\[l :fo [EVESTEGAE’ECEI’V’ 0!" THE 3NTERACTEON OF 105 521 THS F’i’IEFHTEi} ARK? SS‘QK’EH W‘fiRDS Thesis in! he Degree of Ph. D. EAICMGAQ SEW??? HMVERSSTY Kaw'z‘. 323 CC)“: ‘ .0 6% 3 f.- ‘E i q THESIS LIBRARY LIBRA Y ~\l\\[fifififl‘EiflijflflggkjflifluH ‘ Waist? This is to certify that the thesis entitled t AN INVESTIGATION OE THE INTERACTION ‘ OE PRINTED AND SPOKEN WORDS presented by Keith D. Cox has been accepted towards fulfillment of the requirements for Ph.D. Communication degree in .a. ( . ,. K; 4 x? it Saw-u, M Major professor "3 Date November 29, 1965 0-169 AN INVESTIGATION OF THE INTERACTION OF PRINTED AND SPOKEN WORDS By %{ \_.- _l Keith D? Cox A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1965 ABSTRACT AN INVESTIGATION OF THE INTERACTION 0F PRINTED AND SPOKEN WORDS by Keith Cox This study investigated the hypothesis that subjects would learn a list of foreign words in fewer trials if printed and spoken forms of the words were presented together instead of singly. Paired- associate and serial learning techniques were used. This hypothesis assumes that the referential meaning of the printed word is mediated through the spoken word. Experimental evidence supporting this may be found in studies of vocalization and sub- vocalization in good adult readers. Linguistic evidence comes from sound/symbol correspondences. Each letter in the printed word stands for a sound segment of the spoken word. The variables studied were the contrast between printed and spoken words and between audio and visual modes of presentation. These variables are significant in second language learning. They are of special concern because of the extensive development of the oral "audio-lingual" method and of the language laboratory. In the experiment 83 were asked to learn a list of eight fereign words, printed in arbitrary symbols. The experiment was conducted in two phases. In Phase I 83 learned to read the words. In Phase II SS learned to repeat the words. Keith Cox In Phase I 33 were given three types of trials: reading training, reading practice, and a test of reading proficiency. The major independent variable was reading training, varied four ways: letter training, word training, phonetic training, and non-phonetic training. It was hypothesized that Ss with phonetic training would learn to read in fewest trials. This hypothesis was.not supported by evidence, which showed a non-significant advantage favoring word training (simple conditioning trials). In Phase II 83 learned to repeat the words as items in a serial list or as response members in a paired-associate list. It was hypothesized that the PA list would be learned in fewer trials and with fewer errors than the serial list. A slight non-significant advantage favored the PA learning. Presentation was either audio alone or combined audio-visual. It was hypothesized that Ss would learn in fewer trials and with fewer errors under the combined presentation. Instead, the audio alone presentation proved most effective. Part of the 83 receiving the combined audio-visual presentation had been given phonetic training. It was hypothesized that they would learn better than those without phonetic training. No significant differences were found. It was concluded that under the conditions of this experiment phonetic training does not seem to be significant in learning to read or in memorizing verbal materials. Further research with more extensive materials and extended periods of time is suggested. Keith Cox It was concluded that an audio alone presentation of verbal materials was more effective than a combined audio-visual presentation. Further research on the phonetic significance of the visual symbols is suggested. It was concluded that verbal materials are learned equally well serially or as paired associates. A critical area for further research is the nature of the stimulus in serial learning and the total amount of learning involved. ACKNOWLEDGMENTS "No man is an island," and certainly no man's work stands alone. "I am a part of all that I have met"; the concepts developed in this study and their implementation reflect the views and the insights I have acquired during the past several years from members of the Department of Communication and the College of Education. Specifically, I would express appreciation to Dr. David Berlo, Chairman of the Department, who introduced me to a new way of looking at the world; to Dr. Erwin Bettinghaus, who insisted that this was the right way of looking; and to Dr. William Stellwagen, who kept my nose to the calculator until I was willing to look at it that way. In the preparation of this dissertation, I am most grateful to Dr. Bettinghaus, chairman of my committee, who led my thoughts in this direction and finally focused my thinking in this area; who also added counsel at key points in my progress. I am deeply indebted to Dr. Stellwagen, my thesis director, who introduced me to the fundamentals of language behavior, who sharpened the outlines of this study, and who gave many hours of his time to the direction of this experiment and the refinement of the dissertation. Much appreciation is also due to Dr. Byron Van Roekel, whose wealth of experience in the field of reading added dimension to the lines of this study and guided me to conclusions I had failed to anticipate. Thanks are also due to Dr. David Krathwohl, who deepened my insight into the broader aSpects of learning behavior, and to Dr. Gerald Miller, who made me aware of ii the social connotations of language, and repeatedly checked the rigor of my procedures and my expression. I should also like to express appreciation to staff members of the department: to Mrs. Shirley Sherman, who was always there, who always knew who I was, who always had an answer, and who so competently carried out the task of typing the un-typable copy of this manuscript; to Al Talbott, who shared his statistical know-how and gave hours of his time in the analysis of data. My deepest appreciation goes to my family: to my wife, Miriam, who gave unfailing and dynamic support during the small wee hours and the long long months; to Karyl and Gary, who always accepted what they didn't always understand--who kept the long view. Finally I must recognize my faith in a trans-empirical stability, which has deepened during these years of rigorous enlightenment and provided the ultimate support. TABLE OF CONTENTS CHAPTER I RATIONALE AND HYPOTHESES . . . . Studies of Verbal Learning . . Hypotheses . . . . . . . . . Specific Rationale . . . . . II PROCEDURE 0 O O O O O O O O O 0 Reading Training e e e e e 0 Learning Procedures . . . . . Stimulus Materials . . . . . III RESULTS 0 OK 0 O O O O O O O C IV DISCUSSION OF RESULTS . . . . . Reading as Conditioning . . . Learning a Foreign Language as Stimulus Compounding . . . . Conditioning Paired-Associate and Serial Learning . . . . BIBLIOGRAPHY O O O O O O O O O O O O O O APPENDICES O O O O O O O O O O O O O O O 0 iv PAGE 10 10 l” 15 20 23 25 36 36 38 #0 H3 #6 51 TABLE 10 ll 12 LIST OF TABLES Means and standard deviations of the number of trials to criterion in Phase I . . . . . . . . . Analysis of Variance of the number of trials tocriterioninPhaseI............ Means and standard deviations of two learning measureSinPhaseIIeeeeeeeeeeeeee Analysis of Variance of the learning measures in Phase II 0 O O O O O O O O O O O O O O O O 0 Effect of word training and phonetic training on number of correct responses in 5 trials . . . Means and standard deviations of the number of trials to criterion in Phase II . . . . . . . Analysis of Variance of number of trials to criterion using three modes of presentation in two types of rote learning . . . . . . . . . Means and standard deviations of the number of correct responses in five trials in Phase II . . Analysis of Variance of number of-correct responses in five trials in Phase II ee 0 e e e e e e e 0 Effect of previous language training on the number of trials required to reach criterion . . . . Effect of previous language training on the number of trials required to reach criterion by male andfemaleSUbjeCtSeeeeeeeeeeeeee Effect of sex differences on the number of trials required to learn to read words written in anartifiCialalphabete ee ee 0 e e e e e ee ee PAGE 26 26 27 27 28 28 29 30 30 33 33 34 FIGURE 10 11 LIST OF FIGURES Relationships between printed and spoken forms of a word and its referent . Distribution of groups in cell matrix . . Letter training trials . . . . . . Word training and reading practice trials Phonetic training trials . . . . . Non-phonetic training trials . . . Reading proficiency test trials . . Serial learning trials . . . . . . Paired-associate learning trials . Modified Kirundi words used in experiment Comparison of learning curves for serial and paired-associate learning . . . . . . vi PAGE lfl 16 17 18 19 21 22 23 24 32 APPENDIX LIST OF APPENDICES MATERIALS OF LEARNING TASK . . . . . . LETTER TRAINING TRIALS . . . . . . . . DIRECTIONS FOR LETTER TRAINING . . . . WORD TRAINING TRIALS . . . . . . . . DIRECTIONS FOR WORD TRAINING . . . . . PHONETIC TRAINING TRIALS . . . . . . . DIRECTIONS FOR PHONETIC TRAINING . . . NON-PHONETIC TRAINING TRIALS . . . . . DIRECTIONS FOR NON-PHONETIC TRAINING . READING TEST TRIALS . . . . . . . . . DIRECTIONS FOR READING TEST TRIALS . . REVIEW OF WORD LIST . . . . . . . . . SERIAL LEARNING TRIALS . . . . . . . . DIRECTIONS FOR SERIAL LEARNING TRIALS PAIRED-ASSOCIATE LEARNING TRIALS . . . DIRECTIONS FOR PAIRED-ASSOCIATE LEARNING TRIALS O O O O O O O O O O O SERIAL AND PAIRED-ASSOCIATE LEARNING . vii PAGE 51 52 53 51+ 55 S6 57 58 59 60 61 62 63 en 65 66 67 CHAPTER I RATIONALE AND HYPOTHESES The purpose of this investigation was to determine to what extent printed words interact with spoken words in a foreign language learning situation. The principal hypothesis investigated was that 33 would learn a list of foreign words.in.fewer.trials if both the spoken and the printed forms of the words were presented simultaneously. Both serial and paired-associate learning techniques were employed in the study. A description of these two learning techniques will be found in Appendix R. Two aspects of the hypothesized interaction need to be noted. One is the relationship between each of the two forms of the word and the meaning or referent. The other is the correspondence between the sounds of the spoken word and the letters of the printed word. Fig. 1 shows this set of relationships. Referent : Object (hat) I I / / / I ( Printed word: hat Spéken wo :[hat] Fig. 1. Relationships between printed and spoken forms of a word and its referent Certain aspects of these three relationships are of special significance., (1) According to linguistic theory, the spoken word is understood to stand for the referent. (2) In any system of writing, alphabetic or non-alphabetic, the printed word stands for the spoken word. (3) Many theorists,'both linguistic and psychological, would tend to believe that the meaning of the printed word may be mediated through the spoken word. Evidence supporting this last suggestion may be found in the study of lip movements by native speakers during silent reading. McDade (1937) failed to establish a direct relationship between the printed word and the referent without such mediation. He had taught elementary students in Chicago to read by a completely non-oral method. However, Buswell (19u5) reported that McDade's students used lip movements in reading by the time they reached the sixth grade, almost as much as students taught by conventional oral methods. More recently, Bdfeldt (1960) reported that even good adult readers tended to initiate lip movement in reading under conditions of stress. McGuigan, Keller, and Stanton (1965) have also reported evidence in support of this. Vocalization or subvocalization appears to be present in much reading, even at the adult level. Evidence for the relationship of the printed word to the spoken word is theoretical rather than experimental. The spoken word is composed of a series of sounds, just as the printed word is composed of a series of sounds, just as the printed word is composed of a series of letters. Yet the spoken word as a whole stands for the referent as a whole. One part of the spoken word hat does not stand for one part of the object (hat). On the other hand, each letter, or cluster of letters, of the printed word 223 does stand for one part of the spoken word hat; These one-to-one sound/symbol correspondences are sometimes spoken of as the alphabetic principle. This alphabetic principle is basic to linguistic science. Bloomfield (1933) emphasized the fact that language is always oral, and that printed symbols are only a convention for recording and pre- serving speech. Bloomfield (19u2) developed a system of reading in- struction for his own children, based on this principle of sound/symbol correspondences. After his death, his materials were collected and expanded and finally published in book form by Barnhart (1961). The examples of the implementation of the-alphabetic principle cited above are drawn from native speakers of a language. This investigation, on the other hand, is concerned chiefly with second language learning. The variables to be considered are aspects of the above model that apply to learning a foreign language. In addition to the matter of sound/symbol correspondences, the contrast between audio and visual modes of stimulus presentation also seem to be significant. Today much importance is attached in this country to communication in a second language. This can be seen in the current increase in students, teachers, and expenditures. Walsh (1965) reported that the percentage of high school students enrolled in foreign language courses climbed from 14% to 2u% in a period of fifteen years. Hocking (1969) indicated that Illinois had 600 FEES (Foreign Language in Elementary School) programs under way, with 22,000 elementary school children in Chicago enrolled in a foreign language program. Allen (1969) cited federal expenditures of over seven million dollars each year for the additional training of foreign language teachers. Locke (1965) reported that from six to eight thousand language laboratories were in operation by the end of 196% in the secondary schools and colleges of the United States. The language laboratory and the audio-lingual approach are the key to much of this increased interest (Faldman, 196“). Angiolillo (19“?) noted the importance of aural-oral work in the war-time Army Specialized Training Program. Meras (1954) stressed the importance of the oral aspects of language in the direct method. Huebner (1960) listed speaking as a "prime essential in acquiring a foreign language." Mathieu (1961) and Childers (196“) both stress the need for an oral approach first. This is in sharp contrast to the grammar-translation method which was used almost exclusively before World War II (Carroll, 1953). In spite of the emphasis on the audio-lingual method and the use of the language laboratory, criticisms have been levelled at this approach to language learning, criticisms which serve to highlight the variables studied in this investigation. Keating (1963) made a critical evaluation of the language programs in a number of schools in the metropolitan area of New York City. His findings were not particularly favorable to the use of the language laboratory. He reported that students who had been trained without the language laboratory performed as well or better than those who had used it. Subsequent research by Lorge (1964) included a number of variables which had not been part of the Keating study. Her findings were favorable to the use of the language laboratory, but they did not fully answer the questions raised by Keating. Other areas of concern have been noted in regard to the use of the language laboratory. Hocking (1962) reported that a considerable number of students were developing foreign language speech pathologies when they were trained by the language laboratory and the purely audio-lingual method. Carroll (1965) goes on to say, "The audio-lingual habit theory . . . is no longer abreast of recent~ developments." Carroll does not cite research evidence in support of this statement, but earlier (Carroll, 1960) he had noted the need for sounder research in foreign language teaching. The "audio-lingual habit theory" relies on the spoken word. It attempts to produce in the learner a set of purely oral habits which are not at all related to the visual form of the printed word. Carroll contrasts it with the "cognitive code-learning theory," which includes the printed word and is basically the old grammar-translation method. The variables being tested in this study are those that deal with this relationship between the spoken and the printed word. Of the research reviewed for this study, very little deals with specific variables in the language learning process. Most of the research is of a more general type and has to do with the results of methods of instruction and comparisons of different approaches to foreign language learning. Scherer and Wertheimer (196u), for example, compared the results of the audio-lingual method with a traditional _ grammar-translation course in beginning German. In response to Keating's (1963) findings, Lorge (1964) carried out a comparison of various methods and materials in both audio-lingual and grammar- translation courses. Cooper (1960) reported on some of the factors that might be used to predict success in foreign language study. Investigations such as these constitute the bulk of foreign language research. Considering these facts, Carroll (1960) called for a stronger research basis to establish a sound policy for foreign language instruction. One type of such research has been studies of verbal learning procedures. Several studies in this category should be noted. Studies of Verbal Learning Experiments with verbal learning have identified some of the variables pertinent to foreign language learning. Among these have been studies contrasting audio and visual presentations of verbal learning materials. A number of studies have had 83 learn material in one modality and relearn it in the other. Lifton and Goss (1962) found that stimulus materials presented orally were both learned and relearned in fewer trials than stimulus materials presented visually. Pimsleur (1961) also found faster learning with an oral presentation of simulated foreign language words as stimulus items in a paired- associate experiment. Postman and Rosenzweig (1956) reported positive transfer from visual to audio presentations. Weissman and Crockett (1957) reported positive transfer from audio to visual. Hocking (1962) indicated positive transfer from one modality to the other, but faster relearning in the same modality. All of these studies indicate positive transfer effects from one modality to the other. Investigations of audio and visual presentations of reading materials in English have also added depth to this area of research. (Sebesta, 1964; Budoff and Quinlan, 1964; Katz and Deutsch, 1963). Other experimenters have made a presentation using both modalities together. VanMondfrans (1963) compared the results of presenting verbal learning materials in each of the modalities alone, with a compound audiovisual presentation. He found no advantage for the latter. At higher speeds of presentation, the two modalities even seemed to interfere with each other. He explained this in terms of Broadbent's model (1958) which describes the perceptual system as a single channel receptor, able to process material from only one sense modality at a time. Travers (1964) used both modalities simultaneously to present literary materials for reading. The audiovisual presentation showed no advantage over the single modes at slower speeds. At higher speeds, the Ss often blocked out one channel by covering the eyes or ears. This would seem to indicate that the Ss themselves preferred the single modality presentation. At the same time, results were superior to the single channel reception. Travers concluded that each 8 was choosing the channel that was most efficient for him. The audio-visual presentation was a type of compound stimulus. Travers' conclusion was that each S selected only that part of the total stimulus which was most effective for him. Other experimenters have identified similar examples of one selection when the compound stimulus was presented in a single modality, e.g., a trigram and a word (Spear, Ekstrand, and Underwood, 1964). Additional experiments have used color as one of the elements in a compound stimulus. One of the earliest and best known of these was carried out by Weiss and Margolius (1954). They presented each pair of items on a colored card. Later they carried out a set of retention tests, in which some of the groups were given only the verbal stimuli and other groups the color stimuli. Retention was best for the group which was given the compound stimuli, second best for the group which received the color stimuli alone. This might indicate the effectiveness of the compound stimulus, but it might also indicate that each subject was selecting that part of the total stimulus which was most effective for him. Other eXperiments in this general area have been carried out by Underwood, Ham, and Ekstrand (1962), Hill and Wickens (1962), Jenkins and Bailey (1964), and Spear, Ekstrand, and Underwood (1964). Saltz (1963) used a design which included color in both learning and test trials under some conditions and in only one under other conditions. He found learning most efficient when color was present in both, least efficient when it was present in the learning trials but not in the test trials. He also found that if color was present in the test trials and absent in the learning trials, results were better than when no color was used in either set of trials. Stimulus compounding has been investigated largely in the paired- associate learning situation. However, Young and Clark (1964) investigated the possibility that in serial learning a compound stimulus occurs. They suggested that the compound was made up of several of the items preceding each response. Their results did not ‘ give firm support to this hypothesis. Rather they indicated that Ss were using position as a cue. Schulz (1955) and Lippman (1963) also identified position as the functional stimulus in serial learning. Ebenholtz (1964) found that sequence as well as position may serve as a cue. Several conclusions stand out in regard to the nature of the stimulus in verbal learning. (1) It may be presented in either the audio or the visual sense modality. (2) It may use both modalities together as a compound stimulus. (3) A compound stimulus may also be presented in a single modality, e.g., a color and a word are both visual. (4) It may be that 83 tend to select part of the compound stimulus as the functional stimulus. (5) Finally, it may be that the stimulus in serial learning is position or sequence or a combination of both. This verbal learning experiment was designed to investigate some of the variables suggested above. Subjects were asked to memorize eight words in a foreign language, using either paired-associate or serial learning techniques. The words were printed in a set of arbitrary phonetic symbols. Both audio and audiovisual presentation of the words was used. The purpose was to investigate three aspects of the verbal learning situation: (1) the value of phonetic training; (2) the value of a compound audiovisual presentation; and (3) the contrast between serial and paired-associate learning. 10 The experiment investigated the following hypotheses: 1. 83 will learn to read the list of foreign words in fewer trials and with fewer errors if they are given phonetic training, that is, if they are taught what spoken sound is represented by each printed letter. 2. Ss will learn to repeat the list of foreign words in fewer trials and with fewer errors if they are given phonetic training. 3. 85 will learn to repeat a list of foreign words in fewer trials and with fewer errors if both the spoken and the printed forms of the words are presented simultaneously. 4. Se learning under paired-associate learning techniques will learn in fewer trials and with fewer errors than those using serial learning techniques possibly because of stimulus specificity in PA as opposed to serial techniques. Specific Rationale Hypothesis I. 83 will learn to read the list of foreign words with fewer trials and with fewer errors if they are given phonetic training. The theoretical background for this lies in the sound/symbol correspondences which have already been noted. Research evidence may be found in studies of methods of reading instruction. Agnew (1939) reported that third grade students who had received thorough training in phonics had better mastery of reading skills than those who had not. Gurren and Hughes (1965) have summarized research studies on the value of phonetic training, beginning with Agnew's study and continuing to the present time. Of the 22 studies they consider statistically ll rigorous, 19 favor intensive phonetic instruction in beginning reading. Bishop (1964) reported an experiment in which 83 were asked to learn to read a set of Arab words. Part of the Ss were given letter-training in which they learned to associate letter sounds with printed symbols. Other Ss were given word training in which they learned to associate whole spoken words with whole printed words. Results indicated that letter training was superior to word training. Hypothesis II. 88 will learn to repeat the list of foreign words with fewer trials and with fewer errors if they are given phonetic training. This is derived from Hypothesis I on the assumption that if 85 can read the words better, they can learn to repeat them better. Research evidence is lacking in this area, but there is strong theoretical evidence. Marty (1962) and Pimsleur (1964) both stress the fact that if a student is not given the spelling for a spoken word, he will create his own spelling. Pimsleur (1964) made use of this 1>rinciple in his investigation. He used low-discrepancy words, which are spelled as they would be pronounced according to English pro- nunciation standards. For example, the spelling gh_for the sound [o] would be a low discrepancy word, while the French spelling £32 for the same sound would be a high-discrepancy word. Likewise 2225.1“ English would be low discrepancy, while the German spelling hut for the same sound would represent high discrepancy. Pimsleur found that learning of paired-associates took place with fewer errors with the low-discrepancy words. One might interpret this to mean that the words are learned more readily if the 83 under- stand the sound/symbol correspondences that are being used. 12 Hypothesis III. 83 will learn to repeat a list of foreign words in fewer trials and with fewer errors if both the spoken and the printed forms of the words are presented simultaneously. Evidence for this may be derived from several sources: (1) theory, (2) sound/symbol correspondences, (3) transfer effects from one sense modality to the other, and (4) stimulus compounding. Linguistic evidence has already been cited (Bloomfield, 1933, 1942) to show that the printed word symbolized the spoken word. Research studies indicate the value of phonetic training (Gurren and Hughes, 1965), which allows the reader to utilize both the spoken and the printed forms of the word. Transfer studies (Postman and Rosenzweig, 1956; Weissman and Crockett, 1957) show that one modality supports the other in relearning. Studies of compound audiovisual presentations by Travers (1964) showed better results than with a single modality. Other types of stimulus compounding, particularly with color, have also indicated superior results. Hypothesis IV. 83 learning under paired-associate learning techniques will learn in fewer trials and with fewer errors than those using serial learning techniques. Research shows that there is no evidence of transfer from one type of learning to the other. Young (1963) found no transfer from paired-associate to serial learning. Erickson and Young (1963) found that the direction of transfer was determined by the degree of association between the items. Jensen found that transfer from paired-associate to serial would occur if he established a set in favor of such transfer. He suggests that the two types of learning use different techniques. 13 The nature of the stimulus in serial learning is hard to identify. Evidence has already been cited (Ebenholtz, 1964; Schulz, 1955) which identifies both sequence and position as possible stimulus items. On the other hand, the stimulus in paired-associate learning is clear-cut and apparently more stable. This would seem to make it more likely that subjects will learn in fewer trials under that procedure, given low meaningfulness in the serial list. The purpose of this experiment was to investigate some of the variables that operate in a foreign language learning situation. It was hypothesized that Ss would learn foreign language materials more readily if they knew the phonetic meaning of the symbols, that is, if they knew what sound each symbol stood for. It was also hypothesized that the materials would be learned more readily if they were presented visually and audially together. Still another hypothesis was that the use of paired-associate learning techniques would be more effective than serial learning patterns. The following procedures were designed to test the above hypotheses. CHAPTER II PROCEDURE This experiment was designed as a verbal learning experiment to compare serial (8) and paired-associate (P) learning under three different modes of presentation: audio (A), audio-visual (AV), and audio- phonetic (AP). This produced six experimental groups. Fig. 2 shows the distribution of these groups in the cell matrix. Type of Learning Mode of Presentation Audio Audio- Audio- Visual Phonetic Serial Group Group Group SA SAV SAP Paired-Associate Group Group Group PA PAV PAP Pig. 2. Distribution of groups in cell matrix The experiment was conducted in two phases. In Phase I the subjects were taught to read the foreign words printed in a special alphabet. The dependent variable was the number of trials required to learn to read the words. Four training procedures were used and analysis was by means of a simple randomized analysis of variance. In Phase II, which came two weeks later, 83 were asked to memorize the words in one of two ways: (a) as items in a serial learning list, or (b) as response members in a paired-associate list, with 14 15 "meaning" as the stimulus member. Each of the variables, serial techniques and paired-associate techniques, was varied in three ways to produce a 2 X 3 factorial design. The dependent variable was the number of trials required to memorize the list to a criterion of two perfect recitations of the list. Subjects were members of a beginning psychology class at Michigan State University who participated in the experiment to meet a course requirement. Ss were assigned randomly to the different treatments of the research design. Presentation of the verbal learning materials was made by means of an MTA 100 Teaching Machine and a Sony Photo-Sync tape recorder which could be synchronized with the teaching machine. The visual presentation was made on a continuous paper sheet on the teaching machine. The machine was thus able to repeat the list of words an indefinite number of times. The audio presentation was made by the tape recorder, which also activated the teaching machine and advanced it to the next presentation. Placing of words and synchronization on the tape recording was timed by means of a spring-powered metronome, set to beat at .75 second intervals. Timing of stimulus item and/or reinforcement was 1.5 seconds for each presentation. Response period for both types of learning was 3 seconds in length. Phase I: Reading Training Familiarization Training Trials. Before learning to read the foreign words, the subjects were given a series of five training trials to familiarize them with the sounds of the words and the symbols used 16 to represent those sounds. Four types of training trials were used: (1) Letter training, (2) Word training, (3) Phonetic training, and (4) Non-phonetic training. A description of each follows. (1) Letter training. Each letter was presented visually, followed by its pronunciation. Then it was presented again, and the subject repeated it aloud. Stimulus presentation was 1.5 seconds; response period was 3 seconds. This was done for each of the eight letters, and the list was repeated for a total of five times. The pattern for the letter training trials is presented in Pig. 3. The five letter training trials were followed by five reading practice trials as out- lined in Pig. 4. The full schedule for both types of trials is found in Appendix B. Verbal directions are given in Appendix C. Visual Audio Oral B "ah" . . . . E3 "ah" I::’ "ee" .... Same procedure to be followed for each of the eight syllables in the list. Fig. 3. Letter training trials using letters in an artificial alphabet as stimulus items l7 (2) Word Training. Each word was presented visually, followed by its pronunciation. Then it was presented again and the subject repeated it aloud. Period of stimulus presentation was 1.5 seconds; response period was 3 seconds. This was done for each of the eight words in the list, and the list was repeated for a total of five times. These same 323$ practice trials were then repeated five times as reading practice trials. The pattern for these trials is given in Pig. 4. The full schedule for these trials is found in Appendix D. Verbal directions are given in Appendix E. Stimulus Response Visual Audio Oral 4 .v. dimi .... " .‘§7‘ bomi .... .... mada ........ .... mada mada .... 'I:]‘ [:1 mada .... .... dimi ........ .... dimi dimi 4 V dimi Same procedure to be followed for each of the eight words on the list. Pig. 8. Serial learning trials using words in an artificial alphabet as items in the series Paired-Associate Learning. Half of the Ss, chosen at random, were asked to memorize the eight words as response members in a paired- associate list. Two different serial orders of the list were prepared, and these were alternated to avoid serial position effects. This was 23 the procedure: First the two series of words, with their meanings, were presented to the subjects. Then each meaning was presented singly and the S was asked to respond with the correct word. Then the meaning was shown again, and the tape pronounced the correct response. For the audio-visual presentations, the visual symbol accompanied the second appearance of the meaning. Response period was timed at 3 seconds; reinforcement at 1.5 seconds. The list was repeated until the S reached a criterion of two perfect recitations of the list. If the 8 did not reach criterion in 30 trials, he was given an arbitrary score of 30 trials. The pattern for the paired-associate learning trials is given in Pig. 9. The full schedule fer these trials is given in Appendix 0. Verbal instructions are presented in Appendix P. Audio Presentation Audio-Visual Presentation Visual Audio Oral Audio Oral corn .... dimi corn ........ .... dimi corn dimi .... corn ‘(1 ":;7' dimi .... tree .... migu ‘ tree. ........ .... migu tree migu . . . . tree U H migu . . . . Same procedure to be followed for each of the words on the list. Fig. 9. Paired-associate learning trials using words in an artificial alphabet as stimulus items Stimulus Materials The learning materials were eight words derived from Kirundi, a Bantu language of central Africa. The words were shortened and 21+ modified to give them phonetic regularity for the purpose of the ex- periment. The meanings were adjusted to control the semantic range of the stimulus members of the paired-associate list. Finally, each word was transcribed by means of a set of arbitrary symbols which had been assigned phonetic meaning. ‘Fig. 10 gives the list of words with their meanings and phonetic transcriptions. Appendix A gives a list of the words in sentence context, showing the phonetic and semantic modifi- cations that were made. Meaning Word Symbol corn dimi ‘<:j ~1d [:1 t ’ d milk cow amata mada -I::I' books book ibitabo bida D d t 7 d saw eye yabonye bomi b v 11 em ; y7 i likes heart akunda guda L.) d k 7g 3 nC 7 C leopard tree ingwe migu V L4 in 7mi ; we 7u NOTES: 1. The purpose of the phonetic modifications is to reduce the total number of sounds to a minimum and to fit the words into a regular syllabic pattern. 2. The purpose of the modifications in meaning is to confine the range of meanings to nouns with relatively common referents. 52 APPENDIX B Letter Training Trials Visual Audio Oral [::] "ah" .... I::] .... "ah" 1:}. "ee" .... D . . . . "ee" I::I "oh" .... (::} .... "oh" L] "oo" . . . . L__) .... "oo" t:=; "be" .... t::, .... "be" L__i "due" .... LJ . . .. "due" 1::‘ ‘ ‘"go" .... . .... _"go" -;§;:— "ma" .... .—1::]— .... "ma" 53 APPENDIX C DIRECTIONS FOR LETTER TRAINING This experiment is not any kind of intelligence or achievement test. We are simply trying to find out some facts about learning to speak a foreign language. You will be asked to learn to say eight words in this language. These words will not be hard to say, but they will not sound like English words. They will be spelled with letters you have never seen before. In this first part of the experiment you are to learn the sounds of these letters. This will be the procedure: First you will see a letter and hear the sound of it. Then you will see the letter again. This time yop are to sey_the sound of it. Then you will go on to the next letter. Let me repeat the procedure: First you will see a letter and hear the sound of it. Then you will see the letter again. This time yep are to say the sound of it. Then you will go on to the next letter. The list will be repeated a number of times. There are eight letters in the list. 54 APPENDIX D Word Training Trials Visual Audio Oral <1 ‘V 7...“ <1 V /dimi/ TT 1:; /..../ fl 1:] /muga/ :1 d g... :1 'd W., V L4 We, V L; 7...... b. 'V 1..., b v /bomi/ 1.1 [:1 7m, LI :1 a... U 1: mada, 1 l [::j .... /mada/ t>> 1::1 /bida/ .... E E1 0 e e e /bida/ 55 APPENDIX E DIRECTIONS FOR WORD TRAINING This experiment is not any kind of intelligence or achievement test. We are simply trying to find out some facts about learning to speak a foreign language. You will be asked to learn to say eight words in this language. These words will not be hard to say, but they will not sound like English words. They will be spelled with letters you have never seen before. In this part of the experiment you are to learn to pronounce these words. This will be the procedure: First you will see a word and hear the sound of it. Then you will see the word again. This time you_are to sey the sound of it. Then you will go on to the next word. Let me repeat the procedure. First you will see a word and hear the sound of it. Then you will see the wordagain. This time yog_are to sey the sound of it. Then you will go on to the next word. The list will be repeated a number of times. There are eight words in the list. Stimulus Visual unnDLLVVttnnvvnn 56 APPENDIX F Phonetic Training Trials Audio The square stands for the sound of "ah" The triangle stands for the sound of "ac" The circle stands for the sound of "oh" The open square stands for the sound of "oo" The line on the left stands for "b"; this is "bee" The line on the right stands for "d"; this is "due" The vertical line down stands for "g"; this is "go" The horizontal line stands for "m"; this is "ma" Response Oral "ah" Nee" "Ch" "00" "bee" "due" "g0" "ma" 57 APPENDIX G DIRECTIONS FOR PHONETIC TRAINING This experiment is not any kind of intelligence or achievement test. We are simply trying to find out some facts about learning to Speak a foreign language. You will be asked to learn to say eight words in this language. These words will not be hard to say, but they will not sound like English words. They will be spelled with letters you have never seen before. In this first part of the experiment you are to learn the sounds of the letters. This will be the procedure. Each letter will appear on the machine. At the same time you will hear an explanation of its sound. Then the letter will appear again. This time you are to _say the sound of it. Let me repeat. The first time the letter appears you will hear the sound of it. The second time it appears, you are to say it. Then the next letter will appear and the same procedure will be followed. This will be done for each letter in the list. The list will be repeated a number of times. There are eight letters in the list. (After the list was presented three times in this way, it was presented without explanations in a series of two test trials. The following directions were used.) Now we will see how well you have learned these sounds. When the letter appears the first time, tell what it is. When it appears the second time, the tape will repeat it and you can see if you were right. Do this for each letter as it appears. Here is the first one. Stimulus Visual ddonnnvvntnovaa 58 APPENDIX H Non-Phonetic Training Trials Audio The square is one of the letters Another of the letters is a triangle The circle is another of the letters "Ch" "00" Another of the letters is an open square A vertical stem is part of this letter A vertical stem is part of this letter A vertical line down is part of this letter A horizontal line represents part of this letter "ah" "1113" "EC" Response Oral flee" "oh" "00" "ah" "g0" "bee" "ma" "g0" 59 APPENDIX I DIRECTIONS FOR NON-PHONETIC TRAINING This experiment is not any kind of intelligence or achievement test. We are simply trying to find out some facts about learning to Speak a foreign language. You will be asked to learn to say eight words in this language. These words will not be hard to say, but they will not sound like English words. They will be spelled with letters you have never seen before. In this first part of the experiment you are to beeome familiar with the letters and also with the sounds of the words. This will be the procedure. Each letter will appear on the machine, and you will hear a description of it. Then you will hear one of the sounds of the words. You are to repeat this sound. It will not necessarily be the sound represented by the letter. Each time you see a letter, observe its appearance. Each time you hear one of the sounds, repeat it. Do this for each letter and sound. The list will be repeated 3 number of times. There are eight items in the list. Here is the first one. (After the list was presented three times in this way, it was presented two times without explanations. The following directions were used.) In these next trials, you will not hear a description of the letter. Simply observe the letter when you see it; repeat the sound when you hear ite 60 APPENDIX J Reading Test Trials Stimulus Response Visual Audio Oral clv 41V T711: TTCI Syd Eld VH VL—l 1 le bv L4: 1.41:1 1:11: 131:1 Ed SE1“ dimi muga gado O... migu bomi guda mada bida dimi muga gado migu bomi guda mada bida 61 APPENDIX K DIRECTIONS FOR READING TEST TRIALS Now you will be tested to see how well you have learned these words. This will be the procedure: First you will see one of these words. Then there will be a pause. Then you will see the word again and hear it pronounced. Try to say the word during the pause whether you are sure what it is or not. Each time you see a word, try to say it before it is repeated. Then listen to hear if you said it correctly. Do this for each word in the list. Let me repeat the procedure: First you will see one of these words. Then there will be a pause. Then you will see the word again and hear it pronounced. Try to say the word during the pause whether you are sure what it is or not. Each time you see a word, try to say it before it is repeated. Then listen to hear if you said it correctly. Do this for each word in the list. The list will be repeated a number of times. There are eight words in the list. Here is the first one. 62 APPENDIX L Verbal Learning Experiment Review of Word List Circle the symbol for each of the following sounds: “4:14:1ch be ALJDO “ALJDO. g° ”Amaze: ma Circle the correct symbol for each of the following words: l was dVUD T‘rrq; beedah <1 “<7 13d T'TDI deemedvud T‘TEJ', moogahdvud TTEJI meegooutj.’ bv vu goodahud bVVLl qr: bvvu gahd. 1.4:: bv v1.4 What was most helpful in learning these words? DELI; VII—WV DEED EKEIIZIEI '— v_ v* nnnnvvvv alumna What was hardest about learning these words? Would (or did) learning the sounds of the letters help? How? Other comments. Name [LU 63 APPENDIX M Serial Learning Trials Audio Presentation Audio-Visual Presentation Stimulus Response Stimulus ' Response Audio Oral Visual Audio Oral bomi .... b _V. bomi .... 0000 mada 00000000 000. mada I mada Q... U DD mada 000. .... dimi 0000...... .0.0 dimi dimi 0.00 J v dimi .0 .0 00.. gado 0.00000... .0.0 gado ‘ gado .... E Di gado coo. .... muga 00.00.0000 .00. muga muga 0... fl D, muga 000. .0.0 bida 0.00.00... 0... bida I> ' bida .... D bida 0.00 .00. migu 00.000000. 0... migu migu .... V Ll migu .... .00. gUda 00.0000... 00.. guda gUda .000 L4 d gUda 000' 0000 bomi 0000000000 0000 bomi etc. etc. 61+ APPENDIX N DIRECTIONS FOR SERIAL LEARNING TRIALS In this phase of the experiment you are to learn to repeat the words in a specific order. After each word in the list there will be a pause before the next one is heard. During this pause, say the word you think will come next, whether you are sure or not. Each time you hear a word, try to say the one you think will come next. Then listen to hear if you are right. Let me repeat the procedure. During the pause after each word, say the one you think will come next, whether you are sure or not. Then listen to hear if you are right. Do this for each word in the list. The list will be repeated a number of times. This is the order the words will be in: bomi mada dimi gado muga bida migu guda bomi mada dimi gado muga bida migu guda. Now we are ready to begin. Here is the first word. 65 APPENDIX 0 Paired-Associate Learning Trials Audio Presentation Visual corn corn tree tree book book pen pen eye eye heart heart COW COW boy boy etc. Audio dimi migu bida gado bomi guda mada muga Oral dimi migu bida gado bomi guda mada muga Visual COPU corn tree tree book book pen pen eye eye heart heart COW 4v ”Vi—4 bd ad b'IV.’ 1.4:: trd boy boy -T——T- [:q Audio-Visual Presentation Audio dimi migu bida gado bomi guda mada muga etc. Oral dimi migu bida gado bomi guda mada muga 66 APPENDIX P DIRECTIONS FOR PAIRED-ASSOCIATE LEARNING TRIALS In this part of the experiment you are to learn the meanings of the words. Each meaning will appear on the machine, followed by a pause. Then the meaning will appear again and you will hear the word. During the pause, try to say the word that goes with that meaning, whether you are sure what it is or not. Then listen to hear if you are right. Let me repeat the procedure. During the pause after each meaning appears, try to say the word that goes with that meaning, whether you are sure or not. Then listen to hear if you are right. Do this for each word in the list. The list will be repeated a number of times. The first time through the list, the correct words will be given. After that you are to give them. Here is the first one. 67 APPENDIX R Serial and Paired-Associate Learning In serial learning techniques, subjects are asked to memorize a list of items, e.g., words or nonsense syllables, in a specified order. Under the anticipation method of training, when one of the items is given, the subject is expected to respond with the item that is next in the list. The correct item is then given and the subject responds again with the next one. This experiment used a list of eight modified foreign language words. The words were presented by a tape recorder with a pause after each word. During the pause the subject was expected to respond with the next item. This was the pattern of presentation: Tape: bomi .... mada .... dimi .... gado .... muga .... Subject: .... mada .... dimi .... gado .... muga .... etc. In paired-associate learning techniques, two items are presented together, a stimulus item and a response item. Under the anticipation method of training, the stimulus item is given and the subject is asked to respond with the response item. Then the pair of items is given together. This is followed by the stimulus item of the next pair, and the same procedure is followed again. This was the pattern of presentation: Tape: corn .... corn-dimi; tree .... tree-migu; etc. subject: 00.. dimi 000000000 0000 migu 0.000.... .00. "TITTTTITTTTTTT