SOME EFFECTS OF AUDITORY TRAINING 0N SPEECH DISCRIMINATION PERFORMANCE OF HARD OF HEARING ADULTS Thesig for The Degree cfPh. D. MICHIGAN STATE UNIVERSITY H Daniel L. Bode I * 19.6.6 THESIS “MIMI“MIMI“MIMI $1335.; University This is to certify that the thesis entitled SOME EFFECTS OF AUDITORY TRAINING 0N SPEECH DISCRIMINATION PERFORMANCE OF HARD OF HEARING ADULTS presented by Daniel L. Bode has been accepted towards fulfillment of the requirements for All— degree in M. {Major profjr Date August 10: 1966 0-169 ABSTRACT SOME EFFECTS OF AUDITORY TRAINING SPEECH DISCRIMINATION PERFORMANCE HARD OF HEARING ADULTS by Daniel L. Bode The major purpose of this investigation was to study experimentally the effects of auditory training methods and materials on speech discrimination performance, as reflected by three speech discrimination tests. Hearing Handicap Scale self—ratings also were evaluated in relation to obtained performance measures. Thirty—two adults with mild, sensori—neural hearing loss were selected to serve as subjects. These persons were seen initially to record the following measures: pure tone thresholds, speech reception threshold, speech discrimination in quiet (PB—Max), speech discrimination in speech babble, and intellectual performance (estimated by responses to the Vocabulary sub—test of the Wechsler Adult Intelligence Scale). On the basis of the above data, together with case history information, the subjects were assigned systematically to one of four training groups. Groups were matched (means, variances, medians, and ranges) on the above measures. Daniel L. Bode Each training group of eight subjects responded under one of four auditory training conditions: Group lu— write- down material with signal—to—noise ratio (S/N)—varied, Group 2--multiple—choice material with S/N—varied, Group 3—— write-down material with S/N—constant, and Group A——multiple- choice material with S/N—constant. Practice materials, in a variable speech babble background, were delivered to the sound—field at calibrated intensities. Feedback regarding performance and rest breaks were provided at 25-minute intervals during the approximate three hours of training. Speech discrimination tests (W-22, Rhyme, and semi—Diagnostic) were administered in the sound— field pre— and post—training at a 50 dB (re audiometric zero) intensity level in a 45 dB speech babble background. Results indicated the following conclusions: (1) Short—term auditory training results in generally significant positive effects on speech discrimination performance. (2) The effects of concentrated three—hour training proce— dures are reflected by significant changes in total, W—22, and Rhyme test discrimination, but not by the positive changes in Semi—Diagnostic test performance. (3) Discrimination changes reflected by the W—22 test differ significantly from the changes revealed by either the Rhyme or the Semi—Diagnostic tests. Changes on the latter tests do not differ significantly. (4) S/N—Varied and S/N—Constant training methods do not differ in effects on speech discrimination performance. (5) Open mi.- 95:52:39.1: Isa-21.3: arm-It M- ;2'.I1I::.I'—--- .I-' 1JTLF'EI’IE E, Ems: 1m; - I'i"-_‘::I'._'.FL'J‘.=‘IT"‘i-z'ls'z' .mmW - . _ . .g_..._ . - _ , :: .-._....Il - -_,' .I '- .|.. . -. :.- .F-- -- I_..1.-.".- .Iha .IE‘..-l-.-I‘."-'..'."..'.€§II!3 I I In. I . .dl. i. -. .... _ I I. ' .. ' I I .. .. J' I l I l I I Daniel L. Bode Set (write—down) and Closed Set (multiple-choice) training materials do not differ in effects on speech discrimination performance. (6) Discrimination changes differ significantly among four combinations of auditory training methods and materials. If multiple—choice, closed set materials are employed in training, the S/N-Varied method appears to be the method of choice. If write-down, open set materials are utilized in training, either S/N—Constant or S/N—Varied methods apparently can be employed for similar results. (7) Subjects tend to make similar progress in auditory ‘ training irrespective of the degree of reported difficulty reflected by self—ratings on the Hearing Handicap Scale. Age, speech reception threshold, and intelligence appear to be potentially significant factors in auditory training research. Observed trends suggested that increased discrimination was most extensive for those subjects who showed higher intellectual performance, who were older, and who had the least speech reception loss. Those subjects trained on closed set (multiple—choice) materials improved more on the closed set test (Semi—Diagnostic) than on the open set test (W—22). Similarly, subjects trained on open set (write—down) materials improved more on the open set test than on the closed set test. Those subjects rating themselves highest in hearing handicap tended to have slightly lower discrimination scores and slightly greater speech re- ception loss than subjects rating themselves lowest. b I I ‘ZEf-I'H'Ii‘i'ld' l-I'..\_ II.) .__.-.' w- v... - . . . _.- " II_ ..,,' - _I._ _ I ...-- . . " '. :--I " 'I -- T' ' . I . . . -- . . .. ' ‘ ' Daniel L. Bode These findings appear to indicate that basic and applied research in auditory training can enhance audiology's rehabilitation function and provide a framework in which to advance knowledge regarding performance of those persons with impaired hearing. ' I-T .' -.,:' 1‘. If.” 'EQI-n‘ _ -:-...."." , . . I . ... _ . . _. .. .. -. -.u u}.- SOME EFFECTS OF AUDITORY TRAINING SPEECH DISCRIMINATION PERFORMANCE HARD OF HEARING ADULTS By Daniel LT Bode A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Speech 1966 TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . LIST OF FIGURES O O O I O O I O I O I Chapter I. II. III. IV. INTRODUCTION . . . . . . . . . . Statement of Purpose . . . . . Importance of the Study . . . . . Definition of Terms . . . . . . . Limitations of the Study . . . . . Organization of the Report . . . REVIEW OF PERTINENT LITERATURE . . . . Historical Developments . . Speech Reception and Discrimination . Auditory Training: Procedures and Goals Research in Auditory Training . . Principles of Learning and Training Hearing Loss and Hearing Handicap . . Hearing and Listening . Summary . . . . . . SUBJECTS, INSTRUMENTATION, AND PROCEDURES Subjects . . . . . . . . Instrumentation . . . . . . . . Procedures . . . . . . Summary . . . . . . . RESULTS AND DISCUSSION . . . . . . Obtained Performance Measures . Pre— vs Post-Training Performance . . Analysis of Variance . . . . . Descriptive Statistics . . . . Discussion . . . . SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary Conclusions . Recommendations for Further Research ii Page iv Vi 121 I31 I31 136 __. 91:3 ' Page BIBLIOGRAPHY . . . . . . . . . . . . . 140 APPENDIX A LETTER SENT TO PROSPECTIVE SUBJECTS . 149 APPENDIX B RESULTS OF MONAURAL PURE TONE AND SPEECH AUDIOMETRY . . . . 151 APPENDIX C CRITERION DATA EMPLOYED TO ASSIGN SUBJECTS TO FOUR TRAINING GROUPS . 156 APPENDIX D HEARING HANDICAP SCALE . . . . . 159 APPENDIX E SPEECH DISCRIMINATION TESTS;W«22, RHYME, SEMI—DIAGNOSTIC . . . . . 165 APPENDIX F WRITE—DOWN TRAINING MATERIAL . . . 172 APPENDIX G MULTIPLE—CHOICE TRAINING MATERIAL . . 178 APPENDIX H PRE—TRAINING, POST— TRAINING, AND DIFFERENCE—SCORES; HEARING HANDICAP SCALE SELF— RATINGS . . . . . 186 Chapter III. IV. Table 10. LIST OF TABLES Page Means, standard deviations, medians ranges, and descriptive data resulting from group matching procedures (N = 32, 8 . . . . n = . . . . . . . 78 Mean percentage effects of auditory training on speech discrimination per- formance . . . . . . . . . . 100 Mean percentage effects of auditory training as indicated by three discrim— ination tests . . . . . . . . 102 Standard deviations for percentage effects of auditory training on speech discrimination performance . . . . 104 Estimates of error variances within treatment populations . . . . . , 106 Summary of analysis of variance come paring differences between methods, between materials, and among tests , 107 Comparative mean percentage effects of auditory training methods (signal—to— noise ratios) on speech discrimination performance . . . . . . . . . 108 Comparative mean percentage effects of auditory training materials (multiple- choice vs write—down responses) on speech discrimination performance . . 109 Percentage differences between pairs of speech discrimination tests reflecting effects of auditory training . . . 110 Pearson product—moment correlations between pairs of speech discrimination tests reflecting effects of auditory training . . . . . . . . . . 111 Comparative mean percentage effects of four combinations of auditory training methods and materials on speech dis— crimination performance . . . . . 112 iv Table ll. 12. Page Summary of analysis of variance com— paring differences among training groups . . . . . . . . . . . 112 Percentage differences between pairs of treatment means . . . . . . . 113 Figure 1. LIST OF FIGURES Page Schematic representation of physical arrange— ments during preliminary testing and selection of subjects . . . . . . . . . . 72 Schematic representation of physical arrange- ments during auditory testing and training procedures . . . . . . . . . . 9A vi CHAPTER I INTRODUCTION In 1959 two-hundred representatives of professions' concerned with hearing disorders attended a working conference sponsored by the Department of Health, Education, and Welfare.1 This group made a number of recommendations in the area of aural rehabilitation, two of which are pertinent: (1) that the techniques of habilitations and rehabilitations need‘to be re— examined for improved effectiveness and enhanced bene— fits, and (2) that pilot studies should be undertaken to evaluate various habilitation, rehabilitation, and restorative procedures through a controlled comparison of methodologies. Similarly, Oyer in 1966 emphasizes the current state of knowledge: Scientifically based therapeutic approaches are greatly needed for use in programs of aural habilitation and rehabilitation. . . . One of the most neglected areas in the field of clinical audiology is evaluation of the results of habilita— tion and rehabilitation programs.2 lHealth Aspects of Hearing Conservation. Published as a Supplement to the Transactions of the American Academy of Ophthalmology and Otolaryngology, November— December, 1959. 2Herbert J. Oyer, Auditory Communication for the Hard of Hearin (Englewood Cliffs, New Jersey: Prentice— Ha"1'1',"fnc.'1966), p 134 1 Suggested guidelines for research in this area were stated by a subcommittee on hearing problems in adults, appointed by the American Speech and Hearing Association.1 This subcommittee in 1959 posed ques- tions in areas felt to be worthy of further research: (1) What are the critical dimensions, from the social point of View, of auditory function at supra-threshold levels? (2) What new tests of hearing function are needed and what is the meaning of existing tests? (3) What new tests of communication efficiency are needed and what is the meaning of existing tests? The subcommittee defined aural rehabilitation as improve- ment in the capacity of an adult with hearing impair- ment to cope with his environment; they added that it was not possible at that time to give an adequate evaluation of procedures for such rehabilitation. Statement of Purpose Persons with hearing loss often seek rehabilita— tion services from an audiology clinic. Remedial procedures for medically and surgically non—reversible hearing loss may include a hearing aid recommendation, lipreading instruction, counseling, and auditory training. 1Report of Subcommittee on Hearing Problems in Adults, Chapter 9 in Research Needs in Speech Pathology and Audiology, Journal of Speech and Hearing Disorders Monograph Supplement No. 5 (September, 1959). o I I - I - - I. . , ' 4 -‘ 1 4'1- _-_I_ .l, . . J. - , .__. _ . 1'! 1 :J' . These services have developed over the years but, as indicated above, have not received extensive, objective evaluation. The present study was conducted in an effort to obtain and evaluate data relative to a specific aural rehabilitation procedure, namely audi- tory training. More specifically, fundamental questions were posed regarding the effects of auditory training on the speech discrimination performance of hard of hearing adults: 1. Does auditory training result in significant change in speech discrimination performance? Do three speech discrimination tests differ significantly in reflecting the effects of auditory training on speech discrimination performance? Do two auditory training methods differ signifi— cantly in their effects on speech discrimination performance? Do two auditory training materials differ significantly in their effects on speech discrimination performance? Do four combinations of auditory training methods and materials differ significanlty in their effects on Speech discrimination performance? Are self-ratings of social efficiency related to the effects of auditory training? Importance of the Study Silverman in 1957 stated that quantitative informa« tion about the effect of auditory training is fragmen— tary and that investigations suffer from the lack of adequate and valid measuring devices and criteria.1 These statements summarize the position of many indi- viduals professionally involved in aural rehabilitation. Bergman, et al. ,for example, pointed out in 1965 that it has been both written and stated that auditory train— ing can improve speech discrimination, but the literature contains very little quantitative information about its effects.2 Oyer in 1966 further summarized the situation: A great amount of research has been carried out in the field of audition, but a real need exists for research that is aimed at measuring the effective— ness of auditory training methods as they are applied to the hard of hearing.3 o o n 0 a o u o o . a o u u o o a o a o 0 e o . . there is insufficient scientific evidence concerning the change in performance following the administration of auditory training procedures. Therefore, at the present stage of development, it is virtually impossible to attempt to predict the actual behavior of the hearing handicapped after any specific set of procedures used in auditory training has been administered.’1t There is substantial indication that studies of auditory training have been few in number and limited in extent. Most of the generalizations concerning potential benefits were derived from informal observa— lS. Richard Silverman, Handbook of Speech Pathology, ed. L. E. Travis (New York: Appleton-Century—Crofts, Inc., 1957), pp- “30—31. 2Moe Bergman, et al. Auditory Rehabilitation for Hearing- Impaired Blind Persons, ASHA Monographs, No.12 (March, 1.965), p 39 3Oyer, op. cit., p. 12. “Ibid., p. 46. _.1 tion with adults and from studies and observations with deaf children. A review of pertinent literature, re- ported in Chapter II, supported these statements regarding the present status of auditory training. In spite of the lack of extensive research, methodology in auditory training is a minimally con— troversial topic. The majority of sources agreed on the general procedures and objectives of an auditory training program: subjects are given practice in identification of speech stimulus materials under a variety of listening conditions for the express purpose of improving speech discrimination performance in everyday situations. Validity of speech discrimination tests is an area of some confusion and controversy, even though there have been extensive research and discussion on this subject. Attempts to measure or evaluate the efficiency of auditory aspects of communication usually have employed speech stimuli, undoubtedly because of the face validity of this material. As recently as 1965, however, Carhart stated that one can perform a diagnosis of social efficiency with hearing loss only in general and qualitative terms.l Similarly, Davis in lRaymond Carhart,"Problems in the Measurement of Speech Discrimination,"Archives of Otolaryngology, 82 (September, 1965), pp. 253-60. 1963 indicated that the particular tests of the ability to receive auditory communication still need to be validated in the field.1 Among the criticisms that have been directed to— ward conventional monosyllabic speech discrimination tests are: (1) problems of word frequency or difficulty effects (i.e., words appearing most often in a language are familiar and therefore relatively easy to identify); (2) questionable validity of balancing phonetically the frequency of sounds (phonemes) and words (morphemes) within and among alternate test forms; (3) lack of demonstrated correlation between discrimination of mono— syllables and discrimination of continuous discourse; (A) different responses of sophisticated versus naive listeners; and (5) differential effects of level of motivation. Inherent in each of these problems appears the question of validity; that is, criteria for evaluat— ing the diagnostic or predictive value of specific tests and materials. It seems from this review that agreement has not been reached about either the material or criteria for speech discrimination testing, and that the potential lHallowell Davis, Hearing and Deafness, ed. Hallowell Davis and S. Richard Silverman (rev. ed.; New York: Holt, Rinehart and Winston, Inc., 1963), p. 19A. thirst -_.: #:- benefits of auditory training are difficult to evaluate relative to external criteria in the environment of individual subjects. It does seem, however, that specific tests in existence have value relative to criteria in- ternal to the actual training conditions. In other words, progress in training might be evaluated validly and re- liably within the structure of the training program, but the criteria external to the training program still are eluding objective specification. As a consequence of a review of information re— garding auditory training, it is believed that the present study not only answers the basic questions posed earlier, but also contributes to the understanding of one aspect of impaired auditory performance. The potential contribution of the study is enhanced by the use of self—ratings of social efficiency that are evaluated in relation to speech discrimination performance. In addition, test items and training materials were administered within the average loudness range of conversational speech and with a speech babble noise background. Incorporation of these dimensions improves the generalizability of obtained results to a realistic and practical external criterion. One further contribution of this study is the use of subjects whose hearing loss generally is classified as being "mild" in both extent and effects. It is agreed that the measurable extent of loss for these subjects is mild 1'. , ' as”: If . :I ..;,'_-'__' . I-E'__"lr_.' ._I' -:', !'-". .: 1 -._ - .5 ."I'E A: £:.,..:'.:_._ fiftiflfifii'x _I' .7fsgaq’li:.3:f$ iii-:11 Ffiqe" III I. 2 sub: I315 Ijir ': fl... I . .. : - 1:; '.;' . ‘.- _' ." I ,.. . .I. .' . _: I“! .III-II-Ii.'1-fi'if-I.:'.I_"::1 = _3 . , ‘1: If 6‘9“.“ EPI: ’- i - I . - I I _ . I "- I I I. I r - ' III. rrelative to normal and to‘those losses showing greater lack of sensitivity. It is not agreed, however, that the effects are mild. Individuals with so—called mild hearing loss, who report substantial hearing difficulty, represent a fairly large portion of the hard of hearing popula— tion. These persons when seeking professional advice often are told, in essence, that they do not have a problem, that a hearing aid will not help, that oto— logical treatment is not possible, and that they should return to the clinic in about one year when the hearing loss may be worse and services then can be provided. Lipreading, auditory training, and counseling may be suggested, but persons probably do not follow this re— commendation automatically and the encouragement from the clinic to do so may be minimal. It was assumed in this study that these subjects do have a problem, that they usually do not receive professional services for this problem (other than diagnosis), that clinical and research attention should be directed toward this population, and that, in so doing, their problems and the problems of hearing loss in general may be better understood and more adequately handled professionally. Definition of Terms The following definitions are employed in this investigation: 1. Auditory training is a set of procedures and conditions designed to increase the auditory speech discrimination of hearing-impaired adults. 2. Speech reception threshold (SRT) is the sound level in decibels (dB) at which fifty per cent correct identification of spondaic speech mate- rials (CID W—l disc recording)1 is recorded. 3, Sound levels are of two types in this study: (a) the level of pure tones in dB relative to ISO— 1964 standards2 and (b) the level of speech and noise signals in dB relative to ASA-1953 standard reference level of 22 dB sound pressure (SPL).3 From the latter definition, it follows that a 50 dB speech or noise signal re ASA— 1953 standards corresponds to a 72 dB signal relative to the SPL reference of 0.0002 microbar. 4. Auditory speech discrimination is the performance indicated by the percentage of correct responses to speech discrimination tests delivered at speci— fied sound levels in quiet and in noise. 5. Speech discrimination tests are those published, formal sets of speech stimuli presented to sub— lCommercial recording obtained from Technisonic Studios, Inc., 1201 South Brentwood Blvd., Richmond Heights, Missouri. 2Standard Reference Zero for Calibration of Pure Tone Audiometers, ISO/R 389-19EU. 3American Standard Specification for Speech Audiome- ters, ASA Z2“. 13:1953. . . u. u. . . . . .. ... . . .l m . J . 10 Jects for auditory identification of component monosyllables. In this study, the selected tests are: (a) the CID W—22 test,l (b) the Rhyme test,2 (c) the Semi-Diagnostic test.3 6. Noise refers to speech babble and designates the tape—recorded acoustical signal that results from twenty persons reading aloud different speech material at the same time. This type of signal shows primary energy concentration between 150 and 1000 cycles per second (cps) and intentisty flucuations of approximately 3 dB total variation. 7. Hard of hearing adults are those eighteen to sixty year old individuals who, as reported by case history, sustained a sensori-neural hearing loss after having developed normal auditory discrimination of American English. Furthermore, to qualify for this study, sub- Jects demonstrated: (a) an unaided, sound—field speech reception threshold of 5 to 35 dB (mild loss of sensitivity); and (b) an unaided, auditory speech discrimination score greater than ten per cent and less than ninety per cent on the CID w—22 test delivered sound- field in speech babble at a 50/“5 dB signal- to-noise ratio (S/N). 8. Intelligence—estimate refers to the raw score that results from individual administration of the Vocabulary sub-test of the Wechsler Adult Intelligence Scale.“ lIra J. Hirsh et al., "Development of Materials for Speech Audiometry," Journal of Speech and Hearing Disorders, 17 (September, 1952), pp. 321—37. 2Grant Fairbanks, "Test of Phonemic Differentiation: The Rhyme Test," Journal of the Acoustical Society of America, 30 (July, 1958), pp. 596-600. 3Char1es Hutton, E. Thayer Curry, and Mary Beth Arm— strong, "Semi-Diagnostic Test Materials for Aural Rehabilita- tion," Journal ofg§peech and Hearing Disorders, 24 (November, 1959), pp- 319-29. “David Wechsler, Manual for the Wechsler Adult Intelli- ence Scale (New York: The Psychological Corporation, 1955), pp. 2- 3. 53—75. ll 9. Auditory training methods designates the lis— tening conditions Zsignal—to—noise ratios) operating during five successive training sessions. Two methods (or conditions) are evaluated in this study: (a) S/N—constant, where a 5 dB S/N is maintained over the five sessionx 57/52, 54/49, 51/46, 48/43, and 45/40 during the first through fifth sessions in this order; and (b) S/N-varied, where the signal is maintained at 50 dB and the noise is increased in 2 dB steps in each of five sessions: 50/40, 50/42, 50/44, 50/46, and 50/48 during the first through fifth sessions in this order. 10. Auditory training materials refers to the type of material utilized during five successive training sessions. Two types of material are evaluated in this study: 5a) Open Set: mono- syllabic write—down items, and (b) C osed Set: monosyllabic multiple-choice items,2, 11. Self—rating of social efficiency is the numerical score resulting from administration of the Hearing Handicap Scale. 12. Average loudness range of conversational speech is defined as the sound levels at or near 50 dB re audiometric zero. This is an operational definition; in everyday listening situations loudness levels can vary substantially above and below the defined "average" level. lGordon E. Peterson and Ilse Lehiste, "Revised CNC Lists for Auditory Tests," Journal of Speech and Hearing Disorders, 27 (February, 1962), pp. 62-70. 2Laila Larsen, Lists published in Hearing and Deaf- ness, op. cit., pp. 542—44. 3J. C. Kelly, Clinician's Handbook for Auditory Trainin (Dubuque, Iowa: William C. Brown, Inc., 1953), pp. 75-113. “Wallace S. High, Grant Fairbanks, and Aram Glorig, "Scale for Self—Assessment of Hearing Handicap," Journal of Speech and Hearing Disorders, 29 (August, 1964 , pp. 215-30. 12, Limitatiogsvef the Study Only one aspect of the aural rehabilitation pro— cess was examined in this study, namely auditory training. Any generalizations are confined to the population of hard of hearing adults who have the characteristics of, and who perform similarly to, the present subjects. Primary variables of interest included monosyllabic test and training items, signal-to-noise testing and training procedures, conversational loudness range, and average performance of groups. Any variables not designated and/or not controlled in the statistical design were assumed to be normally distributed among subjects within each-of four systematically-matched training groups. Any actual differences that may have existed among these groups were assumed to be non-significant statistically; and, where this assumption was questionable, appropriate caution was exercised in the interpretation of observed results. Organization of the Study Chapter I is organized to give an overall View of the subject matter, the apparent status of knowledge on this subject, the questions and population of interest, and the research approach influencing the conduct of the investigation. In Chapter II literature pertinent to this study is reviewed, interpreted, and summarized under the general headings of historical developments, speech fl: . ‘ . amass-9m In filfiéfi’ififi -'-:.- 4-I’-'f‘-- . -- II I . . f _- hi”; fill-Ilia ' .-.I 5-H :‘r::’f 6-13 13 reception and discrimination, auditory training pro- cedures and goals, research in auditory training, hearing loss and hearing handicap, and hearing and listening. The subject population, the selection of subjects from this population, their assignment to four training groups, the training methods and materials administered, the speech discrimination criterion mea— sures, and the statistical design are described in Chapter III. In Chapter IV results of the investiga— tion are presented and discussed in relation to questions posed earlier in the present chapter. The study is summarized, the conclusions are stated, and the implica- tions for additional research are set forth in Chapter V. m . .u m. .. . NI _. u ..n .3 . .1 n..- .na— . “Fr . .. H q: . .IL .._ H L... “a fin" Wt . n n. .m ..... - 1.". ..... . .m... 1-“ _ ._. a ....m . , .7 _. _ LL .. ._ HI” _ . L mm. | .3. .. m . - ._u . . - lul - L. .. , . . _ .\....n.... . .. . - u-| - 1.... ....lw If .|-,l...uHJ Hum... . .I. .. HUI...|..HIII... |.I.| . I CHAPTER II REVIEW OF PERTINENT LITERATURE Literature pertinent to the purposes, scope, and limitations of this studyugas described in Chapter I.“ is reviewed in this chapter. The literature review is classified under general headings of historical develop- ments, speech reception and discrimination, auditory training procedures and goals, research in auditory training, principles of learning and training, hearing loss and hearing handicap, and hearing and listening. A summary of this review is given at the end of the chapter. For the purpose of introducing the material in this chapter, definitions and objectives of auditory training——suggested by various authors——are given below: The treatment calculated to improve the hearing con— sists in the use of trumpets whereby the nerve appara— tus may be gradually excited as to become sensitive to ordinary sonorous undulations and external stimu— lants.l Stimulation or education of the hearing mechanism and its associated sense organs by sound vibration as applied either by voice or any sonorous instruments.2 lToynbee in 1860, quoted in Max Goldstein, The Acoustic Method for the Training of the Deaf and Hard—of—Hearing Child (St. Louis: Laryngoscope Press, 1939), p. 13. 2Goldstein in 1939, ibid., p. 18. 14 -r 1*.- V- - I'fifi.‘ M's-‘1- e .q ,r udzfisd ”_H 'fitsfi’éfitlmil ~5- 'fltin?’ .4 . ...... ell-"- r-r .: " :' a Lg.“ L. 1 -. - \~" I - 15 . .'. the development of auditory speech percep- tion. . Auditory training implies training with amplified sound. . . . In auditory training the person is taught to deal with distorted sound, to recognize a nevaattern of sound, and to attach meaning to Auditory training for the hearing handicapped is a process by which the hard—of—hearing individual learns to make maximum use of residual hearing.3 Auditory training is the process of teaching the child or adult who is hard of hearing to take full advantage of the sound clues which are still avail' able to him. . . . a series of communication exercises of pro— gressive difficulty leading to greater attention in listening, improved discrimination for the sounds of speech, and improved auditory memory span.5 Thus, it appears that the term, auditory training, in general, designates the activities and processes by which the hearing—handicapped are given the opportunity to make full use of acoustic events impinging upon the auditory mechanism. The development of this concept through the years provides some insight into the current status of auditory training. 1C. V. Hudgins, "Auditory Training: Its Possibilities and Limitations," The Volta Review, 56 (1954), p. 339. 2John J. O'Neill, The Hard of Hearing (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1964), pp. 89—90. 3Herbert J. Oyer, Auditory Communication for the Hard of Hearin (Englewood Cliffs, New Jersey: Prentice—Hall, Inc., 19665, p 45. “Raymond Carhart, Auditory Training, Hearin and Deaf— ness, ed. Hallowell Davis and S. Richard Silverman (rev. ed.; New York: Holt, Rinehart and Winston, Inc., 1963), p. 373. 5J. C. Kelly, Clinician's Handbook for AuditorygTraining (Dubuque, Iowa: William C. Brown, Inc., 1953), p. iv. 16 . Historical Developments Auditory training, from an historical viewpoint, can be viewed conveniently within two broad chronological categories. In the next two sections information is pre— sented regarding auditory training before and following World War II. During the former period auditory training was viewed primarily as a possible educational procedure for deaf children. During the latter period, however, and following the advent of modern amplifiers, auditory training received new stature, both as an important means for educa— ting deaf and hard of hearing children and as one among several procedures included in aural rehabilitation programs for hearing-impaired adults. Pre-World War II One of the most comprehensive descriptions of the history of auditory training before 1940 was provided by Goldstein.l In his review, Goldstein traced auditory train- ing developments from the first century A.D. to the 1920's and 1930's when his "acoustic method" was incorporated into the educational program at the Central Institute for the Deaf in St. Louis, Missouri. Prior to the publication of his book in 1939, most of the accounts of auditory train- ing were to be found in European books and journals. A lGoldstein, op. cit., pp. 11—17. 17 major share of the credit for introducing auditory training to the United States is given to Goldstein. The next few paragraphs trace the pre—l940 history of auditory training, as described by Goldstein. Archigenes, in the first century, recommended the use of sound amplified by a hearing trumpet as a means of stimulating the hearing of deaf persons. Simi— larly, Alexander of Tralles in the sixth century—— followed by Guido Guidi in the sixteenth century—— emphasized stimulation of hearing by presenting to the ear various noises and loud shouting. Ernaud in 1761 demonstrated before the Academy of Sciences in Paris how, by means of practice in differentia— ting vocal sounds, pupils with substantial residual hearing were able to develop some discrimination for words. Pereire in 1767 claimed that practically all deaf subjects showing some remnant of hearing could be trained to hear words. Credit for the first objective study of auditory training is given to Itard, a Paris otologist, who in 1802 suggested that improved hearing performance could be devel— oped in congenitally deaf children. He tested this hypo— thesis in 1805 by experimenting with a class of six deaf pupils. Improvement was observed as a result of practice in discriminating among bells, among musical tones, among the rhythms of a drum beat, among notes produced by a flute, and among vowels and consonants. ens-3:18!” ' 2‘ ail-51312?! 18 Itard's work, following his death in 1832, was continued first by Blanchet and then by teachers at the Institute for the Deaf at Nancy. During this same period Beck, Jager, Wolff, and Frank in Germany, and Toynbee and Wilde in England were giving auditory training to deaf pupils who had residual hearing. Toynbee cited three cases where auditory training was given to adults who showed improvement following train- ing. This perhaps is the only published account in the pre-World War II period where adults were considered as candidates for auditory training. Interest in auditory training reportedly decreased in Europe during the last half of the nineteenth century. During this same period, however, the United States was beginning to consider it seriously as an important form of training for deaf children. Gallaudet in 1884, Gillespie in 1892, Taylor in 1893, and Currier in 1895 reported on their observations. During this period a representative committee studied Itard's work and concluded that the idea of re-stimulating or awakening impressions in the auditory apparatus by means of "sonorous vibrations" had value as an educational procedure. Another important contributor to the development of auditory training was Urbantschitsch, a professor of otology at the University of Vienna. In 1892 he arranged daily practice sessions for a group of deaf pupils. The - a... has." .35- mrz'w warm _ -... Vial-:3;- fi-Be‘fit' -'.:'-"i" Emit-Tm - L . . ... ..f-«I . .- - _I‘ I ' Z 19 following year a public demonstration was given before the Vienna Medical Society showing his results with eighteen of these pupils. The subjects——diagnosed before training as totally deaf-—reportedly showed performance after training ranging from vowel to sen— tence discrimination. Goldstein at this time was doing post-graduate work in medicine in Vienna and observed both Urbant— schitsch's training program and his public demonstra— tion. Upon his return to St. Louis in 1895, Goldstein directed an experimental program at the St. Joseph School for the Deaf while at the same time establish- ing a medical practice. Sixteen girls received daily practice for two years with Goldstein observing and supervising the work two afternoons a week. He pre— sented some encouraging results in 1897 before a meeting of the American Academy of Ophthalmology and Otolaryngology. Medical practice then diverted his attention for several years. It was the founding in 1914 of the Central Institute for the Deaf when Gold— stein began elaborating his Acoustic Method, continuing its development from then until 1939. Post—World War 11 W Newby noted that before World War II, auditory training was given primarily to deaf children as a |'_ - _ a ‘ baamlbwwsifdm' twins.- ‘ semis .v‘aea‘mae-t-u'irwn 21.15303113‘6“ . . fa...-.;_;-_...=_.:i-.. ‘::.-:.?'.z."-'; ins-315 Q'SPWW' _-C. ‘27:?" mfiuvrfiufi 351153. l .' _.: '3 l'. '. 'r diff“?- - ~.- _ ' -'- _ «1' .1 .. .'. . "" .a |._=- ‘. ‘— LB'I' | I ll ' -_. -v ‘ ' a”. .I‘ ”1:51 “SH". I 7. I .'— I '1' i . I I '1'- n n .g i l I' 'l ' 1 I :‘zflsueifi- 233:,- T: _ II E I ~- I 20 means of establishing gross awareness of sound.1 He observed that this was paradoxical since the hard of hearing, with substantial residual hearing,seem to be ideal candidates for potential benefits from auditory training. Newby,2 O'Neill,3 and Oyer“ indicated that war— time aural rehabilitation programs for servicemen and the development of modern, wearable hearing aids pro— vided major impetus for the establishment of auditory training as a rehabilitative procedure for the hard of hearing. Oyer emphasized the developing importance of auditory training since World War II, stating that it has been an integral part of the aural rehabilitation services rendered veterans.5 As a result of these pro— grams, the general public and various professions became aware of the contribution of auditory training to the aural rehabilitation process. Oyer cited Downs' 1961 survey data that indicated that auditory training was provided on the following percentage basis by the 166 hearing centers responding to the survey;6 lHayes A. Newby, Audiology (2nd ed. rev.; New York: Appleton—Century-Crofts, Inc., 1964), p. 291. 21bid. 30'Neill, op. cit., p. 89. “Oyer, op. cit., p. 11. 51b1d. 61bid. 21 University 100.0% Medical Schools or Hospitals 83.3 Hearing Societies 100.0 Federal Agencies 82.2 State Agencies 60.0 Public Agencies 88.9 Private Agencies 100.0 These figures suggest the emphasis and importance placed on auditory training by professional workers in audiology. With reference to auditory training for children, Silverman stated that despite unsolved problems, there is no longer any question about the usefulness of the auditory system in the education of deaf children. 1 Hudgins similarly pointed out that auditory training for deaf children can lead to development of auditory speech perception, better speech production, and broader lang— uage development.2 He suggested that as a result of these improved communication skills, there is a concomi— tant acceleration in the general educational program. Costello summarized the recent status of auditory training for children.3 She indicated that the rewarding results in the past led to this training being used more widely. However, the development of an objective and differentiated rationale has not been established. She l S. Richard Silverman, Deaf Children, Chapter 16 in Hearing and Deafness, op. cit., p. 439. 2Hudgins, o . cit., p. 399. 3M. R. Costello, "Realistic Goals in Auditory Training," Processes 39th Meeting, American Instructors of the Deaf (Wash— ington'D."C.: 133-45. U. 8. Government Printing Office, 1960), pp. -.-- - '.:f:L"-'.‘.!~"--:i a'.."‘-I.‘J;'I-L‘!- 9'59"” . i...:f:-; fall-aid I I I I . . .. 1.... ‘- , :' '-="T. -' . 22 suggested that auditory training is ceasing to be a general exposure to sound, now emphasizing objective approaches. In total then, an historical review of auditory training through the years indicates that the concept has existed for about 2000 years, but that its implica- tions and possibilities have been exploited only recently. Auditory training has become a routine procedure offered to the hearing—handicapped. Speech Reception and Discrimination Research and clinical attempts to evaluate the effectiveness of auditory training must take into con- sideration certain known features of auditory speech perception, as well as limitations inherent in existing measurement procedures. Pertinent literature is exten— sive and includes numerous articles and books that have been written from the professional viewpoints of clinical and experimental psychology, communication and acoustical engineering, experimental phonetics, audiology, linguistics, and other related disciplines. The orientation in the next sections of this chapter is toward clinical audiology. However, some information from other sources is reviewed to delineate the features of speech reception and discrimina— tion most directly involved in the design, conduct, interpre— tation, and conclusions of the present study. Broad areas 23 discussed include measurement, normal and impaired speech perception, and primary versus secondary signals. Of necessity there is some overlap among these topics. Measurement Clinical audiometry generally includes pure tone testing, speech audiometry, and Special tests. The latter category encompasses differential diagnostic tests and is not described here. Pure tone testing is given some description, but the major emphasis is on speech audiometry. The results of pure tone testing usually are plotted on an audiogram where the ordinate represents intensity and the abcissa shows frequency. Tests of this type require that the subject indicate, directly or indirectly, his threshold for each of the different frequencies or tones. Air conduction testing involves presentation of the tones individually to each ear by means of an earphone or receiver (on the ear) that delivers the test tone directly to the external ear canal. Subsequent bone conduction testing involves pre— sentation of the test tones individually to each ear via a vibrator (oscillator) pressed against either the mastoid region or the midline of the forehead. Calibration is such that comparison of air and bone conduction thresholds shows either (a) both types of thresholds are similar, suggesting either normal hearing I . _d «I n 21: or sensori«neural hearing loss; or (b) the air conduction thresholds show less sensitivity than the bone conduc— tion thresholds, suggesting a conductive or middle—ear type hearing loss. Many combinations of air and bone responses can be recorded for individual subjects, and require interpretation by an otologist and an audiolo— gist. Specific procedures and possible interpretations of pure tone audiograms——and of the results of speech audiometry--are given by Newbyl and by Davis and Silverman.2 Speech audiometry usually is administered to record some or all of the following measures: (a) speech reception threshold (SRT), (b) speech discrimina— tion score (DS), (0) detection threshold (DT), (d) most comfortable loudness level (MCL), and (e) discomfort or tolerance level (TL). The first two measures-—SRT and DS——are pertinent to the present study. Speech reception threshold designates the inten- sity level (in dB) at which speech is understandable to the subject about fifty per cent of the time. Spondaic words——stress on each of two syllables--are most often used for this measure. The CID W—l and W—2 disc—recordings were standardized for this procedure.3 lNewby, op. cit. 2Davis and Silverman, op. cit. 3Ira J. Hirsh, et al., "Development of Materials for Speech Audiometry," Journal of Speech and Hearing Disorders, 17 (September, 1952), pp. 321—37. 25 The subject's task is to repeat vocally-~or write down-—the test word as each word accompanied by a carrier phrase is presented above his presumed SRT (suggested by pure tone responses). The average intensity level is decreased until the subject is only able to repeat correctly fifty per cent of the words. The associated intensity level is then operationally defined as that subject's SRT and is compared to normal and to other performance measures for confirmation of pure tone re- sults, for diagnostic purposes, and sometimes for estimating social efficiency. This testing——and speech discrimination testing—— is done for each ear individually and also by sound— field presentation of the test words through a loud— speaker, all at calibrated intensity levels and in a sound—treated test room. The measurement of speech discrimination is a pro— cedure of primary importance to the present study. Historically, speech discrimination testing can be traced from the early studies of telephone systems at Bell Telephone Laboratories.1 A second major development was the extensive research at Harvard Psycho—Acoustic lHarvey Fletcher, Speech and Hearing in Communica— tion (Princeton, New Jersey: D. Van Nostrand Co., Inc., 1953). ..-.l L.,-‘. '- I: -.uL- I.“ 26 Laboratory during World War II.1 A third major contribu- tion to discrimination testing in the clinic occurred in 1 the early 1950's with the standardization and distribu- tion of disc—recorded discrimination tests (the W-22 cal use.2 Speech discrimination testing in the clinic us— ually involves the presentation of 50—item monosyllabic word lists at specified intensity levels above the subject's SRT. A popular discrimination measure called PB—Max (sug— gesting maximum discrimination of phonetically—balanced word lists) requires presentation of the test items some 30 to 40 dB above the SRT. The subject's task is to re— peat the words while an audiologist monitors the responses and records the percentage of words correctly identified. Carhart indicated that the above procedure has many limitations, even for diagnostic audiometry which has an entirely different purpose from the evaluation of audi— tory performance in everyday situations.3 He further stated that existing tests are imperfectly standardized lists) that have enjoyed substantial research and clini- lJames P. Egan, "Articulation Testing Methods," Laryngoscope, 58 (September, 1948), pp. 955—991. 2Hirsh, op. cit. 3Raymond Carhart, "Problems in the Measurement of Speech Discrimination,” Archives of Otolaryngology,82 (September, 1965), pp. 253-60. 27 and lack validation, but that with appropriate re— vision they can become much more definitive clinical tools. This does not deny the importance of either discrimination tests or the research conducted with them. Instead, what is suggested is that these tests have not demonstrated the validity required for inter— pretation or prediction of an individual subject's performance outside a soundatreated test room. Giolas. and Epstein, for example, compared the discrimination of Harvard PB—50 and CID W—22 word lists with the discrimination of continuous discourse (sentence material),and did not find any accurate predictions of the latter from either of the word lists.1 Although validity of discrimination tests has been questioned and subjected to much discussion, reliability of obtained measures apparently is quite high. Ross and Huntington reported reliability coefficients higher than 0.90 for each of the W-22 lists.2 In addition, Elpern compared the four W—22 lists with respect to level and range of difficulty, using approxi- mately 1500 discrimination scores from six different 1Thomas G. Giolas and Aubrey Epstein, "Comparative Intelligibility of Word Lists and Continuous Discourse," Journal of Speech and Hearing Research, 6 (December, 1963), pp. 349—58. 2M. Ross and D. A. Huntington, "Concerning the Re— liability and Equivalence of the CID W—22 Auditory Tests," Journal of Auditory Research, 2 (1962), 220—28. w. ace—iv “Ami z ' . mega-“45' ”‘ 28 Veterans Administration audiology clinics.l He found that differences among lists are not sufficiently great to introduce serious errors when used interchangeably in clinical audiometry. The availability of test material is not a pro— blem. Moser published a systematic listing of mono- syllables in American English according to phonemes, with frequency of occurrence in various cembinations presehted‘ in tabular form.2 Lehiste and Peterson developed CNC lists that reportedly have a simpler phonetic and phonemic composition and a more exact phonemic balance than the PB lists.3 Fairbanks published alternate lists of discrimina— tion test items that require consonant discrimination for correct identification.“ Hutton, Curry, and Armstrong developed Semi—Diagnostic tests (with alternate forms) for aural rehabilitation-—using a multiple—choice format——that is described as sensitive to different kinds lBarry s. Elpern, "Differences in Difficulty among the CID W—22 Auditory Tests," Laryngoscope, 70 (1960), pp. 1560—65. 2Henry M. Moser, One-Syllable Words: Revised and Arranged by Ending Sounds (United States Air Force CCDD Technical Note, NO. 60-58, 1960). 3Ilse Lehiste and Gordon E. Peterson, "Linguistic Considerations in the Study of Speech Intelligibility," Journal of the Acoustical Society of America, 31 (March, 1959), pp. 280-86. “Grant Fairbanks, "Test of Phonemic Differentiation: The Rhyme Test," Journal of the Acoustical Society of America, 30 (July, 1958), pp. 596—600. .4 .4 .u.. .. I. u. 29 of hearing loss while yielding reliable estimates of dis- crimination ability.1 Harris indicated that most tests are too simple and need to be made more sensitive to discrimination defects not obvious by conventional test procedures.2 He rec— ommended further research in assessing an indiVidual's communication efficiency, and stated that some type of masking or distortion should be included in tests de— signed to simulate everyday listening conditions. Egan's observation some years ago (1948) still seems to summarize present measurement problems.3 He stated that all discrimination scores are relativeIand that little trust can be placed in absolute statements about them. Normal Speech Reception and Discrimination Normal speech reception threshold for spondees is approximately 22 dB re 0.0002 microbar with standard test materials, conditions, and procedures.Ll A normal speech lCharles Hutton, E. Thayer Curry, and Mary Beth Arm— strong, "Semi-Diagnostic Test Materials for Aural Rehabili— tation," Journal of Speech and Hearing Disorders, 24 (November, 1959), pp- 319-29. 2J. Donald Harris, Research Frontiers in Audiology, Chapter 11 in Modern Developments in Audiology, ed. James F. Jerger (New York: Academic Press, 1963), pp. 420—23. 3Egan, op. cit. ”Raymond Carhart, "Inconsistency Among Audiometric Zero Reference Levels,” Asha,8 (March, 1966), pp. 63-66. 30 discrimination score is 98 to 100 per cent for the W-22 recordings at 60 dB re 0.0002 microbar (i.e., about 40 dB re audiometric zero).l Most of what has been described in the preceding section of this chapter is applicable to the present topic of normality in speech perception. However, several additional findings and comments of various researchers seem appropriate for this section. In the normal development of speech and language in children, Eeckhout pointed out that this development . involves several different levels of auditory function- ing, such as auditory acuity, listening, memorization, recognition, and recall.2 Licklider described three operations involved in normal speech perception: (1) translation of the speech wave, (2) segmentation of the stream into elements that are recognized and identified, and (3) comprehension of the meaning or significance of the message.3 The complexity of this process was suggested by Ladefoged and Broadbent, who stated that the listener does not deal with each phoneme separately but rather re— sponds to groups of sounds.” lHirsh, op. cit. 2M. J. Eeckhout, "Auditory Imperception," Western Speech, 25 (1961), pp. 180—83. 3J. C. R. Licklider, "On the Process of Speech Perception," Journal of the Acoustical Society of America, 24 (November, 1952), pp. 590-94. “Peter Ladefoged and Donald E. Broadbent, "Percep— tion of Sequence in Auditory Events," Quarterly Journal of Experimental Psychology, 12 (1960), pp. 162-70. ' . _ . . : u '_ E-ii'z-EIEJ'ZCI '51an fi.‘ 91.1.:.JJ_.€;:”‘. _ Ii. i '- 19d?- H‘Q‘.‘ nastiness .fli'fi'i' sari HEW: 2:; m1 ass-:2: £141.} . 19'. 9;}: §.-_.=.':‘..'.-:;=.:. "13.32 shay? -_-.'i _ I' || . ' '...._ -.'4 {1319‘s,} t 31 Various authors have pointed out the relationship between frequency of occurrence of words'in American English and resulting discrimination scores. Rosenzwieg, for example, found that speech intelligibility increases as the frequency of usage in the language increases.1 According to Egan, it is natural for rare words to be less intelligible than frequent words, and for this reason, research in evaluating effects other than fre— quency of usage should employ trained listeners who are familiar with the test material.2 Decker, Rubenstein, and Pollack found that when a message set is unknown (open set), there are substantial word frequency effects, but when it is known (closed set), there are no consistent effects.3 Miller, Heise, and Lichten emphasized the con— tribution of context to the intelligibility of speech.” Words selected from a small vocabulary familiar to the listener are more intelligible than words from a large 1Mark R. Rosenweig, ”Intelligibility as a Function of Frequency of Usage," Journal of the Acoustical Society of America, 28 (1956), p. 759. 2James P. Egan, "Remarks on Rare PB Words," Journal of the Acoustical Society of America, 29, (1957), p. 751. 3Louis Decker, Herbert Rubenstein, and Irwin Pollack, "Word Frequency and Speech Intelligibility for Unknown and Known Message Sets," Journal of the Acoustical Society of America, 30 (July, 1958), p. 67. “G. A. Miller, G. A. Heise, and W. Lichten, "The In— telligibility of Speech as a Function of the Context of the Test Materials," Journal of Experimental Psychology, 41 (1951), pp. 329-335- , -- .'-.:- :'-'_‘ ~75:- C3 gmxhngnfl' I . .. . '....-.--.. _- -~::'.'.ii Mesa ,a‘lfi-‘fi‘.';fi --| _. . =_.|.. _'--.-‘, 32 vocabulary. In other words, those words with a greater probability of occurrence are reflected by higher discrimination scores than words with smaller probability. This indicates that when phrases and sentences are used for testing, the redundancy and the syntactic rules of language allow for correct discrimination even though much of the original signal may be missing. The hard of hearing person receives decided advantages from this feature of language. Impaired Speech Reception and Discrimination Much of what is relevant to the present topic has been suggested in earlier sections of this chapter. Again, however, some description of effects of hearing loss on speech perception seems indicated. The degree of hearing loss——indicated by the SRT measure——can range from near—normal to the maximum limits of the apparatus used to present the test items. Maximum limit on most clinical audiometers is 100 dB re audiometric zero. The effects of the loss on speech discrimination—— indicated by percentage of words correctly identified—- can range, of course, from 0 to 100 per cent. Silverman and Hirsh stated that useful tests of hearing, for clinical purposes, suggested at least one of the following functionsz:L 18. Richard Silverman and Ira J. Hirsh, "Problems Re— lated to the Use of Speech in Clinical Audiometry," Annals %§_Qtology, Rhinology, and Laryngglogy, 64 (1955), pp. 1234— (2) evaluation of medical or surgical treatment or a hearing aid, and (3) estimation of the social adequacy of hearing. They indicated that the first two functions are fairly well satisfied by present monosyllabic test items but that the third function still requires valida- tion studies. They believed that there is a need to think in terms of different discrimination tests for different purposes. Attempts have been made to predict speech recep— tion threshold on the basis of some average of pure tone thresholds.l Among the methods suggested are: (a) a simple average of 500, 1000, and 2000 cycles per second (cps); (b) the average of the two of the above frequencies showing the greatest loss of sensitivity; and (c) a simple average of thresholds at 1000, 2000, and 3000 (cps). Newby indicated that audiological opinion is divided,-and that research has not demonstrated the superiority of any 1 33 (l) diagnosis and prognosis connected with ear disease, \ 2 one of the procedures. ‘ Webster compared the average of 500, 1000, and 2000 cps versus the average of 1000, 2000, and 3000, and conclu— dei that no decision can be made regarding the best pre— A lNewby, op. cit. p. 100. , a \ 2Ibid. 34 diction of impairment until certain terms and concepts are better defined.1 Harris, Haines, and Myers, on the other hand used sentence materials spoken at a faster than normal rate, and suggested that while frequencies above 2000 cps may not contribute significantly to the intelligibility of undistorted speech in laboratory listening conditions, they may be important for dis— crimination of distorted speech in ordinary listening situations.2 Ross e£_gl.studied speech discrimination of hearing—impaired individuals in noise.3 The only factor that appeared to be related to speech discrimina— tion was the extent or configuration of the hearing loss. Those subjects with high—frequency hearing loss demonstrated less relative effect of noise than did subjects with flatter pure tone threshold configurations. It appears that prediction of impairment remains a problem requiring additional research. Carhart suggested that one of the most important considerations in discrimination testing with hard of hearing subjects is the linguistic background of the individual}4 In 1John C. Webster, "Important Frequencies in Noise— Masked Speech," Archives of Otolaryngology, 80 (1964), pp. 494—504. 2J. D. Harris, H. L. Haines, and C. K. Myers, "The Importance of Hearing at 3 kc for Understanding of Speeded Speech," Laryngoscope, 70 (1960), pp. 131—46. 3Mark Ross et al., "Speech Discrimination of Hearing— Impaired Individuals in Noise," Journal of Auditory Research, 5 (1955), pp. 47—72. ‘”“— ”Carhart, "Problems in the Measurement of Speech Discrimination," op. cit. '- rm. .- u i ."‘““- I-.i§:rsjfl'izp , _ “magmii 1' -g. aeiomsppe'fi' innit did:- bszhjagfiii-a-*bfii'iiiflh'i'” ’t. £513.": -.' .1" 3 ":i.":-.1".--.‘. '3 -.'.‘: xiii-d Hath-11413851 and DJ qffinuui sfiiidlpilizdfi%.-,‘ I ‘I L:- - Iv— I n— d - I ' .I- III I' _ .. _ ', ' . .i'::.'| 'I- '_-E Afflirl ..'g I n! l -i | i . _. '—."‘-'r:- .. -— F’— 35 other words, differences in word familiarity limit the generalizability of speech discrimination test results. Some of the effects of word familiarity on test results with hard of hearing subjects were described by Oyer and 2 Primary versus Secondary Signals One of the possible reasons for the lack of demon— strated validity of conventional discrimination tests is that these tests usually are performed without a noise background. Davis pointed out that the proper test for this type of discrimination is one in which a standard background noise is kept at a constant intensity while speech, heard along with noise, is varied in intensity.3 The reverse process also is feasible. Kelly and Steer, using a multiple—choice test format, evaluated the reliability of such measures and concluded that discrimina— tion scores obtained under reduced noise levels are as Doudna.l Schultz reported similar effects. 1 \ reliable as those obtained under high 1evels.Ll \ 1Herbert J. Oyer and Mark Doudna, "Word Familiarity as a Factor in Testing Discrimination of Hard—of—Hearing Subjects," Archives of Otolaryngology, 72 (1960), pp. 351—55. Speech Discrimination," Journal of Speech and Hearing Re— \ ‘ 2 i ' Martin C. Schultz, "Word Familiarity Influences in search, 7 (December, 1964), pp. 395—400. 3Davis, Hearing and Deafness, op. cit., p. 198. 4J. c. Kelly and M. D. Steer, "Intelligibility Test— ing in Three Conditions Involving Masking Noise," Journal of Speech and Hearing Disorders, 14 (1949), pp. 369—72. 36 Egan described the normal—hearing listener's task in noise, indicating that listener performance de— pends on the discriminability of the message and on the listener's criterion for either accepting his response as\forrect or rejecting it as incorrect.1 Hogan and Hanley reported that when either the number, rate, or level of interfering signals is increased, there is an associated decrement in discrimination scores.2 The masking of speech by bands of noise was studied by Hirsh and Bowman.3 They found that a middle frequency band (670 to 1000 cps) is most effective in masking the speech signal. Moser, Dreher, and O'Neill similarly found that prolonged vowel sounds with high concentration of energy between 700 to 1000 cps are most effective in masking words.Ll The masking effects of speech on discrimination were described by several researchers. For example, Hodgson used sixty talkers to produce a noise back— lJames Egan, "Monitoring Task in Speech Communication," Journal of the Acoustical Society of America, 29 (April, 1957), pp. 482—89. ‘ 2D. D. Hogan and T. D. Hanley, "Some Effects on Lis— tener Accuracy of Competing Messages Varied Systematically in Number, Rate, and Level," Journal of the Acoustical Soci— ety of America, 35 (1953), pp. 293-95. 3Ira J. Hirsh and W. D. Bowman, "Masking of Speech by Bands of Noise,” Journal of the Acoustical Society of America, 25 (November, 1953), pp. 1175—80. ”Henry M. Moser, John J. Dreher, and John J. O'Neill, ”Masking of English Words by Prolonged Vowel Sounds," Journal of the Acoustical Society of America, 29 (1957), p. 12V§Ho ‘ 37 ground, and found that this speech noise and white noise are equally effective in masking monosyllables, but that the speech noise is a better masker of continuous dis— course.1 Pollack and Pickett designated different signal—to—noise ratios required for 50 per cent intelli- gibility of PB lists.2 These S/N's range from 12 dB with one background talker to 6 dB with seven talkers. The use of speech babble appears to have potential value when used to sensitize conventional speech discrimination tests, although it is but one type of distortion faced by both normal-hearing and hearing-impaired subjects. This section of the present chapter—-devoted to an overall description of speech reception and discrimina- tion--was presented to indicate the complexity of the speech perception process and to suggest some of the problems involved in the study of speech discrimination. This material should demonstrate some of the difficulties faced by researchers and clinicians when they attempt to evaluate the effects of auditory training. 1William R. Hodgson, "A Comparative Study of the Effects of White Noise, Speech Noise, and Complex Noise on the Intelligibility of Speech" (unpublished Ph.D. dissertation, Ohio University, 1961). 21. Pollack and J. M. Pickett, "Stereophonic Listening against Voice Babble," Journal of the Acoustical Society of America, 30 (February, 1958), pp. 131—33. 38 Auditory Training: IProcedures and Goals Carhart emphasized that sensory habits are as fundamental to life's activities as are the more ob— vious muscular habits.l On this basis, auditory training for very young children is designed to establish auditory habits that allow maximum use of residual hearing and of acoustical stimuli in the environment. In this context, Carhart indicated that there are four major stages in auditory training for children: (1) develop— ment of awareness of sound, (2) development of gross discrimination, (3) development of broad discrimination among simple speech patterns, and (4) development of finer discriminations for speech. Newby similarly rec- commended that training proceed from simple, gross dis— crimination of environmental sounds to the more complex 2 These discrimination of vowel and consonant sounds. efforts are advised even though initially there may not be evidence of the child's response to auditory stimula— tion. Training with deaf children should begin as soon as a hearing loss is suspected. DiCarlo outlined some objectives for this training, including development of: (a) auditory lRaymond Carhart, Auditory Training, Hearing and Deafness, op. cit., pp. 373-77. 2Newby, op. cit., 309—16. recognition, (b) auditory memory span, (c) auditory discrimination, (d) auditory synthesis, (e) auditory scanning, and (f) auditory recall.1 The main objec- tive is the establishment of auditory contact with the environment so that this sensory avenue may be used to its maximum. Auditory training for adults requires a different frame of reference. Here the task, according to Carhart, is the re—education of an impaired ability.2 He further indicated that the subject's task with a hearing aid involves adjustment to loud sounds, development of speech discrimination in noise, improvement in localiza— tion, and so forth. Oyer, as well as other writers, emphasized the im— portance of combining auditory training with other proce— dures——lipreading, speech therapy if indicated, speech conservation, hearing aid orientation, and counseling—— and devoted a chapter to this topic.3 He stressed the need for adequate motivation and suggested means whereby the hearing clinician can stimulate a satisfactory level. Materials for auditory training with both children and lLouis M. DiCarlo, "Speech and Communication for the Deaf," Volta Review, 62 (1960), pp. 317-19. 2Carhart, op. cit., p. 380. 3Oyer, op. cit., Chapter 11. 40 adults were described,1 and the importance and contribu— tion of listening were given emphasis.2 Benefits to be derived from auditory training, according to Alpiner, include improvement in the tolerance threshold for amplified sound, adjustment to the way speech and noise sound when amplified, and possible changes in speech discrimination.3 Browd suggested that a hearing aid alone—-without auditory training—-seldom provides a satisfactory level of hearing.“ He stressed discrimination and interpreta- tion of phonemes and gave secondary emphasis to pitch, accent, rhythm, inflection, and duration. Frankel pro— vided an auditory training program for those who cannot receive daily training at a hearing center.5 Three long— playing records for playback in the home allow practice with or without a hearing aid. Larsen also prepared disc— recordings for practice in the home or clinic.6 1Ibid., Chapter 12. 2Ibid., Chapter 8. 3Jerome G. Alpiner, ”Aspects of Auditory Rehabilita— tion: Part II,” Audecibel, 13 (1964), pp. 47—51. “Victor L. Browd, "Hearing Education without the Use of Hearing Aids," Archives of Otolaryngology, 49 (1949), pp. 511—28. 5George W. Frankel, "A Planned Home Auditory Training Program," The Eye, Ear, Nose and Throat Monthly, 40 (April, 1961), pp. 560—62. 6Laila Larsen, Consonant Sound Discrimination (Bloom— ington, Indiana: The University of Indiana, 1950). FFrm' 41 Adjustment to a hearing aid during early stages of use was described by Haskins.l Subjects are instructed in how to listen again by using the aid first in easy listening situations,then progressing in steps to more complicated conditions. Heineman et a1. discussed the aural rehabilitation program at Walter Reed Army Medical Center.2 The authors emphasized auditory training with and without hearing aids, and stated that the primary goal of auditory training is to teach good listening habits. Some discrimination prac— tice is given for gross sounds, but major effort is ex— pended toward improved speech discrimination by contrast of similar sounding words and by use of recorded practice material. The authors indicated that auditory training is extremely beneficial, even for persons with fairly good low—frequency hearing but poor high—frequency hearing (i.e., the usual non—candidate for a hearing aid). O'Neill suggested several basic approaches to audi- tory training with adults.3 The first approach begins with phoneme—discrimination practice, then moves to syllable drills, and concludes with word-discrimination —fi. lHarriet L. Haskins, "Listening with the Aid of a Hearing Aid," Volta Review, 57 (1955), p. 408. 2Jan Heineman et al., "Listening Through Visual Hearing," Hearing News, 32 (July, 1964), pp. 5—8. 3John J. O'Neill, The Hard of Hearing, 0p. cit., pp. 89-91° 42 practice. The second approach emphasizes discrimination of phrases and sentences. Decision regarding approach depends on the particular needs of the individual subject. Recognition of pairs of words or syllables differing in either the vowels or consonants is a usual training method. A set of twenty lesson plans for adults was de— scribed by Johnson and Siegenthaler.l According to these authors, the goals of auditory training are: (a) to improve personal—social attitudes and relation— ships by understanding and accepting one's hearing problem; (b) to learn the care, operation, and limitations of the hearing aid; (0) to increase tolerance for amplification; (d) to improve sound localization ability; and (e) to improve perception of speech by training with a hearing aid. They reported a general increase in dis- crimination test results for twenty subjects enrolled in this program, but did not show actual data. Hutton suggested that when both auditory and visual stimuli are combined in aural rehabilitation, phoneme intelligibility is high enough for effective communication.2 1A. F. Johnson and B. M. Siegenthaler, "A Clinical Auditory Training Program," Journal of Speech and Hearing Disorders, 16 (March, 1951), pp. 35—39. 2Charles Hutton, "A Diagnostic Approach to Combined Techniques in Aural Rehabilitation," Journal of Speech and Hearing Disorders, 25 (1960), pp. 267—72; 43 Davis stated that speech discrimination cannot be-improved by a hearing aid or by surgery, but that sometimes it can be improved wholly or in part by auditory training.l Hirsh, however, was not so enthusiastic, suggesting that it is unusual for auditory training to change speech discrimination.2 He stated though that the discrimination score might be changed and that he has had reports of some successfully trained subjects. Hirsh indicated that most teachers of auditory training do not propose changes in the auditory systems but hope only to increase use of residual auditory capacities. The use of alphabet letters for discrimination test— ing and for early stages of training was recommended by Kelly.3 He indicated that lists of these letters in various combinations avoided word familiarity problems, were suitable for multiple presentation, were simple enough for children, and could be presented and scored in minimum time. An extensive set of drill material is pre— sented with emphasis placed on group training with talker— listener panels. lHallowell Davis, "The Articulation Area and the Social Adequacy Index for Hearing," Laryngoscope, 58 (August, 1948), pp. 761—78. 2Ira J. Hirsh, The Measurement of Hearing, (New York: McGraw—Hill Book Co., Inc., 1952), p. 300. 3Kelly, Clinician's Handbook, op. cit. 44 Whitehurst also collected extensive drill materials for teenagers and adults.1 These drills stress contrast of phonemes in initial, medial, and final positions within single words. Sentences and supple— mentary material are included. Thus, there are numerous sources available for sug— gestions concerning the form and content of an auditory training program. The general procedures and objectives appear to be similar among these sources. Subjects are given practice in the recognition and discrimination of speech stimulus items with emphasis on contrasts among similar—sounding phonemes, words, and sentences. Improved speech discrimination performance in everyday listening situations is the primary objective. The relative efficiency of various approaches and the effects of individual, task, and environmental variables usually are not described in training manuals. Research in Auditory Training Much of the research in the United States on audi— tory training for children was conducted by Hudgins. In one study he compared a high fidelity auditory training unit with an older—type model, and found that the newer model resulted in substantially more improvement in speech lMary Whitehurst, Auditory Training Manual (New York: Hearing Rehabilitation, 330 East 63rd Street, 1955). 45 perception than the older model.1 In another paper he reported on the positive effects of combining auditory training with visual training for profoundly deaf child— ren, indicating that such success may best be measured in terms of the difference between scores obtained by lipreading and those obtained by both lipreading and auditory stimulation.2 Recent research by Ling also suggested the importance of low-frequency hearing aid amplification in auditory training programs for deaf children.3 Of more immediate concern to the present study are those investigations that relate directly or indirectly to auditory training with adults. Most of what is now considered basic to this training originated from research with normal—hearing listeners during and immediately following World War II. Flanagan ep_gl. summarizedeartime research concerning the learning and reception of codes and speech in noise.” Rosenzweig and Stone reported rapid 1C. V. Hudgins, ”The Response of Profoundly Deaf Children to Auditory Training," Journal of Speech and Hear- ing Disorders, 18 (1953), pp. 273—88. 2Hudgins, "Auditory Training: Its Possibilities and Limitations," op. cit., p. 349 3Daniel Ling, "Implications of Hearing Aid Amplifi— cation Below 300 CPS," Volta Review, Reprint No. 828 (Wash- ington, D. C.: The Volta Bureau, 1537 35th St., N.W., 1965). “J. C. Flanagan, et a1. ,Psychological Research in the Armed Forces, Special Edition, Review of Educational Research, 18 (December, 1948), pp. 528— 655. 46 and extensive improvement in performance following training of both talkers and listeners.1 Steer, Hadley, and Kerr demonstrated that a train— ing program in listening produces significant gains.2 An experimental group of flight students received two hours of instruction and practice in listening to fre— quency—distorted speech. This group and a control group were tested twice with both groups improving in perform— ance on the second test, but with the experimental group showing significantly greater gains. Egan summarized the extensive research programs at the Harvard Psycho—Acoustic Laboratory during the war, and reported that discrimination improvement with normal- hearing listeners is considerable under difficult listen- ing conditions.3 Data are presented that describe learn- ing curves for speaker—listener combinations during eight successive days of practice; scores improved from approxi— mately 59 per cent on the first day to a plateau of about 76 per cent on the eighth day. It is difficult to determine whether this increase was due to improved listening, to 1Mark R. Rosenzweig and Geraldine Stone "Wartime Re- search in Psycho—Acoustics," ibid., pp. 642—5 . 2M. D. Steer, J. Hadley, and W. Kerr, "Listening Training Aids for Pre-Flight and Primary Flight Students,” Cited in Kelly, op. cit., p. 26. 3Egan, "Articulation Testing Methods," op. cit., p. 970. 47 increased speaking skill, or to some combination of the two. Egan showed that discrimination scores are relatively stable among six one-hour training sessions given on the same day. These results with normal—hearing listeners, to- gether with the large number of veterans sustaining hearing loss, led to the incorporation of auditory train— ing into service—connected aural rehabilitation programs. DiCarlo sought to measure the effectiveness of this train- ing at Borden General Hospital.1 Comparison of pre— and post— training discrimination tests (given live-voice) indicated that 472 veterans receiving auditory training improved about 19 per cent in mean performance, while 53 veterans not receiving training improved only about 3 per cent. Research other than armed services sponsored was conducted during this period. Goodfellow in 1942 reported his observations with six hearing—impaired subjects who received four months of auditory training.2 Improvement was observed that Goodfellow attributed to adaptation to frequency distortion, clarification of phonetic concepts, attention to secondary cues contained in the gross pattern lLouis M. DiCarlo, ”Auditory Training for the Adult," Volta Review, 50 (September, 1948), pp° 490-96, 2Louis D. Goodfellow, "The Re—Education of Defective Hearing,” Journal of Psychology, 14 (1942), pp. 53-58, 48 of speech, and changes in the subjects' attitude to— ward their handicap. Silverman in 1944 reported that six out of seven adult subjects (hearing aid users), after receiving systematic auditory training, showed improvement in speech discrimination ranging from 0 to 36 per cent for words and 8 to 52 per cent for sentences.1 Browd in 1949 indicated that thirty—seven out of forty-six subjects showed either no disability or only occasional evidence of it after training.2 He emphasized training without hearing aids and suggested that among persons who are not candidates for hearing aids and who receive auditory training, few fail to attain a satisfactory level of hearing. In 1953 Kelly reported the effects of talker— 1istener drills with two groups of teenagers seen during two summer sessions of one—half hour daily practice for six weeks.3 One group of fifteen subjects showed a signi- ficant mean improvement in discrimination of 21 per cent, and the second group (sixteen subjects) showed a signifi— cant improvement of 17 per cent. Neither group showed lS. Richard Silverman, ”Training for Optimum Use of Hearing Aids,” Laryngoscope, 54 (1944), pp. 29—36. 2Browd, op. cit. 3Kelly, 0 . cit., pp. 26—27. 49 significant improvement in speech reception threshold, although both groups showed some improvement. Hutton in 1960 used the multiple—choice Semi— Diagnostic Test,by live—voice testing,as a criterion measure.l He found a mean improvement of 6 per cent in auditory discrimination for eighteen subjects receiving approximately three months of training. Other researchers were not interested in auditory training per se. Rather, their research interest con— cerned changes in speech and/or pure tone discrimination as a function of test-retest or of duration of practice. Subjects usually were normal-hearing. Thurlow ep_§I.in one study found that PB scores decreased significantly among three tests given over several weeks and months,2 but then in a second study reported significantly better discrimination on test- retest comparisons.3 Moser and Dreher, concerned with listener variability in intelligibility studies, demon— strated that normal-hearing listeners show progressive 4 improvement in performance with training. _r lHutton, op. cit. 2w. R. Thurlow et al., "A Statistical Study of Auditory Tests in Relation to the Fenestration Operation," Laryngpscope, 58 (January, 1948), pp. 43—66. 3w. R. Thurlow et al., "Further Statistical Study of Auditory Tests in Relation to the Fenestration Operation," Egrypgoscope, 59 (1949), pp. 113—29. . “Henry M. Moser and John J. Dreher, "Effects of Train— lng on Listeners in Intelligibility Studies," Journal of the flgoustical Society of America, 27 (November, 1955), pp. 1213-19. 50 Thwing compared listener performance on four succes— sive presentations, one per second, of single PB words under three conditions of noise.1 He found at all sig— nal-to-noise ratios that major improvement (about 10 per cent) occurred with the second presentation of the test word; third and fourth presentations had negligible effects. Campbell and Small studied the effects of prac— tice and feedback on difference-limens for pure tones, and found that these limens decreased as a function of training.2 Changes in pure tone thresholds as a function of practice were investigated by Zwislocki pp_§I.3 Five experimental treatments—-administered to five groups of listeners—-resulted in significant gains in threshold detection with greater improvement at 100 cps than at 1000 cps. Licklider and Miller summarized research in train— ing, stating that there seems to be no better way to teach listeners than to motivate them and have them 1Edward J. Thwing, "Effect of Repetition on Articula- tion Scores for PB Words,” Journal of the Acoustical Society of America, 28 (March, 1956), pp. 302-303. 2Richard-A. Campbell and Arnold M. Small, Jr., "Effect of Practice and Feedback on Frequency Discrimination," Journal of the Acoustical Society of America, 35 (1963), pp. 1511—14. 3Jozef Zwislocki et al., "On the Effect of Practice and Motivation on the Threshold of Audibility," Journal of the Acoustical Society of America, 30 (1958), pp. 254—62. 51 listen.1 They indicated that subject variability among normal listeners is considerable, even after extensive training. It appears that information from. studies in learning and training has general relevance to auditory training as a specific type of sensory learning or re-learning. Studies thus far reviewed indicate positive effects of training on speech discrimina- tion. Principles of-Learning and Training Wolfle suggested a convenient dichotomy between learning and training, one which serves as a guide for the discussion in this section. He stated that if an investi- gator's primary interest is in the processes by which knowledge or skill is acquired, then his studies are classified under the heading of learning; if his primary purpose is to investigate the teaching of knowledge or skill, his work is classified under the heading of 2 The latter category was of main interest in training. the present study. World War 11 gave psychologists an Opportunity to give practical tests to theoretical hypotheses about learn— 1J. C. R. Licklider and G. A. Miller, The Perception of Speech, Chapter 26 in Handbook of Experimental Psychology, ed. S. S. Stevens (New York: John Wiley and Sons, Inc., 1951), p. 1068. 2Dael Wolfle, Training, Chapter 34 in Handbook of Experimental Psychology, ipidl, pp. 1267-862 52 ing behavior, particularly those changes in behavior that result from training. Wolfle described some of the use— ful parts of learning theory applicable to training, and suggested the following guidelines for training programs.1 First, the distribution of practice should be suitable for the task to be learned. Second, active participation by the learner is superior to passive receptivity. Third, practice material should be varied so that the learner can adapt to realistic variation and so that motivation during drill is improved. Fourth, accurate performance records must be kept in order to evaluate progress and effects of training. Fifth, the most useful single con— tribution of psychology to training was the provision for immediate knowledge given to learners regarding their performance. Postman and Egan--writing shortly after World War II ——provided suggestions and guidelines for research in training.2 In general, the more distributed the practice trials, the better the learning, although this depends to some extent on the particular task. Meaningful materials lDael Wolfle, "Military Training and the Useful Parts of Learning Theory," Journal of Consulting Psychology, 10 (1946), pp. 73—75. 2Leo Postman and James P. Egan, Experimental Psycho- logy (New York: Harper and Brothers, Inc., 1949), pp. 395—462. " 53 are better retained than nonsense items. Ideally, subjects learn best by performing tasks with full awareness of the principles guiding successful per- formance. Of potential relevance to the present study is the authors' statement regarding sensory capacity. They stated that the beneficial effects of past ex— perience probably are due to the transfer of observa— tional techniques and attitudes.1 McGeoch and Irion summarized a great deal of the 2 A number research and theorizing in learning behavior. of principles, in addition to or complementary to those mentioned above, are pertinent to the present study. The importance of adequate motivation was emphasized. The subject must initiate and sustain his practice of a given task in order to reach a satisfactory performance level. The beneficial effects of spacing practice—- particularly with motor—type skills-—can be demonstrated in a large number of situations. The authors stated that the conditions necessary for learning is a subject of considerable theoretical dispute.3 llbid., p. A26. 2John A. McGeoch and Arthur L. Irion, The Ps cholo of Learning (New York: Longmans, Green and Co., 1952). 3Ibid., p. 296. 54 Further relevant information was suggested by Kingsley.l He stated the principle that the smaller the interference between practice sessions, the more substantial the learning. In addition, it appears that the shorter the practice period, within limits, the greater the learning that occurs. The question of the importance of spaced practiCe was discussed by Mednick, who indicated that Spacing only wastes time when the responses to be learned are familiar but have to be given new meaning or connected 2 This condition seems to exist to a different stimulus. in auditory training with hard of hearing adults. Their task is one of attaching new meaning to distorted acoustic signals. Therefore, a concentrated training program may have some advantage over excessively-spaced sessions. The relations between learning and practical de— mands of instruction and training were described by Hilgard.3 He indicated that the decline in learning 1Howard L. Kingsley,The Nature and Conditions of Learning (2nd ed.; Englewood'Cliffs, New Jersey: Prentice— Hall, Inc., 1957), ppo 237-57. 2Sarnoff A. Mednick, Learning (Englewood Cliffs, New Jersey: Prentice—Hall, Inc., 1964), p. 87. 3Ernest R. Hilgard, Theories of Learning (2nd ed.; New York: Appleton-Century—Crofts, Inc., 1956), pp. 485- 87. 55 ability with age, in adult years, depends on what it is that is being learned. A comment of interest is that a motivated learner acquires what he learns because he practices more than an unmotivated person. Furthermore, excessive motivation may be detrimental, particularly for tasks requiring difficult discriminations. Hilgard emphasized the importance of providing knowledge of performance during training. This brief review of training principles suggests the following applications to auditory training. An adequate level of motivation should be maintained. Spaced practice may be of minimal importance. Knowledge of performance is indispensable in improving skill. Variation in practice is needed to maintain interest and to prepare the subject for variations in situations where the training is to be applied. Interference between training sessions should be anticipated. The subject should have an active rather than a passive role in the training° The subject should be aware during training of realistic objectives and should be informed regarding principles aiding successful performance. These and other basic principles guided the formulation of the training procedures evaluated in the present investigation. Hearing Loss and Hearing Handicap In recent years effort was made by various pro— fessions to delineate the meaning of three different terms: 56 impairment, disability, and handicap. The distinguishing features among these terms are fairly well specified at this time. Hamilton described the distinctions: A disability is a condition of impairment, physical or mental, having an objective aspect that can usually be described by a physician . . . A handicap is the cumulative result of the obstacles which dis- ability interposes between the individual and his maximum functional level.1 Wright further recommended that a handicap must be evaluated in terms of the demands of the situation in which the person finds himself, and that disability cannot be equated with handicap.2 A person may feel physically handicapped even though from the medical point of View, his physical limitations are not disabilities. In addition, the obstacles that the physical disability inter- poses may be as much social in character as physical. These considerations led Wright to conclude that (a) a physical attribute is a physical handicap only when it is seen as a significant barrier to the accomplishment of particular goals, and (b) a physical attribute may become handicapping not because it is physically limiting but be- cause it adversely affects social relationships. Oyer indicated that measurements of deficit and handicap are not the same, stating that a person might lK° W. Hamilton, Counseling the Handicapped in the Rehabilitation Process (New York: Ronald Press, 1950), p. 17. 2Beatrice A. Wright, Physical Disability: A Psycholo— gical Approach (New York: Harper and Row, 1960), pp. 9—10. H, "I 57 have a disorder but not be handicapped by it while another person may have several handicaps because of only one dis- order.1 He stated that hearing handicap varies as a function of demands placed on the hearing-impaired in- dividual.2 Several attempts have been made to assess hearing handicap. Davis' Social Adequacy Index was given some clinical use.3 Only two measures were required——SRT and Harvard PB-Max--to enter a prepared SAI table in which values had been correlated with subjects' reports of the degree of handicap they encountered in various listening situations. Davis, however, recently indicated that the SAI has not been too successful: . . . we do not yet seem to know quite enough about the relation of the hearing and understand- ing of connected speech in words and sentences to its component frequencies, phonemes, and sylla— bles. Speech is a very dynamic, variable affair and we can define it only in broad statistical terms or in terms of word- and soundfipatterns that have not yet been standardized. Carhart agreed, stating that SAI merely ranks subjects along a scale whose practical significance has not been clarified throughout its entire range.5 1Oyer.,Auditory Communication . . ., op. cit., p. 126. 2Ibid., p. in. 3Davis, "The Articulation Area . . .," op. cit. “Davis, and Silverman, Hearing and Deafness, 9p. cit. p. 194. 5Carhart, "Problems in the Measurement . . .," op. cit. t W 58 Davis devoted an entire chapter to this complex prob— lem, and described some of the considerations and controver- sies surrounding military standards and medico—legal ruling.l Consideration of this problem led High, Fairbanks, and Glorig to state that systematic investigation of hearing handicap has lagged far behind the development of techniques for the measurement of hearing impairment.2 Beginning with this apparently valid assumption, these researchers developed a Hearing Handicap Scale for assessment of one aspect of hearing handicap——the subject's self-report of the degree of difficulty he experiences in hearing-related activities. The authors reported that significant correlation coefficients (about 0.70) were obtained between this scale and all measures of hearing sensitivity for the subjects' better ear. The scale did not correlate significantly with either better or poorer ear speech discrimination. One of two alternate forms of this scale was used in the present study and is described further in Chapter III. (A copy is included in Appendix D). Attempts have been made to develop sentence—type tests that would have high face-validity for representing everyday speech. Silverman and Hirsh specified criteria as described 1Davis and Silverman, op. cit., Chapter 9. 2Wallace S. High, Grant Fairbanks, and Aram Glorig, "Scale for Self—Assessment of Hearing Handicap," Journal of Speech and Hearing Disorders, 29 (August, 1964), pp. 215-30. 59 by a panel of experts,1 and Davis and Silverman published a set of sentences that reportedly meet these criteria.2 Standardization studies, however, have not been published. The complexities involved in this type of investigation apparently have delayed specification of the operations required for a comprehensive validity—study. Hearing and Listening The distinction between hearing and listening is not often very clear in the literature. It is easy to assume that if a person hears an acoustical event, he is also listening to it. This actually may or may not be happening. In fact, audition possibly might not be either a necessary or sufficient condition for the listening act. For example, O'Neill discussed "visual listening" in describing the attitude of a lipreader required for success in interpreting the visual cues offered by the positions and movements of a speaker's lips, facial musculature, and gestures.3 Listening and attention may be somewhat analogous terms. Kingsley stated that attention is an indispensable condition for learning, and defined attention as a process that selects stimuli for perception.)4 O'Neill suggested lSilverman and Hirsh, "Problems Related to . . .," op. cit. 2Davis and Silverman, Hearing and Deafness, 0p. cit., pp. 5u8-52. 3O'Neill, op. cit., p. 85. “Kingsley, op. cit., p. 359. 6O a broader definition of listening, indicating that it may be considered to be an analysis of the impressions re— sulting from concentration where an effort of will is required.1 Oyer recently devoted an entire chapter to the discussion of listening in relation to auditory communication.2 He defined listening in this context as an attitude, a posture, or a mental set that prepares a person in his attempt to receive and use information transmitted by acoustic events.3 The growing interest in listening as a phenomenon worthy of academic, scientific, and general interest is indicated by two recent publications edited by Duker. The first one presents some fifty articles by over thirty- A The second book is an annotated biblio- five authors. graphy of over 1000 entries, all dealing with listening.5 The relevance of listening to studies in audition was suggested by Farrimond,who stated that at low intensity levels of signals, listening techniques appear to affect performance.6 Oyer has emphasized the contribution of lO'Neill, Op. cit. 2Oyer, 0p. cit., Chapter 8. 3Ibid., p. 81. “Sam Duker, Listening: Readings (New York: The Scare— crow Press, Inc., 1966)} 5Sam Duker, Listening Bibliography (New York: The Scarecrow Press,’Inc.,‘I964). ' 6T. Farrimond, "Factors Influencing Auditory Percep- tion of Pure Tones and Speech" Journal of Speech and Hearing Research, 5 (1962), pp. l9A-2OU. 1t 7 »~ 61 listening to auditory training.1 He designated possible barriers to successful listening,and suggested means of improving listening performance by clinical procedures. The importance of listening to auditory training also was indicated by O'Neill: Listening may be active, passive, or partial. The good listener must be involved in what is being said ———not a mere spectator or eavesdropper. The person who has had a hearing loss for any length of time will have lost the ability to be an active listener.2 The responsibility of the audiologist, therefore, is to establish or re-establish optimum listening performance in the hard of hearing subject receiving auditory training. There seem to be many directions in which both clinical endeavors and related research might investigate, describe, and relate listening behavior to auditory be— \ havior. The contingencies among these behaviors and other aspects of human communication remain challenges for systematic study. Summary ‘ Auditory training designates the procedures by which the hearing—handicapped are given the opportunity to make 1 full use of acoustic signals impinging on the auditory mechanism. The rationale for these procedures has been known for almost 2000 years, but it was not until the 19th lOyer, op. cit. 2O'Neill, op. cit., p. 83. century that it was given major consideration as a means of educating deaf children. Max Goldstein, a St. Louis otologist, is considered as one of the leaders who introduced and established auditory training practices in the United States. The introduction of modern, wearable hearing aids, and the aural rehabilitation programs during World War II, gave impetus to the incorporation of auditory training in rehabilitation programs for hard of hearing adults. In recent years auditory training has become a standard practice in practically every audiology clinic offering services to the hearing—handicapped. Research or clinical attempts to evaluate the effective- ness of auditory training require recognition of basic known features of auditory speech perception, particularly the necessary conditions for measurement and the complexities involved therein. Auditory training for adults generally includes super— vised practice in the recognition and discrimination of distorted speech,with emphasis on contrasts among similar- sounding phonemes, words, and sentences. Improved speech discrimination performance in everyday listening situations is the primary objective. Research in auditory training has suggested that lis- tener performance can be improved to some degree by practice. The extent of this improvement and the relative efficiency :'T ’63 of various procedures apparently have not been studied extensively with a substantial number of subjects or statistical controls. Generalizations concerning potential benefits were derived from informal observations with adults and from studies and observations with deaf children and with normal-hearing listeners. Basic principles of learning and training have applica— tion to both research and clinical efforts in auditory training. Clarification of terms such as impairment, dis— ability, and handicap is advisable when describing auditory training processes and objectives. Diagnosis of hearing "handicap" on an objective basis remains a challenge to the audiology profession. Dimensions of "listening" may be critical in determining the effectiveness of auditory training. CHAPTER III SUBJECTS, INSTRUMENTATION, AND PROCEDURES In this chapter specifics are given regarding the selection of subjects, their assignment to four training groups, the actual training administered to them, and the overall research design. In brief, subjects were selected from a population of mildly hard of hearing individuals. These subjects were seen first for prelimi- nary audiological testing and then were assigned systemati— cally to one of four training groups, each group being matched on speech reception threshold, discrimination in noise, intelligence, discrimination in quiet, age, education, sex, and duration of hearing loss. Each of the four groups participated in auditory training under one of four combinations of methods and materials. Methods involved experimental manipulation of signal—to-noise ratios and materials consisted of open set (write—down) and closed set (multiple—choice) types of stimulus materials. Three discrimination tests were administered both before and after training to determine the relative effectiveness of the methods and materials. The criterion measures were obtained by means of the W-22, Rhyme, and Semi—Diagnostic tests. 64 65 Subjects Interest was focused on adults who have mild, sensori-neural hearing loss. The term "sensori-neural" indicates the absence of a significant difference between air and bone conduction threshold responses, suggesting that the hearing loss is not amenable to surgical or med— ical treatment. Mild hearing loss was operationally de— fined as a speech reception threshold between 5 and 35 dB sound—field. This population was further limited to those subjects who reported that their hearing loss was sustained after acquisition of normal auditory discrimina- tion of American English (i.e., after approximately five years of age). In addition, the population was defined in terms of persons who were referred to or sought the assistance of an audiology clinic. The relationship between this clinical population and the total hearing loss popula— tion in the United States was unknown. This population generally reports a primary hearing difficulty in noise—background situations. Significant hearing impairment for speech usually is not shown for quiet listening conditions. Listening to speech at a distance from the speaker presents discrimination diffi- culties for this group. A hearing aid often does not alleviate the discrimination problem. Pure tone testing may show one of the following configurations: (a) flat 66 responses across frequencies, (b) gradual decrease (or increase) in responses, (c) abrupt decrease in responses, or (d) some combination of these between ears . Selection of Subjects Names of subjects meeting the above criteria for consideration were obtained from the files of (a) the Speech and Hearing Clinic, Michigan State University, and (b) the Speech and Hearing Department, Rehabilitation Medical Center, Edward W. Sparrow Hospital in Lansing, Michigan. Both clinics are staffed and supervised by faculty members in the Speech and Hearing Science area of the Department of Speech, Michigan State University. Files at the above clinics were surveyed,anui data on each potential subject were recorded. These data included information about educational background, reported duration of loss, occupation, hearing aid status, therapy status, address, phone number, pure tone air and bone conduction threshold responses, speech discrimination scores for each ear and by sound-field testing, and the date of the most recent audiological evaluation. Subjeet Contact Approximately fifty potential subjects were contacted by letter, a copy of which is shown in Appendix A. Some thirty—six persons agreed to participate in the study, and were seen for preliminary testing to establish their ' is?) "J's: AESTLJ £6? imfli 67 candidacy for this study. Four of these persons were eliminated from consideration because they either failed to meet selection criteria or decided not to participate in the subsequent training program. All subjects were informed regarding the overall purposes and procedures of this investigation,but were not given information concerning specific questions of interest. Instrumentation The following instrumentation was utilized during preliminary audiological testing and during the subsequent testing and training procedures. Tape—recording instru- ments and calibration apparatus also are listed below. Preliminary Testing Commercial test-room (Suttle) Commercial dual—channel audiometer (Allison, Model 22) microphone (Electro—Voice, Model 636) stereo tape transport (Viking, Model 87) phonograph (Bogen, Model B62) binaural ear-receivers (TDH—39, MXAlAR) Sound—field—speaker (Electro-Voice, Model SPl2) Recording Apparatus Microphone (Electro-Voice, Model 65A) Dual—channel tape-recorder (Ampex, Model 602) Tape—recorder (Ampex, Model 601) High—fidelity tape (Scotch Low Noise 203) Analysis of Speech Babble Tape—recorder (Ampex, Model AG350) AF Spectrometer (B & K, Model 2112) Graphic level—recorder (B & K, Model 2305) 68 Calibration Equipment Microphone (B & K, Model 4132) Artificial ear (B & K, Model 4152, 6 cc coupler) Sound level meter (B & K, Model 2203) Testing and Training (Same as apparatus described above under Prelimi- nary Testing). The specific employment of this instrumentation is described in subsequent sections of this chapter. Procedures Subjects were seen individually to obtain perfor— mance measures that then were used to establish four train- ing groups with eight subjects in each group. The proce— dures described in this section were preliminary steps conducted prior to the auditory training procedures. When subjects were seen for this testing, case history information in clinic files was verified. The mean age of subjects was 44 years (median: 44 years) with a range of 21 to 60 years. Mean educational background was 13.4 years (median: 12 years) with a range of 8 to 20 years of formal schooling. The reported duration of hear— ing loss was 2 to 30 years with a mean of 12.4 years. There were 26 males and 6 females, and 9 hearing aid users. (See Appendix C for individual data and Table 1, page 71 for summary statistics). 7+” I A 69 Aggielogical.Measures Audiometric measures for each subject included pure tone air conduction thresholds, speech reception thresh— olds, and speech discrimination scores in quiet and in noise. All testing was conducted in a commercial test— room (Suttle) and with a commercial dual—channel audi— ometer (Allison, Model 22) The audiometer and accessories were capable of delivering input signals at calibrated output levels through either binaural ear—receivers (matched set of TDH—39 phones with MX41AR cushions) or sound—field speakers (Electro—Voice, Model SP12). Only one of the two available sound-field speakers was used in this study (for both the testing and training conditions). The following audiometer input-channels were used: (a) microphone (Electro—Voice, Model 636), used for giving instructions to the subject; (b) pure tone input channels; (0) tape transport (Viking, Model 87% used for playback of the speech babble recording; (d) white noise generator, used for calibration of the sound-field speaker; and (e) phonograph (Bogen, Model B62), used for playback of the CID W—l and CID W-22 disc—recordings.l \ lCommercial recordings obtained.from Technisonic Studios, Inc., 1201 South Brentwood B1vd., Richmond Heights, Missouri. 70 The output of each ear—receiver and of the sounde- field speaker was calibrated prior to the testing pro— cedures and was checked periodically during the several weeks of preliminary testing. Pure tone calibration of ear—receivers was done according to recommended ISO pro- 1 Instrumentation included an artificial ear cedures. (B & K, Model 4152, 6 cc coupler), a microphone (B & K, Model 4132), and a sound level meter (B & K Precision Sound Level Meter, Model 2203). Coupler sound pressure levels (re 0.0002 microbar) resulting from these procedures were transformed to equivalent standard values in audio— metric decibels for each frequency. Output of the sound-field speaker was calibrated at the anticipated head—level position of the subjects.2 White noise was used as a calibrating signal in such a way that zero-reference on the audiometer attenuator— dial corresponded to approximately 22 dB sound pressure level. Subsequent attenuator readings thus were with reference to this audiometric—zero. All remaining inten- sity—level data in this study are expressed with this reference, speech babble levels included. 1Standard Reference Zero for Calibration of Pure Tone AudiOmeters, ISO/R 389-1964. 2American Standard Specification for Speech Audio- meters, ASA 224.13—1953. ,-3-¢‘a 3553-5 3 551.6535 si_+i—n5: -s.u.E 2!.J.fl: --bttia5"' _a :-:..-'i-_-:-'.'-:.'2:.’vat-"5.1 w- -=. ..:'...'i.;.:.;.-;i' {Terr-m1 r.- _ ' ' ' _’_.“: ':‘,‘. 1:": =_- '— ' ' ' "I! "I L'fl' l'l‘lialwr 71 r..— The experimenter was seated at the audiometer con- sole in a room adjacent to the test-room and could observe and communicate with each subject during the individual evaluation. Each subject was seated in the test—room, facing directly toward the sound—field speaker with his head approximately 56 inches from the speaker. The test arrangement is shown in Figure 1. Pure tpne testing.—-Pure tone air conduction thresh- olds were determined by the Hughson—Westlake ascending technique.l Frequencies tested were 500, 1000, 2000, 3000, and 4000 cps. The better ear, as indicated by pre— vious test results, was tested first and apprOpriate masking , if necessary, was delivered to this ear during poorer ear testing. The majority of the subjects had similar pure tone sensitivity between ears. (Pure tone results for each subject are reported in Appendix B). Speech reception threshold.—-SRT measurements were obtained for each ear and for the sound—field con— dition. The CID W—l disc—recording was the source of Spondaic speech stimulus items. A conventional descending- ascending, "bracketing" procedure was employed. In other words, the intensity of the recording was varied near threshold until the intensity was found at which the sub- lWalter Hughson and Harold Westlake, "Manual for Pro- gram Outline for Rehabilitation of Aural Casualties Both Military and Civilian," Transactions of the American Academy of Ophthalmology and Otolaryngology Supplement, E8 (1944), pp. 1415. 72 chair "SOOQ-I 5611 speaker two«way window audiometer Figure l.w—Schematic representation of physical arrangements during preliminary testing and selection of subjects. 73 ject repeated 50 per cent of the test words correctly. The better ear was tested first, followed by the other ear, and then concluded with the sound-field measurement. (Results of this testing for each subject are reported in Appendices B and C). Speech discriminationfiin quiet.--Three of the four different CID W-22 disc-recordings were used as the source of monosyllabic testing items, a different test for each ear and for the sound-field measure in this test order. Each list was presented at SRT plus 35 dB. This repre— sented a compromise between conventional 30 and 40 dB levels above SRT. The experimenter monitored subject responses and recorded the number of correct items for each list. (Percentage of correct responses for each sub— ject on these selection tests is reported in Appendices B and C). Speech discrimination in noise.-—The last audiologi- cal measure was speech discrimination in noise, and the remaining W-22 list was utilized. This list was delivered to the sound-field at 50 dB,with speech babble (described in the next section) delivered simultaneously through the same speaker at an average 45 dB intensity-level. The subject's task again was to identify correctly the mono- syllabic test item. Number of items correct was recorded for each Subject. (See Appendix C for percentage correct data). 74 Speechfpabble.-—A speech babble signal was employed to sensitize the discrimination task so that it would be more difficult than a similar task in quiet. This proce- dure, though arbitrary, not only enhanced inter—subject differences not easily detected otherwise, but also approximated the type of listening condition in which many hard of hearing subjects report their greatest hear- ing difficulty. The source of the speech babble--both during the pre— sent testing and during subsequent training conditions—- was a tape-recording of twenty male and female college students reading aloud different speech material at the same time. The recording was prepared by Higgins, who reported that a microphone (Electro—Voice, Model 605) connected to a tape—recorder (Ampex, Model 601) was placed in the middle of a classroom during the simultaneous oral l A segment of this recording showing a minimal readings. amount of intensity fluctuation (no more than 3 dB) was made into a loop. A thirty—minute recording of this loop then was made (i.e., the signal was transferred from an Ampex 601 recorder to a Wollensak T—1500 recorder). lDoris Mary Higgins, "The Effects of White Noise and Speech Babble on the Intelligibility of Phonetically Balanced Lists of Monosyllabic Words" (unpublished Master's thesis, The University of Tennessee, 1965). 75 Both Higgins' analysis and present analysis (B & K Spectrometer, Model 2112 and B & K Level Recorder, Model 2305) of this recording showed primary energy concentration for the filter-frequencies between 150 and 1000 cps, and an.amplitude—time display for filter- frequencies within this range showed minor intensity fluctuations (3 dB range). WAIS Vocabulary Testing The Vocabulary sub—test of the Wechsler Adult Intelligence Scale was given to estimate the intellec- tual performance of the present subjects. Since this sub—test reportedly correlates approximately 0.83 with the Full—Scale score (about 0.90 with Verbal score), and because its reported split-half reliability coef— ficient is about 0.95,1 it was believed to be suitable for present purposes. This estimate was needed as a control factor for intelligence. For an assumption of "equivalency” among training groups on factors important to amount of learn— ing in a training situation, some estimate of intelligence (i.e., ability to learn) was necessary. The Vocabulary sub—test of the WAIS should meet this requirement rela— 1David Wechsler, The Measurement and Appraisal of Adult Intelligence (4th ed.; Baltimore: Williams and Wilkins, Inc., 1958), pp. 99—103. 76 tive to gpppp performance. This procedure is not followed or recommended in a clinical situation where individual performance is evaluated. Any conclusions in this study relative to the intellectual factor were interpreted with due caution. Recommended procedure for individual administra— tion and scoring was followed.1 Each subject was given a printed listing of the forty words included in the Vocabulary sub-test. The experimenter then asked the subject the meaning of each successive word. Responses were recorded by the experimenter for later scoring. Scoring of all responses of the thirty—two subjects was done with each response scored according to criteria given in the WAIS manual. The subject could receive 0, l, or 2 points for each word, making a range of possible scores from O to 80 points. During the scoring the experimenter did not know whose responses were being evaluated. Scores assigned to items were re-evaluated on a second occasion with no significant change in origi— nal scores. (See Appendix C for raw scores on each subject). Assignment of Subjects to Four Training Groups On the basis of test performance and case history information——described in previous sections of this lDavid Wechsler, Manual for the Wechsler Adult Intelli- gence Scale (New York: The Psychological Corporation, 1955), pp. 42—43, 63—75. 77 chapter——the thirty—two subjects were assigned systemati— cally to one of four training groups. Group means, standard deviations, ranges, and medians were matched among} groups. This was done to establish relative "equivalency" among groups so that potential effects of auditory train- ing can be attributed to the training rather than to possible initial differences among groups. Substantially more confidence then can be given to conclusions drawn regarding the effects of the specific treatments under experimental control. The results of the above matching-among—groups pro- cedure are shown in Table 1. Mean performance of all thirty—two subjects was as follows: (a) sound—field discrimination—in—noise score of 47.0 per cent, (b) sound- field speech reception threshold of 22.2 dB, (0) WAIS Vocabulary sub—test score of 49.1 points, and (d) sound- field PB—Max score of 78.1 per cent. Mean age was 44.0 years, mean educational level was 13.4 years, and mean duration of hearing loss was 12.4 years. There were one female in each of two groups and two females in each of the two remaining groups. There were two hearing aid users in three groups and one in the remaining group. As can be noted, there were no major discrepancies among groups on these measures. However, these measures now having been specified can be consulted, if necessary, and results re— ported in Chapter IV qualified accordingly. 78 TABLE 1.--Means, standard deviations, medians, ranges, and descriptive data resulting from group matching pro- cedures (N = 32, n = 8). i—: fir Group 1 Group 2 Group 3 Group 4 Total DS/Noise (%) 48.3 46.5 45.5 47.5 47.0 stnd. dev.: 16.6 15.7 18.1 19.2 range: 28-86 18-70 24—76 16-72 SRT (dB) 21.3 21.9 23.1 22.5 22.2 stnd. dev.: 9.3 10.9 9.7 9.0 range: 5-35 5—35 5-3 5—35 WAIS Vocab. 50.4 49.6 47.5 49.0 49.1 stnd. dev.: 13.7 13.1 15.2 13.1 range: 28-72 26—72 18-67 31-67 PB-Max (%) 76.5 80.8 75.3 79.8 78.1 range: 52—96 64-94 48-98 52-96 Age (years) 43.4 46.0 43 3 44.3 44.0 range: 28-54 21—60 26-60 35—55 median: 45 46 42 42 Educ. (years) 13.5 13.1 13.3 13.9 13.4 range: 9—20 8—20 8-19 11-20 median: 12 l2 13 12 Duration of Loss (years) 12.6 13.3 9.0 14.8 12.4 range: 4—20 3—25 2—20 4—30 Males 6 7 7 6 26 Females 2 1 l 2 6 Hearing Aid Users 2 3 2 2 9 u' 79 Six factors were of interest in this investigation: (a) comparison of pre- and post-training discrimination performance; (b) comparison of Speech discrimination tests (W-22 vs Rhyme vs Semi-Diagnostic); (c) comparison of auditory training methods (S/N-constant vs S/N-varied); (d) comparison of auditory training materials (open set monosyllables vs closed set multiple—choice); (e) compari- son of differential training effects among the four train— ing groups; and (f) comparison of Hearing Handicap Scale self-ratings and effects of auditory training on speech discrimination. These factors are described in the following sections. Speech Discrimination Tests The selected tests of speech discrimination were: (a) CID w—22 lists 3 and 4,1 (b) Rhyme lists 1 and 2,2 and (c) Semi-Diagnostic forms A and B.3 The disc-record— ings of the W—22 lists were used, while tape-recordings of the Rhyme and Semi-Diagnostic tests were made for this study. (The test—items for each of the three discrimina— tion tests are given in Appendix E). lIra J. Hirsh et al., "Development of Materials for Speech Audiometry," Journal of Speech and Hearing Disorders, 17 (September, 1952), pp. 321-37. 2Grant Fairbanks, "Test of Phonemic Differentiation: The Rhyme Test," Journal of the Acoustical Society of America, 30 (July, 1958), pp. 596-600. 3Charles Hutton, E. Thayer Curry, and Mary Beth Arm- strong, "Semi—Diagnostic Test Materials for Aural Rehabilita- tion," Journal of Speech and Hearing Disorders, 24 (November, 1959), pp. 319-29. 80 The selection of these particular tests was some— what arbitrary; however, several criteria were employed: (1) the tests are published and thus easily available to both researchers and clincians; (2) the W—22 lists have received extensive research and clinical use; (3) the Semi-Diagnostic test was developed specifically for evaluating aural rehabilitation candidacy and progress, but has not been evaluated extensively; (4) the Rhyme Test similarly has not received extensive study with hearing loss populations,and is noteworthy because of its emphasis on consonant discrimination; (5) some information was available regarding correlation between the W—22 and Semi—Diagnostic tests; and (6) most important of all, for the purposes and conditions of this study, these tests were believed to have satisfactory validity and reliability for specifying the pre— and post—training discrimination performance of the present subjects. All subjects received the same test—items through counter—balanced alternate forms of each test. That is, half the subjects in each group received one form of each test pre—training and the other form post—training, while the remaining four subjects in each group received the reverse order of presentation. Each subject served as his own control, receiving both forms of all tests on the same day. All tests were administered at a 5 dB signal— to—noise ratio with the discrimination tests at 50 dB sound—field and the speech babble at 45 dB sound—field. :fiafifi ? '- .--.r 5.: Millage til-“:55- «Em-'75 m: sometime: ' '- amt.” snail 57-25 9...; {'7' :--.ni.Eb'.‘i-L_,.=~ 13.315 .. _ .-_.-...'.' --_'~..'.:.--.-.:‘«-1 Edda-2981.551”- '5 . - -- !: i-.-'-* _-€+2L4n:g£ffi;i: ‘5 us 3-H-ili‘r'3I'7L : .".-... cursi- 371}! .' '1.- - ' ..5‘. ..-.. _.:.._._ IA. 1 ._ a l 1 1 n 81 The subject's task for the W—22 test was to repeat vocally the monosyllable heard in the noise background, with the experimenter monitoring the responses. The Rhyme Test required a written response from the subject. He was given a response sheet of 50 word— stems with each stem preceded by a blank space in which he entered one letter to complete the spelling of the word (e.g., __ot, __ay, __op, __ake). The subject was told that he would hear 50 words and that he was to write in the appropriate letter as he heard each word. This test re— quired discrimination of consonants as they appear at the beginning of monosyllables. The test was thought to be an ideal one for the present subjects since their primary discrimination problem was with consonant sounds, partic— ularly in a noise background condition. The multiple—choice format was employed with the Semi—Diagnostic test. The subject's response sheet showed four words that differed only in vowels or consonants; his task was to circle the correct word. Fifty such foils, with fifty associated test items, constituted a single form of this test. In this study, two of the four W—22, two of the five Rhyme, and two of the four Semi—Diagnostic comparable lists were employed. ' MWQNWWH I II seitfieQEss add nnrntsingh mn"t sansqsn. £33361. u bcwiupet E-_s '5: a ;:g 1-. g - -. ..c;.., :.rn'gg ass. sf! . . .2'.. ' _- "‘ ' "J'J 'Ei'1fi'5' fifliwll . Lfissnnbugflf .Tfim ..§;9fl .::.n .' '1. ac- 82 AuditoryITrainingiMethods Out of a number of possible methods of training subjects, two were selected that seemed to be most per— tinent to the present objectives. One of the main interests concerned the effects of auditory training on the discrimination of speech that was delivered at the loudness—level of average conversational speech, defined here as 50 dB re audiometric zero. In addition, the discrimination task involved listening to speech in a speech babble noise background of 45 dB re audiometric zero. These considerations led to the selection of two training "methods" that seemed to have relevance to the required discrimination task. Two methods were selected for evaluation: (a) signal— to—noise ratio constant with the overall intensity-level decreased from conversationally—loud to conversationally— soft speech as a function of five training sessions, and (b) signal-to—noise ratio varied with the signal held constant at the intensity—level of conversationally— average speech and the noise level increased as a function of the five training sessions. Each subject participated in five training sessions over an approximate three—hour period (not including pre— and post—training discrimination testing). Those sixteen subjects who received the S/N—constant method listened under the following five conditions as training progressed: Session 1: signal at 57 dB, speech babble at 52 dB Session 2: signal at 54 dB, speech babble at 49 dB Session 3: signal at 51 dB, speech babble at 46 dB Session 4: signal at 48 dB, speech babble at 43 dB Session 5: signal at 45 dB, speech babble at 40 dB Those sixteen subjects who received the S/N—varied method listened under the following five conditions as training progressed: Session 1: signal at 50 dB, speech babble at 40 dB Session 2: signal at 50 dB, speech babble at 42 dB Session 3: signal at 50 dB, speech babble at 44 dB Session 4: signal at 50 dB, speech babble at 46 dB Session 5: signal at 50 dB, speech babble at 48 dB As can be observed above, the first method maintained a 5 dB S/N with overall intensity decreased during training, and the second method maintained a 50 dB signal throughout while S/N was varied from 10 to 8 to 6 to 4 to 2 dB during successive sessions. Therefore, both methods differed in intensity—levels and signal—to-noise ratios but were similar in that the difficulty of the listening task increased as training progressed. It was believed that, all thing being equal, success early in training with a simpler task served to motivate the subjects and to prepare them to accomplish more as listening conditions became more difficult. It was presumed further that these accomplishments would reflect themselves in post—training discrimination,and that if differences actually existed between the S/N methods, these differences would be indicated by a statistically significant difference between the mean gains for each method. .Il 55,1 if! .33. (fine. -‘1 g I '3‘ .sh.ni is pffidfio firsswa .55 i! 53. -..--.r:~.1 -.-«--.I_-r.-.\--:i'- '. z. ' """ ”"WQHMI . . 's-t"':.'-:' .. -. . ' ..-'i=sr'F-“' '35-'15“ " :fiéfi! rm": E's-31‘1" I 'II 4:. -'-* .. _.._'.'" . 153E. 84 Auditory Training Materials Since most of the subjects in the present popula— tion experience primary difficulty in consonant discrim— ination, training materials should be of the type where speech discrimination is dependent on consonant identifi— cation rather than on contextual influences. This was one reason why monosyllables-—instead of multisyllabic or sentence-type materials——were employed in this study. It was decided to include pypg of practice material as a factor of interest. Multiple—choice (closed set) materials and write—down (open set) materials were selected. The subject's task was considerably different, depending on which type of material was utilized. The multiple— choice task merely required the ability to select one out of only a few possible alternatives, each of which was con— veniently listed in front of him. The write-down task, however, required that the subject select one out of numerous choices. This question of multiple—choice versus write—down training materials seemed to be a basic considera- tion in the planning and evaluation of auditory training. The interest in this study was to evaluate objectively this question and to determine if either type of material was most effective in auditory training. By the line of reasoning suggested above, interest was drawn away from what seemed to be a less fundamental question at this time; that is, evaluation of specific material published by a particular author. Though this topic 85. has some importance, most collections of training materials were not derived from auditory training re— search,but were either borrowed from other sources or selected arbitrarily. In addition, it was difficult to note any real differences among most of these materials. For these reasons then, it was decided not to evaluate specific materialg but to select those published materials that met the multiple—choice or write— down criteria. Monosyllabic write—down material.--The ten 50-word CNC lists published by Peterson and Lehiste were selected as write—down material.1 The lists were compiled by the authors for the purpose of intelligibility testing; however, these lists were employed in this study as training items. They have a simple phonemic composition and are among the most common monosyllables in spoken American English. The sixteen subjects who used the write-down training material were required to listen as each monosyllable was presented in speech babble and to write the word they believed was spoken. The interval between items was approxi- mately seven seconds. Subjects were instructed not to be too concerned about spelling,but to spell their responses well enough so that later scoring by the subjects would be 1Gordon E. Peterson and Ilse Lehiste, "Revised CNC Lists for Auditory Tests," Journal of Speech and Hearing Disorders, 27 (February, 1962), pp. 62—70. 86 possible. Each subject scored his own responses after each of the five practice sessions. This procedure served to encourage the subjects' active participation in the training and to give them periodic feedback regarding their performance. One—hundred words were included in each of the five practice sessions, making a total of five-hundred stimulus items during the entire training procedure; (See Appendix F for a listing of the write-down training material). Multipleechoice material.--Two sets of multiple- choice items were employed in this study. The first set was Larson's discrimination drill material,1 and the second set was selected from Kelly's manual (lists VA to VJ).2 There were two response alternatives on the Larsen lists and three on the Kelly lists. This means that the subjects selected from either two or three possible choices, depending on which particular lists were being presented. The pairs and the triplets differed only in consonant sounds (e.g., filed,_ppild, 311d). Accuracy with this ma— terial required a high level of emphasis on consonant dis- crimination. '3'.- -:.-l « lLaila Larsen, Lists published in Hearing and Deafness, ed. Hallowell Davis and S. Richard Silverman (rev. ed.; New York: Holt, Rinehart and Winston, Inc., 1963), pp. 542—44. 2J. C. Kelly, Clinician's Handbook for Auditory Train— in (Dubuque, Iowa: William C. Brown, Inc., 1953), pp. 7 —113. f. 'I I}. I I I - h'sv'zss swbhugaiflé“ t. '_ :.'.'t 111 no: 391591115: 9171355 "I‘M-'nsfil'iir-T pajamas: sisal-.591 2:15-51 :21; mum-t évbg-G$-'JI_ .S'J . ':= =" -.'~ " - -“'..‘Air'.' .....'.I"2["'.-' :.Lnfld“tfl9-'" .. 'arn..2.'::':-:.1a fibtéw - ' Eli-Jim? Ems-15f -,:- :'-' : ." -.= '!.-'..-1"’£"" 817 Each of the five practice sessions with the multiple- choice material included 48 words from the Kelly lists and 52 words from the Larsen lists. (See Appendix G for a listing of the multiple-choice training material). Total training consisted of 240 words from the Kelly lists and 260 words from the Larsen lists. Therefore, the sixteen subjects who used the multiple-choice material responded to the same number of words during training as the sixteen subjects who used the write-down material. The subjects who used the multiple-choice training material scored their own responses after each practice session. Again, active participation and periodic feed- back regarding performance were the primary purposes for this procedure. Poussirsésissfiroups Each group of eight subjects responded under one of four training conditions: Group l--write—down material with S/N varied, Group 2--multiple—choice material with S/N varied, Group 3——write-down material with S/N constant, and Group 4——multiple—choice material with S/N constant. These specific treatment combinations were assigned ran- domly to the four groups. It was believed that comparisons among mean responses of these four independent groups would answer one of the questions posed in this study. That is, do four combina- tions of auditory training methods and materials differ his war! asstutmf . 5.101 fi-zibnsfiafl 553) {stati- iisrwnsai with.s:£_stlsnnnsiq;§imp, ." l"; ;.l' .7‘.-."."." ! '----‘:' (Apt. 5!: 11936181“. I =‘ ;.$3'ens'5lqt‘ -..-i=..:. .sm 5.3939155}! '! 5:13 Li..L.-. .' 51.1 03 .. . L1 \' 'l .'. - ..I F: 88 significantly in their effects on speech discrimination performance. It was thought that the answer to this ques— tion might have direct clinical applicabilitm and also might be of theoretical importance. Hearing Handicap Scale One of the questions posed in this study concerned the possible relationship between self—ratings of social, difficulty (i.e., hearing handicap) and speech discrimina— tion performance in auditory training. It was of interest to know whether subjects who rated themselves high in de— gree of hearing handicap showed greater or lesser response to training than subjects who rated themselves low. Self— concept and/or degree of motivation may be different between these two types of subjects,and these possible differences might reflect themselves through responses in auditory training. To answer the above question, a recently published scale for self—rating of degree of hearing difficulty was administered to the present subjects immediately before the auditory training procedures. The Hearing Handicap Scale was developed and described by High, Faribanks, and Glorig.l The final form of the Scale consists of two 20— lWallace S. High, Grant Fairbanks, and Aram Glorig, "Scale for Self-Assessment of Hearing Handicap," Journal of Speech and Hearing Disorders, 29 (August, 1964), pp. 215430. 89 item parallel forms, A and B. Form A was employed in the present study. (See Appendix D where this form is shown). Items on the Hearing Handicap Scale refer to common auditory experiences or situations, particularly those encountered in an urban environment. It requires that subjects indicate on a 5-point scale of relative frequency how often they experience difficulty in each of the speci— fied listening situations. Using a sample of fifty hearing—impaired men and women, the authors found an in- ternal reliability coefficient of 0.96.1 The response . scaling is designed so that the higher the numerical value, the greater the degree of reported hearing difficulty. Additional comments regarding the Hearing Handicap Scale and the topic of hearing handicap in general are found in Chapter III, pages 55-59, Recording of Test and Training Materials Three master tape—recordings were required: (a) a Discrimination Test Tape, (b) a Multiple-Choice Training Tape, and (c) a Write-Down Training Tape. The commercial disc-recording of the W—22 test was reproduced directly by the audiometer's phonograph during pre— and post—train— ing discrimination testing. The Rhyme and Semi—Diagnostic 1Ibid., p. 222. 90 tests, however, as well as the training materials, were pre-recorded using the experimenter's voice. Recording Procedures.-—The experimenter was seated in a quiet room approximately six-inches from a high- fidelity microphone (Electro-Voice, Model 654). The speech signal transduced by this microphone was fed directly into Channel A of a dual-channel tape—recorder (Ampex, Model 602) located in an adjacent room. This recorder was situated in front of a two—way window so that the experi- menter could monitor his production of a carrier phrase preceding each stimulus word. An electronics technician—— located in the adjacent room with the recording apparatus—— monitored the overall recording procedures. He also was responsible for monitoring the record—level of the speech babbleIthat was being delivered simultaneously from a second tape—recorder (Ampex, Model 601) into Channel B of the master recorder. All recordings were made at 7 % ipS using high-fidelity tape (Scotch Low Noise 203). The carrier phrase for each stimulus word was one of the following: "The first word is ____." "The next word is ." "The last word is ." The experimenter attempted to maintain a 0 dB VU—meter de— flection during his production of the two words underlined above. The stimulus word was produced in a natural manner without any deliberate effort to "peak" at 0 dB. Such 91 an effort would have distorted both the "real—life" relative intensities among stimulus words and the relative intensities of the consonant-vowel or vowel—consonant relationships within words. An assumption underlying this procedure was that the cumulative average intensity of the stimulus words and phonemes approximated 0 dB on the VU—meter. If this assumption were not valid, comments regarding specific signal—to-noise ratios——during either testing or training—-would be inaccurate. A regular time interval was maintained between carrier phrases, 5-seconds for the test words and 8- seconds for the training words. The latter interval served two purposes. First, those subjects using the write-down training material needed sufficient time to write their responses. Second, duration of training among all thirty—two subjects was not a factor of interest in this study,and thus was kept constant for all subjects, whether they wrote or circled their responses. Discrimination Test Tape.——The Test Tape contained the following order of test stimuli, although only half of the subjects received this particular ordering: a. No signal on track A (The W-22 test was delivered directly from the audiometer's phonograph). Six—minutes of speech babble on track B. b. Rhyme Test 1 on track A. Speech babble on track B. c. Semi—Diagnostic form A on track A. Speech babble on track B. mammals-[wow wc- LsflfiV-iflafiOl-ROD _'-- -‘-'--:5-.1'.I-':.' 'r-zi-ElquEa... u". .535ch 111.1313! - aa-Lumiae' .:5.‘_T§n’uv-' m _ . - ': : '..':.Ti2"153'” gniiaer 92 d. Rhyme Test 2 on track A. Speech babble on track B. e. Semi-Diagnostic form B on track A. Speech babble on track B. The dual—track recording procedure provided play- back of either or both tracks from the audiometer's stereo— playback recorder. Intensity of each track was controlled independently by means of separate attenuators on the audiometer console. Therefore, the tests could be adminis- tered to the subjects in any desired order with average intensities and signal-to—noise ratios under experimental control. Separate VU—meters allowed direct monitoring of input-calibration throughout the testing procedures. Training Tapes.--Two master tapes were recorded, a Multiple—Choice Training Tape and a Write-Down Training Tape. The Larsen and Kelly multiple—choice items and the CNC write—down items were pre—recorded in a manner and with equipment identical to that used for recording the dis— crimination tests. The speech stimulus words and the speech babble were recorded simultaneously on separate tape—tracks of the dual-channel recorder. Either tape could be played back during the training sessions with the input of either track on either tape under experimental control. Training of Subjects Subjects were seated in a commercial test—room (Suttle) for both discrimination testing and training. Training materials were delivered at calibrated intensity—levels #5155! =.nhtmsfiqgg£ - - 3' ' - .."3 -'.2--.'ZL';J‘I 93 through a single sound-field speaker (Electro—Voice, Model SP12). Input to this speaker was controlled by the experimenter, who was seated at the audiometer console (Allison, Model 22) in a room adjacent to the training room. Two subjects from each group participated simul- taneously in training and were seated facing toward the 'sound-field speaker with head-level approximately 56 inches from the speaker. The training arrangement is shown in Figure 2. Two—way communication and observation was possible during the training sessions. Subjects participated in the following format of pre—test, training, and post—test procedures during the approximate three and one-half hour duration of the test- ing and training: a. The Hearing Handicap Scale was completed by the subject. b. Pre-training speech discrimination tests were administered. c. After a 5-minute rest period, instructions were given regarding subsequent training sessions. d. Five 25—minute practice sessions were given with lO—minute rest periods between sessions for feed- back regarding performance. A longer rest period (20-minutes) was provided between the 3rd and 4th sessions. chair 56H 561! I two-way window 1_ chair HOOD- speaker audiometer Figure 2.-—Schematic representation of physical arrangements during auditory testing and training procedures. 95 e. Following the 5th training session, and after a final rest period and feedback condition, post-training speech discrimination tests were administered. (bneral instructions to subjects were as follows: You are going to listen to single words delivered through this loudspeaker. There also will be noise along with the words, so you must listen carefully and attempt to ignore the noise. There will be five 25-minute practice sessions during the next few hours. Your task will be the same for each session: listen to the words and write-down (circle) what you hear. Here are sheets of paper on which you are to write (circle) the words. Each session will be slightly more difficult than the previous one, so you will have to listen more closely. Do not worry if you miss some words or do not understand others. There is no penalty here for mistakes. These are practice sessions designed to give you a chance to improve your listening. After each 25-minute session, we will have a rest break and you can step into the reception area to see how well you are doing. You will check your own work as I tell and/or show you the words which were spoken. After the third session, there will be a longer rest break and you can relax and have coffee or coke. You then will complete the remaining two practice sessions, there will be a final rest break, and three tests similar to the ones you just had will be given to see how much your listening has improved. Feel free to ask questions at any time. Do you have any now? Statistical Design This investigation used a 2 x 2 x 3 (methods x ma- terials x tests) analysis of variance, factorial design suggested by Winer.l The last factor, discrimination tests, 1B. J. Winer, Statistical Principles in Experimental Design (New York: McGraw-Hill Book Co., Inc., 1962), pp. 337—49. ti ,nuidiéfiéu'xaAUhasi has tfifi“ hiss: noi3sdim.” sit fiasaqn afllflllain . b51511}. Tfih 2L.35Ifii o: incidnu13§fii . ':5 uni' 11 .2115 'qu-- ' =.;; 75”? HQHHWHOITSE . - 5:95.713 I ...... Inns T '--: 96 had repeated observations (i.e., all subjects received all three discrimination tests). Statistical comparison among training groups involved a one-way analysis of variance. Dependent measures for all of the above statistical pro— cedures were difference—scores between pre— and post— training speech discrimination performance. Related or paired p—tests were used to evaluate the pre— versus post- training comparisons. Descriptive statistics were em- ployed where necessary to specify particular aspects of subject performance. These statistical procedures provided an objective basis for answering the questions posed by this study. Summary Thirty-two subjects were selected for this study. Each potential subject was seen first for preliminary testing, at which time the following measures were re— corded: pure tone thresholds, speech reception thresh- olds, speech discrimination scores in quiet (PB—Max), speech discrimination score in noise (sound—field at a 5 dB S/N), and WAIS Vocabulary raw score. On the basis of these measures, together with case history information, subjects were assigned to one of four training groups. Means, variances, medians, and ranges of recorded data were matched systematically among groups in order to support a pre—training "equivalency" assumption. ---.J'1q .1251: 57:1...-. Enema flakiéahihfl f eonsiuhv it tfsmisns E -. Zun- 97 Two subjects participated simultaneously in training in a commercial sound—treated room. Practice materials were delivered to the sound—field in a variable speech babble background. Feedback regarding performance and rest breaks were provided at 25—minute intervals during the approximate three hours of training. Speech discrimination tests were administered in the sound—field pre— and post— training at a 50 dB (re audiometric zero) intensity level in a 45 dB speech babble background. Discrimination tests included alternate forms of W—22, Rhyme, and Semi—Diagnostic tests, administered in this order,but with counter—balanced presentation of the alternate forms. The commercial disc—recording of the W—22 test was reproduced directly during pre— and post- training testing. The Rhyme and Semi-Diagnostic tests, as well as the training materials, were tape—recorded using the experimenter's voice. Each training group of eight subjects responded under one of four auditory training conditions: Group 1—— write—down material with S/N—varied, Group 2-—multiple— choice material with S/N—varied, Group 3——write—down material with S/N—constant, and Group 4--multiple-choice material with S/N—constant. Instrumentation provided for administration of discrimination tests and training materials in any desired order,and with average intensities and signal—to-noise 98 ratios under experimental control. The speech stimulus words and the speech babble were pre—recorded simultan— eously on separate tape—tracks using a dual-channel recorder. Primary factors of interest included discrimination changes among tests, among training groups, between methods, and between materials. An introspective report of hearing difficulty--the Hearing Handicap Scale ——a1so was adminis- tered for evaluation in relation to obtained performance measures . CHAPTER IV RESULTS AND DISCUSSION This chapter presents the major results obtained by the procedures described in Chapter III. Thirty-two subjects were assigned systematically to one of four groups in such a way that means, variances, ranges, and medians of extraneous variables were matched among groups. This was done in order to support an equivalency-among— groups assumption prior to the training. Somewhat greater confidence then can be given to conclusions regarding the effects of specific factors under experimental control. The results of these matching procedures are shown in Table 1, Chapter III, page 78, The subjects participated in three and one—half hours of practice responding (write-down or multiple—choice) to stimulus words delivered to the sound—field in a variable speech babble background. Each group of eight subjects responded under one of four possible auditory training conditions. Pre- and post— training speech discrimination tests were administered in the sound—field in a speech babble background as a means of evaluating differential effects of the training procedures. 99 100 Following presentation of statistical results and associated conclusions, these results are interpreted and discussed in terms of the original questions posed in this investigation. Obtained Performance Measures In Table l is presented a summary of the mean per- centage effects of the present auditory training proce— dures on the subjects' speech discrimination performance. TABLE 1.--Mean percentage effects of auditory train- ing on speech discrimination per— formance. Multiple—Choice 3’ Write-Down S/N-C S N— S/N—C S/N-V Tests (Group 4) (Group 2) (Group 3) (Group 1) W—22 4.0 8.7 9.3 8.8 Rhyme 2.8 6.0 2.8 2.3 Sem'i—Diag. —o.5 4.8 2.3 —o.5 The dependent variable, represented by the means in Table l, was the difference—score for each subject. It , was calculated by subtracting pre-training from post— training discrimination test performance. Alternate forms of each test were used with a counter-balanced order of presentation among subjects within each group. (Pre- II has Sesssqiu3§t iii smfsdiit 1. 'é hi fishnq aiLIJEFLfi fafifjtwfi OH?”%0' -.noEt I'II' . - .'-_ 25-55.:- ' visas 1;; 3333555 ml— safihb 101 training, post—training, and difference-scores for each subject on each test are given in Appendix H). The mean percentage of discrimination change for any group on any test under any combination of auditory train— ing methods and materials may be observed in Table 1. As can be noted, ten out of twelve indices of performance change showed positive effects; that is, subjects improved in discrimination test performance. The remaining two indices—-both recorded on the Semi—Diagnostic test—— showed negative effects in that subject performance de- creased. It should be remembered, however, that means in each column in Table 1 are not independent (i.e., subjects in each group received all three discrimination tests). Therefore, even though slight negative effects were re— corded for two groups on the Semi-Diagnostic test, these same two groups showed positive effects on both the W—22 test and the Rhyme test. The data indicate then that the auditory training procedures resulted generally in in— creased discrimination test performance. Pre— vs Post—Training Performance In Table 2 are shown the mean difference—scores on each test and on total discrimination performance. The thirty—two subjects improved 7.7% in W—22 discrimination, 3.5% in Rhyme discrimination, and 1.5% in Semi—Diagnostic discrimination, for an overall mean improvement of 4.2% in total discrimination. ' " Iféi-"fiim" ~51. 1c? sauna: 4"L635.mlfi$?tb lo' ' .51. ,":f*'bi o 50:5nn’irnc gas 193nm 555:5 _ -._ um slat-153m: Ems .. . I. I‘m; 1 - . ll. '2: I ”91°“ '5.“ - ' P -~ "fifii ' . ,1“:ng 5?. para-#5 '-:.-. -' -~: :: :1 2b .1 It ' l. . l l_ } I":1 102 Calculation of p—tests for paired observationsl indicated that: (a) the W—22 increase was significant be— yond the 0.0l-1evel, (b) the Rhyme increase was signifi- cant beyond the 0.05—1evel, (c) the Semi—Diagnostic in- crease was non—significant at the 0.05—level, and_(d) the total discrimination increase was significant beyond the 0.05—level. TABLE 2.——Mean percentage effects of auditory training as indicated by three dis— crimination tests. W—22 Rhyme Test Semi Diagnostic Total 7.7a 3.5b 1.5 4.2b aSignificant beyond the 0.0l—1eve1, df = 31. Significant beyond the 0.05—1evel, df = 31. Assuming that test—retest carry—over effects were negligible,2 these results indicate that the auditory traihing procedures brought about significant increase in speech discrimination as measured by total, by W—22, and by Rhyme test performance, but that these significant 1William L. Hays, Statistics for Psychologists (New York: Holt,Rinehart and Winston, Inc., 1963), pp. 333—35. 2Egan's data @Articulation Testing Methods,” cited in Chapter III) suggests that listener performance on discrimina— tion tests is relatively stable during six one—hour testing sessions on the same day; in fact, scores tended to decrease during each of the two three—hour testing segments. f " {ad apes ‘ EH"; sass-ass: smq‘flfi 52:5 (.5) ...I .an' . ~i“lngia a w -:2 :‘JSQLEF"—’Pfifl :3: .5) ggpjg;_so,g. ':I: _- ;"::'v.: lib .[Ifiifl_ I0 ‘ *W 103 effects were not similarly reflected by performance on the Semi-Diagnostic test. One possible explanation of these results is that the Semi—Diagnostic test may be a fairly stable measure of speech discrimination,and there— fore resistive to the effects of auditory training, at least with a mild hearing loss population. Perhaps the Wu22 and Rhyme tests are structured in such a way that they are more sensitive to discrimination changes than the Semi-Diagnostic test. Analysis of Variance Of primary interest in this study were the compari- sons between and among levels of each of the main factors or independent variables. Therefore, two auditory train— ing methods, two types of auditory training materials, three discrimination tests, and four groups of subjects were comparisons of interest. The statistic involved was mean change in speech discrimination performance (i.e., mean difference—score). For multiple statistical tests of means, a factorial design with associated analysis of variance procedures provided for the above comparisons. Normality of Distribution One of the assumptions for the fixed-effects model analysis of variance is that the distribution of errors for any treatment population is normal. According to Hays, this is equivalent to the assumption that each population “I. ..'.. . -_' :- I.. _ I. '19 air w T s 3d gsm 1353 vittonaslfietmwfi eds'a '.fi$P' has,nr":u-uiW-uir '13?§E We hwhnaom -'-. - .-‘:. ~' ' .6.. .- ' '. 2:1- ':.-:1..' L:.3'5’;rl' r.- {.51 "'1 5: rid-1w :1 -: _-'-."E 121.315 525-! . 1'1. ' "1 I 'r:= a zsdf 1 . A - -- "2 52-11? .5, 104 has a normal distribution of dependent scores.l He stated that, other things being equal, inferences made about means that are valid in the case of normal populations are also valid when the forms of the population distributions depart considerably from normal, provided that the number of subjects in each sample is relatively large. In the present study, dependent scores were the differences between pre- and post-training discrimination test performance. For descriptive purposes, the standard deviations within each set of dependent measures are given in Table 3. As can be noted, calculated standard devia— tions range in size from 5.1% to 10.2%. TABLE 3.——Standard deviations for percentage effects of auditory training on speech discrimina- tion performance. Multiple—Choice Write-Down S/N-C S/N-V S/N-C S/N-V Tests (Group 4) (Group 2) (Group 3) (Group 1) W-22 8.5 7.9 5.1 6.3 Rhyme 9.5 10.2 6.1 6.9 Semi-Diag. 5.1 6.3 6.7 6.1 lIbid., pp. 378—79. ...—__— 1 - _t‘, :_ no ‘-‘ '-' hwy-q '. Err-r6! "$.63 ”953:: “#H'J‘ . 111' , .IJ-EL125 1 . 2;.Icqou nn= in'amiofi 5&5“. ' -';. L. a .. 'u.'n" 'T“UI§1:539P .... .i 5355' 105 Distribution of raw scores within each group also was inspected. This inspection did not reveal any gross departures from normality, suggesting normal distribution of errors in the treatment populations. It was assumed, therefore, that normal distributions existed within the populations of interest,:and that if departures from nor— mality actually did exist, they were not sufficient to invalidate analysis of variance procedures. Homogeneity of Variance A second assumption of the present analysis of variance was that error variance among treatment popula— tions was the same; that is, no statistically significant differences existed among these error variances. Hays stated that, other things being equal, this assumption of homogenous variances can be violated without serious risk, provided that the number of cases in each sample is the same.1 There were equal numbers of subjects in each of the present treatment groups, supporting a homogeneity assump— tion. However, as a statistical test of this assumption, comparisons among error variances (estimated from sample values) were made. The estimated error variances are shown in Table 4. Two tests were used to evaluate the homogeneity of variance hypothesis.2 The first, Hartley's F—max test—— lIbid. 2B. J. Winer, Statistical Principles in Experimental Desi n (New York: McGraw—Hill Book, Co., Inc., 1962), pp. 92-9 . _- 6m? , sea-1'3 ' vita-"15353:! 33.21%" Km ....j-Ir.J-:'t-Ha;f:a- .s.r:*.-'-.--.: guises-33:11: “Eat ' ' ..-.- ..,, .'a "".""-', jug-35:13:55“: fl “.3qu ‘5 i=5.»- . . _ - 1»... 7' ‘le'- : I ' .1 ."..-.- "-'.v-1-_'_.'! 5115113 r _..I. -.-' ' ' . “..I 106 TABLE 4.-—Estimates of error variances within treat- ment populations. Multiple—Choice Write—Down S/N-C S/N-V S/N-C S/N-V Tests (Group 4) (Group 2) (Group 3) (Group 1) W—22 83.4 70.8 29.6 45.6 Rhyme 103.9 118.9 42.2 54.2 Semi—Diag. 29.4 45.1 50.8 42.0 simply the largest variance estimate divided by the smallest—« gave a value of 4.04 with a critical value of 24.0 required for rejection. The second, Cochran's C test——the largest variance estimate divided by the sum of the estimates—— resulting in a value of 0.166 with a critical value of 0.31. Thus, for a 0.01-1eve1 test——number of variances equal to l2,degrees of freedom equal to 7——the hypothesis of homogeneity of variance was not rejected by either test. Statistical Independence The fixed—effects model also requires-—according to Hays—— an assumption of independence of errors among the dependent measures.1 The present study was conducted in a manner de— signed to satisfy this requirement. Measures were obtained in such a way that each subjects's score was independent lHays, o . cit. 107 from any other subject's score. It was noted that this assumption of independence was not necessary for the com- parison of discrimination tests. Even though each sub- ject received all three discrimination tests, the selected design was one in which the relatedness or dependency among the three tests was built into the calculation of mean square values. F-Tests and Individual Comparisons The present analysis of variance procedure involved a three—factor fixed—effects design with repeated measures on one of the factors.1 The main factors were auditory training methods, auditory training materials, and dis— crimination tests, the latter having repeated measures on the same subjects. A summary of the results of this analy— sis is given in Table 5. As can be noted, the only factor TABLE 5.—-Summary of analysis of variance comparing dif— ferences between methods, between materials, and among tests. Source of Variation df Mean Square F Between Sub'ects 3i MaterialsiA) l 1.04 Methods(B) l 63.38 A x B 1 187.04 2.26(NS) error (b) 28 82.9 Within Subjects 64 Tests C 2 326.0 6.86(S)* A x C 2 48.67 B X C 2 1.5 A x B x C 2 8.09 error (w) 56 47-49 *Significant beyond the 0.01.1eve1. lWiner, op. cit., pp. 337-349. 108 showing statistical significance was the tests factor. Levels within methods, levels within materials, and in— teractions were non—significant. Auditory Training Methodsv—Mean discrimination changes recorded for each of the two groups of sixteen subjects receiving either S/N-Varied or S/N-Constant methods are shown in Table 6. Mean discrimination change was 4.995% for the S/N—Varied condition and 3.415% for the S/N-Constant condition. The mean difference of 1.58% between these two auditory training methods was non—significant, as indicated by analysis of variance. TABLE 6.-—Comparative mean percentage effects of auditory training methods (signal-to—noise ratios) on speech dis— crimination performance. S/N-Varied S/N—Constant Difference 4.995 3.415 1.58* *Non-significant. These two methods of training subjects apparently have similar significant effects on speech discrimination performance. This conclusion can only be generalized to the population from which present subjects were selected (i.e., adult, mildly hard of hearing individuals with the particular characteristics specified earlier). 109 Auditory Training Materials.--Mean discrimination changes recorded for each of the two groups of sixteen subjects using either multiple-choice (closed set) or write-down (open set) materials are shown in Table 7. Mean discrimination change was 4.285% for the multiple- choice condition and 4.125% for the write-down condition. The mean difference of 0.16% between these two types of auditory training materials was non-significant, as tested by analysis of variance. TABLE 7.—-Comparative mean percentage effects of auditory training materials (multiple—choice vs write—down response) on speech discrimination performance. *5 -3 1 - 1 1 - ..1 Multiple-Choice Write-Down Difference 4.285 4.125 0.16% '-*Non-significant. I These two types of training materials seemingly have similar significant effects on speech discrimination perfor- mance. It should be remembered that this conclusion—~as well as the previous one regarding methods——represents a summary statement of overall effects. Means for materials and means for methods were obtained by summing across levels of the other factor. In other words, the materials do not differ when both are used in conjunction with both methods. Similarly, the methods do not differ when both are employed with both types of material. 110 Speech Discrimination Tests.——The analysis of variance indicated that the W—22, Rhyme, and Semi—Diagnostic tests differed significantly (0.01—1eve1) in reflecting the effects of auditory training. Mean percentage discrimination changes recorded for the thirty—two subjects on each test were 7.7% on the W—22 test, 3.5% on the Rhyme Test, and 1.5% on the Semi-Diagnostic test. Differences between each pair of tests are shown in Table 8. The use of a priori indivi— dual comparison procedures (k = 3, df = 56)1 showed critical differences to be 4.14 (0.05-leve1) and 5.21 (0.01—level). TABLE 8.-—Percentage differences between pairs of speech discrimination tests reflecting effects of auditory training. W—22 Rhyme Test Semi-Diagnostic w—22 —— 4.2a 6.2b Rhyme —— -— 2.0 aSignificant beyond the 0.05-leve1. bSignificant beyond the 0.01—leve1. These results indicate, with reference to speech dis- crimination changes brought about by auditory training, that: (a) the W—22 test differs significantly from both the Rhyme test (0.05—level) and the Semi—Diagnostic test (0.01—1eve1), and (b) the Rhyme test and the Semi—Diagnostic test do not differ significantly. Pearson correlations were calculated lWiner, ibid,. p. 85 111 between each pair of tests and are shown in Table 9. These coefficients were 0.57 between W—22 and Rhyme, 0.01 between W—22 and Semi—Diagnostic, and 0.02 between Rhyme and Semis Diagnostic tests. These indicate that some linear predic— tion between W—22 and Rhyme test performance is possible when describing the changes that occur in auditory training with a mild hearing loss population. This prediction is not perfect and considerable error exists. Linear predictive power between the Semi—Diagnostic test and either of the other two tests was not demonstrated in this study TABLE 9.--Pearson product-moment correlations between pairs of speech discrimination tests reflecting effects of auditory training. w—22 Rhyme Test Semi—Diagnostic W—22 -— 0.57 0.01 Rhyme —— —— 0.02 It should be remembered that statistical comparisons (pages 101-103) of pre— versus post—training performance indicated significant positive change in both W—22 and Rhyme test discrimination, and non—significant change in Semi— Diagnostic performance. Training Groups.——A comparison of interest was the differential effects among four training groups of combinations of methods and materials on speech discrimination. Mean percentage change for each group of eight subjects is shown in Table 10. meow-396.1650 ' sonny: IM‘ 4’33} has email". nebular..- 3.0.0 1'3“...th wadfi’: Wmun lad: ~+nsihfll EHSHT .eiaadt I'.2.'.I'1""- --'e_-_'1.'*'.' ..‘ns. BIS-5‘5. . . I . .. 5' f! I -_- -.i"' firildt'lbfiflb w 'mfi-MPIH £“H:1w !‘ fl 3fiwf59q ' '. .' . '_'.'\:'-_-J 1 1 1 112 TABLE 10.--Comparative mean percentage effects of four combinations of auditory training methods and materials on speech discrimination performance. S/N-Varied S/N-Constant Multiple-Choice 6.5 2.08 (Group 2) (Group 4) Write—Down 3.5 4.75 (Group 1) (Group 3) Specific mean values were as follows: (a) multiple- choice and S/N—Varied: 6.5%, (b) multiple—choice and S/N- Constant: 2.08%, (c) write—down and S/N—Varied: 3.5%, and (d) write—down and S/N-Constant: 4.75%. A one—way analysis of variance1 showed that the four means differed significantly beyond the 0.05—1eve1. A summary of the analysis is given in Table 11. TABLE 11.——Summary of analysis of variance comparing differences among training groups. Source of Variation df Mean Square F Between Subjects 3 84.5 3.89(S)* Within Subjects 28 21.70 Total 31 *Significant beyond the 0.05—1eve1. lIbid., pp. 48—56. "Hl‘flafi'v I . 3;, 9'1 {59.15.13 ".5115” 113 Differences between each pair of training com— binations are shown in Table 12. The use of a priori individual comparison procedures (k = 4, df = 28)1 showed a critical difference to be a 2.74 at the 0.05—1eve1. The difference of 2.67 between Group 3 and Group 4, al— though non-significant by strict statistical definition, was given a liberal interpretation in this study. TABLE l2.——Percentage differences between pairs of treatment means. Group 2 Group 3 Group 1 Group 4 Group 2 -— 1.75 3.0a 4.42a Group 3 —— —— 1.25 2.67b Group 1 —— —— —— 1.42 ’ aSignificant beyond the 0.05—1eve1. bSignificant near the 0.05-1evel. The information in Table 12 may be summarized schema~ tically as follows: Group 2 Group 3 Group 1 Group 4 6.5 4.75 3.5 2.08 Those group means underlined by a common line did not differ significantly from each other; group means not underlined by a common line did differ from each other (beyond or near the 0.05—level). l1bid., p. 85. hum-hi“ “3’ ' - 'I'CBS . 15, ‘13. a )1}- 33513650914. . ..--u.; an; r 56.: 5 ad on consislltbuflfipij .. . -' ‘ ‘r‘ _ ----- 1"‘5‘ -'-' '\'.".7 '1: “J _ . . "ql=ffiirgis—noa' I 45613 any £.I 114 These results suggest the following rank—ordering of the combinations of auditory training methods and materials: (a) Multiple-Choice, S/N-Varied (gain: 6.5%) (b) Write—Down, S/N-Constant (gain: 4.75%) (c) Write-Down, S/N-Varied (gain: 3.5%) (d) Multiple—Choice, S/N—Constant (gain: 2.08%) Since (a) and (b), (b) and (c), and (c) and (d) do not differ significantly, it seems as though the combination (a) (b) would be the auditory training procedures of choice, with somewhat greater emphasis given to (a). If multiple— choice items are used in training, the S/N—Varied method should be the method of choice. If write—down items are used, either S/N—Constant or S/N—Varied methods apparently can be employed for similar training results. These con— clusions may only be applicable to the mildly hard of hear— ing population from which present subjects were selected. Descriptive Statistics In addition to the inferential statistics employed in previous comparisons, several descriptive statistics were calculated. This was done in an effort to gain additional information regarding factors that might have influenced present results,and that may be important to future clinical activity and research in auditory training. Type of train— ing, the Hearing Handicap Scale, age, speech reception threshold, and intelligence are discussed in the following sections. 1-‘1‘7 . 3..- 5.1412,! tr-F-Ihfri-wai'rg-i-U" -!'I ' '- '1'Fil: I . .f_— 115 Type of Training vs Type of Test Response A trend was noted in subject performance on the three pypgg of tests. Those subjects trained.®n*closed set (multiple—choice) materials improved more on the closed set test, the Semi—Diagnostic, than on the open set test, the W—22. The opposite tendency was noted for those subjects trained on open set (write—down) materials. This overall trend is more clearly shown as follows: Subjects Trained Subjects Trained with Closed Set (n=16) with Open Set (n=16) w—22 6.37% 9.0 Rhyme 4.37 2.5 Semi—Diag. 2.25 0.87 On the open set test (W-22), subjects trained with open set materials (write—down) showed 9.0% improvement while subjects trained with closed set materials (multiple-choice) showed only 6.37% improvement. Conversely, on the closed set test (Semi—Diagnostic), subjects trained with closed set materials showed 2.25% increase in comparison to only 0.87% for those subjects trained with open set material. These results support the hypothesis that subjects should receive training similar to the type of performance ex— pected of them either on criterion tests or in criterion situations. It was noted that performance on the Rhyme Test seemingly is enhanced more by closed set training than by open set training, at least with the present subjects. ' I . =5 "1:: .- 'i‘:- :'. scia‘ nu flimfimTi‘q ...swur. IIii 5|: a-‘h l" .'. urn-.1 . . I I . lI -" J“""] '41- '. h: :. -. .- i,d: : and! Fit? :2" "1:1" =."" Iii-II?) .-. rhu -'. I -. '.. ,- r . ..,.-. . u-I’T .... 'Il.l 116 The Hearing Handicap Scale Subjects rated themselves on degree of handicap experienced in everyday listening situations. On the Hearing Handicap Scale the higher the numberical score, the greater the degree of reported difficulty. (See Appendix D for a copy of the scale and Appendix H for raw scores). On the basis of these ratings, subjects were viewed arbitrarily in three categories: (a) eleven with "high handicap" ratings, (b) ten with "middle handicap" ratings and (c) eleven with "low handicap" ratings. Mean ratings were 68, 58, and 43, respectively. Percentage improvement, discrimination in noise, and speech reception threshold for these three groups were as follows: % Improvement 3.8 4.5 4.3 Discrim./Noise (%) 66.0 73.0 69.0 SRT (dB) 26 18 22 As can be noted above, there were no obvious trends among groups with respect to these particular performance measures. Comparing the high and low groups, however, those subjects rating themselves high in handicap showed slightly less improvement than those rating themselves low. The high handicap group had a lower discrimination in noise score than the low handicap group. Speech reception thresh- olds also were greater (i.e. worse) for the high than for the low handicap group. It appears then that Hearing m-‘ibamflm '- .r 5 .... 5 -.~ 11:35:15 shin-Nah “#45“ I' _ .-"'--5'_'= 'iin‘l’fiié in?! fi-Efl'lfls-fiuflfifi' I I r .s— 55de ,"Is. I - " SITE" AT. 3.285561%: .-_,-..=_.:-.. 35:“ - . ..me ...-.4, 1.5:; 117 Handicap Scale ratings had some correspondence to objective measures (DS/Noise and SRT) with the present mild hearing loss population. Spearman rank-order correlations between Hearing Handicap Scale (HHS) ratings and discrimination in noise {—0.18} and between HHS and SRT (0.02),suggested that predictive power is negligible with this population. Similarly, an obtained product—moment correlation of —0.02 between PB—Max and per cent improvement suggests that a linear prediction between these two measures is not possible. Since the present subjects were relatively homogeneous with respect to range of hearing loss, these are descriptive statements. Correlations might be higher with a more exten— sive range in degree of hearing loss than represented by subjects in this study. The Hearing Handicap Scale seems to have potential value as one means of complementing objective performance measures. Continued evaluation of its relevance to aural rehabilitation in general and auditory training in particular appears to be a worthwhile research and clinical endeavor. Validity of objective measures might be determined by these studies. Age An additional observation was that the older subjects showed more discrimination improvement than the younger ones. Those fifteen subjects having a mean age of 54 years in— 118 creased 5.73% while those fifteen subjects with a mean age of 34 years improved 2.3%. This trend was more noticeable when the thirty—two subjects were viewed in three groups on the basis of age. That is, ten subjects with a mean age of 56 years showed 7.2% increase, twelve subjects with a mean age of 45 years showed a 4.1% in— crease, and ten subjects with a mean age of 31 years showed only 1.3% increase. One possible explanation of this trend is that as age increases, subjects with mild hearing loss may not function near their optimum on pre—training tests,and therefore have more room for improvement as a result of training. Listening habits may decrease in effectiveness as a function of age and/or duration of hearing loss° In any event, present results suggest some optimism for those older subjects enrolled in auditory training. Subsequent research might investigate the specific effects of age on response to different training procedures. It may be found that some procedures are most appropriate for older subjects,while other methods and materials are most effec— tive for younger ones. Speech Reception Threshold The trend with respect to SRT and per cent improvement in discrimination was for subjects hearing test and training materials at a loud level to show more improvement than subjects responding to a low level. Since the testing and 119 training conditions,involved sound5field intensity levels around 50 dB (re audiometric zero), those subjects with a slight speech reception loss heard the signals some 30— HO dB above their SRTs. Subjects with greater speech re— ception losses, however, responded to the signals only 15—25 dB above their SRTs. Those nine subjects with 5—15 dB SRTS improved 5.18% in discrimination, those twelve subjects with 20—25 dB SRTs‘improved 4.83%,and those eleven subjects with 30-35 dB SRTs improved 2.66%. This trend suggests that initial gains from auditory training increased as a function of sensation levels at which testing and train— ing materials were presented. Subjects seemingly should receive training that is related to what has been assumed here to be a realistic and practical criterion——the loud- ness of average conversational speech in a noise background. All of the present subjects were trained without hear— ing—aids. In a clinic, persons with 30-35 dB SRTs probably would receive most of their auditory training while using a hearing aid. Therefore, the criterion average loudness level is still a feasible training condition since a hear— ing aid would increase the effective sensation level to a point where these individuals could show discrimination improvement similar to that recorded for the present 5—15 dB SRT group. 120 WAIS Vocabulary The range of possible scores on the WAIS Vocabulary sub-test was 0—80 points. Arbitrarily viewed, the top sixteen subjects (with a mean score of 60 points) improved 5.83% in discrimination while those sixteen subjects scoring lowest (with a mean score of 38 points) improved 2.58%. In addition, the ten subjects scoring highest showed 6.86% improvement while the ten subjects scoring lowest improved 2.13%. The trend was for intelligence (as estimated by the WAIS sub—test) to be a relevant factor in auditory training research. However, the relationship within the present population did not provide substantial linear prediction of individual performance in auditory training. A product— moment correlation of 0.21 was obtained between WAIS Vocabulary performance and per cent discrimination improve— ment. Use of a larger sample than employed in this study and administration of the entire WAIS might show a more adequate relationship for predicting response to auditory training. Research in auditory training should take into account intelligence as a possible factor influencing re— sults. Present training groups were matched (means, variances, and ranges) with respect to intelligence as estimated by the Vocabulary sub—test of the WAIS. This procedure was assumed to be satisfactory for a training group size of eight subjects. 121 Discussion Several questions were posed at the outset of this study regarding the effects of auditory training on the speech discrimination performance of a population of mildly hard of hearing individuals. These questions are re—stated in the following sections,with an effort made to answer them in terms of the present results. One of the preliminary questions asked was whether auditory training results in significant changes in speech discrimination. Research was cited in Chapter III that suggested an affirmative answer to this question. However, there did not appear to be specific research that exercised control of extraneous variables and used a sufficient number of subjects. Ten out of twelve measures of discrimination change showed positive increase following auditory training. This increase was significant on total, W—22, and Rhyme Test performance,but non—significant on the Semi—Diagnostic test, although in the positive direction. It appears that auditory training results in generally significant positive effects on speech discrimination with a mild hearing loss population. It is interesting to speculate on what further effects might be observed in a longer—term training program. If effects are positive following a concentrated three and one— half hour program, could these effects be enhanced even further by additional training? On the basis of usual learn— ing curves, one could expect that performance would continue 122 to increase to a plateau level. This assumes that sufficient motivation is maintained,and that any inter- ference effects are overcome by positive gains. The location of the plateau for hard of hearing subjects remains to be determined. Individual variation around this plateau is of clinical and research interest. Since auditory training seeks to assist individuals to make "maximum use of residual hearing," the objective specifica— tion of performance plateaus seems to have practical and theoretical importance. Training goals and information regarding auditory capacity are potential benefits to be derived from further speculation and research on this aspect of auditory training. A second question asked whether the W—22, Rhyme, and SemiuDiagnostic tests differed in reflecting the effects of auditory training with the present population. It was found that the W—22 changes differed significantly from those revealed by either the Rhyme or the Semi—Diagnostic, and that the latter tests did not differ significantly in changes. A product—moment correlation of 0.57 between W—22 and Rhyme tests indicates that some slight linear predic— tion between these two tests is possible with regard to discrimination gains in auditory training. One instead of both tests might be used as a criterion measure of auditory training progress. The question of which test(s) to use in evaluating auditory training was not an easy one to answer, especially 123 since a mild hearing loss population was studied here. The answer undoubtedly depends on the tester's purpose. If he wishes to measure progress in responding to Open set stimuli, the W—22 test seems best. If he wishes to measure progress in responses to closed set stimuli, the Semi—Diagnostic test might be suitable,although initial gains with the present population probably would be mini— mally reflected by this test. The Rhyme Test might be a better choice. It not only reflects the significant effects of auditory training but seems to be a closed set indice (p. 115), and also stresses consonant discrimination. The latter emphasis appears to be a valuable contribution of the Rhyme Test to discrimination testing with a mild hear— ing loss population. Subjects in this population usually do not experience much difficulty discriminating vowels. Their problem manifests itself by impaired consonant dis— crimination, and it is this type of discrimination training (and testing) that should be emphasized in auditory training. It is noted that present conclusions regarding dis— crimination tests apply only to the signal—to—noise testing conditions employed in this study. An S/N of 5 dB appears satisfactory as a means of sensitizing the W—22 and Rhyme tests. A different S/N——perhaps O dB——might have altered present results with the Semi—Diagnostic test. The closed set, four-choice response form of this test perhaps was too easy for present subjects, reducing the net gain to be ex— pected on a pre— vs post—training test basis. A number of I - ' in Jr.. “.33: qua-q 5’55 :Q J? gfiébnfig£nfi'n. saawaqu_ ... 5.5-"- .555 '55'.'5 551:5.- -- __ ._ __ --.-.-.-1r. 55-5-2355“ __ . . I _. . .I: . .‘..I-gl, fmf. “wiga jiiflfl = m -1fi 124 the test items require only vowel discrimination that, as indicated earlier, does not present much of a task for a mild hearing loss population. It is suggested that this test might be better sensitized by either decreasing the S/N ratio or scoring only those items that emphasize con— sonant discrimination. This entire problem of discrimination testing in auditory training remains to be solved. Do significant changes in test performance also indicate significant changes in everyday speech discrimination performance? The question now is finding how much change in test per— formance is needed before concomitant changes in every— day listening are observed by the subject, his associates, and the clinical audiologist. The question is not whether auditory training helps but rather how much it helps and what its limitations are in reducing communication break— down. From research and clinical viewpoints, continued evaluation on a large scale of the tests used to evaluate auditory training seems mandatory. A third question concerned the differential effects of manipulating signal—to—noise ratios during auditory training. A comparison of S/N—varied versus S/N—constant showed non—significant differences between these training methods. This finding suggests that either method can be used for similar results. Present subjects were both tested and trained with a speech babble noise background. One wonders what could be expected if subjects were trained 125 with speech babble and then tested with some other noise background (e.g., music, traffic, and the like) or vice versa. What transfer of training effects might be expected? How does an audiologist duplicate complex and variable listening conditions for training purposes? Subjects seemingly should be trained in conditions highly similar to those that exist when the results of training are to be applied. This suggests that a factory worker should be trained with a tape-recorded background of unique machinery noises, an office worker with a back— ground of typewriter noises, and so forth. It would be desirable to have one, two, or three standard noise back— grounds suitable for a wide range of application. Perhaps listening with noise, filtered speech, variation in talker— types, and/or variation in materials might be shown eventu— ally to enhance general discrimination performance. In any event, speech babble appears to have value as one means of partially duplicating the type of situation in which subjects with mild hearing loss report their most signifi- cant hearing difficulty. A fourth question was posed regarding the effects of one type of training material versus another type. This study failed to demonstrate a significant difference between effects of open set (write—down) and effects of closed set (multiple—choice) materials. Either type of stimulus—response training apparently can be used effectively 126 in auditory training. It was observed that subjects trained with open set materials performed better on the open set test (W—22) than on the closed set test (Semi- Diagnostic). The opposite result also was observed: subjects trained with closed set materials improved more on the closed set test than on the open set test. Sub— jects trained on specific skills logically should perform better on tests that demand these types of skills. This is consistent with other research in learning,and with the suggestion that subjects should be trained in condi— tions similar to those that exist when the results of training are to be applied. An initial preference for closed set (multiple—choice) materials is suggested, since most listening situations (i.e., listening to sentences) involve contextual, closed set types of listening skills. However, this preference is not too logical when one remembers that all words in sentences are not bound by contextual or closed set limits. Proper nouns and less familiar words in a language, for example, are not easily identified by context and do not obey closed set rules enough to be of direct assistance to the hard of hearing listener. A certain level of open set speech discrimination is required before sentence dis— crimination is possible. For these reasons then, auditory training should not be limited to closed set, multiple— $15 127 choice items,but also should include open set items. This topic is discussed further in the following paragraphs. A fifth question was asked concerning the effects of specific combinations of methods and materials on speech discrimination. Discussion thus far has considered methods and materials separately. Now, attention is drawn to the combined effects of these two variables. A fairly definite rank-ordering of the four combinations was observed: (a) Multiple-Choice, S/N—Varied (b) Write—Down, S/N—Constant (c) Write—Down, S/N—Varied (d) Multiple-Choice, S/NuConstant The above (a) through (d) ordering represents a descending order of combination effectiveness. The most obvious and significant characteristic of these findings concerns the two extreme combinations. Multiplenchoice materials were used with both; the only apparent factor distinguish— ing these combinations was the S/N method employed. Subjects making the most improvement participated in training under a S/N—Varied listening condition,while those making least improvement participated with S/N—Constant. Theoretical explanation of these findings is limited. Perhaps the discrimination required for multiple—choice items in the S/N—Constant condition is so simple that de— mands are minimal and subjects do not expend the effort required to learn sufficiently from the training. Perhaps 128 an optimum listening condition (i.e., training condition) exists when multiple—choice items are used with S/N—Varied so that the task is not too difficult and yet not too simple. Replication of this particular feature of the present study might provide additional information of both theoretical and practical importance. Additional research is needed to determine if these findings are fortuitous (i.e., peculiar to the specific conditions of this study) or more general and basic to speech discrimination per— formance in auditory training. Clinical application of these results is as follows. If multiple—choice items are used in training, the S/N-Varied method seemingly should be employed for most substantial gains. If write—down, open set materials are used, the S/N—Constant method could be slightly more bene— ficial than the S/N-Varied method,a1though either combina— tion gives similar initial results. These suggestions might only be applicable to a mild hearing loss population. Ideally, until additional information is obtained, clinicians perhaps would be functioning best if they used both (a) multiple—choice materials with S/N—varied and (b) write—down materials with S/N—constant. These combinations did not differ significantly but their combined effects possibly are better than any other two of the procedures. These suggestions are consistent with the earlier suggestion that training should include both closed set and open set material. 129 The sixth and last primary question of this study involved Hearing Handicap Scale self—ratings. The subjects reported substantial hearing difficulty. This supports one of the basic premises of this study that a "mild" hear— ing loss population does not always experience "mild" hearing difficulty. The mild measurable loss does not indicate categorically mild effects. Some correspondence between objective measures and Hearing Handicap Scale ratings was found in this study. Subjects rating themselves high in handicap (relative to other subjects) had a lower discrimination—in—noise score than subjects rating themselves low. Similarly, speech reception thresholds, on the average, showed greater loss of sensitivity for the high than for the low handicap sub— jects. Predictive power, however, between ratings and both discrimination—in—noise and SRT was low for the present subjects. Correlations might be higher with a less homo— geneous hearing loss population. Those subjects rating themselves high in handicap showed slightly less discrimination improvement than those subjects rating themselves low. Self—concept perhaps could influence progress in auditory training’although this question was not answered in the present investigation. In this same context, PB-Max scores did not provide linear prediction of response to auditory training. In any event, continued evalua— tion of the Hearing Handicap Scale——as one means of correlating measured performance with everyday listening performance—- 130 seems to be a potentially important means of increasing the validity of discrimination tests. In addition to self—reports of hearing difficulty, ratings by family and associates may provide information of importance in evaluating the effectiveness of auditory training. Supplementary observations suggested that age, speech reception threshold, and intelligence are factors to be controlled in auditory training research. An un— expected trend with the present subjects was for older individuals to show greater gains from auditory training than the younger ones. A second trend was for responses to auditory training to improve as the sensation level of test and training materials increased. The third trend was for training response to improve as intelligence (estimated by WAIS Vocabulary scores) increased. The specific effects of these factors on auditory training remain topics for future research. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The basic purpose of this research was to evaluate experimentally the effects of auditory training on speech discrimination performance of adults who have mild, sensori— neural hearing loss. Questions were posed concerning the relative discrimination changes resulting from specific auditory training methods and materials, as reflected by three published speech discrimination tests. Self—reports of hearing difficulty——indicated by Hearing Handicap Scale ratings——also were obtained for evaluation in relation to speech discrimination performance in auditory training. Summary Thirty—two subjects were selected for this study. The subjects had been seen for audiological testing on previous occasions. Each potential subject was seen for re~testing, at which time the following measures were recorded; pure tone air conduction thresholds, speech discrimination in quiet (PB—Max), speech discrimination in speech babble (sound—field at 5 dB S/N), and WAIS (Wechsler Adult In— telligence Scale) Vocabulary score. 131 132 On the basis of the above test data, together with case history information, subjects were assigned systemati- cally to one of four training groups. To support an equivalency—among—groups assumption, groups were matched (means, variances, medians, and ranges) on recorded measures. Each training group of eight subjects responded under one of four auditory training conditions: Group 1——write— down material with S/N-varied, Group 2——mu1tip1e—choice material with S/N—varied, Group 3——write—down material with S/N-constant, and Group A——multiple—choice material with S/N—constant. Two subjects participated simultaneously in a commer— cial sound—treated room. Practice materials, in a variable speech babble background, were delivered to the sound—field at calibrated intensities. Feedback regarding performance and rest breaks were provided at 25—minute intervals during the approximate three hours of training. Speech discrimina— tion tests (W—22, Rhyme, and SemieDiagnostic) were administered in the sound—field pre— and post—training at a 50 dB (re audiometric zero) intensity level in a 45 dB speech babble background. Results indicated that the auditory training procedures resulted in significant increase in speech discrimination performance as measured by total, W522, and Rhyme tests, but that these effects were not reflected by the Semi—Diagnostic test. Analysis of variance indicated that: (a) discrimination changes differed significantly among the three tests, (b) . '. "III. " . up — 7‘5 a}.'=.5'ti'."r-:'u' h ......jc; :5 15"" 121'; "9'13 .' :_'.". f’lr'f'hiflfi'fl :' |I . (I 5.3": 52".;- 5:5:3-4 ' .._ .- . . 133 discrimination changes differed significantly among the four training groups, (c) discrimination changes did not differ significantly between S/N—Varied and S/N—Constant methods, and (d) discrimination changes did not differ significantly between Multiple—Choice (closed set) and Write—Down (open set) materials. The use of a priori individual comparison procedures indicated that W—22 discrimination changes differed signi— ficantly from both Rhyme Test and Semi—Diagnostic changes, and that the latter tests did not differ significantly in discrimination changes. Similar critical—difference pro— cedures indicated the following rank—ordering of training combinations: (1) Multiple—Choice with S/N—Varied, (2) Write—Down with S/N Constant, (3) Write—Down with S/Nu Varied, and (A) Multiple—Choice with S/N—Constant. Differ— ences were not significant between (1) and (2), between (2) and (3), or between (3) and (4). Descriptive statistics with Hearing Handicap Scale selfuratings indicated that subjects rating themselves high in handicap showed slightly less discrimination improvement than those subjects rating themselves low. It also was observed that the high handicap group had a slightly lower discrimination—in—noise score than the low handicap group. In addition, speech reception thresholds showed slightly greater loss of sensitivity for the high than for the low handicap group. 13A Several trends were noted in the data. -Those subjects trained on closed set (multiple—choice) materials improved more on the closed set test (Sem15Diagnostic) than on the open set test (W—22). Similarly, subjects trained on open set (write—down) materials improved more on the open set test than on the closed set test. Perfor— mance on the-Rhyme Test seemingly was enhanced more by closed set than by open set training. Additional trends were for (a) older subjects to show greater discrimination changes than younger subjects, (b) discrimination changes to improve as sensation levels of test and training materials increased,and (c) discrimination changes to im— prove as intelligence (estimated by WAIS Vocabulary scores) increased. Conclusions Within the limits imposed by the present selection, testing, and training procedures,and by the use of adult subjects with mild hearing loss, the following conclusions seem warranted; 1. That short-term auditory training results in generally significant positive effects on speech discrimina— tion performance. Stability of discriminationugains remains to be determined. 2. That these effects, following a concentrated three—hour training program, are reflected by significant changes in total,W-22,Iand Rhyme Test discrimination,but not by the positive changes in Semi—Diagnostic test performance. 135 3. That W—22 discrimination changes differ signifi— cantly from changes revealed by either the Rhyme or the Semi—Diagnostic tests. Changes on the latter tests do not differ significantly. 4. That some linear prediction between W—22 and Rhyme Test discrimination changes is possible,though this predic— tion is not perfect and considerable error exists. Linear predictive power between Semi-Diagnostic changes and changes on either W—22 or Rhyme Test was not demonstrated. 5. That S/N-Varied and S/N—Constant training methods do not differ in effects on speech discrimination perfor— mance. If a difference does exist, it was not demonstrated in this study. 6. That Open Set (write—down) and Closed Set (multiple— choice) training materials do not differ in effects on speech discrimination performance. If a difference exists, it was not shown in this study. 7. That discrimination changes differ significantly among four combinations of auditory training methods and materials. If multiple—choice, closed set materials are employed in training, the S/N—Varied method seemingly should be the method of choice. If write—down, open set materials are utilized in training, either S/N—Constant or S/N—Varied methods apparently can be employed for similar results, i.e., if a difference between these alternatives exists, it was not demonstrated in this study. 136 8. That some general correspondence exists between objective performance measures and Hearing Handicap Scale self—ratings of degree of hearing difficulty. The exact specification of these relationships remains to be deter- mined. Subjects tend to make similar progress in auditory training irrespective of the degree of reported difficulty. 9. That age, speech reception threshold, and intelli— gence appear to be potentially significant factors in the conduct of research in auditory training and seemingly should be experimentally controlled when other factors or effects are being investigated. If these and other ex— traneous variables had not been controlled in this study, the above conclusions would be weakened substantially. Recommendations for Further Research The long—term effects of auditory training should be studied. This, however, represents a major undertaking because of difficulties regarding measurement criteria. Measurement problems perhaps could be clarified through research in auditory training where speech discrimination performance can be observed over a period of hours, days, weeks, and months. Subject variation under a variety of liStening conditions could be evaluated periodically and related to reports by the subjects and their associates. The area of self-reports and rating scales for the most part has been ignored, presumably because of re- liability problems. However, this type of research is needed before the validity of speech discrimination tests E _ _a. - .2. :1"! .u .' II LIIII II-lII - 'II ‘_.'-,.--I .. '.5-52555.55“ ”5.5 ' ' ' l . r .__. 5.1%.? 1:}- $2.0 , 137 can be established. Future research in auditory training should not just study tests or just evaluate rating scales. The two types of measures need to be studied concurrently with as much time and effort spent evaluating the ratings as is spent in the selection and administration of discrimina— tion tests. Discrimination tests are numerous with many combina— tions possible. Acoustical instrumentation is at a high level of sophistication. The diagnosis of peripheral and central auditory lesions has received and is receiving extensive study in a number of laboratories. But where are the audiological rating scales, the listening tests and scales, the reports of hearing performance at work and in social situations, and the like? How many laboratories and clinics are actively and consistently engaged in this re— . search and publishing their findings? The development of the Hearing Handicap Scale represents one of the few major efforts in this direction. It and similar scales need evaluation and refinement on a large—scale basis. Psycho— metrics, research design, and statistical analyses have enough sophistication to make these investigations both feasible and worthwhile. The thesis of this discussion then is to recommend the following; 1. Research in auditory training should be conducted on as high and extensive a level as possible, and not inde— pendent from other interest areas in audiology. Both basic 138 and applied research with persons enrolled in auditory training can enhance audiology‘s rehabilitation function and provide a framework in which to advance knowledge regarding impaired auditory performance. 2. Self—reports and rating scales, selective lis— tening, hearing aid characteristics, anxiety and auditory behavior, sensory facilitation between audition and vision, monaural and stereophonic effects in training, and so forth are a few of the directions for auditory training research with both adults and children. 3. The present results need to be substantiated by independent researchers. For comparative purposes, other hearing loss populations should be studied, employing similar and dissimilar tests and procedures. The effects of train— ing on speech discrimination at other than the present average—loudness levels need investigation. Training with materials delivered Via ear—receivers and/or transduced by hearing aids also might be studied for effects on speech discrimination. A. Since auditory training has as its major goal the development of maximum use of residual hearing, long— term studies should be conducted to determine plateaus in speech discrimination performance. Age, sensation levels of training materials, and intelligence are factors that need study in relation to performance in auditory training. 139 5. The Rhyme and Semi—Diagnostic tests, and/or similar tests, might be modified so that separate forms are available for evaluating vowel, consonant, voiced consonant, and unvoiced consonant discrimination. 6. The "sentence" as a unit for testing and train— ing has value for further research, i.e., the study of closed and open set monitoring performance in auditory training. 7. Variation in talker-types (e.g., dialectal vari— ants, rate of speaking, talker intelligibility, foreign accent, and the like) could be studied for relevance to auditory training. 8. Programmed instruction seems to offer a number of possibilities for evaluating the effects of various types of training procedures. Tape—libraries, emphasizing a¢coustical and linguistic variations, might be developed and studied for potential use in self—instruction and home— training. In essence, auditory training represents an area of study where the profession of audiology is uniquely qualified and responsible. The exercise of these qualifications and responsibilities remains a challenge for future researchers. BIBLIOGRAPHY 1140 BIBLIOGRAPHY Books Browd, V. L. The New Way to Better Hearing. New York: Crown Publishers, Inc., 1951. Carhart, R. "Auditory Training," Hearing and Deafness.. H. Davis and S. R. Silverman, editors. Revised edition. New York: Holt, Rinehart and Winston, Inc., 1963. Davis, H. and Silverman, S. R. (eds.). Hearing and Deafness. Revised edition. New York: HoIt, Rinehart and Winston, Inc., 1963. Fletcher, H. Speech and Hearing in Communication. Princeton, New Jersey: D. VanNostrand Co., Inc., 1953. Goldstein, M. The Acoustic Method for the Training of the Deaf and Hard—of—Hearing Child. St. Louis: Laryngoscope Press, 1939. Harris, J. D. Research Frontiers in Audiology, Chapter 11 in Modern Developments in Audiology. Edited by J. F. Jerger. New York: Academic Press, 1963. Hays, W. L. Statistics for Psychologists. New York: Holt, Rinehart and Winston, Inc., 1963. Hilgard, E. R. Theories of Learning. Second Edition. New York: Appleton—Century-Crofts, Inc., 1956. Hirsh, I. J. The Measurement of Hearing. New York: McGraw—Hill Co. Inc., 1952. Kelly, J. C. Clinicians's Handbook for Auditory Training. Dubuque, Iowa; William C. Brown, Inc., 1953. Kingsley, H. L. The Nature and Conditions of Learning. Second Edition. (Revised by R. Garry). Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1957. 141 142 Licklider, J. C. R. and Miller, G. A. ItThe Perception of Speech," Handbook of Experimental Psychology,’ Edited by S. S. Stevens. New York: John Wiley and Sons, Inc., 1951. McGeoch, J. A. and Irion, A. L. The Psychology of Learning. New York: Longmans, Green and Co., 1952. Mednick, S. A. Learning. Englewood Cliffs, New Jersey: Prentice—Hall, Inc., 1964. Newby, H. A. Audiology. Second Edition. New York: Appleton-Century—Crofts, Inc., 1964. O‘Neill, J. J. The Hard of Hearing. Englewood Cliffs, New Jersey: Prentice—Hall, Inc., 196 . Oyer, H. J. Auditory Communication for the Hard of Hearing. Englewood Cliffs, New Jersey: Prentice— Hall, Inc., 1966. Postman, L. and Egan, J. P. Experimental Psychology. New York: Harper and Brothers, 1949. '" ’ Silverman, S. R. Handbook of Speech Pathology, Chapters 10 and 11”L. E. Travis editor. New York: Appleton— Century_Crofts, Inc., 1957. Wechsler, D. The Measurement and Appraisal of Adult Intelligence. Fourth Edition. BaltimoreEI‘Williams and Wilkins, Co., 1958. . Manual for the Wechsler Adult Intelligence cale. New York: The Psychological CorpOration, 1955. U1 Whitehurst, M. Auditory Training Manual. New York: Hearing Rehabilitation, 1955. Winer, B. J. Statistical Principles in Experimental Design. New York: McGraw5Hill Co. Inc., 1962. Wolfle, D. Training, Chapter 34 in Handbook of Experimental Psychology. Edited by S. S. Stevens. New York: John Wiley and Sons, Inc., 1951. Wright, B. A. Physical Disability: A Psychological Approach. New York: Harper and Row, 1960. 143 REPORTS Bergman, M., et al. Auditory Rehabilitation for Hear- ing Impaired Blind Persons. ASHA Mono ra hs. Washington, D. C.: American SpeecH'an% Hearing Association, 1965. Costello, M. R. Realistic Goals in Auditory Training. In Processes, 39th Meeting, American Instructors of the Deaf. Washington, D. C.: U.S. Government Printing Office, 1960. Flanagan, J. C., et al. Psychological Research in the Armed Forces. Special Edition. Review of Educational Research, 18 (December, 19485. Health Aspects of Hearing Conservation. Published as ‘ a Supplement to the Transactions of the American Academy of Ophthalmology and Otolaryngology, (November-December, 1959). Moser, H. M. One-Syllable Words: Revised and Arranged by Ending Sounds. United StateS'Air FOrce CCDD“ Technical Note, No. 60558, 1960. Report of Subcommittee on Hearing Problems in Adults. Chapter 9 in Research Needs in Speech Pathology and Audiology. Journal of Speech and Hearing Disorders Monograph Supplement,No. 5, (September, 1959). ’ ' ‘ ' PERIODICALS Alpiner, J. G. "Aspects of Auditory Rehabilitation: Part II," Audecibel, 13 (1964), 47—51. Bangs, T. ”Methodology in Auditory Training," The W 56 (April, 1955), 159—64. ‘— Browd, V. L. "Hearing Education Without the Use of Hearing Aids," Archives of Otolaryngology, 49 (1949), 511—28. Campbell, R. A. and Small, A. M., Jr. ”Effect of Practice and Feedback on Frequency Discrimination," Journal of the Acoustical Society of America, 35 (1963), 1511—14. 144 Carhart, R. "Problems in the Measurement of Speech Discrimination, " Archives of Otolaryngology, 82 (September, 1965), 253—60. . "Inconsistency Among Audiometric Zero Reference Levels," Asha, 8 (March, 1966), 63—66. Davis, H. "The Articulation Area and the Social Adequacy Index for Hearing," Laryngoscope, 58 (August, 1948), 761— 78. Decker, L., Rubenstein, H. and Pollack, I. "Word Frequency and Speech Intelligibility for Unknown and Known Message Sets, " Journal of the Acoustical Society of America, 30 (July, 1958), 67. DiCarlo, L. M. "Auditory Training for the Adult," The Volta Review, 50 (September, 1948), 490—96. . "Speech and Communication for the Deaf, " IHe Volta Review, 62 (1960), 317—19. Eeckhout, M. J. ”Auditory Imperception, " Western Speech, 25 (1961), 180 83 Egan, J. P. ”Monitoring Task in Speech Communication," Journal of the Acoustical Society of America, 29 (April, 1957), 482—89. . ”Articulation Testing Methods," Laryngoscope, 58 (September, 1948), 955 91. . "Remarks on Rare PB Words," Journal of the Acoustical Society of America, 29 (11957), 751. Elpern, B. S. "Differences in Difficulty Among the CID W— 22 Auditory Tests, " Laryngoscope, 70 (1960), 1560— 65 Fairbanks, G. "Test of Phonemic Differentiation: The Rhyme Test, " Journal of the Acoustical Society of America, 30 (July, 1958). 596— 600 Farrimond, T. "Factors Influencing Auditory Perception of Pure Tones and Speech," Journal of Speech and Hearing Research, 5 (1962), 194—204. Frankel, G. W. "A Planned Home Auditory Training Program," Eye, Ear, Nose, and Throat Monthly, 40 (1961), 560—62. 145 Giolas, T. G. and Epstein, A. "Comparative Intelligibility of Word Lists and Continuous Discourse," Journal of Speech and Hearing Research, 6 (1963), 349-58. Goodfellow, L. D. "The Re-Education of Defective Hearing," The Journal of Psychology, 14 (1942), 53—58. Harris, J. D., Haines, H. L., and Myers, C. K. "The Importance of Hearing at 3 kc for Understanding of Speeded Speech. " Laryngoscope, 70 (1960), 131- 46. Haskins, H. L. "Listening with the Aid of a Hearing Aid," The Volta Review, 57 (1955), 408. Heineman, J., Muth, E., Creston, J. E., and Shutts, R. E. "Listening Through Visual Hearing," Hearing News, 32 (July, 1964), 5-8. High, W. S., Fairbanks, G., and Glorig, A. "Scale for Self Assessment of Hearing Handicap," Journal of Speech and Hearing Disorders, 29 (AuguSt, 1964), 215—30. Hirsh, I. J., and Bowman, W. D. "Masking of Speech by Bands of Noise, ” Journal of the Acoustical Society of America, 25 (November, 1953), 1175— 80. Hirsh, I. J., et a1. "Development of Materials for Speech Audiometry, Journal of Speech and Hearing Dis— orders, 17 (September, 1952), 321— 37. Hogan, D. D. and Hanley, T. D. "Some Effects on Listener Accuracy of Competing Messages Varied Systematically in Number, Rate, and Level, " Journal of the Acoustical Society of America, 35 (1963), 293— 95. Hudgins, C. V. ”Auditory Training: Its Possibilities and Limitations," The Volta Review, 56 (1954), 339—49. ”The Response of Profoundly Deaf Children to Auditory Training," Journal of Speech and Hearing Disorders, 18 (1953), 273— 88. Hutton, C. "A Diagnostic Approach to Combined Techniques in Aural Rehabilitation," Journal of Speech and Hearing Disorders, 25 (1960), 267—72. Hutton, C., Curry, E. T. and Armstrong, M. B. ”Semi— Diagnostic Test Materials for Aural Rehabilitation, " Journal of Speech and Hearing Disorders, 24 (1959), 319- 29. 146 Johnson, A. F. and Siegenthaler, B. "A Clinical Auditory Training Program," Journal of Speech and Hearing Disorders,l6 (March, 1951), 35- 39 Kelly, J. C. and Steer, M. D. "Intelligibility Testing in Three Conditions Involving Masking Noise," Journal of Speech and Hearing Disorders, 14 (1949), 369- 72. Ladefoged, P. and Broadbent, D. E. "Perception of Sequence in Auditory Events, " Quarterly of Experimental Psychology, 12 (1960),162—70. Lehiste, I. and Peterson, G. "Linguistic Considerations in the Study of Speech Intelligibility," Journal of the Acoustical Society of America, 31 (March, 1959), 280— 86. Licklider, J. C. R. "On the Process of Speech Perception," Journal of the Acoustical Society of America, 24 (November, 1952), 590594 Miller, G. A., Heise, G. A., and Lichten, W. "The Intelligi- bility of Speech as a Function of the Context of the Test Materials, " Journal of Experimental Psychology, 41 (1951), 329— 35 Moser, H., Dreher, J., and O'Neill, J. "Masking of English Words by Prolonged Vowel Sounds, " Journal of the Acoustical Society of America, 29 (119575, 1254. Moser, H. M. and Dreher, J. J. "Effects of Training on Listeners in. Intelligibility Studies, " Journal of the Acoustical Society of America, 27 (November, 1955), 1213—19. Oyer, H. J. and Doudna, M. ”Word Familiarity as a Factor in Testing Discrimination of Hard—of— Hearing Subjects," Archives of Otolaryngology, 72 (1960), 351-55. Peterson, G. E. and Lehiste, 1. ”Revised CNC Lists for Auditory Tests," Journal of Speech and Hearing Disorders, 27 (February, 1962), 62-70. Pollack, I., and Pickett, J. M. "Cocktail Party Effect, Journal of the Acoustical Society of America, 29 (1957), 1262 147 Pollack, I., Rubenstein, H., and Decker, L. "Analysis of Incorrect Responses to an Unknown Message Set, " Journal of the Acoustical Society of America, 32 (1960), 454-57 Resenzweig, M. R. "Intelligibility as a Function of Frequency of Usage, " Journal of the Acoustical Society of America, 28 (1956), 759. Ross, M. et a1. "Speech Discrimination of Hearing Impaired Individuals in Noise " Journal of Auditory Research, 5 (1965), 47- 7 Ross, M. and Huntington, D. A. "Concerning the Reliability and Equivalency of the CID W—22 Auditory Tests," Journal of Auditory Research, 2 (1962), 220—28. Schultz, M. C. "Word Familiarity Influences in Speech Discrimination, " Journal of Speech and Hearing Research, 7 (December, 1964), 395— 400. Silverman, S. R., and Hirsh, I. J. "Problems of Speech Audiometry, " Annals of Otology, Rhinology, and Laryngology, 64 (1955), 1234— 44. Silverman, S. R. "Training for Optimum Use of Hearing Aids," Laryngoscope, 54 (1944), 29—36. Thurlow, W. R. et al. "Further Statistical Study of Auditory Tests in Relation to the Fenestration Operation," Laryngoscope, 59 (1949), 113—29. Thwing, E. J. "Effect of Repetition on Articulation Scores for PB Words," Journal of the Acoustical Society of America, 28 (March, 1956), 302—03. Webster, J. Cn "Important Frequencies in Noise—Masked Speech," Archives of Otolaryngology, 80 (1964), 494—504. Wolfle, D. "Military Training and the Useful Parts of Learning Theory," Journal of Consulting Psychology, Zwislocki, J., Marie, F., Feldman, A. S., and Rubin, H. "On the Effect of Practice and Motivation on the Threshold of Audibility," Journal of the Acoustical Society of America, 30 (1958?, 254—62. 148 UNPUBLISHED MATERIAL Alpiner, J. G. "Relations Between Speech Intelligibility and Alterations in the Frequency Spectrum of Speech Noise Masking." Unpublished Ph.D. disser— tation, Ohio University, 1961. Higgins, D. M. "The Effects of White Noise and Speech Babble on the Intelligibility of Phonetically Balanced Lists of Monosyllabic Words." Unpublished Master's thesis, The University of Tennessee, 1965. Hodgson, W. R. "A Comparative Study of the Effects of White Noise, Speech Noise, and Complex Noise on the Intelligibility of Speech." Unpublished Ph.D. dissertation, Ohio University, 1961. APPENDIX A LETTER SENT TO PROSPECTIVE SUBJECTS 149 150 MICHIGAN STATE UNIVERSITY East Lansing College of Communication Arts\Department of Speech This letter is to describe an Auditory Training Research Program to be conducted soon at the Speech and Hearing Clinic, Michigan State University. We are very eager to have you take part in this program. Auditory training, basically, is a set of procedures that gives a person with a hearing loss the chance to make better use of his hearing through practice. This practice includes listening to speech in a variety of situations so that the person improves in the ease and ability with which he is able to understand speech in these types of situations. Most people with mild, moderate, and even severe hearing loss can benefit from this train- ing. The amount of improvement depends on the type and degree of hearing loss, the age, motivation, and listening ability of the person, the amount of practice, and so forth. If you would like to take part in this program, the following is required. First, your hearing must be re— tested before the training. This testing, to be done some— time during the next few weeks, will take about one hour and will be scheduled at your convenience. Second, you must be able to attend a one—half day training program. At this time, the training sessions are being planned for the month of April. You will be contacted by phone within the next week to answer any questions you might have and to find out,if you will Join us for what we believe will be an enjoyable experience for you. There will be no charge to you for either the testing or the training. In addition, the results of this research will be used to increase our ability to help people, like yourself, who have a hearing loss. Sincerely, Daniel L. Bode, U.S. Office of Education Fellow Herbert J. Oyer, Ph.D., Director Speech and Hearing Clinics APPENDIX B RESULTS OF MONAURAL PURE TONE AND SPEECH AUDIOMETRY 151 152 GROUP 1 Subject Ear 500 cps 1000 2000 3000 4000 SRT PB—Max 1 R OdB 15 50 75 8o 5dB 72%@40dB L 5 15 50 7o 60 10 84%@45dB 2 R 15 25 35 40 3o 20 96%@55dB L 20 25 35 35 25 2o 92%@55dB 3 R 25 55 50 5o 45 50 60%@85dB L 25 4o 40 45 4o 35 76%@70dB 4 R 35 25 20 35 3o 20 92%@55dB L 45 45 45 55 45 4o 92%@75dB 5 R 35 65 6o 55 50 35 76%@70dB L 20 65 55 6o 50 35 80%@70dB 6 R 15 35 55 75 65 25 68%@60dB L 5 5o 45 6o 55 35 88%@70dB 7 R 25 25 6o 75 65 15 76%@50dB L 40 35 65 70 65 40 85%@75dB 8 R 25 35 70 75 6o 25 48%@60dB L 30 45 75 80 6o 35 28%@70dB 153 GROUP 2 Subject Ear 500 cps 1000 2000 3000 4000 SRT PB—Max 9 R 40 35 45 65 60 25 64%@60dB L 45 35 6o 65 7o 35 44%@70dB 10 R 15 40 65 75 65 25 60%@60dB L 10 55 65 75 7o 35 64%@70dB 11 R 40 45 55 60 6o 45 64%@80dB L 10 35 65 65 7o 30 68%@65dB 12 R 15 20 55 65 55 20 68%@55dB L 5 5 50 55 6o 15 60%@50dB 13 R 55 4o 45 30 20 4o 72%@75dB L 45 40 30 3o 15 35 76%@70dB 14 R 40 10 2o 35 4o 20 52%@55dB L 15 5 20 30 3o 20 52%@55dB 15 R 40 5o 55 55 55 4o 80%@75dB L 35 40 45 6o 55 35 88%@70dB 16 R o o 5 6o 65 5 96%@40dB L 70 7O 85 75 7O _ ________ * *No response at maximum output of the audiometer. 154 GROUP 3 Subject Ear 500 cps 1000 2000 3000 4000 SRT PB-Max 17 R 30dB 25 20 15 25 25dB 96%@60dB L 40 40 35 45 40 40 88%@75dB 18 R 40 65 65 60 55 45 40%080dB L 30 70 55 65 55 4O 56%@75dB 19 R 20 35 50 65 7o 25 76%@60dB L 25 45 50 55 55 25 64%Q60dB 20 R 5 15 55 80 80 10 52%@45dB L 5 20 70 85 80 20 44%@55dB 21 R 15 40 85 95 95 35 40%07003 L 5 40 70 80 85 35 48%@70dB 22 R 20 35 35 55 55 30 72%@65dB L 20 50 50 55 55 35 88%@70dB 23 R 0 10 50 60 55 5 88%@40dB L 0 5 45 65 75 5 80%@40dB 24 R 30 30 60 50 45 30 80%@65dB L 85 90 85 _ - - ........ * *No response at maximum output of the audiometer. 155 GROUP 4 Subject Ear 500 cps 1000 2000 3000 4000 SRT PB-Max 25 R 55 55 60 55 60 55 92768063 L 50 35 30 40 35 25 92766063 26 R 30 40 85 95 100 35 56767063 L 35 40 55 55 50 30 80766563 27 R 15 55 50 55 60 35 84767063 L 10 55 65 65 60 30 76766563 28 R 50 35 40 6o 65 40 76767563 L 25 20 25 50 45 15 84765063 29 R 30 5 85 100 105 10 52764563 L 30 20 95 100 — 35 20767063 30 R 30 45 60 70 70 35 60767063 L 20 55 7o 70 70 20 44765563 31 R 40 50 45 30 20 35 84767063 L 45 45 65 80 75 55 60769063 32 R 5 0 0 30 50 5 98764063 L 0 5 10 45 65 10 96764563 APPENDIX C CRITERION DATA EMPLOYED TO ASSIGN SUBJECTS TO FOUR TRAINING GROUPS 156 157 .mpcmsmhdmwme onHmlocsom mam xmzlmm cam .Bmm .wwH02\mQ* 02 x mH HN :m mN m ow wH mow x 0H mm Nm H: mm :2 mH 02 x cm ow :m NF ON mm :H mm» x NH m: mm mm mm mm MH 02 x wH m: :m cm m mm NH oz x w um :m m: mN mH HH 02 OH NH 3 NH i N em OH mm» x NH mm mm m: mN w: m N mbomw mm» x NH N: Nm s: om mN w mow x m mm :m mm mH N: 5 02 x NH oz ow w: ON mm m 02 x ON Hm 2m Na om mm m 02 R H N mm we ON mm 3 62 x NH 3 ow NN mm m: m 02 x NH om mm mm mH mm N 02 x mH 2m mm mm m Nm H UH< OHNEmm mHmS Amnmmzv Ampmmzv ARV thoom smav Ampv ARV pomnnsm wcHmem .osom mw< *xmzlmm .hmoo> mH<3 *Bmm *mmHoz\wm H mbomw 158 .mpcoEOHSmmOE OHOHMIchom mam xmzlmm Ohm .Bmm .mmH02\mQ* 02 x ON m: Om OO O NO Nm 02 x NH mm OO Hm mm NO Hm 02 x NH Om OO NO ON ON om 02 x HH mm Nm Nm ON OH ON 02 x NH :m OO O: OH OO ON 02 x ON mm NO NO om O: ON mow x NH Om OO H: om OO ON mmw x NH Om :O NO ON Om mN O NOOOO 62 O OH mm OO OH om OO :N mm» x OH Om OO mm m :m ON mmm x NH O: OH Om ON Om NN 02 x OH Nm :O O: om ON HN oz 3 NH OO O: HO OH ON ON 02 x O OO OO NO ON NO OH 62 x mH ON OO em Om ON OH 02 x OH OO OO :O ON ON OH OH< mHmEmm mHmz Amhmmmv Amhmmzv HOV Ammoom SOLO Amuv HOV pomnhsm OOHmem .osem 6O< *xmzumm .Omoo> OH<3 49mm *63H62\OO m mbomw APPENDIX D HEARING HANDICAP SCALE 160 HHS Questionnaire (Form A) (Copyright Wallace S. High, Grant Fairbanks, and Aram Glorig, 1964) Name Today's Date Instructions: Following are a series of questions dealing with activities that depend upon hearing. In each question we want to know how often you are able to perform the activity referred to by the question. To answer each question, please check the scale from one to five. If you wear a hearing aid, answer the questions in terms of your experiences as they would be without the hearing aid. Please be sure to answer all the questions: 1. If you are six to twelve feet from the loudspeaker of a radio,do you understand speech well? 1. Practically always 2. Frequently . As often as not 4. Occasionally 5. Almost never 141 2. Can you carry on a telephone conversation without difficulty? 1. Practically always 2. Frequently 3. As often as not 4. Occasionally 5. Almost never 3. If you are six to twelve feet away from a television set, do you understand most of what is said? Practically always Frequently As often as not Occasionally Almost never U'IJE‘UUfUl-J lllll 161 4. Can you carry on a conversation with one other person when you are on a noisy street corner? U‘I-EUUIU 1 l U'I-L‘JL/ONF‘ 6. If there are noises traffic, music, etc. speaks to you? Hi! \_J‘IJ‘—'LJUF\)I--J 7. Can _____ l. .———__— 2. __ 3- 4. 5. 8. Can are chewing crisp foods, Hlll you understand a 1 o Practically always Frequently As often as not Occasionally Almost never Do you hear all right when you are in a street car, airplane, bus,or train? Practically always Frequently As often as not Occasionally Almost never from other voices, typewriters, can you understand when someone 3 Practically always Frequently As often as not Occasionally Almost never person when you are seated beside him and cannot see his face? U'lJl‘LUMH Practically always Frequently As often as not Occasionally Almost never you understand if someone speaks to you while you such as potato chips or celery? Practically always Frequently As Often as not Occasionally Almost never 10. 11. 12. 13. 162 Can you carry on a conversation with one other person when you are in a noisy place, such as a restaurant or at a party? Ill! UTJl‘UUNI-J Practically always Frequently As often as not Occasionally Almost never Can you understand if someone speaks to you in a whisper and you can't see his face? Hll U'IJZ'UUI'UH When you talk with a etc., can you understand U'l-P—‘UUM bus Practically always Frequently As often as not Occasionally Almost never driver, waiter, ticket salesman, all right? Practically always Frequently As often as not Occasionally Almost never Can you carry on a conversation if you are seated across the room from someone who speaks in a normal tone of voice? U‘lI—‘WMH Can you understand women Ill}! U'IJZ'LAJNH Practically always Frequently As often as not Occasionally Almost never when they talk? Practically always Frequently As often as not Occasionally Almost never 14. 15. 16. 17. 18. 163 Can you carry on a conversation with one other person when you are When you are out-of—doors and it is reasonably quiet? Practically always Frequently As often as not Occasionally Almost never would you know the speaker was talking if you could not see his lips moving? IHI U1J‘:UU|’\)I—' Can you follow the large dinner table UW-EUJNH Practically always Frequently As often as not Occasionally Almost never conversation when you are at a or in a meeting with a small group? Practically always Frequently As often as not Occasionally Almost never If you are seated under the balcony of a theater or auditorium, can you hear well enough to follow what is going on? U‘l-IL'LAJNH When you are Practically always Frequently As Often as not Occasionally Almost never in a large formal gathering (a church, lodge, lecture hall, etc.) can you hear what is said when the speaker does not use a microphone? Hill UW-ELUNH Practically always Frequently As often as not Occasionally Almost never 19. 20. 164 Can you hear the telephone ring when you are in the room where it is located? UTJL'WNI-J Practically always Frequently As often as not Occasionally Almost never Can you hear warning signals such as automobile horns, railway crossing bells, U‘IEWNH or emergency vehicle sirens? Practically always Frequently As often as not Occasionally Almost never APPENDIX E SPEECH DISCRIMINATION TESTS: W-22, RHYME, SEMI-DIAGNOSTIC 165 J‘: 12. 13. 14. l5. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. OKOCDNGU'I List 3 camp chair cute done dull ears end farm glove hand have jar king knit lie 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. W-22 TEST ITEMS may nest 1'10 owes pie raw say shove smooth start tan ten this three though tie use we west when wool year! 166 J: 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20; 21. 22. 23. 24. 25. \ooofioxm 21315.11 aid all bread can chin clothes cook darn dolls dust jump leave men 26. 27. 38. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. near net nuts of ought our pale save shoe so stiff tea tin than they through toy where who why will wood yes yet \O 11. 12. l3. l4. 15. 16. l7. 18. 19. 20. 21. 22. 23. 24. 25. oowoxmzw List 1 hot 227 top peel wake Iaw vile neat look £111 tire male sent moon kick same wide rip sore Pans men park coil Rig rage 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. RHYME tend 167 TEST ITEMS J: oxooowcnm 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. List 2 got may 30p reel take saw mile seat gook 3111 hire tale rent noon sick fame tide 21p bore hang gen bark foil H18 cage 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 11. 13. 15. 17. 19. led laid lad* lied pie* high by pen pain* pan pine wife* white wipe wise tea 168 SEMI—DIAGNOSTIC LIST 1A an* 2. add at am tea* 4. see she key luck 6. lake lack like* can 8. cap cat* catch net 10. not night nut* chains 12. chair chained change* plate 14. plays place* play bear* 16. pair dare wear pen 18. ten* then hen wrote 20. boat note vote* I'd eyes* ice I've caught cat coat cut* mean mine men* man pen then* ten when low row* go bow bow1* boil ball bull shame came tame same* far fire fur* fair wide word wood* what wise rise dies 1ies* barn burn“ born been we me- knee* be sun 23. 25. wide* died ride lake look luck* lock let led* leg less 24. light lay“ bite way might ray night' day let pole yet’ pull met pile get pool“ *Test—item. 11. 13. 15. 17. 19. ray way day lay* ten* then pen hen fur far* fair for plate place* plays play lied laid led 1ad* dad died dead did* such some* song it eat* at ate dare pair wear bear* note boat vote* wrote 170 SEMI—DIAGNOSTIC LIST 1B same* 2. came shame tame wool 4. we'll well will* bow 6. go row* low night* 8. might bite light be 10. knee* me we they 12. bay day* gay key 14. see he* tea fair 16. far fire fur* ball 18. bowl* bull boil yet* 20. met let get win with* will wish led* less leg let wide* died ride lied pull pole pool* pile man mine mean men* lock luck* lake look by pie* high tie wipe wife* white wise cut* coat caught cat I've eyes* I'd ice seek* seen seed seat word wood* what wide played* play plane plate won* fun run gun key see she tea* fell feel full fill* lake luck like* lack am at add an* not net nut* night win wing* wind will 171 SEMI—DIAGNOSTIC LIST 1B——Continued 21. cap is 22. when burn* cat* it* ten been catch ill then* I born 7 can if pen barn 23. cool* seed 24. pine seat pool seen* pen set 1 tool seem pain* sit* fool sing pan sat 25. chair lies* chains rise change* dies chained wise *Test—item. APPENDIX F WRITE—DOWN TRAINING MATERIAL 172 f\) l-l |_l H 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. O\OOJ\10\UTJ‘:UU Jar boil tough tooth goose toad rout mess kite Jug pad salve van home cape shore wreck shirt knife hull yearn wheel fit patch List 1 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. CNC (REVISED) make dime bean thin seize hate wood check ditch rose merge lease loop king dead chore boat wish name pick ripe fall lag gale 173 LISTS 1. rail 2. vine root 4. fake 5. cob 6. moon 7. talk 8. fern 9. this 10. nose ll. ship 12. leak 13. nurse l4. hash 15. lead 16. jet 17. south 18. dire 19. beg 20. pan 21. much 22. dodge 23. weep 24. wag 25. sap _Li_s:_2 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 45. 46. 47. 48. 49. 50. hide choice met red goal should car pave love which bought soul gain germ beam ring dam tire tall late coat suck choose puff hill ._l [_l l—' 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. O\OOD\]O\U14:W List 3 Jail toss soon faith sung keg vote size numb what room kid dike mate well rig four bush dip perch sheep house 26. 27. 28. 29. 30. I) 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. fade lake gull rouge bar tone chin piece purge bell work life shine toll joke head with keen more leave hut noise yam 174 J: 10. 11. 12. 13. l4. l5. l6. l7. 18. 19. 20. 21. 22. 23. 24. 25. \oooxioxw sock pool chief pause give lap write serve bone said tower wig chum thumb loan take birch dose him deal Job wail read shake List 4 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. rice cash hire gas phone can mop rage long nice till youth when pack war mill hoof void date shut loud mirth foot keep 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. CDNONU'l-E'LU O\0 List 5 veil worm half gaze limb Juice light zeal town chalk bathe food mean boot yoke tease hot then rough raid dawn pull luck nudge 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. good nag wire robe thought beach dim purse tell coal cup dock care sore five myth match sing sail knit shop lean hush back 175 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. List 6 whip bud shone rug cheese chain look dull pope calf fire turn raise SOLII' tube veal get pace night hiss shock wing 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. door niece cat move cool web knock Jot cage mode search gone rush pole dig bad live map Wife fan birth team howl hike Jam [..I 12. l3. 14. 15. 16. l7. l8. 19. 20. 21. 22. 23. 24. oxoooflmmzw List 7 note doom coke hole join third mouth sure vague pearl loot save side heat fish have mole 26. 27. 28. 29. 3o. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. reach face bet caught laugh shall geese tape sack ridge cheek dumb top young rib pass wit call neck such lose tar 176 I: 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. O\OCI>\'IO'\U'I List 8 moss daze loathe road muff vowel tip thing week wheat foam poor wet seek lash hail page lock gear hoop( learn guide fuss jerk p086 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. rot touch calm gin some real bite near 888 cheap wake hurl tin noose dive bath there cough shawl 10. 11. 12. 13. 14. 15. l6. l7. l8. 19. 20. 21. 22. 23. 24. 25. \OWNQU'IJ-‘Jw List 9 lack 26. watch 27. power 28. mire 29. nail 30. thine 31. word 32. tool 33. mob 34. hen 35- got 36. sane 37- shout 38. pill 39. both 40. shade 41. jazz 42. lathe 43. catch 44. white 45. chair 46. loaf 47. pun 48. ham 49. lip 50. wrong yes curve haze girl time book reap fudge voice mud ball deck need cheer soap feet tick roof beat dish 177 H 12. 13. l4. l5. l6. l7. l8. 19. 20. 21. 22. 23. 24. 25. OkOGJNONUl-D'UUNH sub din death chill coin cause burn loose palm judge wash fine while chat bit nick neat hair safe jade hurt pile List 10 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. shack cone sell your term mood deep meek rope witch ride bake gore fool guess mouse lung load path peak run sag cave thatch towel APPENDIX G MULTIPLE—CHOICE TRAINING MATERIAL 178 shovel pear ton anger stag whisker choose ladder* buzz* Jump* pose* fence hard cable gain* ankle stung* weed* lunge* wait degree* cash bunch* beach* wig flash neck* mint surface stack* vender drab* shuffle* bear tongue anchor* stack whisper shoes latter bus Chump post sense* heart* gable* cane angle* stun wheat lunch wake decree gash* punch peach wing flesh* knack minute service* slack fender grab 179 hovel tear* tub* hanker tack* whimper* shoot* batter bun hump boast tense harp able pane tangle spun weep lump wade* debris dash hunch teach wick* lash nick meant* surplus stag sender* crab KELLY'S LISTS 2. 4. 4. gauge* peril rival cheer* singe ether* glue muzzle worse eyes* slide leave chief bag* game steam town truck* veal prove phone list* laws seat rag* pie* gem* rain sang tribe* grime grim cage barrel* rifle jeer cinch* either clue* muscle* worth* ice slight* leaf cheek* back came* steed down* trunk fee1* proof* foam* lift loss seed rack pink jam range sack trite crime grin age carol trif1e* sheer inch heater blue tussle work light light lease* cheap bank aim steep* gown drug seal prude home left loft* seek* rank pick gym rage* sag* tripe grind* brim* *Stimulus-Word. crop green trill ridge* lime* lose* bug* drag skim breathe graft* broom skill vowel vault spar simple vase vale preach paint* bride* crew* phase* dues think* grime packing* view* taste dime* safe dunk list* come zeal* drop* greed drill* rich line loose buck brag* skin breeze* craft groom* spill* fowl* fau1t* star* symbol* face* fail* breach* faint' pride deuce* thing crime* backing few 180 haste dine sane duck* live gun feel prop greet* thrill rib lion loot bunk track spin* breed grass prune still howl salt scar single fate sale reach taint bright brew fade tooth thick grind tacking new stab1e* scene raise* seize rice lamp* truce blame rudder still ample fleas* zinc herd mean* stir goat p1aque* chair* string* vast* smug search scare van smack* post cone* back bigger plank coal five* bounce* bulb zoo staple seam* race cease rise land truth claim* runner* spill amble* fleece sink* hurt meet spur coat* black share spring fast snug* surge* spare* fan* snack boat comb bag* bicker* blank* goal* fife pounce bulk* sue* table seed raid tease* right* lamb roost* plane rubber skill* camel fleet sing her* bean slur* coke plank air sprig fat mug purge stair sand stack boast* code pack picker clank go vise bouts pulp due *Stimulus—word. pill* moon pays true knit chuck* place* dumb* raise blend fan* door sack lose* vast dent* news sum* brief cane* graze* nice* burn* plume broad bored trench* spoke parrot bale* job* chair mice raid class* pride bill noon* pace* drew* mit* chunk plays done race* b1ed* fad tore* sag* loose fast* tent noose* sun grief* game grace knife bird bloom* brought* poured* drench smoke* parent* pale chop share* nice* rate* glass bride* 181 till mood paste grew lit chug plate dug rage lend sad bore sank loot last bent moose sung reef pane gray night burr loom bought toward wrench soak pennant rail shop care ice rake last bright 6. 2. 2. 6. 2. tame* budge* start neat* knob blank* fame chin view grab his prize* crutch tack racer* rag* sky* cause* etch blame brim case cave smear jade splint scant pin joke* last gape trance* clean case push* plus* dame bunch stark* meet mob* black vein shin* few* crab* hiss* price crunch* tag* razor rack spy gauze edge plane* grin* cape gave* sneer* shade* sprint* scamp bin* choke lest* cape prance gleam gaze* bush pluck game punch spark beet mop plank same* gym feud drab hid pride crux tank raider rank sty cost itch* fame grim cane* pave spear chain print camp* din show list cake* plants lean* gave wish plunk *Stimulus—word. crag blaze* crew grudge* girl ridden* flute groove wick* patch tug* graze* shack tangle* waif bug* crack* place grew* crutch curl* written fluke* groom wink batch* tuck grace shank* tango wave* buck 182 crank plate brew drudge cur ribbon flew group* wig catch tub great shag dangle wade bunk peak pan blunder flash motto* came* triple* brink might loose crow dug* dare tooth sag meal beak* ban* p1under* slash* model game cripple drink* night lose grow* duck bare* toot* sack kneel* beet fan thunder clash bottle gain trickle brick light* loom* groan dunk pear tube sank* deal *Stimulus—word. 183 LARSEN'S LISTS f and ch p and b m and 1 sh and f fin* chin pin bin* mine* line show foe* few* chew pie* buy mast* last shore* four filed child* pole bowl* moan* loan shade fade* calf* catch cap* cab name nail* cash* calf four chore* rope* robe home* hole leash* leaf f and k b and m n and v d and n fit kit* bill* mill nice* vice dot not* four* core boast most* nurse* verse die nigh* find . kind* bake make* nine vine* deed* need cliff click* robe roam* loans* loaves ode* own laugh* lack tab* tam lean leave* did* din k and g m and v t and th 9 and f coal* goal mice vice* tie thigh* pour* four came game* ham have* tin* thin pile* file coat goat* glum* glove trill thrill* par far* luck* lug mine* vine mit* myth cap* calf rack* rag mile vile* pat* path cup cuff* v and z 1 and v 1 and z v and f live lies* lane* vane lip zip* five* fife have has* lie vie* loan* zone vase* face rave* raise lace vase* lisle lies* leave leaf* View 200* lull* love dole does* View few* wives* wise rail* rave male maze* loaves* loafs ' Vine* fine 1 and n b and d s and sh f and b 1ame* name bid did* 1ease* leash fun* bun light night* big* dig sew* show fig* big loan known* buy die* sign shine* fan ban* dial* dine rob* rod sip ship* cuff* cub pail* pain bell* dell save* shave calf* cab rail rain* robe* road lass 1ash* graph grab* k and t m and n b and v kick tick* mine* nine bet vet* kite* tight mew knew* bow* vow code* toad time* tine bile* vile Shirk shirt* dime dine* bigger vigor* park* part dumb done* robe rove* kin tin* loam* lone boat* vote *Stimulus—word th and v than van* thy* vie that* vat thine* vine loathes loave d and 5 door* gore dot* got doe go* date* gate drove* grove bud bug* dye* guy dad gag* th and s theme seam* thin* sin thumb sum* truth truce* path* pass myth miss* thing* sing thank* sank word endings store stores* close* closes will* wills m and l mine* line mast* last moan* loan name nail* home hole* 184 f and t k and 2 four* tore pike pipe* fall tall* car* par fan tan* core* pore fill‘ till coke poke* s* free* tree cock* cop fry try* crock crop* cry pry* coal* pole s and z t and p f and s ice eyes* tore* pore fine* sign seal* zeal tine pine* fur* sir sip* zip tail pail* four soar* loose lose* cat* cap flat s1at* bus* buzz cut* cup cuff* cuss lice lies* tar* par knife nice* juice Jews* toll pole* lift list* fuss* fuzz coat cope* loft* lost ch and sh th and f chop shop* thin* fin chip* ship thirst first* chair* share three* free chew* shoe Thor for* watch wash* thought fought* catch* cash throw* fro which wish* thrill frill* cheap sheep* f and ch 9 and b stored fin chin* pin bin* closed few* chew pie buy* willed filed* child pole* bowl calf catch* cap cab* four chore* rope* robe sh and f f and k b and m show* foe fit* kit bill mill* shore four* four* core boast most* shade fade* find kind* bake make* cash* calf cliff* click robe* roam leash* leaf laugh lack* tab tam* *Stimulus —word. n and v nice* Vice nurse verse* nine vine* loans* loaves lean* leave t and th tie* thigh tin* thin trill thrill* mit myth* pat* path d dot die dee ode did po pi pa ca cu 185 and n not* nigh* d* need own* * din p and f ur four* 1e file* r* far p calf* p* cuff coal came coat* luck* rack* k and g mice ham* glum* mine mile goal* game* goat lug rag m and v *Stimulus-word. Vice* have glove vine* Vile* APPENDIX H PRE—TRAINING, POST—TRAINING, AND DIFFERENCE—SCORES; HEARING HANDICAP SCALE SELF—RATINGS 186 .Amgoom 3mpv mmsHumHIwHOm mHmOm OOOHcsmm mcHmem** .OUSpm mwsp CH mmhdmmms psmuscamo map who: mmaoomlmosmhmmem* 187 OH H O O OO OO NO NO HO HH OH HO NH OHI O OO OO NO OO OO HO OH OO O N I O I OO ON NN OO ON ON HH OO O N OH OO OO OO OO OO OH OH OO N I OH OH OO OO HN OO NO OO NH NO OH O N OO OH HO OO OH NO HH ON O NH OH ON OH NO OH HO NH OH OO HH HN NN HO HO NO OO OH OH O N NOOOO OO O I O ON HO OO Om OO HO OH O OO OH O H I OO OO OO OO NH HO H OH H N O OO OO OO HO OO NO O OO OHI N O OH OO HO OO OO OH O OO N O O OO OO NO OO OO HO H HO O OHI O OO HO OO OO HO NH O HO N H I O OO HO HO OO OO OO N OO HO I HHH HHH HOO HHO ONO ONO HON OOH H .OOHOIHsmm OEOOO NNI3 .OOHOIHEOO OEOOO NNIO .OOHOIHEOO oemgm NNIz OOOOOOO **mmm *GMOOWIOOCweHmQMHQ MCHCfithIPwOnH MQflQHMhBImhm H mbomw 188 .Ampoom SOLO mwsflpthOHmm mHmom OOOHUQOm wcfihmmm** .OOSpm mflnp :H wmhdmmms psmusmgmw map mum: mmmoomlmosmmcmmHm* ON N N I O OOH OO NO OO OOH HO NO OO H O I O I OO OO OO NO HO NO HO HO N I ON HH OO NO HO NO NO OH OO OO O I N I O OO HO OO OO OO OO ON HO O I H I N OO HO OO HO OO OO ON OH O OH OH OO NO NO HO HO HH ON OO H O O I OO NO NO HO NO OO ON OO N I N I H I OO OO OH NO OO HH ON H OOOOO OH H O OH OO OO OO NO OO OH HN OO O O NH NO OOH NO OO NO OO ON OO OH OHI N OO NO NH OO NO OH NN OO H I H N OO OO OH HO OH OO HN HO O OH O OO OO OH OO OH NO ON HO H O OH HO NO OO OO OO NO OH OO NHI O O OH OO NO OO NO HN OH OO ON ON I OOH OOOH OOO ONO OOO OOO OOO OH .OOHOIHEOO msOnm NNIz .OOHOIHEOO OEOOO NNI3 .OOHOIHEOO OEOOO NNIO OOOOOOO *Ommm *mhoomlmocmymmmHQ mcHCHwLBIpmom MQHCHOABIOON m Mbomo I l ! flICHIG STRTE UNIV. L 11111 1 O 18 111171 31 IBRRRIES 111111111111 8 61 1 1. 3 mflh‘F 3:. ....