticed in many records was the confusion either between or among the pictures, or between or among the blocks. In several instances, this confusion persisted for several successive trials, before it was eliminated and in the case of a few of the children, the confusion reappeared after the correct. response had been made once, twice, or more times. The responses before which a child showed considerable hesitation were indicated on the record sheet. A large number of the children named aloud the pictures or the blocks, or both, while the examiner was showing the material to the child or while the child was hand- ing the blocks to the examiner. Such general items as these and others that refer to specific children only were recorded on their respective score sheets. CUMULATIVE RECORD CARD A cumulative record card was used in assembling data from the various trials of each child. This record ecard contains spaces for recording the name of the subject, date of birth, chronological age at the time of the learning experiment, date of Stanford-Binet ex- amination with chronological and mental ages, intelligence quotient, name of examiner, and dates on which the learning experiment is begun and finished. The forms and methods by which the material may be learned are listed. They are checked on each child’s record card to indicate the form and method used on his trials, the number of the trial on which the first perfect score is attained, and the num- ber of the trial on which the third successive perfect score is achieved. The card contains also a space for recording the num- ber of days absent and a list of psychological tests whose scores are recorded for purposes of correlation with the number of trials necessary for learning the associations of pictures and blocks. POSSIBILITIES FOR FuRTHER STUDY The material used in this investigation lends itself readily to adaptation to various forms of presentation and to various methods of learning. In addition to the forms of presentation used in this investigation, the block-picture, or standard, form of learning the material, the picture-block form, or check upon learning, the re- 18 IOWA STUDIES IN CHILD WELFARE versed, or interchanged associations, and learning only Series I or Series II, other forms of presentation have been planned but not used. One of these variations in presentation of material for learn- ing will consist in having the child learn the material according to the picture-block form of presenting the blocks one at a time to the child and having him respond with the pictures. The check in this case will, therefore, consist of one trial according to the stand- ard form of presentation of material. The material might be pre- sented in another form by arranging the blocks in the same order on the table as for learning according to the standard form, but shuffling the cards before each presentation so as to have the pic- tures mixed instead of in the same order for each trial. Conversely, the pictures could be presented in the original order as for the standard form of learning, and the blocks placed upon the table im mixed order. These suggested methods of procedure should result in Information upon the effect on learning of presentation and sequence of material. It has been planned also to devise a means of presenting the material according to a nonverbal form so that it may be used with foreign-born and deaf children. This form could be used with Series I only or Series II only with very young children. In further work with this material, other varia- tions of methods of procedure will suggest themselves, no doubt, as means of studying various phases of the learning process. CHAPTER II SUBJECTS OF THE EXPERIMENT The material of this experiment was given to 203 children ac- cording to one or two of the forms of presentation and one or two of the methods of learning. The records of twenty-three of the total number of children could not be included in the analysis of data because the learning had been discontinued on account of absences (ten children) ; or because the child had failed to learn completely the material by the end of the twentieth trial and the experiment was, therefore, discontinued (eleven children) ; or be- cause the record of complete learning was not of value since the children with whom they were to have been paired were absent the entire time that the learning experiment was being conducted (two children). The 180 children on whose records the study is based were secured from two sources, the Preschool Laboratories of the Iowa Child Welfare Research Station, State University of Iowa, and five public school kindergartens of a small city in Iowa. CHILDREN IN THE PRESCHOOL LABORATORIES Seventy of the 180 children were in the Preschool Laboratories of the Iowa Child Welfare Research Station. The Preschool Labora- tories comprise two groups of children, one group, designated the Preschool Laboratory, includes children whose chronological age range is from two to five years, and another group of children, known as the Junior Primary Group of the University Elementary School, includes children whose chronological age range is from five to six years. All of the junior primary children who were subjects of the experiment fell within the five and one-half-year age group, that is, the chronological age range of these children was from five years and no months to five years and six months. CHILDREN IN Five KINDERGARTENS One hundred ten children were in five kindergartens of the public schools of Mason City, Iowa. The kindergarten children who learned the material according to the standard form of pre- sentation were selected so as to be comparable in chronological age 19 20 IOWA STUDIES IN CHILD WELFARE to the junior primary children. The experimental work on learn- ing was done with the kindergarten children in the spring shortly before the close of the school year. As these children, with few exceptions, had passed the fifth birthday before entering kinder- garten, approximately seven months previously, it was necessary to select the children who were younger chronologically at the time of entering since those who were older were a few months too old to be grouped with the junior primary children. This factor of selection may possibly have eliminated the pupils who were not so bright, since the brighter pupils are usually the younger. Some of the kindergarten children who were subjects of the variant form of learning one series of the material only were a few months older than those selected for learning the material according to the standard form, but no child was older than six years and five months, the upper limit of the six-year-old group. RANGE OF CHRONOLOGICAL AGES In this study, two years of age refers to children who range in chronological age from one year and six months to two years and five months; three years, from two years and six months to three years and five months; four years, from three years and six months to four years and five months; five years, from four years and six months to five years and five months; and six years, from five years and six months to six years and five months. The chronological age distribution of the 180 children who com- pleted. the learning was from two years and two months to six years and five months. There are too few subjects at the lower ages for statistical analysis so this is limited to the children of the five and one-half-year age group, that is, from five years and no months to five years and six months, at which age the majority of children are found. One child only falls in the two-year age group and three children only fall in the three-year age group. RANGE OF MENTAL AGES The mental age referred to here is that derived from the Stan- ford-Binet mental examination. Each of the 132 children who learned the material according to the standard form of presenta- tion by various methods of learning was given an individual Stan- ford-Binet mental examination. The children in the Preschool Laboratories are given Stanford-Binet examinations as a part of LEARNING PROCESS IN YOUNG CHILDREN 21 the regular routine. The kindergarten children used as subjects had not been given Stanford-Binet examinations previous to this study. The examiner therefore gave Stanford-Binet examinations to the sixty-four kindergarten children to whom the material was TABLE 1 Distribution of 180 Children according to Form of Presentation of Material and Method of Learning and Relearning Form of presentation Relearning Variant Single ee eas Standard Standard form of : Sos} om A | presen- tz . a & 2 2 tation ° ° as o = os Children Groupings 5 & 5 2 E E E q4 3 Method of learning ie) fo) Be eee Ee ok Al @ S Za a 3 s —~| Sm Ham| Sm Sn Sn E m Sse eo Se |) 2B ee Se g| Re <0| ne Do Dd qo Preschool Individual 32 32 fi 1 Junior primary Individual 16; 15 1 8 Junior primary Paired 22 Eee L 7 Kindergarten A Individual 35| 35 Kindergarten B Individual 15; 15 Kindergarten B Paired 16 8 8 Kindergarten C Individual 12 12 Kindergarten C Paired 14 7 7 Kindergarten D Paired 12 6 6 Kindergarten E Paired 6 3 3 Ce0vsei a0 wie ako 24 24 6 7 presented according to the standard form. The forty-eight to whom the material was presented according to the variant form, known as the single series form, were not given Stanford-Binet examinations. The only mental test records of these forty-eight children were scores on the Detroit kindergarten test. The only two-year-old child in the entire group of subjects who completed the learning had a mental age of three years and eight months. It is the lowest mental age found among the children. The highest mental age, eight years and eight months, was that of a junior primary boy whose chronological age was five years and eight months. This mental age range from three years and eight 22 IOWA STUDIES IN CHILD WELFARE months to eight years and eight months includes the 132 children who learned the material according to the standard form of pre- sentation by the various methods of learning. DISTRIBUTION ACCORDING TO PRESENTATION OF MATERIAL The material was presented on successive days according to the standard form to 120 children, thirty-two preschool, twenty-six junior primary, and sixty-two kindergarten children. The ma- terial was presented on alternate days according to the standard form to twelve of the junior primary children. The material was presented on successive days, according to one of the variant forms, to forty-eight kindergarten children; twenty-four were given Series I only and twenty-four Series II only. The other variant form, learning of interchanged learned associations, was presented to seven. preschool children and eight junior primary children, all of whom had learned the material according to the standard form on successive days. One preschool and twelve junior primary chil- dren relearned the material after an interval of one year, according to the method of learning it originally, seven on alternate days and Six on successive days. Table 1 shows the distribution of the chil- dren who completely learned the material. The number of children on any one form, or method, is not suf- ficiently large to allow standardization of results according to chronological or mental age groups, except at the five and one- half-year age on the standard form of presentation of material on successive days. CHAPTER III RESULTS OF EXPERIMENTAL STUDY An understanding of how the children formed the required as- sociations involved in this association reaction experiment was based upon an analysis of the results in regard to the cues used by the children in forming associations, the order of difficulty, whether the confusions were due largely to confusion between pic- tures or to confusion between blocks, and whether there was in- terference or transfer when the learned associations were inter- changed. Factors such as distribution of work and rest periods, learning only one half of the material, and relearning were found to have considerable effect upon the length of time required for learning. The quantitative results derived from the children’s records were supplemented by comments of the children during the experiment and by observations of the examiner at the time of the child’s responses. Statistical results are based mainly upon the standard form of presentation, since the number of children on each of the variant forms is insufficient to justify statistical treat- ment of the data. The ages of the children are scattered over such a range that at only one age, five and one-half years, are there enough children to formulate a tentative norm for an age group. Each of these factors and the interpretation of the part played by them in the association reaction type of learning will be taken up and discussed separately. CRITERION OF LEARNING In giving the directions for learning the material used in this experiment, it was stated that the criterion of learning is three consecutive perfect scores of 20. If the child did not attain this record by the end of the twentieth trial, the experiment was dis- continued. Many of the children attained one score or perhaps two consecutive scores of 20, but failed to score 20 on the following trial. Sometimes the decrease was a matter of 1 point only, but in the majority of instances the decrease in score was 2 points, and in a very few cases, 3. The reason for the drop of 2 points was that the decrease in score was due to a confusion of two of the 23 24 IOWA STUDIES IN CHILD WELFARE parallel forms. For example, some of the confusions that oc- curred most frequently were responding with the six-pointed star when the five-pointed star should have been given, and later, re- sponding with the five-pointed star instead of the six-pointed ; simi- larly, the regular pentagon and regular hexagon were confused. In some of the studies found in the literature on the psychology of the learning process, a subject has been considered to have learned given material when he was able to respond with one per- fect trial. The results of the present investigation show that one perfect performance is inadequate as a criterion of learning. In many instances, at the time the subject is able to achieve one per- fect trial the learning of the material has just reached the threshold of learning. The associative bonds involved in the process of learning are not sufficiently strong to warrant a similar perform- ance on successive trials. It was decided arbitrarily before he- einning the work on this investigation to set three successive per- fect trials as one of the criteria of learning. Results indicate that when the child has given three successive perfect responses the material may be regarded as having been completely learned. Table 2 shows the number and distribution of children who at- tained three successive scores of 20 (perfect) and those who at- tained one score of 20, or two successive scores of 20, then a score less than 20, and later three successive perfect scores. One child who had scored 20 on two successive trials made a score of 18 on the following trial, and maintained this record for six consecutive days, followed by a score of 19, and then attained three consecu- tive perfect scores. The continuance of score 18 was due to a per- sistence of the confusion of the regular pentagon and regular hexa- gon blocks. This child secured her first perfect score on the ninth trial, but it was not until the twentieth trial that she had attained the third successive score of 20. Another child who had attained one score of 20, on the following day scored 17 only; this was fol- lowed by a score of 20 on each of the next two days, then 18 on the two following days, after which three successive scores of 20 were achieved. In this case, the unstable associations were in connec- tion with blocks 10 and 20 and 8 and 17. ‘The first score of 20 had been secured on the eighth trial, but it was not until the sixteenth trial that the third successive score of 20 was secured. The third successive perfect score was followed by the check up- on learning. So few of the children failed to make a perfect score LEARNING PROCESS IN YOUNG CHILDREN 25 on the check that it is believed that an increase in the number of successive perfect scores fixed as one of the criteria of learning would not improve the results. On the other hand, two successive TABLE 2 Distribution of 180 Children according to Attainment of Perfect Scores | Perfect scores Three successive | followed by 3 perfect. scores decrease S Form of presentation S Variant Variant = “he Neale Dace eae Leiba are i eed Children Groupings x : = & = F = H F te | ° om 3 o) ots B 5 4H qe Sp ort SY LS op on 3 beside hice (aeons é| 3 sae a ss 2 ee RE qa & yy S| See bp a| $ gs 8 £ gs 4 TD ea ane: TD nD. ww Preschool Individual 24 6 8 1 Junior primary Individual 15 8 1 Junior primary Paired 15 3 7 Kindergarten A Individual 30 5 Kindergarten B Individual 14 a! Kindergarten B Paired 14 2 Kindergarten C Individual 12 Kindergarten C Paired 13 i Kindergarten D Paired 11 1 Kindergarten E Paired 6 * Children to whom the material was presented according to the standard form inter- changed had learned it according to the standard form. perfect scores were quite frequently followed by a decrease in score. Therefore, the validity of the criterion of three successive perfect scores, although decided upon arbitrarily, seems justified by re- sults, ABSENCES AS A FACTOR IN LEARNING In discussing the method of presentation of material used in this study of learning, it was stated that each child’s record was be- gun on a Monday in order to have the same number of trials be- fore Saturday and Sunday intervened. Theoretically, this plan would give the same relationship between work periods and rest periods for each child so that the records among the children in each of the groupings would be comparable. As would be ex- 26 IOWA STUDIES IN CHILD WELFARE pected, a certain number of absences occurred during the time when some of the children were being used as subjects of the experiment. Since the effect of absence on the child’s record would be a factor that can not be measured, and since in any learning situation the distribution of work periods and rest periods is of great importance and must be taken into consideration in an analysis of the data, particularly in the comparison of the record of one child with that of another, it was arbitrarily decided before beginning this investigation to discontinue work with any subject who was absent for more than three consecutive days. Some of the children had a total number of absences that amounted to more than three days, but no child’s record that was used in this report shows an absence of more than three consecutive days. In one of the two investigations that are the only published work on learning of young children,’*® the children were given trials once a day five days a week, just as in the case of the children who learned the associations of pictures and blocks according to the standard form of presentation of this material. The authors of this investigation, a card sorting experiment, state, ‘‘For about one-third of the children there was an interval of approximately one month between the tenth and eleventh trial which resulted in a decrease of score in some eases.’’ In investigations of learning in which a special study has been made of the influence of distribu- tion of practice periods, particular emphasis is placed on the con- sideration that must be given to intervals between work periods. The comparison of the learning accomplished by one third of the children who were subjects of this experiment who had an interval of approximately one month between the tenth and eleventh trials (or just at the middle of the total length of practice time, since each child was given twenty trials), with the learning accomplished by two thirds of the subjects who did not have this one month in- terval between these two practice periods ean hardly be expected to give reliable results. In regard to the effect of a month’s absence on the practice curve for card sorting, the authors admit that there was a decrease of score in some eases, but conclude the discussion with a statement that ‘‘this decrease was in general no greater than when only an occasional day was lost.’’ The actual amount of decrease is not stated; neither is it known to how many of the total number of fifty-six children the reference is made by the statement that the one month’s absence ‘‘resulted in a decrease of LEARNING PROCESS IN YOUNG CHILDREN 27 score in some cases.’’ In fact, the data given regarding effect of ab- sences on the learning involved is too general; and in the tabulated results, no indication is made as to which of the children had this prolonged absence so that one does not know whether or not to interpret all the decreases in scores from the tenth to the eleventh trial as due to this absence, or to specific conditions that influenced the child’s learning at that particular time so as to cause a de- erease in score. Of the 180 children who formed completely the twenty associa- tions involved in the learning experiment herein reported, 126 had not one day’s absence. Fifty-three children were absent from one to five times and one child was absent seven times, which was the maximum, that is, among the total number of subjects under dis- cussion, there are almost two and one-half times as many children who did not miss a trial due to absence as the total number of children who had from one to seven days’ absence. The influence of absence on progress in learning has, therefore, been reduced to a minimum by discontinuing the experiment with the subjects who were absent more than three successive days. The complete record of number and distribution of absences for each form in which the material was presented is shown in Table 3. QUANTITATIVE AND QUALITATIVE ANALYSIS OF DATA MATERIAL The material devised for this experiment appealed to the chil- dren and interested them very much. The examiner always re- ferred to the material as a ‘‘game’’ when speaking to the chil- dren, and to the experimental situation as ‘‘playing the game.’’ Many of the children said that they liked to play with blocks and that they liked ‘‘the block game’’ as they ealled it, although the examiner did not so name it. It is interesting to note that every child who referred to the learning material as a whole used the blocks as a means of designating it. Often when the examiner went into the group room, a child who had already had trials on the material jumped up and asked the examiner, ‘‘Am I going to play with your blocks?’’ or ‘‘Can I play the block game now?’’ During the time when the material was being used, the children asked many questions about the blocks, such as ‘‘Who made these blocks?’’ ‘‘Where did you get these blocks?’’ ‘‘Can I take these IOWA STUDIES IN CHILD WELFARE 28 i G G 3 GlS= Ge TIGL 08ST I G 9 sABP DAISSIDONG selloes d[sUTg pe1eg Q uo,esiopuryy SL SL SsABP AAISSBDDNG sales YL SUTE pelegd q uozedsiopuryy 2 e 8 cal shBp oAtssaoong sdillos o[sulg poled Q uoJIedsiepuryy p 8 SL skBp oAIsseoong pivpuejZg [VNpIArIpul QO uUoJVsiopuryy Ye e OL SABP VAISSBDONG Soles djsUIg peed gq uoWediopury G 6 Go sABep dAtssaoong pivpurjg [enpiaripuy gq UeJIESIepuryy $ 1 cE skVep dAIssaoong pivpuejyg [VNplAipuy YW wozyiessiopuryy B 6 IL skep 0} BUI, V parepury¢ polled Arewtid izorne + L LE skep oAtssaoong paevpuryg ported Aiewtid ioe T T skVvp 0}BUI0}[V prepueyg [enprarpuy Aieuiid ion T be ral CT sABp oAISse0ong piepueyg [enprlarpuy Aieuiid iomne L I ZS i 9 c CL Ze skep oAIssaoong pivpueyg [Bnpriarpuy [ooyosolg L 9 G v & 6 L 0 uorp[ryo beds Aon oe FO 10q Sululve, JO poyyeyy | -ord fo wuo0g | sdurdnoin WOIp IYO sep ‘soouosq Vy -UNU 1240, Sururvs'yT JO poyyoyY pus [erozepY FO woryeqzuoserg JO WOT 0} SULpLosoe UoeIp[TyD OST FO soouesqy fo aoryNqrysiq pus requny e T1aviL LEARNING PROCESS IN YOUNG CHILDREN 29 blocks home?’’ ‘‘I wish I had some of the blocks at home.’’ ‘‘How did you make these blocks?’’ ‘‘Who gave you these blocks?’’ When the children were not actually seeing the material, the pictures seemed to play a much smaller part in their idea of the ‘‘game’’ than the blocks. None of the children while not in the examining room referred to ‘‘the picture game.’’ It was always “‘the block game.’’ This is due probably to the fact that during the learning, the child responds by picking up each block and hand- ing it to the examiner. The child, therefore, handles the blocks and his attention is directed more to them. The pictures brought out comments during the various trials. These comments were more likely to be in regard to the pictures as connected with the blocks, however, than to the pictures as pictures; whereas, in the ease of the blocks, the blocks as blocks interested the children. A few of the children asked the examiner, ‘‘Did you draw these pic- tures?’’ ‘‘Who made these pictures?’’ or commented upon the object represented in the picture. When the pictures were men- tioned apart from the similarity of blocks and pictures the com- ments made were generally in regard to the esthetic effect or af- fective tone produced by the picture upon the child. The comment was made very frequently in regard to picture 18, ‘‘That’s a pretty flower,’’ ‘‘TI like that flower.’’ The picture of the windmill, picture 19, was another pleasing one. The picture of the flower and the picture of the windmill called forth more comments of this type probably than any other of the pictures. ‘‘I think that’s a pretty little windmill,’’ ‘‘I like that windmill,’’ or a similar remark was made quite frequently. Almost every child made some comment upon the similarity of certain of the pictures and their corresponding blocks. The re- semblance between pictures and blocks was evident enough for even the youngest of the children to observe and comment upon cer- tain of the resemblances. Frequently a child pointed out a re- semblance and then added a comment that no doubt helped to fix the association in his mind. For example, several of the pre- school children who were learning the material shortly after Christ- mas remarked upon the similarity of block 3 and the coal car of the train picture with which this block is associated, and remarked, ‘*That’s like my choo-choo train too’’ or ‘‘I got one like that for Christmas.’’ With many children there was a tendency to repeat during each successive trial comments made on the first few trials 30 IOWA STUDIES IN CHILD WELFARE regarding the similarity of certain blocks and pictures. The only picture that seemed to be of an object unrecognizable to some of the children was the picture of the old-fashioned writing desk. When a child asked the examiner what the picture was, it was considered permissible to tell him. The fact that this object was more likely than any other to be unknown to the children did not interfere with their ability to form the required association between the pic- ture of the desk and block 12, since very few errors were made on this. In this case the obvious resemblance that serves as a basis for the child’s formation. of the associations seems to be a good index to the fact that the children were able to pick out correspon- dence between block and picture, even in case the object repre- sented in the picture was not recognized by the child. One small boy during each of his daily trials ran his finger along the sloping part of block 12 and immediately after ran his finger along the corresponding part of the picture, the slanting writing surface of the desk, and commented that they were ‘‘just alike,’’ or ‘‘just the same.’’ After several days of this repetition the examiner pointed to the picture and asked the child, ‘‘ What is that? What is that a picture of?’’ The child responded without any hesitation, ‘‘ An ice bex.’’ Nevertheless, the association of the correct bloek with this picture had been made from the first trial and the child con- tinued to respond correctly each time until learning had been com- pleted. FORMS OF PRESENTATION OF MATERIAL AND METHODS OF LEARNING Standard Form The 182 children who learned the material according to the standard form of presentation were distributed as follows: Number Children Groupings Method of learning of children Preschool Individual Successive days 32 Junior primary Individual Successive days 15 Junior primary Individual Alternate days j Junior primary Paired Successive days 11 Junior primary Paired Alternate days 11 Kindergarten A Individual Successive days 35 Kindergarten B Individual Successive days 15 Kindergarten C Individual Successive days 12 132 Among the 120 children who learned the material on successive days were eleven children who were paired off for the purpose of LEARNING PROCESS IN YOUNG CHILDREN 31 comparing their learning with that of the children who learned the material on alternate days. In the analysis of the data, the records of the children of the pairs who learned the material by the standard form of presentation on successive days may be included both with those of the children who learned the material on succes- Sive days and were not paired, and with those of the children with whom they were paired who learned the material on alternate days. The thirteen children who relearned the material after an interval of one year are included in the total 182, six having learned it on successive and seven on alternate days. Learning on Successive Days: Children Studied. Indwidually. — The 120 children who are discussed here include these groupings: Number Children Groupings Method of learning of children Preschool Individual Successive days 32 Junior primary Individual Successive days 15 Junior primary Paired Successive days 11 Kindergarten A Individual Successive days 30 Kindergarten B Individual Successive days 15 Kindergarten C Individual Successive days 12 120 The few children who learned the material on successive days and were paired with children who learned it on alternate days were included in this group since the form of presentation and method of learning are identical with those of the children who were not paired. For the thirty-two preschool children who learned the material according to the standard form of presentation and the method of learning on successive days, the range of trials necessary for complete learning was from five to twenty; for the twenty-sx jun- ior primary children the range was from four to fourteen; and for the sixty-two kindergarten children the range was exactly the same as for the junior primary children. The junior primary group and the kindergarten group each includes one child only who learned the material in four trials and one only who required fourteen trials in which to complete the learning. In each of these eroups, the majority of children required from five to nine trials in order to form correctly the twenty associations. Table 4 shows for each group of children the average number of trials required for learning for various forms of presentation and for various methods of learning. The records on the standard IOWA STUDIES IN CHILD WELFARE 32 GL’S 8h 6L°S $3 } Shep OAISsodONS UO SUIUIvETT “[]T satIag 19'S oS skep oatssooons uo SuIuIBey “[ sotsag worzeyUeserd JO WaAoF YURTIVA CLG ral skBp 07BUI0}[e UO SULUIBErT 20'L 39 CPL 98 VEIL 38 skep oAIssodons UO Surutvory woryeyuesoid FO WIoF prepuryg +e re) +B O t+ 5 fe) +8 —oo as og aB og as roe as om © a = ® de a “e © de a> ® do as 53 BR ae Bw x @ Ba i 6 BS Fh iy Fh 29 SUP dATSSODONG I soteg SOLIOS V[SUIG WIOF JUBLIBA 2] SOlog U0 SuluIVI] pus [ sotteg uO SutuIvE"T 0°06 0°06 0°06 S6T 0°06 O96T GST VET SIL sep 0}euioz[V prepueys 0°06 0°06 0°06 L6T 006 S6T O96T S6T OGL T6T 88l GOT 6FT GOL SABP BATSSADONG prepueys WIoF prspueyg :sXkVp oyeUI0}[eV UO Sutuivos, pus skep oAtssooons uo Surusvs'y] a1099 ue (GIG AL. dhe © ta et ee eet Dk Le ee Surure9y FO poyywoyy worzezuesord Sr acter ee ae SIG IT oe Sa Se eR ee yO WLIO SUIUIvEPY pus Suruivod8yT pues ‘Sururvoy fo poy ‘Teozey FO UoT}EZ,USSeIgG JO WO YoU uo spey, Aq se100g osvioAY JO suostivdu0g 9 HIAVL LEARNING PROCESS IN YOUNG CHILDREN 37 Figure 12 shows this comparison graphically. It is interesting to note that in each group of children, the learning curve reaches the perfect score of 20, but does not remain there for three suc- cessive trials. The curve that represents learning on successive days shows more fluctuations than the curve that represents learn- ing on alternate days. The curve representing learning on alter- nate days has reached and maintained three successive perfect scores of 20 by the ninth trial but the curve representing learning on successive days does not reach 20 until the tenth trial, after SCORE 16 Ce tba dak a a yf) ep 4 € aT es ECE 10 |] SRS Se ee eee Fig. 12. ee curves of Pan on successive days ( ) and learning on alternate days (----) when the ma- terial was presented according to the standard form. 14 TRIALS which there is a decrease before three successive perfect scores are attained on the fourteenth trial. Although the total number of cases, twenty-two, is small, re- sults on these records indicate that for this type of learning, pre- sentation of material on alternate days results in greater economy of learning than presentation on. successive days. Variant Forms Standard Form Interchanged: Children Studied Individual- ly.—Fifteen children, seven from the Preschool Laboratory and eight from the Junior Primary Group, were subjects on the phase of the investigation involving the interchange of learned associations. Each of the children had completed the learning ac- cording to the standard form of presentation of material, followed by a check on learning. The material was then presented according to the directions for learning, when the learned associations are in- IOWA STUDIES IN CHILD WELFARE 38 0é 06 06 OG ST LT GI 6T 06 0G 0G GT ST ST OT OL LT ST 0G 06 06 0é 06 06 06 0G 06 06 OT 06 0G 06 0G 61 0@ 0G 06 06 6I 6I io So +I On wk WO DWH Dass 0é 06 06 0G GI LT LT 6 03 06 06 8T 06 9T LW aes 03 0@ 06 06 0G GT ST ST 8ST 8T 8T 8T 02 0G OT ST FI ST ST ST 0@ 0G 0G 0G LT OT GT OT OT OT OT goes 9T 0é 0G 0G 0G BT 8ST 06 0G LT OG ST LT ST FT IT 0@ 06 06 0G ST 6T OT &T ST SS dose ST «0G 0G 0G 0G 8ST 0G ST ST 6T ST ST LT ST WoIp[tyo Jooyoseig S[BII} 81009 ee ool eee SS a Ee tee TWOTJVIIV A ee UesOL Okra bee Loe UV let bes Leite ORG 278s 2-2 0er Go mip [e101 [eu Plo be Ss bI Gt OF Ceo posuvyoi1eyUL posuevyo10zuUt posueyo10yUL posuvyo10yUL posueyoiozUr posuvyoiezuUt posusyo.1eqUt WAIOF PIBpUBYS I ‘MOF Prvpunyg WLOF pPlvpury¢g ‘WO, PAVPULIG WAOF PlVpuRyE, ‘WIoy Plvpuvyg WIOF pPlVpuRyg ‘M1OF PABPUBIG WAIoF PIVpuRyg ‘mI0F prepueyg W1OF Plepuryg ‘WUIOF PIBpUTI_ wWIoF prepuryg ‘WIOF PLVPULIG wotyeyuosaid JO WOT posuByoLoJUT ULIO,T PLVpURIG OY} 0} SutIpioooe Suturvsy 10F pus [BVlIo}V]Y FO WOI4VJUOSeIG FO WOT prvpueyg oy} 0} Surpcoose Surusvsy] 1oF woip[yH woejtq7 Fo spetry, Jo equnyy fo uostreduo0g L wavy, 9) ‘ e LEARNING PROCESS IN YOUNG CHILDREN ‘ZULUIBVOT UO YOY WO eIODS eJOUSP SOITBIT x a aaa a a a G 06 0% 03 03 ST IT | Pesueyosezur ‘m0; paepuryg t— 9 0% 02 03 0S GL STL ST WLOZ plepusyg 8 fi 0¢ 0% 03 03 SL FL OL L | posuvyosozut ‘mos prepueyg T+ 9 0@ 02 03 02 LI OT OL ULOF PIVpUByY L c 0@ 0% 0% 03 ST FI | pesuvyosozur ‘m10f prepueyg t— 9 08 03 03 03 ST YL ST WLOF PlBpuezg 9 G 0% 03 03 03 GI IL | pesuvyosoqzurt ‘w10F prepueyzg t— 9 02 03 03 03 LI SL BI WLOoy Plepueyzg c g 08 03 0% 08 GI ET | pesuvyorozur ‘ur10F prepuryg Ozc- ¢ 08 0% 08 0% 9T &I W1OF prvpuryzg P 9 02 03 02 0G FI FI OL | pesuvyoiozur ‘m0; prepueyg Or 9 0% 03 03 03 8L 8L 9 WIOF PLVpULyg ¢ ¢ 02 03 0% 03 ST 2 posuvyor1ezut ‘tof prepuryzg t— 9 08 03 08 03 SL 9T 9 WAOF PLBpuezG Zz L 0% 0% 0% 03 ST OT ST SL | Pesuvyorejur ‘u10y prepuryg a+ G 0% 0% 03 03 6L 6 UWLLOF PLBpULyG T ueip[Tys Areutsid Ione 40 IOWA STUDIES IN CHILD WELFARE terchanged, as explained in Chapter I. Each child’s record on learning when the associations were interchanged was compared with his record on the original learning. Table 7 shows this com- parison. The problem involved in this part of the investigation is to find out whether there is interference or transfer among the various as- sociations that have been formed, when these associations are broken down and others of a somewhat similar nature are built up. Re- sults indicate that there was very little interference. The learning of the material according to the standard form seemed to facilitate the forming of the interchanged associations. Of the fifteen chil- dren, nine children learned the interchanged associations in fewer trials than were required for the original learning, four required the same number of trials, one required one more trial, and two re- quired two more trials. Using the total number of trials required for original learning and the total number of trials required for learning the interchanged associations, the coefficient of correlation was computed by Pearson’s product-moment method. It was found to be .83 + .06, which is sufficiently high to be interpreted that the learning of the associations according to the standard form assists rather than interferes in the learning of the interchanged associa- tions. The children seemed to enjoy very much the interchange of asso- ciations. It amused most of them, particularly on the first trial. This trial was given just after there had been complete learning ac- cording to the standard form, when the child was very sure which block belonged with each picture. During the first trial of learning the interchanged associations, when the examiner held up a picture, several of the children glanced toward the block that they had learned to associate with it, and when the examiner picked up a dif- ferent block, the block that paralleled the one formerly associated with it, the children frequently laughed aloud. It seemed to amuse them to be told that another block ‘‘went with’’ the picture instead of the one that they were accustomed to associate with it. Often, when the examiner said: ‘‘Remember, now, this is the block that goes with this picture,’’ the child added, ‘‘ And this is the one that used to go with it.’’ If the child responded this way repeatedly, the examiner usually said, ‘‘ Yes, but now there is a different block that goes with each picture,’’ which is a repetition of a part of the direc- tions that are given just before beginning the first trial of learning LEARNING PROCESS IN YOUNG CHILDREN 41 the interchanged associations. Only two children commented upon the direct interchange of the blocks. One child said, ‘‘This one [block 1] used to go with the clock and now it goes with the owl, and this is the one [block 11] that used to go with the owl and now it goes with the clock.’’ These two children pointed out the direct interchange of the blocks in a number of instances. The other children knew that different blocks were to be associated with the pictures, but apparently did not observe, or at least did not com- ment upon, the direct interchange of relationships between blocks and pictures. Of the fifteen children who learned the material ac- cording to the interchange of associations, seven made a higher score on the first trial when learning the material according to the inter- change of associations than they had made on the first trial of the original learning by the standard form of presentation; six made a lower score; and two made exactly the same score on the first trial when learning the material either by the standard form or accord- ing to the variant form, interchange of learned associations. Single Series: Series I and Series II: Children Studied by Pars.—Sinee Series I and Series II of the association reaction ma- terial are so comparable, it was decided to give Series I to one child of a pair and Series II to the other child of the pair and compare results. Since none of the kindergarten children had been given a Stan- ford-Binet examination. before the examiner began to work with them on the learning experiment, and there was not time enough to give each child a Stanford-Binet, the children used in pairs on the single series variant form of presentation, learning on Series I and learning on Series II, were paired on the basis of their records on the Detroit kindergarten test. Since this basis of comparison may be not so accurate as comparison on the basis of similarity in mental age (Stanford-Binet), the children of each pair were selected so that there was not more difference than one month in chronological age between the two nor more than, 1 point in score on the Detroit kindergarten test. In some pairs there was no difference in chron- ological age, in others no difference in Detroit kindergarten score, and in other pairs, no difference in either chronological age or De- troit kindergarten score. With so much similarity of records, the mental ability of the children may be expected to be comparable. In the school system in which these four kindergartens are found, each kindergarten teacher administers the Detroit kindergarten test IOWA STUDIES IN CHILD WELFARE 42 8 ORO be OC Ole 6 seo ele Sa ts T &T mre res W 88 L Of OL. UTR GleaGy s0L& 52-5 él Ly See erry Gi 8 9 OF BOT Ul OL -0 > Laver ST Garg G W GL LZ Ole OL OLS OTS 6) Shree Se 39 ST 1 cle W L ¥ Oe OD e0Le Ob 2 LT RE a W B89 S OF OL OT OT 6 9 I LI PO ae if 9 v Of OT OL. 0D-2G IG ce ee W eg v CO OL Lae 0 2G T 4 6G Epos, | iL ¢ v Die OSs Tei ae T GG Lea W SP v OF OL OL OT 8 I 96 ERS gare | 14 6 OL OTRO Ce O86 0D) ch S282 OL: oF I ih él O aay. TL BE OT Die OT a Lee) Petia as ero Ors ete eee él 6 ae W & L Oe eat) Laer Ls Gt ce are aed, 6 DL meer ul BG 9 Gree Les Ee sO Lie Gage Oe: as T OT UD Blogs Ria W G 9 Chole Ot OLS8 = 65-20 U OL ee pays a AL my. 9 aOR OTS OL OL 8 Se 829 T LT | Rates he 2 wl I cee een A A ae A A i SS EEE: Ee I Seen ee O'S OS ied Le aaa oe ete et kL q Uozresiopury Se an iE US ean Dl I A le i eae oe Se ei eee ee a Be 81009 Hi aS V100G os S 5 et Cis Ge eu aol ona Ose. ho ome Cas AT e1009 = 4 0 eae a xog ila Wall e SooUoloyft [®70.L ae BS aN: o8B [BOISOT[OMOIYH queuttiedxe SUIUIve, UO P1090 480} WoJIVSIOPUTY JLOIJOG WO p10d0x II Selleg pus T sotsog Fo SururvsyT PITGO LOF (4S0J, UoJIVSLOpuTyy yLOIZO OY} WO Soe10dg FO SIseg oY} UO polIV_) WoLp[IYO uozrvSsropury yy Stry-43.10,7 JO satoog Fo wostivdu0y 9 a1av 43 ‘JUIUABIT UO YOoOYD UO 9IOOS O}JOUEP SOLVIT LEARNING PROCESS IN YOUNG CHILDREN G Ois-0L UL Uh Ss aes. 91 Osan F iT e9 & OF OL OT OL 9T Ogee ar oF ci 9 8 O10 Waa bea Les hoi oe ae bee 8 ol [izes 4, W Sy g Ole Ota Ula Ub La ss IL | eer a W G G OL ce 0 le. 0 Leica, 3 LI Guar are LD BP 9 OF 220 Ti Ue 0 Bal On 60 a LT Se ae WN v 9 (i Ole 020 G8 a Oar. LT 0 eee W BE 3 Ub UL Ol Oto ee I 8T Oa ae TL 6 4 Oise ee beau ek 61 Do ae TL BG S Oree0 120 laeiles Sec UE i 06 SES Ie8 W G L Oper OL 0 ULAeG aa Nees 8 06 Eaton W Sr v CLAUS O T= 8 u [ IG Ley ul I C UoJlVsilopuryy 9 Gb O Ls UG 20s bea Sas 8 61 Pe CoG AL BL 9 Cis ie EU be Geer ose I 61 aaron ce L 6 ie bee Gee Oar Boe een cee pee Be AL él Cte W %9 L Oke We Onn Oe 1G oe Oa BY I T $1 Late WN 9 L Ole Ube is OU Leeeee Ge a ak I FL ie eran IN 35 L IR CRUD RO) Coed ss os ees Sete U CT Piece hae re ¢ g OE AUP ea La 6 es 81 | Ee neg oH WN CP 9 GLU LOL Ole se be eS 8T 1 Ae orate Ge P L Ole AO Loy Ota Oe ag 0G tee saat T BE 8 WEDS AT ES Gag eS a Ts Oa ae sb aa I 06 eer ae) ul & S OLA Le 0 be Lahee ae 06 Nae oe J BG G ree Le OL Ose oe T 1G ae IN G 9 Pe 09 Rotana, 1) eae 9 poe ee IG Stas eG W er 8 Cie Ot Ole Oi -62" 65. 8 bts oF a IG Gore & ul T ) 19} AVSIOPUTST IOWA STUDIES IN CHILD WELFARE 44 6 ULOeOL WOT OF ¥G a re 78 WN BE v OLS 0s OL OT -9 VG Ui esieer, oS ul & € OL Ute L iu 66 | Se W BG v Glee Lae aU bed 66 Oe ae W 6 g Ole Oem Lao 2G 8T Oe aes W ST G Uae. UbeeOa 9 8- Pe See W L Se a a a a ec eg ee te ae ee WY UdZIVS1OpUryT SN ea nn on a ee ae ae -| 8100G osy = s a.1009 5 “ —_————. a xg | PITHO [ely sooueLo Iq, 958 [BOISO[OU0IYO quomtledxe SutuIvay uo pi0d0yy 480} UoJIVSIOpUIY YIOIYoq UO P1000z TI Sollog puv JT sotog jo Suruivs8T LOF (}S9], WoJLeSLopUryy F1OIJaC_ oY} WO saL0dg Fo siseg oy} WO poled) WoIp[yH uozresiopury 4yYyS1y-4710,7 FO saro0dg Fo wostreduroy (ponurzu0)) g aTAVL, LEARNING PROCESS IN YOUNG CHILDREN 45 to the pupils in her own classes. In order to keep constant as far as possible any error due to the personal equation of the examiner be- cause of the fact that the kindergarten teachers are not specially trained in the technic of mental testing, each child was paired with another child who was in the same class and had been given the De- troit kindergarten test by the same teacher. The first child of each pair was given Series I only and the second child of each pair was given Series IT only. Table 8 shows the basis for pairing the forty-eight kindergarten children whose records are used for this section of the study, and the scores obtained by each child on each trial of the learning ex- SCORE eT De nor eee IR ae or tae by ceria a] Dey ACSC ea ae 6 4 7. bg a | ot TRIALS Fig. 13. eee curves * Pa on aes LN learning on Series II (----). ) and periment and the total number of trials required for learning. Re- sults show a marked similarity between the number of trials re- quired by the first child of each pair to learn the ten associations of Series I and the number of trials required by the second child of each pair to learn the ten associations of Series II. The coefficient of correlation between the number of trials required for learning Series I and Series IT was .981 + .054. This was computed by Pear- son’s product-moment formula. As shown in Table 4, the average number of trials required for learning Series I was 5.61 and for learning Series II, 5.79. It is interesting to note that the average number of trials required for forming ten associations on successive days, found by averaging the average number of trials required for learning Series I and for learning Series II, was 5.75, while the average number of trials for learning the twenty associations ac- cording to the standard form of presentation and on alternate days IOWA STUDIES IN CHILD WELFARE 46 SG GG vG SG jl ah ead PS Gkb 2 08t 3 T € 9 pomeg | woyresiepuryy 9 9 GL poreg dq weyeo1opury L L vL pe1eg Q Uoyedsiopury ol 6L [enpiarpuy = Voyesopury 8 8 9T poeg q woJIesiopury ST GI enpiaipuy q weyEsopury GE cg - [BNPIAIpuy «=-W_—«MeyABSopury G T EL 9 IL rad peueg Aivurid rorunge 8 ti Te ei ee OL jenpratpuy Axreuid sone T 9 Giiet AG oe [BNPIAIpuy [ooyosoig xi eee ee Rama RH Manw & Ro ak He Pia Wed aes can OF POO th age © th ay ®t QO, Pe Oo il oe Ors o, Pe or 2 eS2 8isSe 8|/sSe 8 |e Se 8/8 Se 8/8 Se 8] x ae Pee are Pan eee ree Rr ee 5 2) °o ° i) °o 2) ao z 2 3 2 3 : 3 sep dAtssavong sep sep sfep sXkvp o soutdnoiy weIp[IyO 94VUI0}[V @AISSBDONG o7BU10}[V @AISSODONG “ SUIULBOT FO POY! E. a B B TI sotseg I sor10g posuvyo10zUt paepureyg wWIoF prvpuryg SOLIS I[SUIG qUBIIBA Ate eS ee i aS PR an Se alae an ee eee taamaiee Nate ee. A AE SRS worzeyUesoid Fo W407 | Re ass Ri ete CRIA Resa NG DSc ie aa Re Nc neal eee enn cee cee ee Sone en een SN eS SOLODG JOOFIOT OAISSODING odaIyT, FO JUOWUTeZV 19}JB SUTUIvIT UO YooyD FO s}[nsoy 6 WIavL LEARNING PROCESS IN YOUNG CHILDREN 47 was also 5.75. Figure 13 shows the average curves of learning Series I and learning Series IJ. Perhaps the children in the two groups were not exactly comparable but the results seem to have significance when viewed from the standpoint of effect on distribu- tion of practice periods upon learning. CHECK ON LEARNING Three successive perfect scores of 20 were followed by the check on learning. Results show that for the material used in, this in- vestigation three successive perfect scores is sufficient evidence of complete learning. As shown in Table 9, of the 180 children who completely learned the material, 164 attained a perfect score on one irial of the check and five children only attained less than a perfect score on the check. Eleven of the children who had completed three successive perfect trials had no check on learning because they were absent, the examiner had left for experimental work in another town, or a vacation began just after the child had completed the learning. In most instances, the children located the pictures very readily during the check on learning. It was only occasionally that a child walked up and down scanning the pictures while trying to find the one with which to respond. Check on Learning of Standard Form Of the 182 children who completely learned the material accord- ing to the standard form of presentation, 119 made a perfect score on the check on learning, four scored less than 20, and nine children had no check. Of the thirteen children who relearned the material according to the standard form after an interval of one year, every one seored perfect on the check. Check on Learning of Variant Forms Of the fifteen children who learned the material according to the variant form, standard form interchanged, that is, interchange of learned associations, each child made a perfect score of 20 on the check. Of the forty-eight children who learned the material according to the variant form, single series, by which the child learned either Series I or Series II only, one child only failed to make a perfect score of 10 on the check. RELEARNING AFTER AN INTERVAL OF ONE YEAR The work with the association reaction material was begun in the spring of one year with a small number of children and continued IOWA STUDIES IN CHILD WELFARE 48 v 0¢ 03 03 03 6L SUTUIvOTOY, 6 9 0g 02 03 21 ST SI suluiveyT = y 03 0S 02 0S “6E SUIUIBO[OY 3 9 OS. 02.022 Slee beni suluIveyT 9 S 0@ 03 03 03 8L #1 Surureajoy T 9 0@ 03 03 08 6L ST 8 suluiveyT g v 03 02-5067 0C= SL SULULBE[OY T g 0S -0cmUceuLats SUIUIveT F g 02 03 08 O08 SI S8TI SULUIBO[OY 0 g Li -03--06 06 “FE 29 sululvatT ¢ 4 03-06- 022-06 = 61 SULUIBO [OY T G 03-02 06-06 26h SI SuuIveyT fg ¥ 02-08 03° 0% OT SUIUIBO[OY T Gg 06 0S (0S 0G- een OL Sululee'T [ sXkep 0}8U10}[8 WO Surured8'T pA Alin este aan NOL ea NE A Ae ea TAT. St OTA eae te Pe, NE eR ona SUIUIBIT “ax 10J 81009 fiessooou | S[Bit} FO SUIUIBIT pro S[BlI} FO ROCUET Deas bee Y Lee ts ete bee DEO Line Gian mers ty ety ain ante Pokey Coen teem | JO WIT qequinu Ut Te30,.L esRel0e(q, [etty, IBIX 9UGC FO [VALOUT UV 1o}Ze Sulureo[ey pue Surusvs9y 10F spety, Aq setoog jo uostuvdwm0g OT A1avL 49 LEARNING PROCESS IN YOUNG CHILDRIEN “SUIUIBO, UO YIOYD UO 91008 9j0UEP SOI[VIT y SE LP aE PPR a a TR a aga” Ws secgeeks i ia Nn Nig Sah RNS Sea en, g 0g 02 03 O08 SULUIBOTOY SL CT 06 US— 06 OCESL-06.8be SL 6L. Sh Sl wl Slo Pieerio or suluiveyT eT 9 04 06 06602 Leslee LL SUIUIO[OY 3 8 03-2037 08 06s: bb Sie Lec SululveyT ZL ¢ 0 03 .03 02 SUIUIBOTOY T F Dee 0-06 BOCELT SULULBOT cE P 028-02 SUZ acest SUIUIVIIOY SZ 9 0 03 03 03 GL OT LZ sululveyT OT P 02 03 08 O08 8I SUIUIBOTO P 8 08. 08 0% <02 {6b 56 erPeert 9 SUIUIvOT 6 g 0F 02 0G 02 SUTUIvE[OY 8 IL 0é 08 -06- 08. 61 208. 02 S6laxig-4 0a ete er SuUIUIveT 8g sXep OAISsa.ONS UO SUTUIvOT i a a Ri nina enersncareececereoeesan ese nee aceeenmm enema mee re oe a SE SD ns oo NE eS! LSS 50 IOWA STUDIES IN CHILD WELFARE through the whole of the following school year. It was planned to repeat the experiment one year later with the children who were still available. Thirteen children who had learned the material by the standard form of presentation exactly one year previously were given it again in order to compare their scores and rate of learning on the original learning with their scores and rate of learning after an interval of one year. Of the thirteen children, seven had learned the material originally on alternate days and six had learned it originally on successive days. For this study of relearning, the ma- terial was presented to the child by the method of learning that had been used originally. In several instances, the child had been paired the previous year with a child who was no longer in the school and therefore not available as a subject for relearning of the material. This, however, did not prevent the child of the pair who was still in the school from being a subject for relearning since for the study of relearning each child’s record is compared with his own record of learning one year earlier, regardless of whether or not he was one of the children studied as one of a pair or individually. Table 10 shows a comparison of individual records by trials for learning and for relearning after an interval of one year. Subjects 1 to 12 were in the Junior Primary Group at the time of the original learning and in the first grade at the time of relearning; subject 13 was in the Preschool Laboratory at the time of original learning and in the Junior Primary Group at the time of relearning. With the exception of Child 3, who required five trials for learn- ing the material on alternate days and five trials for relearning the material by the same method, each child showed a decrease in the number of trials required for relearning as compared with the num- ber of trials required for learning. No child showed an increase in the number of trials required for relearning the material. In Table 7, the column giving the decrease in number of trials for relearning shows another interesting point. The range of difference in number of trials on relearning as compared with learning for the thirteen children is from 0 to 12. When the two groups of, children are con- sidered separately, those who learned and relearned the material on successive days and those who learned and relearned the material on alternate days, a still more interesting fact stands out. Of the seven children who relearned the material on alternate days, one required the same number of trials for relearning as for learning, four showed a decrease of one trial, and two a decrease of two trials. Of the six LEARNING PROCESS IN YOUNG CHILDREN 51 children who relearned the material on successive days, one showed a decrease of one trial in the number of trials required for relearn- ing, two showed a decrease of two trials, one showed a decrease of four trials, one a decrease of eight trials, and: one a decrease of twelve trials. The average of the decrease in number of trials for relearning of those who learned and relearned the material on alter- nate days is 1.14 trials and for those who learned and relearned the material on successive days is 4.83 trials. The children who learned the material on, alternate days had required fewer trials than those who learned it on successive days, and showed a smaller decrease in the number of trials for relearning. The explanation of this may be that the children who learned the material on alternate days SCORE J aa eee S| 20 = Ne gm to Va ba ie ea on | jal ae eet TRIALS Tae 6 AHO ea eed Fig. 14. Average curves of learning by the standard form of presentation ( ) and relearning after an interval of one year (----). made such rapid progress that they soon reached their physiological limit beyond which there is little room for improvement. The de- erease in number of trials required for relearning is therefore very small, since the original learning was accomplished at approximate- ly maximum speed of forming the required associations. The chil- dren who learned the material oni successive days made slower progress on the original learning and required so many trials for learning that there was still much room for improvement. The de- erease in number of trials required for relearning was therefore greater. Table 6 shows a comparison of the average scores by trials for learning and relearning after an interval of one year. By the sixth trial, the relearning curve has reached a score of 20 and maintained D2 IOWA STUDIES IN CHILD WELFARE it for three successive trials. The learning curve has reached 20 for the first time on the sixth trial and after maintaining this score for two trials, falls back before attaining three successive scores of 20. Figure 14 presents the average curve for scores by trials for learning and relearning after an interval of one year, and Figure 15, four in- dividual curves. All but one of the children accomplished re- learning after an interval of one year in fewer trials than were re- quired for the original learning; in the one exceptional case, learn- ing and relearning were accomplished in the same number of trials. As soon as the child finishes the first trial of relearning, the ex- aminer asks him three questions. First, ‘‘Did you ever play this game before?’’ If the child responds ‘‘Yes,’’ the second question SCORE 20 20 10 re 5 10 15 20 5 10 15 20 Child 1 Child 3 Child 2 Child 4 Fig. 15. Individual curves representing learning (heavy ) and relearning after an interval of one year (light ) on the standard form of presentation. Child 1 and Child 4 also learned the material according to the standard form in- terchanged (----). Child 1, it will be noted, scored 20 on three successive trials on relearning. is asked, ‘‘When did you play the game before?’’ After the child has responded, the examiner asks the third question, ‘‘Did you re- member how to play the game?’’ One child of this group only LEARNING PROCESS IN YOUNG CHILDREN D3 said ‘‘No’’ to the first question and one child said, ‘‘I don’t know. I don’t remember.’’ Lest : 9 a g Gea f T Vee bee OT Gee eae snevozeid fo 1oquinyy v OL €¢ 1 74 rt rm ri CIMMmnAOONDAMAAOS ra Ol O8T 9 sfep dAtssooong SOTLOS O[SUTG poled WW woesiopuryy SL sAep OAISSoDONG SOLIS [SUL poeg q wowediopury cae skep dAIsso00ng SOIIOS [SUG poleg Q woyVs1opuryy SL skep dArssaoong plepuvjg [BNpPIAIpuy O UoJLVs.opuUrysy OL sAep oAtssooong SOLLOS O[.dUIG peuegd q woWesiopursy CL skep OAISsooong pivpusyg [eNprlAIpuy gq Woeyesiopury GE skep dAtssooong piepuvyg [VNpPIAIpuy YW wUo,IVsd1opuryy yg skVp OFVULOI[V paiepuryg polled Areutad rzo1un¢ tL skep dAIssaoong pivpuryg poired Arvwtid 1ormne T skvp o}euse}[V pavpueyg yenpraipuy Aarewrd somne CT skep OAIssa0ong prepusyg yenptaripuy Airvwtid somne BE skBp OAISsoDONG piepuey4g [eNnpLlarpuy [ooyosolg waIpTIyo T1O1}B} UWS JO 10q Suraivsy FO poyyyy | -ord yo wai0oy | sdurdnory uwoIp[IyO -uMu [BIOJ, MOIP[IYO OSL FO soainH SutaivosyT ut snvozv[q FO worynqiaystq pues roquiny IL Davy, LEARNING PROCESS IN YOUNG CHILDREN 5d interpreted? Is the plateau an essential of all types of learning or is it peculiar to certain types only? The individual learning curves plotted for the children in the in- vestigation brought out some interesting, and it may be significant, facts in. regard to plateaus as found in this type of association re- action learning of young subjects. Of the 180 children (Table 11), 111 (61.6 per cent) have no plateaus in their learning curves and sixty-nine have from one to five plateaus. Of the 120 children who learned the material according to the standard form of presentation on successive days, the largest number of children who can be grouped together on the basis of a common factor, seventy-five (62.5 per cent) have no plateaus and forty-five have from one to five plateaus. It is an interesting fact that the difference in percentage between children having no plateaus on all forms of presentation considered collectively, and for the standard form of presentation, the form by which the largest number of children learned the ma- terial, is only 0.9. Another interesting point is the fact that the greater number of plateaus occurs among the curves of the younger children. Of the 150 children of the junior primary and kinder- garten groups, 104 (69.3 per cent) have no plateaus, forty-three (28.6 per cent) have one plateau, and three (2.0 per cent) have two plateaus. Among the preschool children, however, the situation is very different; of the thirty-two, nine (28.1 per cent) have no plateaus, and twenty-three (71.9 per cent) have from one to five plateaus. A smaller number of preschool children have no plateaus than have one plateau, and this situation is not found in any other croup of children in this study. From these facts it may be con- cluded that plateaus are not a necessity in this form of association reaction type of learning and that whether or not they occur seems to be a matter of individual differences and of maturity. An, interpretation of the plateau that seems applicable to the learning curves under discussion is that sometimes when new asso- ciative bonds are formed, particularly if several new bonds are formed at almost the same time, they are not formed with sufficient security to be fixed and a basis for the upbuilding of further asso- ciations. The fact that a plateau or even a decrease in score almost always follows a very large increase seems to suggest a relation be- tween. the period of rapid progress and the period when apparently no progress is taking place. Progress, in one sense, probably is oc- curring, that is, new associative bonds that were formed at the time 56 IOWA STUDIES IN CHILD WELFARE of the very rapid progress are becoming fixed and automatized. Until automatization of these bonds has taken place, there can be no further progress. Figures 16 and 17 show the plotted records of eight children who with one exception had a trial on the learning material according to the standard form of presentation and on suc- cessive days; Child 2, Figure 16, was given the material on alternate 0 TRIALS 5 10 15 20 5 10 15 20 Child 1 Child 3 Child 2 Child 4 Fig. 16. Individual learning curves of four children who learned the material according to the standard form of pre- sentation ( ). Child 4 also learned the material according to the standard form interchanged (----). The method of learning by Child 1, Child 2, and Child 3 was by trials on successive days; by Child 4, on alternate days. These four curves show plateaus following on large increases in the number of associations formed. days. The learning curves of the children designated Child 1, 2, 3, and 4, Figure 16, and Child 1 and Child 3, Figure 17, represent a large increase in number of associations formed, followed by a plateau, and the curves of Child 2 and Child 4, Figure 17, show a decrease in score following upon a large increase. These are a few random samplings, but in almost every learning curve in which one or more plateaus occur, a large gain is to be found immediately be- fore each plateau. LEARNING PROCESS IN YOUNG CHILDREN 57 Recently, there has come into usage in psychological literature the term ‘‘shock absorber’’ to denote a preliminary test that is given to subjects who are not accustomed to psychological test conditions in order to familiarize them with the type of test on which they are to be scored. The difference in response on the first trial between children accustomed to psychological tests and those unaccustomed LO Se eee Lae ee ee DA Fe VP pas (|| 4 eS 8 ee ALS a a ad ES |RSS fae PAN ea oe se |) ASS Bara Se oe ee ae ey 3 ie a (ee Wey || La [Pe LE a a ee ee ae : REESE EA RS REPRE TRIALS 5 is 10 15 20 Child i Child 3 Child 2 Child 4 Fig. 17. Individual learning curves of four children who learned the material according to the standard form of pre- sentation ( ) by trials on successive days. Child 3 and Child 4 also learned ae material according to the standard form interchanged (----). The learning curves of Child 1 and Child 3 show plateaus following on large increases in the number of associations formed, and the learning curves of Child 2 and Child 4 show a decrease in score following a large increase. to them is well demonstrated in the results obtained on the junior primary and kindergarten groups, in which the children are com- parable in age and school status. In the Iowa Child Welfare Research Station Laboratories, special psychological problems are being investigated constantly so that every preschool and junior primary child had had several psycho- logical tests previous to being a subject for the learning experiment, | and many of the children who had been in the preschool group the 58 IOWA STUDIES IN CHILD WELFARE year or years previously had taken a large number of psychological tests and were thoroughly habituated to psychological test condi- tions or to ‘‘play a game,’’ as the examiners always suggest when a child is asked to leave the play group to go into the examining room for a mental test. The kindergarten children were not accustomed to taking psycho- logical tests. With a few exceptions they had been given the De- SCORE 20 15 10 5 20 15 if BECHER EEE ae : PER hs | he a 2 |e a ae Sk BE TRIALS 5 10 15 20 5 10 15 20 Child 1 Child 3 Child 2 Child 4 Fig. 18. Individual learning curves of four kindergarten children who learned the material according to the standard form of presentation by trials on successive days. These curves show the tendency to begin with a low score followed by a rapid rise. troit kindergarten test seven months previously by their teachers, but some had never had a psychological test of any kind. The majority of junior primary children began with a score of 9 or more and several of them began with a score of 15 or more. The majority of the kindergarten children, however, began with a very low score. Several of the very bright kindergarten children scored 3 or 4 points on the first trial and a number scored 5 or 6 points. Those who began so low invariably showed a very rapid rise, after LEARNING PROCESS IN YOUNG CHILDREN 59 which their learning curves progressed very much the same as the learning curves of the junior primary children (Figures 18 and 19). In no case did a junior primary child make so large an initial gain as was made by many of the kindergarten children. The children were not afraid of the examiner, their codperation was excellent, and they seemed interested in the blocks and pictures. As far as could be observed, there were no extraneous factors that TRIALS 5 10 1s 10 ite 20 Child 1 Child 3 Child 2 Child 4 Fig. 19. Individual learning curves of four kindergarten children who learned the material according to the standard form of presentation by trials on successive days. These curves, like those in Figure 18, show the tendency to begin with a low score followed by a rapid initial rise. might detract from the children’s attention, or in any way account for the low score on the first trial made by all of the kindergarten children, unless it was necessary for the children to become accus- tomed to psychological test conditions. Since the kindergarten children complete the learning within the same number of trials as the junior primary children with whom they are comparable and the number of the trial on which the child attains the third consecu- tive perfect score is the measure of learning for this experiment, the factor of a low score on the first trial is hidden in the results. 60 IOWA STUDIES IN CHILD WELFARE SCORES A tabulated record was kept of the score made by each child on each trial. For certain comparisons, such as a comparison of the orzo xKR DOA OOTs ws a / % [ve 0 Soh pea "O 2 [elu 3 YOM. ST VCD Soe “a 1 17 |337] 3 Cuae HEN ee Wes PET uly, LON Ca REE OR er wick eC YO 2 3 | / vas 7b ea so 7 9 3 24 ey fa Wo a 3 Leie Pie o, iy dad te Raa Aa ix baer M237 3 67 We que 23 4 40 ccd OEE we Tt PRE 9 a Sane Big amet oe ey Ae 7291 4249 “D4 i is | Saat ey FOYesERtINS 3 4 20 "OH 5) qb 6 yeti 72 7 18 28 4 12 fa STAC EuTWak cae oy! 2 Yar 6 41 3 / P97 Ti 4 ¥ 332502 204 WK Fig. 20. Summary of the distribution chart of the analysis of successes and errors on the 434 trials of the thirty-two pre- school children who learned the material according to the standard form of presentation on successive days. Beginning at the top left corner, the order of the drawings of the blocks on both the perpendicular and horizontal represents the order in which the child is to respond. The method of reading the chart is as follows: Select any block in the perpendicular column; read across to the column below any other block; the number at the point of intersection is the number of times the block on the horizontal was given for the block on the perpendicular. The figures on the diagonal show the number of times each block was given correctly. records of several children or of the records of children whose scores are almost the same for several trials, the tabulated records of individual scores on each trial present the data more effectively than the individual learning curves. Among the tables that show tabulated individual scores are Tables 5, 7, 8, and 10. ANALYSIS OF ERRORS In order to analyze the data obtained from this experiment, it is necessary to know not only how many errors are made, but exactly what these errors are. A child’s correct response is indicated on LEARNING PROCESS IN YOUNG CHILDREN 61 the score sheet by a plus sign upon the corresponding figure. When a child responds with the incorrect block, instead of simply indicat- ing an error by a minus sign, the examiner draws the incorrect block upon the score sheet beside the outline of the block. By means of distribution charts, successes and errors have been tabulated and charted for the various groups of children according to each form of presentation and each method of learning. In this way, it may be seen how many times each one of the blocks has been given cor- rectly and how many times each one of the blocks has been confused with each of the other nineteen blocks. Figure 20 shows the frequency of correct responses and of errors of the thirty-two preschool children who learned the material ac- cording to the standard form of presentation on successive days, with a total of 484 trials. A study of this chart reveals certain in- teresting facts: The six-pointed star, block 20, was given forty times instead of the five-pointed star, block 10, and the five-pointed star, block 10, was given. fifty times instead of the six-pointed star, block 20. The regular hexagon, block 15, was given thirty-three times instead of the regular pentagon, block 5, and the regular pentagon, block 5, was given fifty-eight times instead of the regular hexagon, block 15. In these instances, the large number of errors was due to confusion between two similar blocks. In other in- stances, the errors were due to confusion between two pictures of somewhat similar objects. For example, block 17 was given twenty-three times instead of block 13, and block 13 was given thirty-three times instead of block 17. Block 18 is to be associated with the picture of a flying bird and block 17 with the picture of a duck. Sometimes, the confusion was not between block and block, or picture and picture, but between block and picture. This is illustrated best by the confusion of the round block, block 1, and the picture of a six-spoked wheel, picture 15. The round block, block 1, was given sixty times incorrectly in, response to the picture of a six-spoked wheel, picture 15. On the other hand, block 15 was given six times only for picture 1: This shows definitely that the errors were due neither to a confusion of one block with another, nor to a confusion of one picture with another, but to the confusion of a block with a picture. Figure 21 presents the records of the correct responses and errors of sixty-two kindergarten children who had a total number of 435 trials. A comparison of Figures 20 and 21 brings out an interesting 62 IOWA STUDIES IN CHILD WELFARE fact. The number of kindergarten children is almost exactly twice the number of preschool children, yet the total number of trials re- quired by the sixty-two kindergarten children is 485 and the total number of trials required by the thirty-two preschool children is 434. This is a striking example of decrease in number of trials re- quired for learning with increase in age. ODD AONSWTOROOTGAC I SEoOS& lelo 0009000003 00000006 00 opayvwr26 701036346 70000 th o fobs! | S/00 002231 2200) Ooo (b243537/2433 ¢NWl127 "O61 0 FB 1 OS REI 20820227 3 23290/BH37014303 4987007 HS o35/ 1 3Bil000 226d Ie OOT SEDO Page d BIA 20 ebb ts ak eS 00200 11 CBPAS 66022727 9 92300204 HPAO60 0214/5439 OF OO 390/22 Sf os 2 PO Oallal 1012957 toate Mh of0021)000 O6W12 22/21 000 m7o00203 23 0/27 PA// R14 2 fos 344 00326 2.) 390 © 7 70a eee % 220303540026 707) OB) ¢¥ 522 “O73 43246214 1 ¢ SBT 32 0 ys 6102232305933 32 9 3 "09200626126 00007 0 Obs 50d 0d ea Out 0s0 7 Orb el Olfe/ a2 Olli sr2324011/12603 Fig. 21. Summary of the distribution chart of the an- alysis of successes and errors on the 435 trials of the sixty- two kindergarten children who learned the material according to the standard form of presentation on successive days. This chart is read by the same method as the chart shown in Figure 20. A comparison of the confusion shown in Figures 20 and 21 shows a remarkable similarity of errors made by preschool and kinder- garten children. For example, by the kindergarten children (Figure 21) the six-pointed star, block 20, was given. sixty-four times instead of the five-pointed star, block 10, and the five-pointed star, block 10, was given fifty times instead of the six-pointed star, block 20. The regular hexagon, block 15, was given thirty-one times instead of the regular pentagon, block 5, and the regular pentagon, block 5, was given thirty-five times instead of the regular hexagon, block 15. Block 17 was given eighteen times instead of block 13, and block 13 was given thirty-seven times instead of block 17. Block 1 was given LEARNING PROCESS IN YOUNG CHILDREN 63 twenty-two times when picture 15 was shown, but block 15 was not given once for picture 1. Figure 22 shows the successes and errors of the twenty-two junior primary children. LHleven of these children learned the material according to the standard form of presentation on successive days, with a total of ninety-five trials, and eleven on alternate days, with SPOOR CILC of oo reO OO ++ Of Spek OrORpOsOr Ory ooo=-7-s0 008 Yzjow~ Dro OD 0.079 0A OOD fas] BER~OOOSOOOcr-*[PBIOOC SOS COCO CH OG =—'O2OnO CO 9 O10 79 OF) O~ F4O~ DO DO “FOSS w~0-0- 07 nn & 9S so Does "9 =O CO sv sOorR OP OSMhDQ DOO~~WD OD OwOCDOdD0 ae Or OH MO DO AHO DO OOK © ODO O~ ego CO ES _~— =) --2 O]g2]-0 O o- be Orb O SO pe Ove Os DONO O on ~s Ore OO~ 0 CO se -%& OO 0 0 HOO O °°) SEC Ome SO) OL Or Ge On=OTgO O-= On OOo oR WHSOH# OO O++9O 0 349-90 DOx9CO OO Of 9S DS FRO~D BO SDO OO OS- 905 6 H/exX1o0 oO 9 450-0745 ooon--corO 0-0 000 07r9 O00 8S =4 0 CO u~8h OO 0 0 0 O;| o= O'O°O 5-0 O°O-O"== OF 0 ei rp~ O 0-0 +9600 Ot o- GO O P+ FF UHBO=KS DOD O7O4O°5O. OO o~. Le) 9S 0 O #*o -KFO~ EOE eo R= Q-D—-S O- is Fig. 22. Summary of the distribution chart of the an- alysis of successes and errors of the twenty-two junior primary children who learned the material according to the standard form of presentation, on successive days for eleven children, who required ninety-five trials, and on alternate days for eleven children, who required sixty-four trials. This chart is read by the same method as the chart shown in Figure 20. The upper number of each pair of numbers at the intersection represents the records of the children who learned the material on successive days. a total of sixty-four trials. Since these children were so comparable on three variables, mental age, height, and weight, the reduction of thirty-one in the number of trials required for learning the material according to the standard form on successive days when compared with the number of trials required for learning the material accord- ing to the standard form on alternate days is probably the most significant fact shown by this chart. A chart was made representing a summary of the distribution 64 IOWA STUDIES IN CHILD WELFARE ‘CHILDREN cuTLDnEN CHILDREN AONE IO) es O 432 20 383 NPs | Bod 3O 354 D 186 © tal CUE SAP EL DENG P PON Tern SEP saat 0 ga) 3H ALON SH Ayan vayWEY) oer aie y aay Sy om {13 © 33! $a pe x ITT a fe A spd) go => 169 BSG HS, >, 3.27 ine 7 lod omnes 79 "do 317 0 a O 365 BOS IS 2 ea ales ant | nZ7 304 Oo |b QO 363 wO 302 @) if O55] sO 300 O 153 0 342 6O 268 m3 e QQ 338 Teekay L752 8 Oe374 Bx 246 © $4y bt =i 9X3 227 ew 14s te 369 wry 223 uw 140 x 304 By ih) 7ae sce Fig. 23. Order of difficulty, from least difficult to most difficult, in forming the required associations of twenty blocks and twenty pictures, found from an analysis of the records of thirty-two preschool children, twenty-six junior primary chil- dren, and sixty-two kindergarten children, each of whom learned the material according to the standard form of pre- sentation on successive days. LEARNING PROCESS IN YOUNG CHILDREN 65 chart of the analysis of successes and errors of the twelve junior primary children, five of whom learned the material according to the standard form of presentation on successive days, and seven on alternate days, with the frequency of correct responses and of errors for each group together with the number of successes and failures for each group when relearning the material one year later accord- ing to the standard form of presentation on successive or alternate days, according to the method that was used in the learning of the material one year earlier. Charts showing the summary of the analysis of successes and errors on each of the two variant forms of presentation of material were made also. These three charts are not included because of their general similarity to Figures 20, 21, and 22. ORDER OF DIFFICULTY IN FORMING THE REQUIRED ASSOCIATIONS The figures on the diagonal through each of the charts represent- ing a summary of correct and incorrect responses show the number of times each block was given correctly, since it is at these points that the drawing of each block on the perpendicular intersects with the drawing of the same block on the horizontal. In order to ascer- tain the order of difficulty in forming the required associations, the figures on each chart that are found on the diagonal were arranged in. order from highest to lowest. This represents the order of diffi- culty from least difficult to most difficult in forming the required associations. Figure 23 represents the order of difficulty from least difficult to most difficult in forming the required associations of twenty blocks and twenty pictures as found from the records of thirty-two preschool children, twenty-six junior primary children, and sixty-two kindergarten children each of whom learned the ma- terial according to the standard form of presentation on successive days. There is a very close similarity between the order of difficulty for the children in the three groups. As would be supposed, greater similarity was shown in the order of difficulty between the junior primary and the kindergarten children than between the preschool and the junior primary children or between; the preschool and kindergarten children. In order to study the apparent similarity, coefficients of correlation were computed by Pearson’s rank differ- ence method for each combination of two of the groups of children: Tene Is Preschool land sunior primary, oe eee 84 +.04 Preschool and kindergarten ............-...----.-s0e0c++-« 86 +.04 Junior primary and kindergarten .................... ci avoanrd Wf 66 IOWA STUDIES IN CHILD WELFARE The coefficient of correlation in each ease is a sufficient number of times larger than the probable error to be significant. COMMENTS OF THE CHILDREN DURING THE EXPERIMENT The pictures used in this experiment were designed purposely to resemble in some way the blocks with which they are to be asso- ciated, but an attempt was made to have the pictures not too similar to the blocks, in order to avoid the possibility of the situation’s re- solving itself into merely a matching of block and picture. Even the youngest child, aged two years and two months, observed and commented upon certain of the resemblances. The material elicited a great deal of comment from almost every child to whom it was presented. In addition to remarks that stated the resemblance of certain parts of the picture to certain parts of the block or to the block as a whole, or of the block as a whole to the picture as a whole, or to certain parts of the picture, many of the children gave to cer- tain blocks the name of the object represented in the picture with which the block corresponds. Many of the children when looking for a block asked, ‘‘ Where is the train?’’ ‘‘Where is the drum?”’ ‘Where is the flower?’’ or, after having found the required block, exclaimed, ‘‘ Here’s the train,’’ ‘‘I found the drum,’’ or ‘‘ This is the flower,’’ in each ease identifying the block by the name of the object represented in the picture that corresponds to the block. The blocks seemed to be the actual objects to the children. This may be due in part to the fact that few of the children knew the names of the var- ious shaped blocks. Some of the children referred to the round block as ‘‘the round one’’; some referred to the five or six-pointed star as ‘‘stars’’; only a very few of the children referred to the square block as ‘‘a square’’; and none of the other blocks was re- ferred to by name except when the block was called by the name of the picture to which the block corresponds. A large majority of the children tried to fit the blocks upon the pictures. Some of them responded in this way from the first trial; others, following the di- rections a little more closely, responded at first by handing each block to the examiner, but later, as resemblances stood out more clearly in their minds probably, began to fit certain blocks, or each block, upon corresponding pictures. Some of the children, when trying to fit a block on a picture to which there is an obvious re- semblance seemed annoyed or disappointed when it was too large or too small to fit the picture exactly. Many of the children commented upon this, particularly in regard to block 8 and picture 8, saying, LEARNING PROCESS IN YOUNG CHILDREN 67 ‘‘This block is too fat’’ or ‘‘Why doesn’t that go farther in?’’ re- ferring to the coneave sides of block 8. The fact that there was no stem attached to block 18, which corresponds with the picture of a flower on a stem, picture 18, annoyed some children who commented, ‘*It [the block] is just like it [the picture], but it hasn’t got this’’ [pointing to the stem on the picture of the flower]. More often the children asked a question in regard to the dissimilarity between pic- ture and block than made a statement in regard to the dissimilarity. Generally it was in the form, ‘‘Why doesn’t this [the block] have this [referring to some part of the picture that is not a part of the block] on it?’’ For example, the regular pentagon, block 5, seemed incomplete to many of the children because it has nothing attached to it that resembles the tail of the kite in the picture that’is to be associated with it, although in other respects the block and picture are identical in shape. This occurred in several other associations besides in connection with those mentioned. It seemed that the children after having found a similarity between block and picture looked for identity between the two. One outstanding fact of the learning situation as a whole was the general optimism of the subjects as expressed in their comments. It was a usual occurrence for a child to say, ‘‘I got them all right to- day,’’ as soon as one of the trials had been finished. Often, this was after one of the trials near the beginning of the child’s learning of the material when his score was far from perfect. Sometimes the child would not seem so positive of his absolute correctness and after stating, ‘‘I got them all right today,’’ added, ‘‘Didn’t I?’’ and looked questioningly at the examiner. Sometimes before beginning a trial a child said, ‘‘I got them all right the last time and I’m going to get them all right today,’’ although on neither the earlier per- formance nor on the later one had a perfect score been made. One preschool boy remarked, ‘‘I know them, don’t I?’’ at the end of the first trial. His score was 8. One time as the examiner was taking a child back to the kinder- garten room, she met another kindergarten child, Mildred, just out- side the door of the examining room. The examiner had not in- tended having Mildred next but when she remarked, ‘‘I came next because I wanted to play the game. Can I play next?’’ the exam- iner let her have her turn then. This was Mildred’s third trial and she made a score of 18. When she had finished this trial, she re- marked, ‘‘Maybe I’ll get them all right tomorrow.’’ Then an in- 68 IOWA STUDIES IN CHILD WELFARE stant later she added, ‘‘Maybe I won’t. We can’t tell yet. Who do you want next? I’ll go and get them.’’ Mildred showed more self- criticism than most of the children and realized that she had not made a perfect score. The next day, as Mildred entered the exam- ining room, she remarked, ‘‘I think I’ll get them all right today.’’ Many of the children remarked upon the ease in associating the correct block with a certain picture; fewer of the children remarked upon the difficulty. Comments such as ‘‘That’s an easy, easy one”’ usually referred to block 1 and picture 1, the picture of the face of a clock, and to block 18 and picture 18, the picture of a flower. Al- though many of the children were unable to recognize the picture of the old fashioned desk, picture 12, the block that is associated with it, block 12, is so similar that the response for this association also, brought out the comment, ‘‘That’s easy.’’ A great many of the children said that they hiked the game or made a remark such as, ‘‘This is a nice game.’’ There was only one child among the 205 with whom the experiment was tried who seemed to tire of the game. This was a boy who was the dullest child en- rolled in the Preschool Laboratories. After about fifteen trials he seemed somewhat reluctant to come to ‘‘play the game.’’ With this one exception, all of the children were interested in ‘‘the game’’ and the materials used appealed to them and held their attention. They did not tire of the repetition. Perhaps the same thing in a child’s make-up that causes him to want to be told the same story again and again, or to be shown the same pictures over and over, or to have the same songs sung repeatedly, makes the child like to ‘‘play the block game’’ for many trials without tiring. More of the individual comments of the children will be given in Chapter IV, where a special study is made of twelve interesting sub- jects of the experiment. STATISTICAL ANALYSIS It has been stated that the number of children at each age and on each form of presentation of the material is too small to yield re- liable statistical results except for the five and one-half-year age group of children who learned the material according to the standard form of presentation on successive days. Almost all of the statistical results given here are based therefore upon this one age group and according to this one form of presentation and one method of learn- ing. The correlations have been worked separately for junior pri- mary and kindergarten children. LEARNING PROCESS IN YOUNG CHILDREN 69 RELIABILITY OF THE EXPERIMENT The reliability of the experiment was found by correlating the seores on one half of the material (Series I) with the scores on the other half of the material (Series II) on all records of children five and one-half years of age who had learned the material according to the standard form of presentation on succcessive days. The re- lability coefficient of one half of the material correlated with the other half is .88 + .02 for the junior primary children and .83 + .0O1 for the kindergarten children. The reliability of the experi- ment was found in another manner, by correlating the scores on the odd and even items of the material on all records of children five and one-half years of age who had learned the material accord- ing to the standard form of presentation on successive days, it is 87 — .01 for the junior primary children and .85 + .01 for the kindergarten children. When the Spearman-Brown prophecy formula? was applied for the two halves of the material, the result was .91 for both the junior primary and kindergarten children. When the Spearman-Brown prophecy formula was applied for the odd and even items of the material, the result was .93 for the junior primary children and .92 for the kindergarten children. The correlation between Series I and Series II is based upon the records of the children who learned all of the material of both series together as one series. A correlation of the scores on Series I and Series II of paired kindergarten children, one of each pair having learned Series I only and the other of the pair having learned Series II only, gives a reliability coefficient of .98 + .05. A summary of these statistical results is as follows: Spearman- Brown Spm S88 the py ey sual Prophecy Children Same child Paired children Formula Scores on Series I and Series II Junior primary 83 +.02 ALM Kindergarten Oot. OL 98 +.05 OL Scores on Odd and Even Items Junior primary Hey (ame i! 93 Kindergarten 85 +.01 92 Nr 1 <7 -Dr, 70 IOWA STUDIES IN CHILD WELFARE CORRELATIONS BETWEEN THE NUMBER OF TRIALS NECESSARY FOR LEARN- ING ON THE STANDARD FORM OF PRESENTATION AND RESULTS OF STANFORD-BINET EXAMINATION The number of trials necessary for complete learning is the basis of scoring the results of this experiment. The number of trials necessary for complete learning was correlated with Stanford-Binet mental ages for various age groups, boys and girls combined, and for age five and one-half years, boys and girls separate. Of the 132 children who completely learned the material accord- ing to the standard form of presentation, 113 were given an individ- ual Stanford-Binet examination by the examiner who carried on the learning experiment. Nineteen of the children were given the Stan- ford-Binet examination by one of three other examiners of the Iowa Child Welfare Research Station. One of these examiners gave the Stanford-Binet examination to four of the preschool children, an- other, to one preschool child and five junior primary children, and the third, to nine junior primary children. Each of the sixty-two kindergarten children included in the total of 182 children was given the Stanford-Binet mental examination by the examiner who carried on the learning experiment. The factor of the personal equation of the examiner is therefore reduced considerably. All of the sixty-two Stanford-Binet examinations of the kindergarten children were given during the period in which the learning expert- ment was being carried on. A few of the preschool and junior pri- mary children were subjects of the learning experiment a few months after having been given a Stanford-Binet mental examina- tion. The mental age was corrected up to the time of the learning experiment in all cases in which there was a difference of one month or more in chronological age between the time of the Stanford-Binet examination and the time when the child was a subject of the learn- ing experiment. Mental Age (Stanford-Binet) by Groups Of the thirty-two preschool children who completed the learning of the material according to the standard form of presentation, one child was in the two-year age group, three were in the three-year age group, seventeen in the four-year age group, and eleven in the five-year age group. Twenty-six junior primary and sixty-two kindergarten children were in the five and one-half-year age group. The one child in the two-year age group and the three children in LEARNING PROCESS IN YOUNG CHILDREN 71 the three-year age group were disregarded, and a correlation was found between the number of trials necessary for complete learning on the standard form and the results of the Stanford-Binet exam- ination of the fifty-four four, five, and five and one-half-year-old children of the Preschool Laboratory and the Junior Primary Group. By the method of partial correlation for eliminating the influence of mental age, the correlation between chronological age and number of trials required for complete learning for these fifty- four children was —.27 + .08; by the method of partial correlation for eliminating the influence of chronological age, the correlation between mental age and number of trials required for complete learning was —.47 + .07. These correlations show that the number of trials necessary for complete learning correlates better with mental age than with chronological age. The negative correlations must be interpreted as follows: on the Stanford-Binet, the higher the mental age in relation to the chronological age, the better is the ehild’s performance; whereas, for the learning material, the fewer the number of trials required for complete learning, the better is the child’s performance. Thus, instead of having an increase in score to denote superior performance on the learning material, as is the usual method of scoring psychological tests, a superior perform- ance on the learning material is denoted by a decrease in the number of trials required for complete learning as compared with the num- ber of trials required for complete learning when the performance is either average or below average. The correlation —.47 + .07, when the influence of chronological age is eliminated through the method of partial correlation, shows that there is a relation to that extent between the two variables, mental age and number of trials, and that the higher the mental age, the fewer the number of trials re- quired for complete learning of the association reaction learning material. Correlations were found also between the number of trials re- quired for complete learning and mental age for the two groups of children of the chronological age five and one-half years (Table 12). For twenty-six junior primary children the correlation is —.48 -— .10; for the sixty-two kindergarten children, —.52 + .06. The cor- relations, therefore, are very similar to each other, and are similar also to the correlation —.47 + .07 found between the number of trials and the mental age of the fifty-four preschool and junior primary children. IOWA STUDIES IN CHILD WELFARE 12 So’ = Or— 4G GP STIL uspiesiopury pues Arvutid 101une¢ LO Ase shog uopiesiopury pue Arvutid soins 90° = 6g — “AG 29 UdpPLVS.IOPULyy Or + 8r— “1G 93 Arewtid ron | jooyosoig pourquios sexag “Ta Tad eq! [eorgojouoryO “wd # e380 a = woipItyD Bow quejsuod os W Fb SUOTJVUMMUVX JUTE -plozuUeyG JO Sj[Nsey puB [eLLo}VPT FO WOTye}UOSeIG JO WIO prvpuezyg uo SuruiveyT loz AressodoN S[VLL, FO IoqUINNY WosAjod SUOI}E[eII0H CL ATAVL LEARNING PROCESS IN YOUNG CHILDREN 73 Mental Age (Stanford-Binet) by Sexes Correlations were computed by sex for the eighty-eight junior primary and kindergarten children in the five and one-half-year age group. It had not been determined how nearly evenly the sexes were divided until the records were being separated for computing these correlations for each sex. The sex distribution of this group is as follows: Boys Girls Total RI ALIERG Dap T LYS Ver trl eee tan pene 8 18 26 CONG Ces hte Nat Cap ie, Skeid pap Saas, EAE Tins 35 27 62 43 45 88 The correlation between the number of trials required for complete learning and the mental age of the boys and girls of the five and one-half-year age group is —.58 + .07 for the forty-three boys, and —.40 + .08 for the forty-five girls. The difference between the correlation between the number of trials and mental age for the boys alone and the correlation between the number of trials and mental age for the girls alone is greater than the difference between the correlations for the junior primary and kindergarten children of the five and one-half-year age group, when the sexes are not con- sidered separately, and the correlation for boys is higher than for girls. All of the correlations based upon the number of trials required for complete learning of the association reaction learning material and mental age are found to be negative and must be interpreted as has been explained. The correlations are not very high, but in each case the correlation. is a sufficient number of times larger than the probable error for the correlation to be regarded as significant. Table 12 shows the correlations between the number of trials nec- essary for complete learning on the standard form of presentation of the association reaction learning material and the Stanford- Binet mental ages of the various groupings of children. CORRELATIONS BETWEEN THE NUMBER OF TRIALS NECESSARY FOR LEARN- ING ON THE STANDARD FORM OF PRESENTATION AND VARIOUS PSYCHO- LOGICAL TESTS ON CHILDREN OF THE FIVE AND ONE-HALF-YEAR AGH GROUP Of the twenty-six junior primary children of the five and one- half-year age group, records were available on the Detroit kinder- garten test and on two of the Pintner-Paterson performance tests (the Goddard-Seguin form board and the manikin test) for twenty- 74 IOWA STUDIES IN CHILD WELFARE two, and a record on the Montessori cylinders test for nineteen of the children. Each of these eighty-five individual tests had been given to the junior primary children by the examiner who carried on the learning experiment. Here again, the factor of the personal equation of the examiner may be regarded as of negligible influence TABLE 13 Correlations between Number of Trials Necessary for Learning on Standard Form of Presentation of Material and Scores on Various Psychological Tests on Children of the Five and One-Half-Year Age Group. © £8 Children Test etc Tepe bck =r A'S Junior primary Detroit kindergarten 22 —26 +.13 Junior primary Goddard-Seguin form board 22 Ba bie ih 8 Junior primary Manikin 22 —,05 +.14 Junior primary Montessori cylinders Board 1 19 aA aay) Ye" Board 2 19 02 7.14 Board 3 1h) 26 +.14 Kindergarten Detroit kindergarten 54 —.33 ==.08 when correlating the scores on these various psychological tests with the number of trials required for complete learning of the material of the association reaction learning experiment. Of the sixty-two kindergarten children of the five and one-half- year age group, a record was available on the Detroit kindergarten test for fifty-four children. These children were from three kinder- gartens so three different persons had administered the Detroit kindergarten test to the children in this group, since each kinder- garten teacher gave the test to her own pupils. The personal equa- tion of the examiner is not, therefore, a constant factor in these records. Table 13 shows the correlations between the number of trials re- quired for complete learning according to the standard form of presentation of the learning material and the scores on the Detroit kindergarten test for junior primary and kindergarten boys and girls of the five and one-half-year age group, and on the Goddard- Seguin form board, manikin test, and Montessori cylinders test for the junior primary boys and girls of the same age group. Detroit Kindergarten Test As shown in Table 13, the correlation between. the number of LEARNING PROCESS IN YOUNG CHILDREN 75 trials required for complete learning and the scores on the Detroit kindergarten test for the fifty-four kindergarten children is —.33 — .08. This correlation indicates that the children who are above average in mental age require fewer than the average number of trials for completing the learning. The correlation is a little more than four times the probable error so that the correlation may be regarded as significant. A different situation is found when the number of trials required for complete learning is correlated with the scores on the Detroit kindergarten test for the twenty-two junior primary children. The correlation in this case is —.26 + .18. The correlation is only twice the probable error and therefore it may not be regarded as signifi- cant. Two Performance Tests (Pintner and Paterson) The Goddard-Seguin form board was given to twenty-two of the twenty-six junior primary children. The number of trials re- quired for complete learning was correlated with the shortest time required for one of three trials on the Goddard-Seguin form board test. For this test, the shorter the time, the better is the perform- ance, just as in the ease of the learning material, the fewer the num- ber of trials, the better is the performance. The correlation in this case is positive, but it is too small to be of any significance. The cor- relation between the number of trials required for complete learning and the shortest time required on one trial of the Goddard-Seguin form board test is .12 + .14. The manikin test was given to the same twenty-two junior pri- mary children to whom the Goddard-Seguin form board test had been given. The number of trials for complete learning was corre- lated with the score on the manikin test. In the manikin test, the higher the score, the better is the performance, whereas in the learn- ing experiment the fewer the number of trials, the better is the per- formance. The correlation is negative again. It is, however, too small to be of any significance, —.05 + .14. Montessori Cylinders Nineteen of the twenty-two junior primary children were given the three parts of the Montessori cylinders test. The number of trials required for complete learning was correlated with the score on each of the three parts of the Montessori cylinders test. On this test, the higher the score, the better is the performance. The corre- lations, however, were positive. In no ease is the correlation of any 76 IOWA STUDIES IN CHILD WELFARE significance. For board 1 the correlation is .21+ .15, for board 2, .o2 —& .14, and for board 3, .26 + .14. The correlations between the number of trials required for com- plete learning according to the standard form of presenting the as- sociation reaction material and the results of the various psycholog- ical tests included here are of no significance with the exception, perhaps, of the correlation between the number of trials and the scores on the Detroit kindergarten test mn the ease of fifty-four kindergarten children, when the correlation is —.33 + .08. CHAPTER IV STUDY OF SPECIAL CASES In the statistical analysis of data, interesting features of individ- ual records are hidden. In order to bring out some of these inter- esting points, a selection has been made of the records of twelve children who learned the material according to the standard form of presentation. In one case the learning was on alternate days; in all other cases, on successive days. These records include those of four children who failed to learn the material within twenty trials and of eight children who completed the learning. The eases of fail- ure demonstrate the part played in the learning process by factors such as immaturity, emotional instability, and memory defects. The records of successful learning demonstrate unusual rapidity of some children in forming associations; the ability of the youngest child to complete the learning because of her superior mental development ; consistent alternation of gains and plateaus, in one ease; persistency for six trials in the confusion of two blocks after two successive per- fect scores of 20, in one case; and marked resemblance in learning the standard form and learning when associations are interchanged, in a number of cases. ANALYSIS OF Four CASES OF FAILURE IN LEARNING THE MATERIAL Eleven children who were given the material for learning in this experiment failed to make a perfect score within the limit of twenty trials. Although the records of these eleven children had to be dis- earded in the analysis of the results of the experiment since only records of completed learning were included, these cases of failure are very interesting from the standpoint of analysis of individual reactions to the material, and for the light which they throw upon the learning process in eases in which a certain amount of learning is achieved but the learning is not completed within the require- ments of the experimental! situation. Of the eleven cases of failure, nine occurred among the preschool children, that is, among the youngest of the subjects of the experi- ment. The material is undoubtedly too difficult for two-year-old children as a group, although one exceptionally bright child of this 77 78 IOWA STUDIES IN CHILD WELFARE age completed the learning within the number of trials set as the limit. It is interesting to note that of the eighty-eight children of kindergarten age, the total number of junior primary and kinder- garten children combined, there were two eases of failure only and these may be accounted for by special defects in each ease. The records of four of the children, two boys and two girls, who failed to complete the learning within twenty trials, have been se- lected for the purpose of individual analysis. Figure 24 shows the learning curves of these four children. SCORE 20 15 10 20 15 10 0 TRIALS 5 10 15 20 4s 10 ais) 20 Child 1 Child 3 Child 2 Child 4 Fig. 24. Individual learning curves of four children who failed to learn the material within twenty trials according to the standard form of presentation on successive days. Child 1 and Child 4 are girls; Child 2 and Child 3 are boys. CHILD 1 Sara, who belonged to the Preschool Laboratory group, was chronologically two years and eight months old at the time of the experiment. Her mental age then was three years and three months, and her intelligence quotient 123. Fail- ure in this case’ was due, undoubtedly, to immaturity. The task of learning twenty associations within twenty trials was beyond the stage of development of her ability at that time. The learning curve for this child shows, however, sev- eral interesting features. For trials 1 through 7, scores include three scores of LEARNING PROCESS IN YOUNG CHILDREN it. 1, two scores of 0, one score of 2, and one of 3. No doubt, during these trials at the beginning of the experiment, the whole situation was too large and too complex for the child to grasp, with the result that there was probably a buzz- ing confusion in her mind as to just what was expected of her. Successes dur- ing trials 1 to 7 may be regarded as due to chance with the possible exceptions of the association of block 1 and the picture of the clock and block 18 and the picture of a flower, since from the early trials these occurred correctly more times than any other association. After the seventh trial, the situation evi- dently began to become clearer in Sara’s mind, and from the seventh through the thirteenth trials there was a steady increase of score that shows a daily increase of 2 points during four of the six intervals between these trials and an increase of 1 point between each of the trials for the two other intervals. In plotting Sara’s learning curve (Child 1, Figure 24) no plateau was found up to the thirteenth trial, when a plateau occurs that extends over two trials; this is followed by an increase of 1 point after which there is another plateau extending over two trials, then another increase of 1 point followed by a plateau lasting for three trials. On the succeeding trial, the score increases by 2 points, reaching a total of 15. This is on the twentieth trial, however, when the experiment was discontinued. Although this curve represents failure as judged by the criteria used here for complete learning, it shows learning quite plainly and definitely. From the continued period of increase in the amount learned and from the fact that the shape of the curve from the seventh trial through the twentieth is very similar to many of the curves of completed learning, it seems justifiable to assume that this child would have completed the learning within perhaps five more trials. Sara seemed interested in the material and paid close attention while the ex- aminer showed it to her. Almost every time a picture and its corresponding block were presented to her, she nodded her head as if she understood the directions, but evidently she either did not understand them or, because of immaturity, was unable to form the necessary associations. The latter seems the more plau- sible reason. During the early trials, her responses consisted mainly in picking up blocks at random, especially the blocks directly in front of her, and frequent- ly she picked up the blocks in the order that they had been placed upon. the table. Beginning with the seventh trial, when Sara began to know some of the associations, she sometimes responded with an incorrect block, put it down quick- ly, and reached for the correct one. From this trial on, she really looked for the correct blocks and showed less of the tendency to pick up one block after another as they lay before her. She showed perseverance in her responses, par- ticularly with blocks 1 and 3. In the beginning, she was more interested in the pictures than in connecting them with the blocks. On the second trial she named several of the pictures, as ‘‘trees,’’ ‘‘star,’’ ‘‘little birdie’’ [flying bird], ‘‘round wheel,’’ ‘‘house’’ [windmill], and during almost every trial thereafter she named these and others of the pictures. After a few trials she responded by placing the blocks upon the pictures as if trying to fit them to- gether. On the twelfth trial, Sara made a comment that showed she was defin- itely comparing blocks and pictures; when the examiner showed her the picture of the maple leaf and block 10, the five-pointed star, Sara pointed to the star and said, ‘‘Too small.’’ On the next trial she made a similar comment by say- 80 IOWA STUDIES IN CHILD WELFARE ing in regard to block 14 and the picture of the sail boat, ‘‘ That’s too little.’’ After this she commented ‘‘ Just like it’’ for several pictures and their cor- responding blocks. CHILD 2 John, in the Preschool Laboratory, was chronologically three years and seven months old at the time of the experiment. His mental age then was three years and nine months, and his intelligence quotient 104. John’s learning curve (Child 2, Figure 24) is the most eccentric of the total number of 203 curves plotted. When the examiner showed it to one of the other workers in the Laboratory and to the child’s teacher, each of whom had known, worked with, and observed the child over a much longer period of time than the examiner, one of the immediate remarks was, ‘‘ Well, isn’t that just typical of John. He’s such an erratic child.’’ Their further remarks and the examiner’s observations seem to justify the statement that the learning curve of this child may be regarded as typical of his reactions to many other situations and may be looked upon as a profile of the child without overestimating and overinter- preting what the curve represents. One day John tries to hit every other child in the room; the next day, he plays so quietly at the sand table, or with some other equipment in the labora- tory, that one acquainted with the group assumes at first that he is absent. Not infrequently he incurs the displeasure of the children in the group because of striking them and because of his bullying attitude, so that later when he goes around the room and asks each child, one at a time, to play with him, he is re- fused by every child. At times he is extremely noisy, shouting aloud and bang- ing toys or furniture or anything that happens to be near at hand; at other times he is very quiet. He and two or three others came to school daily in a taxicab. It was a usual occurrence for this little boy to hit a certain little girl who came in the taxicab also. The driver had John ride on the front seat with him for several days; when John was allowed again to ride with the other chil- dren, he again began hitting the little girl who previously had been the object of his blows, but there were days intermittently when he would ride with the chil- dren without creating any disturbance. John’s learning curve begins with a plateau on a score of 2 points. This is followed by two rapid rises, so that by the fourth trial he has scored 11 points. The record of the next trial drops, however, to 3 points, a greater decrease than in any other of the total of 203 learning curves plotted. The score of 11 ob- tained on trial 4 is not reached or surpassed until trials 18 and 19, each of which has a score of 12 points. Between trials 4 and 18, there is a continual fluctuation in score. On the twentieth and last trial, there is a very unusual drop in score from 12 to 5. John did not seem bored with the material nor was he ever reluctant to come to ‘‘play the game’’ with the examiner. In fact, he is rather the type of child who enjoys adult attention and came very willingly to the examining room each day. During every one of the twenty trials he was talkative and usually kept up a running conversation during the entire experi- ment. Several times, on the first trial and on a few successive trials, after John had made a response, he picked up another block and asked, ‘‘ Where’s the picture that goes with this one?’’ Beginning with the second trial, when LEARNING PROCESS IN YOUNG CHILDREN 81 he named ‘‘the star’’ and other blocks and called the picture of the bell ‘‘a dinner bell,’’ he very often named either the pictures or the blocks. He seemed to observe resemblances carefully while the examiner showed him the blocks and pictures. On the fourth trial when the examiner showed him the picture of the six-spoked wheel and held up block 15 he said, ‘‘ This one [picking up block 1] should go with that.’’ Actually the shape of block 15 is more like the shape of the picture of the wheel than that of the block that is to be associated with it. In responding, however, John gave the correct block. The next day, when the examiner showed the picture of the six-spoked wheel and held up block 15, John again picked up block 1 and asked, ‘‘Why wouldn’t this be the right one?’’ and this time responded with block 1. Frequently, he fitted the block upon its corresponding picture, particularly if it were an association that had been definitely formed. Sometimes he asked, as he placed a block upon its cor- responding picture, ‘‘Does this fit with this one?’’ or ‘‘ Will it fit on that pic- ture?’’ ‘Will it fit on there?’’ Sometimes he tried to fit the wrong block on a picture as, for example, when he responded with block 3 instead of block 16, but turned block 3 in a perpendicular position instead of horizontal, the way in which it is used. In this turned position it more nearly resembles block 16 and could be made almost to fit over picture 16. This child seemed at all times interested in the ‘‘game’’ and responded quickly. During the first trial he said that he liked the blocks and volunteered the information that he had some blocks at home. The next day when the ex- aminer asked him to come out with her to play a game, he asked, ‘‘ Going to play with the blocks?’’ and seemed pleased by the prospect. Throughout his twenty trials this child seemed interested and paid close attention to all of the directions, codperated well, and responded without urging, but continued to make poor records. CHILD 3 One of the two kindergarten children who failed to learn completely the ma- terial used in this experiment is George. At the time of the experiment, he was chronologically five years and five months old, mentally five years and four months, and had an intelligence quotient of 98. He is a dull appearing child. During the learning experiment, George nodded his head each time the ex- aminer showed him a block and the corresponding picture. While he was re- sponding with the blocks, he worked quietly, looked carefully for each block, and evidently was trying very hard to make correct responses. His learning curve (Child 38, Figure 24) is a rather unusual one. It shows three plateaus, each of which extends over a period of three trials and another plateau that extends over two trials. At five points on the curve there is a decrease in score and at two of these points the decrease comes immediately after a plateau that extends over three trials. Fluctuations of this type in George’s learning made the examiner suspect a memory defect. The examiner asked his teacher for in- formation concerning him and was told at once that the child’s memory seemed to be defective. When carrying on a project that could not be completed in one day most of the children in the group would start in the next day with the work where it was discontinued. George seemed never to remember from one day to the next, however, so that it was necessary each day to give again to 82 IOWA STUDIES IN CHILD WELFARE him all of the directions that had been given the previous day. This and sim- ilar instances of inability to retain material had made the kindergarten teacher feel doubtful of promoting him to the first grade. It would be interesting to follow this child for a few years to note whether or not his learning on school work follows the form of the curve that represents his record on the association reaction learning. CHILD 4 The second of the two kindergarten children who failed to learn completely the material of this experiment was Gladys, who was, at the time of the learn- ing experiment, five years and four months old chronologically and mentally five years and six months. Her intelligence quotient was 103. Gladys seems to be of the emotionally unstable type. She showed a marked tendency toward indecision; in almost every trial she changed several responses. Frequently she changed a response two or three times. When Gladys was men- tioned to the teacher, she at once began to comment upon the child’s defective memory and told of several situations in which the quantity and quality of the child’s work fluctuated markedly from day to day. At the time the examiner was collecting data on the learning experiment the kindergarten teacher was having the children learn the names of various birds by means of large colored pictures of each bird. The names of seventeen different birds had been taught to the class and many of the children could identify the entire series. The teacher said that on one day Gladys would be able to recognize as many as fourteen of the bird pictures and on the following day only three, or perhaps four. Gladys’ learning curve (Child 4, Figure 24) shows the same number of points at which there is a decrease in score as is found in George’s learning curve. In the case of Gladys’ record, however, the amount of each decrease is larger gen- erally and at one time there is a drop of 5 points. Each time the examiner presented the material to her she nodded, but seldom made comments. Almost all of the children of Gladys’ age learn the material in less than nine trials. After the twelfth trial, the examiner asked Gladys if she were tired of playing the game. She smiled, shook her head vigorously, and said, ‘‘No, I like this game’’. Gladys did not show at any time a disinclination to come ‘‘to play the game’’. In this case, as in the preceding, the child’s teacher mentioned the noticeable memory defect and the tendency for both children to show marked fluctuations between good work on one day and very poor work on the following day. Un- doubtedly, memory defect is the most important factor in determining the failure of these two children to learn the required associations. Learning is di- rectly dependent upon memory, upon the fact that past experiences are pre- served within the psychophysical organism in the form of bonds ready to act when a given stimulus is presented. Unless the bonds have not been formed sufficiently strong, a repeated presentation of the stimulus will not call forth the correct response. ANALYSIS OF Four CASES OF SUCCESS IN LEARNING THE MATERIAL In addition to the curves discussed in Chapter III that follow the typical curve for the learning of the material used in this experi- LEARNING PROCESS IN YOUNG CHILDREN 83 ment, there are certain individual curves of successful learning that are especially interesting. The curves of four children who com- pleted the learning of the association reaction material according to the standard form of presentation (Figure 25) have been selected for analysis. All of these children learned the material by pre- sentation on successive days. Rak ACSI ON ae eS 7 [dsl i LV a Hpi een edey EEE Etech | bas Eau leads] Seeds pads be alts [sab va (Cohn a AP i i ne WAI og Laat faa feet £6 ead neay Hf ini Feo a ag bebe ee “on Sat ont -} SS Ea ey eA EEE EE «gh beal ke Gl tha el oo Bl eh heal a et on oO to Oo = oO = Oo TRIALS 5 10 ees 20 5 10 15 20 Child 1 Child 3 Child 2 Child 4 Fig. 25. Individual learning curves of four children who succeeded in learning the material within the limit of twenty trials according to the standard form of presentation on suc- cessive days ( ). Child 2 and Child 3 also learned the material according to the standard form interchanged (----). Child 1 and Child 2 are girls; Child 3 and Child 4 are boys. CHILD 1 The record card of Alice, in the Preschool Laboratory, shows that at the time of learning the association of blocks and pictures she was three years and eleven months of age, had a mental age of four years and six months, and an intelli- gence quotient of 117. Her learning curve (Child 1, Figure 25) is very inter- esting. It begins with a score of 2, reaches 11 on the second day, and rises to 15 on the third day. At this point, the large increase between trials ceases; there is an increase of 2 points for two successive trials, followed by an increase of 1 point that brings the score to 20. After having attained one perfect trial, the score falls back 3 points. These are regained on the next trial, however, and 84 IOWA STUDIES IN CHILD WELFARE this score of 20 is maintained for two more successive trials and thus the learn- ing is completed. The curve has risen from score 2 to the first perfect score of 20 in six trials, and undoubtedly the instability of certain of the associative bonds is due to the rapidity with which they were formed. This accounts for their breaking down, as shown in the decrease of 3 points in score after one perfect performance. CHILD 2 The learning curve of Elizabeth (Child 2, Figure 25), in the Preschool Laboratory, who had had her fifth birthday a few days before beginning the learning experiment, shows some unusual features. At the time, her mental age was five years and six months, and her intelligence quotient 108. Her record begins with a score of 3. It increases to 5 points on the second trial, and then has a rapid rise to 12 on the third trial. The next score is 15 points. A plateau occurs at this point, followed by a decrease in score. An increase of 4 points occurs on the next trial, but this is followed by a decrease of 2 points. The score then increases 4 points, which brings it to 20, where it remains for two successive trials. Then it drops 2 points and remains at 18 for six consecutive trials before rising to 19, then 20, and remains there for three consecutive perfect scores. This child completed the learning on the twentieth trial, the maximum number of trials allowed for learning, although trials 9 and 10 yielded a score of 20 points also. The plateaus and the several decreases after large increases in score seem to uphold the theory that both the plateaus and the decreases in score are due to the instability of certain of the associations that have been formed so rapidly and that time is needed for the automatiza- tion of the bonds. The outstanding feature of this learning curve is the plateau of six trials on score 18 that follows two successful perfect trials. The plateau is due to a persistence of confusion between blocks 5 and 15, the pentagon and the hexagon. An incorrect association for these two blocks, that consisted in a confusion between the two, had been formed early in the experiment and it was difficult for the newer, correct association to eliminate the earlier formed incor- rect associations. After Elizabeth had three successive perfect scores she was given the check on learning, on which she made a perfect score. During the check, she re- sponded very quickly and without even one hesitation. She held her hands be- hind her and each time brought her right hand forward to touch the picture. Just before completing the learning, it seemed that she had localized the posi- tion on the table of certain of the blocks and this seemed to carry over to the situation in which the material was presented as a check on learning. Elizabeth walked up and down as she made her responses and seemed to know whether to go toward the left or toward the right when a picture was needed. The ma- terial was presented to Elizabeth according to the standard form interchanged. The learning of the interchanged associations was accomplished in less than one half of the number of trials required for the original learning. The learning curve for the learning of reversed associations is shorter, but resembles in gen- eral the learning curve for original learning. Both show decreases in score fol- lowing upon large increases. Although the curve for learning the interchanged associations shows decreases it does not show plateaus. LEARNING PROCESS IN YOUNG CHILDREN 85 Observational notes on Elizabeth’s records state that she always paid close attention and showed much interest while the examiner presented the material to her. Beginning with the second trial, she looked carefully back and forth between picture and block as the examiner held them side by side before her. When it was Elizabeth’s turn to respond, she searched carefully for each block. At the end of the second trial, she remarked that she liked the game and when the examiner said that she would ‘‘play the game again,’’ Elizabeth seemed pleased and inquired, ‘‘Tomorrow?’’ At the beginning of the fourth trial, she remarked again that she liked the game, and added, ‘‘ These blocks aren’t so hard, are they?’’ She looked over the blocks carefully before picking up each one. By the fifth trial, she had formed fifteen correct associations. Several of the incorrect associations are interesting. They demonstrate the child’s ability to observe resemblances, although at times the wrong cue is followed and the wrong response results. For example, blocks 5 and 15, blocks 8 and 17, and blocks 10 and 20 were confused. Almost all of Elizabeth’s errors were due to a close but incorrect resemblance of blocks. Beginning with the fifth trial, she responded quickly and seemed to be very sure of her responses. She picked up each block and firmly laid it down directly in front of the examiner. Some- times while the examiner was going through the series of blocks and pictures, Elizabeth named each picture, as ‘‘That’s a clock,’’ ‘‘That’s a train,’’ and later when she responded with the blocks she called the block by the name of the picture, as ‘‘kite,’’ ‘‘drum,’’ ‘‘bell,’’ ‘‘wheel.’? This was done by a number of the children. When learning had been completed and the material was presented to Eliza- beth for interchange of learned associations, she responded with ten correct blocks on the first trial. She was very thoughtful in her responses. Several times during this first trial she was somewhat slow in responding, but after these hesitations she usually gave correct responses and appeared to be thinking very attentively during her silences. On this first trial of interchanged asso- ciations, she showed only one persistence of a former association. This was in the case of the picture of a flower and block 8. She worked quietly, but asked a few times, ‘‘Is this the one that goes with that now?’’ or ‘‘Is this the right one?’’ When she came to the examining room for the fifth trial, she said, ‘‘I got them all right yesterday, didn’t I?’’ The score on the day before had been 16, but she scored 20 that day. When she had almost finished this trial she re- marked, ‘‘I’m doing it quicker. Aren’t I doing it quicker?’’ Just as on the original learning, so on the learning of interchanged associations, Elizabeth re- sponded each time by reaching over both rows of blocks, and by placing the block before the examiner. On the sixth trial, as soon as she sat down before the blocks, she picked up blocks 11, 12, and 13, and said, ‘‘ These are the ones,’’ evidently referring to the ones that come at the beginning of the material when it is presented for learn- ing of interchanged associations. ‘‘ Does this go with the train?’’ and she held up block 13. ‘‘Does the tree go next?’’ and she reached over and picked up block 14. She responded quickly and accurately, and completed the learning of interchanged associations in nine trials. Twenty trials had been necessary for her to learn the material completely by the original form of presentation. 86 IOWA STUDIES IN CHILD WELFARE CHILD 3 Paul’s learning curve (Child 3, Figure 25) shows that he learned the twenty associations quickly for a child of his age. He was in the Preschool Laboratory. At the time of the experiment he was four years and eight months old chron- ologically. His mental age was five years and four months, and his intelligence quotient 114. After Paul had completed the learning by the standard form he was given the material to learn by the method of interchange of associations. He learned the required associations by this method even more quickly than he had learned them according to the standard form. Paul’s scores for learning by each of the two methods are remarkably similar. Nine trials were required to complete the learning when the material was presented according to the standard form of learning, and eight trials when the associations already learned were interchanged. There are no crossings of these two curves, as is generally observed. Seven of the nine scores on learning according to the standard form are paralleled by the scores on learning the interchange of asso- ciations. Since learning on interchange of associations is accomplished in one less than the number of trials required for the original learning, there is really one point only at which the learning curve for the original learning according to standard form of presentation is not paralleled by the learning curve for the interchange of learned associations. The latter curve is above the former at all points until the two curves coincide when complete learning has been attained. Paul saw resemblances between pictures and blocks on the first trial and tried to fit several of the blocks upon their corresponding pictures. In the meantime he made remarks, such as ‘‘ That’s the drum’’ and ‘‘ That just fits,’’ as he re- sponded by placing block 6 upon the picture of the drum. Once when he started to make an incorrect response he said, ‘‘ No, that doesn’t fit it.’’ He remarked, when responding with block 13 to the picture of the flying bird, ‘‘That’s the wing,’’ indicating that he had observed the similarity between the points on the picture of the bird’s wings and the points on the corresponding block. The next day when Paul came into the examining room for the second trial and saw the blocks, he said at once, ‘‘Here are those blocks that I played with before. I remember the game.’’ Then he added thoughtfully, ‘‘ But I don’t remember what is the picture that goes with each block,’’ and began picking up blocks and naming them. Paul paid very close attention while the examiner showed him the series of blocks and pictures. He nodded his head and said ‘‘ Yes’? al- most every time the examiner showed him a picture and the block that corre- sponded with it. He responded by placing nearly every block upon a picture, and said, ‘‘IT put that one on,’’ ‘‘That doesn’t cover it all up,’’ or ‘That just covers the picture up.’’ When the examiner showed him the picture of the evergreen trees, he said, as he picked up the triangular block, block 3, ‘‘ ’Cause that’s a Christmas tree, and that’s why that [triangular block] goes with that ’cause it’s this way,’’ and he ran his finger up and down the two sides of the triangle and on the corresponding part of the picture of a tree. He made re- marks about certain of the blocks, such as blocks 12 and 18, about each of which he said, ‘‘That’s an easy one’’ as he handed the block to the examiner. On the first trial with the interchanged blocks and pictures, Paul said several times, ‘‘ This used to be the one,’’ and then gave the correct response according to the method of interchange of associations. When the first trial was over LEARNING PROCESS IN YOUNG CHILDREN 87 by this new method, he smiled and said, ‘‘I just like this funny game,’’ and seemed amused by the fact that a certain block that formerly was paired with a certain picture was now paired with a different one. On the third trial, as soon as he entered the examining room, he picked up block 11 and said, ‘‘ This is the one and this [picking up block 1] used to be the one that went with the clock.’’ He responded very quickly and accurately. His score was 18, although it was only his third trial on the interchange of associations. Paul is an alert, talkative child who seemed very much interested in ‘‘the game’’ at all times. He responded by looking carefully for the blocks and from the first trial did not pick up the blocks at random. He showed a great deal of interest in his own success and frequently asked the examiner, ‘‘ Have I got most of them?’’ or ‘‘ Have I got them all yet?’’ Sometimes he said to the examiner, aS when he was half finished with trial 2, ‘‘I’m getting them all right today.’’ His score on this trial was 11. CHILD 4 Fred had the highest intelligence quotient of any boy in the Preschool Labor- atory. At the time of giving him the learning material, he was two years and eleven months of age, and had a mental age of four years and six months. His intelligence quotient was 153. Fred is very talkative and has a remarkable vocabulary for a child of his age. Fred’s learning curve starts at 1, jumps to 5, and after this rise falls back. Next comes a slight increase, followed by a large increase, 6 points, but a de- crease follows it again. Then the curve goes upward for four trials, at which it reaches the score of 20, but falls back again. On the next trial the score of 20 is regained and maintained for three successive perfect trials. Each time that a decrease occurs, it is after a large increase. When the examiner came into the group play room the day after Fred’s second trial Fred came up, took the examiner by the hand, and said he wanted to go to ‘‘play the game.’’ As soon as Fred saw the first picture and before the examiner had begun the di- rections, Fred picked up block 1 and placed it upon the picture. Fred was very much interested and enthusiastic while the examiner showed the blocks and pic- tures, and while he was responding with the blocks. After he had responded correctly with block 10, Fred picked up blocks 13, 18, and 19, and asked for each, ‘‘ We haven’t come to this one yet, have we?’’ By the fourth trial, cer- tain associations had been formed correctly, but at first when Fred did not know which block to pick up he responded with any block near at hand, with- out looking around. When Fred came to the examining room the day of the sixth trial, he picked up the triangular block, block 4, and said, ‘‘Let me see the Christmas tree’’ and then picked up the five-pointed star, block 10, and said, ‘*Tet me see the pretty leaf.’’ This was before the examiner began to show him the pictures and blocks. His interest in the material lasted throughout the learning experiment. On his fourteenth and last trial, he announced to the ex- aminer, ‘‘All ready for the game,’’ as soon as he came into the examining room. This may be considered as typical of Fred’s attitude; he is always ready to enter into any project. He is one of the youngest children in the preschool group, but is very mature for his age and talks like a child much older. He is a good example of a very young child’s completion of the learning because of superior mental ability. Child 4, Figure 25 shows Fred’s learning curve. 88 IOWA STUDIES IN CHILD WELFARE SUMMARY The eases selected for discussion in this chapter are only a few that have outstanding features that make them worthy of special at- tention. It would be interesting to know how far the curves of each child represented here will be typical of his learning reactions in other situations. Will Fred, the very young child who successfully learned the association reaction material, continue to stand out as superior to others in the group? Will John, because of his emo- tional instability, continue to show marked fluctuations from day to day in his ability to learn so that he will become the exasperation of the various teachers in whose classes he will be? In like manner, will those children who learned the material with unusual rapidity continue to be at the head of their classmates? CHAPTER V REACTIONS OF A SELECTED GROUP OF ADULTS TO THE LEARNING MATERIAL The comments of the children gave many clues as to how they formed the necessary associations of block and picture during the course of the experiment. The children, of course, could not be ex- pected to introspect purposely, but many of their comments take on the nature of introspections. Shortly after beginning to use the learning material, the examiner asked a few of the junior primary children at the end of a trial on which the score was 20 or almost 20, ‘‘How do you know which block goes with each picture?’’ or, hold- ing up a certain picture and its corresponding block, asked, ‘‘ How do you know this is the right block that goes with the picture?’’ The children could not tell when asked the question directly. Sev- eral children who voluntarily pointed out resemblances of blocks and pictures could not explain when asked to tell how they knew the eorrect blocks when shown the pictures. After a few attempts only, the examiner decided not to ask the children these questions and to rely wholly upon the information volunteered by the children in regard to how they associated certain blocks with certain pictures. It was thought that if the material were presented to a few adult subjects who were well qualified to give introspections as to how they formed the associations, and to state what to them seemed to be the order of difficulty in forming the associations, an interesting com- parison might be made between the comments of the children and the introspections of the adults. Seven adults trained in introspec- tion agreed to act as subjects. Subjects A, D, E, and G are grad- uate students who hold appointments as research assistants in psychology ; subjects B and C are graduate assistants in psychology, and subject F' is a widely known professor of psychology. Two of the graduate students received the degree of doctor of philosophy a few weeks later, and each of the others has had more than the re- quired work for the degree of master of arts. Subjects A, B, D, and E are women; subjects C, F, and G, are men. The material was presented once only to each adult according to the standard form of presentation. After the subject had responded 89 90 IOWA STUDIES IN CHILD WELFARE to each of the twenty pictures, the examiner showed one at a t:me each block and its corresponding picture, and asked the subject to state exactly how he formed the association between block and pic- ture in each ease, and to introspect on the formation of the associa- tion, as fully as possible. Although the scores of the adults are not taken into account in analyzing their introspective reports, it is in- teresting to note that of these seven highly selected adults, two only scored 20, one scored 19, one 18, one 17, one 16, and one 14. It seems, therefore, that the material is such that even for these adults there is an opportunity for learning to take place. REPORTS OF INTROSPECTION OF ADULTS A few general statements will be made before taking up responses to specific blocks and pictures. Both adults and children have cer- tain factors in common in the process of forming the required asso- clations, as shown from a comparison of comments of children and reports of introspections of adults. The adults mentioned specifically certain similarities between blocks and their corresponding pictures, and the cases in which the similarity seemed most obvious to the adults were also those men- tioned most frequently by the children. Almost every child at some time during the experiment responded by placing certain of the blocks upon their corresponding pictures and tried to fit the blocks to the pictures. As the directions say nothing about a resemblance between block and picture, the response of trying to fit the block to the picture is due obviously to the child’s perception of the existing similarity. Some children tried to fit almost every block to its pic- ture and showed quite plainly that they were pleased when the blocks and pictures were very similar in outline and that they were an- noyed when certain of the blocks did not correspond so well to their respective pictures. Each of the seven adults mentioned ‘‘a tend- ency to try to fit the block to the picture,’’ or, a tendency to say in- wardly ‘‘The same’’ or ‘‘The same; the block fits the picture.’’ In the case of the adults, it was a mental fitting of block to picture, but in the case of the children, it was an actual placing of the block upon the picture. Each of the adults expressed a feeling of annoyance in the cases of most dissimilarity between block and picture. In the ease of the children, annoyance over these dissimilarities was most generally shown by facial expressions. In certain instances, the child said, ‘‘Why doesn’t that fit?’’ ‘‘That’s not exactly the same’’ LEARNING PROCESS IN YOUNG CHILDREN 91 or ‘‘That’s larger,’’ ‘‘That’s smaller,’’ or ‘‘That’s too large,’’ or ‘Too small,’’ accordingly. Many of the children named aloud all blocks that were of shapes with which they were familiar, such as ‘‘The star,’’ ‘‘The round one,’’ ‘‘The square one’’ or ‘‘The long one,’’ as the rectangle was frequently referred to, or ‘‘The pointed one,’’ as several children ealled the triangle. Undoubtedly, the names of the geometrical forms of many of the blocks were unknown to the children, since it would be difficult for most adults to call each of the twenty forms by its correct geometrical name. It would seem that the ability or inability to name the shape of the block would have its influence upon rapidity in forming the required association for the block. Two of the adults mentioned specifically a tendency to name each block as presented and mentioned the experiencing of a slight an- noyance in each case when it was not easy to name the form of the block instantly. One of these two adults mentioned also a tendency to name the pictures and thus, in part, her formation of associations was based on association between name of block and name of pic- ture, as ‘‘triangle-Christmas trees,’’ as well as on similarity be- tween the shape of the block and outline of the picture. Each of the seven adults mentioned that in certain cases the sim- ilarity of the block as a whole to the picture was noted, and in other eases the association was between a small part of the block and a small part of the corresponding picture. The same holds true of the children in many instances, where comments such as, ‘‘ This block is just like this picture’’ or, when one small element of the shape of block or picture was selected as, for instance, the sloping side of block 12 over which one child daily ran his finger and then ran it over the corresponding slope in, the picture of the old-fashioned desk and remarked that they were ‘‘just alike’’ or ‘‘just the same.’’ By both children and adults, block 8 was associated with the picture of the goblet because of the curve on each side and not by the resem- blanee as a whole. This holds true for several other blocks and their corresponding pictures. Four of the adults mentioned that in cer- tain instances the resemblance was seen first as a whole and then as a resemblance between elements. S'x of the adults mentioned that when a part was selected and responded to, there was a tendency to ignore all other parts. : Two of the adult subjects mentioned localization as an aid to learning. The position of the middle blocks or end blocks was fre- 92 IOWA STUDIES IN CHILD WELFARE quently thought of and noted as the experimenter picked up the block to show the subject with which picture it was to be associated and noted again as the experimenter replaced the block upon the table. It was found that a very few of the children, when all or nearly all of the correct. associations had been built up, responded in part through localization of the blocks. Six of the adults mentioned ‘‘subvocal verbalization’’ or ‘‘verbal inner speech’’ as an aid to learning; six mentioned kinaesthetic imagery in connection with certain of the associations, and four mentioned visual imagery. There is, of course, no data available as to these processes in. the preschool subjects. | One adult subject mentioned that in the ease of her responses it was ‘‘all or nothing.’’ She either knew immediately or did not know at all which was the correct response. Her score was 16. Two other subjects stated that when they did not know which one was the right block they looked around for the block most like the picture and in some instances this resulted in arousing the associa- tion that had been in mind when the examiner was demonstrating: which block and picture were to be associated, and therefore the cor- rect response was given. Comments made by the children and introspections on certain as- sociations made by the adults in regard to specific blocks and pic- tures should be of interest in addition to the summarized statements that have just been given. The association with which there was the most affective tone was between the picture of the flower and the corresponding block, block 18. Subject E stated, ‘‘The block was just the same general shape as the picture of the flower. The curves of the block and of the picture had a sort of soothing, restful effect.’’? Subject F said, ‘“That was a real joy.