P ~ c e p i u aand l Motor Skills, 1978,47, 859-862. @ Perceptual and Motor Skills 1978
SPEECH DISFLUENCIES AND DELAYED AUDITORY FEEDBACK REACTIONS OF STUTTERING AND NON-STUTTERING CHILDREN1 BEVERLY A. TIMMONS AND JAMES P. BOUDREAU' University of Victoria, Victoria, B. C. Summasy.-25 male stutterers and 25 male non-stutterers matched by age and speaking task, read or recited under conditions of normal and 113-, 2 2 6 , 3 0 6 , 413-, and 520-msec. delayed auditory feedback. Disfluency counts were correlated with delayed auditorp feedbadt reactions which were changes in disfluencies under delay conditions. Pearson product-moment correlations were negative and significant for the combined group of stutterers and non-stutterers under all delays used. Correlations for stutterers were negative and significant for 113, 226, 306, and 413 msec. delay. For the total group of non-stutterers, all correlations were negative and significant. Correlations for age groups within the stuttering and non-stuttering groups were also presented.
Are the number of disfluencies in speech under conditions of normal feedback related to reactions of stuttering and non-stuttering children under delayed auditory feedback? Compared with normal feedback, the speech disfluencies of non-stuttering adults have been shown to increase under conditions of delayed auctitory feedback (Fairbanks & Guttman, 1958; Chase, Sutton, Rapin, Standfast, & Harvey, 1959). However, marked variability in the effect of delayed auditory feedback on non-stutterers' disfluencies has been reported (Beaumont & Foss, 1957; Atkinson, 1954; Yates, 1963; Van Riper, 1971). Atkinson (1954) reported that those adults with high intelligibility in their speech under normal speaking conditions were less affected by delayed auditory feedback than those with low intelligibility under normal conditions. Delayed auditory feedback has been reported to facilitate fluent speech in adult stutterers (Chase, Sutton, & Rapin, 1961; Lotzmann, 1961; Goldiamond, 1966; Soderberg, 1969). Variable effects of delayed auditory feedback have been noted within groups of adult stutterers (Ham & Steer, 1967; Neelley, 1961; Lotzmann, 1961; Chase, Sutton, First, & Zubin, 1961). Ham and Steer (1967) indicated that those stutterers judged severe "reduced blocks during any type of feedback alteration" while mild and moderate stutterers varied in their reactions. The intelligibility of both stutterers and non-stutterers has been observed to be related to reaction to delayed auditory feedback, but correlations between disfluencies under normal ,and delayed auditory feedback have not been reported. Timmons and Boudreau (1978) indicated that groups of stutterers and non-stutterers did not differ in their reactions to delayed auditory feedback. Intelligibility, as measured by disfluency counts, irrespective of speech group'This study was supported by a grant from the Spencer Foundation, Chicago, Illinois. W e acknowledge work by Ann Tamboline, Research Assistant, in collection of data.
860
B. A. TIMMONS
&
J. P. BOUDREAU
ing, might be related to reactions to delayed auditory feedback. A re-exarnination of those data, using correlational methods, seems warranted, since delayed auditory feedback is being used in therapeutic programs for stutterers (Goldiamond, 1966; Van Riper, 1971; Ryan & Van Kirk, 1974).
METHOD Twenty-five male stutterers and 25 male non-stutterers constituted the sample. Diagnosis of stuttering was made by speech therapists employed by the Tacoma, Washington, public schools from which the sample was drawn. Stutterers were matched with non-stutterers by age and speaking task. The pairs ranged in age from 5 to 13 yr., with a mean of 9 yr. The speech of each subject was recorded under conditions of normal feedback and with delays of 113, 226, 306, 413, and 520 msec. Delays were presented in random order. , T h e apparatus used to record speech and to produce delayed auditory feedback was a Korting Constellation 8 with a moveable playback head. A Korting 284 dynamic microphone picked up speech which was returned through a Sharpe 660 headset of 8 ohms nominal impedance. The overall band pass of the instrument was 30 to 19 kHz. The feedback intensity was approximately 75 d b above background noise with slight variations produced by the subjects and limitations in the equipment. The materials used as stimuli for verbal output were taken from the SRA Elementary Reading Laboratory (Parker, 1958). Reading materials were at least 1 yr. below the independent reading level of the subjects. A nursery rhyme was used with non-readers or those reading below the third grade level. The two groups were measured under conditions of normal and delayed feedback, using number of disfluencies (omissions, repetitions, substitutions, and additions). Disfluencies were counted by two independent observers, whose reliability was .92 as determined by a Pearson product-moment correlation. Difference scores were obtained for the number of disfluencies under normal and delayed feedback as a measure of reaction to delayed auditory feedback. Pearson product-moment correlations were calculated. RESULTSAND DISCUSSION Pearson product-moment correlations were calculated, using disfluency counts under normal feedback and difference scores indicating reaction to delayed auditory feedback (see Table 1). Negative correlations for stutterers were statistically significant under 1 1 5 , 226-, 306-, and 413-msec. delay conditions ( < 0 1 When age groups within the stutterers were investigated, correlations for 5- to 8-yr.-olds were negative and statistically significant. Correlations for 1 1 5 , 226-, 306-, and 520-msec. delays were significant ( p < .01); the correlation for 413 msec. was significant ( p < .05). For 9- to 13-yr.-olds, correlations for 1 1 5 , 226-, and 306-msec. delay were significant ( p < .01), while the correlation at 413 msec. was significant ( 9 < .05).
SPEECH DISFLUENCIES AND DELAYED AUDITORY FEEDBACK
861
TABLE 1 PEARSON PRODUCT-MOMENT CORRELATIONS BETWEEN DISFLUENCIES IN SPEECH UNDERNORMAL C O N D I ~ O N AND S DIFFERENCE SCORES SHOWINGREACTION TO DELAYED AUDITORYFEEDBACK Group
D
113 Stutterers 5- to f~-~r.-olds 9- to 13-yr.-olds Non-sruaerers 5- to 8-yr.-olds 9-to 13-yr.-olds All Subjects 'p < .05. t p < .Ol.
25
10 15 25 10 15 50
-.81t -.88t -.89t -.71t -.64*
-.88t -.81t
Delay of Feedback (msec.) 226 306 413 -.84$ -.86t -.87t -.78t -.60* -.86t -.82t
-.77t -.83t -.73t -.60t -.01 -.86t -.72t
-.Got -.65* -.59* -.31 .36 -.76t -.48t
520 -.28 -.83t -.40* .57 -.53t -.40t
Negative correlations were also found for the non-stuttering group. Correlations were statistically significant (p < .01) for 113-, 226-, and 306-msec. delays, while the correlation for 520 msec. was significant ( p < .05). For the 5- to 8-yr.-olds within the non-stuttering group, the correlation for scores at 113-msec. delay was significant ( p < .05 ) . None of the other correlations differed significantly from zero. In the 9- to 13-yr.-old group, correlations at delays of 113, 226, 306, and 413 msec. were significant ( p < .Ol), while the correlation for 520-msec. delay was significant ( p < .05). Age differences in reaction to specific delays were more noticeable for the non-stuttering group than for the stuttering group, with non-stuttering 5- to 8-yr.-olds being significantly affected by only one delay. Age differences in reactions of non-stutterers to delay have been noted in the literature (Chase, Sutton, First, & Zubin, 1961; MacKay, 1968; Buxton, 1969; Timmons & Boudreau, 1976). When both stutterers and non-stutterers were considered together, correlations for scores at all delays were negative and statistically significant ( p < .01). These data confirm the observations that for both stutterers' (Ham & Steer, 1967) and non-stutterers' (Atkinson, 1954) reactions to delayed auditory feedback are negatively related to intelligibility under conditions of normal feedback. However, use of difference scores, as Cronbach and Furby (1970) pointed out, may contribute to the high negative values observed here. Delayed auditory feedback seems to be useful when dealing with severe problems of disfluency, whether or not they are diagnosed as stuttering. Further investigation of this therapeutic tool is necessary to demonstrate its effectiveness with speech difficulties other than stuttering. However, there may be a point (related to degree of disfluency) at which diminishing returns make the delays used in this study inappropriate for therapy. The sample tested is not considered large enough to determine whether, or at what point, these delays become of little use.
B. A. TIMMONS
&
J. P. BOUDREAU
REFERENCES ATKINSON,C. J. Some effects' on intelligibility as the sidetone level and amount of sidetone delay are changed. Iowa Academy of Science, 1954, 61, 334-340. BEAUMONT,J. T., & FOSS, B. M. Individual differences in reacting to delayed auditory feedback. British Journal of Psychology, 1957, 48, 85-89. BUXTON, L. F. An investigation of sex and age differences in speech behaviors under delayed auditory feedback. Unpublished doctoral dissertation, Ohio State Univer.. 1969. CHASE, R. A., SUITON, S., FIRST, D., & ZUBIN, J. A developmental study of changes in behavior under delayed auditory feedback. Journal of Genetic Psychology, 1961, 99, 101-112. CHASE, R. A., S u r r o ~ ,S., & RAPIN, J. Sensory feedback influences on motor performance. Journal of Auditory Research, 1964, 3, 212-223. CHASE, R. A., SUTON, S., RAPIN, J., STANDFAST,S., & K~RVBY, S. Sensory feedback influences on motor performance. Communications Lab., Columbia Universiry and Department of Biometrics Research, State of New York, Research Report, 1959. CRONBACH,L. J., & FURBY, L. How we should measure "changew---or should we? Psychological Bulletin, 1970, 74, 68-80. FAIRBANKS,G., & GU'ITMAN. N. Effects of delayed auditory feedback upon articulation Journal of Speech and Hearing Research, 1958, 1, 12-22. GOLDIAMOND, I. Stuttering and fluency as mani ulatable operant response classes. In L. Krasner & L. P. Ullmann (Eds.), ResearcK in behavior modification. New York: Holt. Rinehart &Winston, 1966. Pp. 106-156. H m ,R., & STEER, M. D. Certain effects of alterations in auditory feedback. Folia Phoniatrica, 1967, 19, 53-62. LO'IZMANN,G. Zur Anwendung variierter Verzogemngszeiten bei Balbuties. Folia Phoniatrica, 1961, 13, 276-310. MACKAY, D. G. Metamorphosis of a critical interval: age-linked changes in delay in auditory feedback that produces maximal disruption of speech. Journal o f the Acoustical Society of America, 1968, 19, 811-821. NEELLEX, J. N. A study of speech behavior of stutterers and non-stutterers under normal and delayed auditory feedback. Journal o f Speech and Hearing Disorders, Monograph Supplement, 1961, 7 , 63-82. PAR=, D. H. SRA Reading, laboratory, elementary edition, teacher's handbook. Chicago: Science Research Assoc., 1958. RYAN, B., & VAN KIRK, B. The establishment, transfer, and maintenance of fluent speech in 50 stutterers using delayed auditory feedback and operant procedures. journal o f Speech and Hearing Disorders, 1974, 39, 3-10. SODERBERG, G. A. Delayed auditory feedback and the speech of stutterers: a review of studies. Journal o f Speech and Hearing Disorders, 1969, 34, 20-29. TIMMONS,B. A., & BOUDRBAU,J. P. Age, sex, and delay time as factors affecting reaction to delayed auditory feedback. Perceptual and Motor Skills, 1976, 43, 27-30: TIMMONS,B. A., & BOUDREAU,J. P. Delayed auditory feedback and the speech of stuttering and non-stuttering children. Perceptual and Motor Skills, 1978, 46, 551555. VAN RIPEX, C. T h e nature of stuttering. Englewood Cliffs, N . J.: Prencice-Hall, 1971. YATES, A. J. Delayed auditory feedback. Psychological Bulletin, 1963, 60, 213-232.
Accepted August 23, 1978.
SPEECH DISFLUENCIES AND DELAYED AUDITORY FEEDBACK REACTIONS OF STUTTERING AND NON-STUTTERING CHILDREN1 BEVERLY A. TIMMONS AND JAMES P. BOUDREAU' University of Victoria, Victoria, B. C. Summasy.-25 male stutterers and 25 male non-stutterers matched by age and speaking task, read or recited under conditions of normal and 113-, 2 2 6 , 3 0 6 , 413-, and 520-msec. delayed auditory feedback. Disfluency counts were correlated with delayed auditorp feedbadt reactions which were changes in disfluencies under delay conditions. Pearson product-moment correlations were negative and significant for the combined group of stutterers and non-stutterers under all delays used. Correlations for stutterers were negative and significant for 113, 226, 306, and 413 msec. delay. For the total group of non-stutterers, all correlations were negative and significant. Correlations for age groups within the stuttering and non-stuttering groups were also presented.
Are the number of disfluencies in speech under conditions of normal feedback related to reactions of stuttering and non-stuttering children under delayed auditory feedback? Compared with normal feedback, the speech disfluencies of non-stuttering adults have been shown to increase under conditions of delayed auctitory feedback (Fairbanks & Guttman, 1958; Chase, Sutton, Rapin, Standfast, & Harvey, 1959). However, marked variability in the effect of delayed auditory feedback on non-stutterers' disfluencies has been reported (Beaumont & Foss, 1957; Atkinson, 1954; Yates, 1963; Van Riper, 1971). Atkinson (1954) reported that those adults with high intelligibility in their speech under normal speaking conditions were less affected by delayed auditory feedback than those with low intelligibility under normal conditions. Delayed auditory feedback has been reported to facilitate fluent speech in adult stutterers (Chase, Sutton, & Rapin, 1961; Lotzmann, 1961; Goldiamond, 1966; Soderberg, 1969). Variable effects of delayed auditory feedback have been noted within groups of adult stutterers (Ham & Steer, 1967; Neelley, 1961; Lotzmann, 1961; Chase, Sutton, First, & Zubin, 1961). Ham and Steer (1967) indicated that those stutterers judged severe "reduced blocks during any type of feedback alteration" while mild and moderate stutterers varied in their reactions. The intelligibility of both stutterers and non-stutterers has been observed to be related to reaction to delayed auditory feedback, but correlations between disfluencies under normal ,and delayed auditory feedback have not been reported. Timmons and Boudreau (1978) indicated that groups of stutterers and non-stutterers did not differ in their reactions to delayed auditory feedback. Intelligibility, as measured by disfluency counts, irrespective of speech group'This study was supported by a grant from the Spencer Foundation, Chicago, Illinois. W e acknowledge work by Ann Tamboline, Research Assistant, in collection of data.
860
B. A. TIMMONS
&
J. P. BOUDREAU
ing, might be related to reactions to delayed auditory feedback. A re-exarnination of those data, using correlational methods, seems warranted, since delayed auditory feedback is being used in therapeutic programs for stutterers (Goldiamond, 1966; Van Riper, 1971; Ryan & Van Kirk, 1974).
METHOD Twenty-five male stutterers and 25 male non-stutterers constituted the sample. Diagnosis of stuttering was made by speech therapists employed by the Tacoma, Washington, public schools from which the sample was drawn. Stutterers were matched with non-stutterers by age and speaking task. The pairs ranged in age from 5 to 13 yr., with a mean of 9 yr. The speech of each subject was recorded under conditions of normal feedback and with delays of 113, 226, 306, 413, and 520 msec. Delays were presented in random order. , T h e apparatus used to record speech and to produce delayed auditory feedback was a Korting Constellation 8 with a moveable playback head. A Korting 284 dynamic microphone picked up speech which was returned through a Sharpe 660 headset of 8 ohms nominal impedance. The overall band pass of the instrument was 30 to 19 kHz. The feedback intensity was approximately 75 d b above background noise with slight variations produced by the subjects and limitations in the equipment. The materials used as stimuli for verbal output were taken from the SRA Elementary Reading Laboratory (Parker, 1958). Reading materials were at least 1 yr. below the independent reading level of the subjects. A nursery rhyme was used with non-readers or those reading below the third grade level. The two groups were measured under conditions of normal and delayed feedback, using number of disfluencies (omissions, repetitions, substitutions, and additions). Disfluencies were counted by two independent observers, whose reliability was .92 as determined by a Pearson product-moment correlation. Difference scores were obtained for the number of disfluencies under normal and delayed feedback as a measure of reaction to delayed auditory feedback. Pearson product-moment correlations were calculated. RESULTSAND DISCUSSION Pearson product-moment correlations were calculated, using disfluency counts under normal feedback and difference scores indicating reaction to delayed auditory feedback (see Table 1). Negative correlations for stutterers were statistically significant under 1 1 5 , 226-, 306-, and 413-msec. delay conditions ( < 0 1 When age groups within the stutterers were investigated, correlations for 5- to 8-yr.-olds were negative and statistically significant. Correlations for 1 1 5 , 226-, 306-, and 520-msec. delays were significant ( p < .01); the correlation for 413 msec. was significant ( p < .05). For 9- to 13-yr.-olds, correlations for 1 1 5 , 226-, and 306-msec. delay were significant ( p < .01), while the correlation at 413 msec. was significant ( 9 < .05).
SPEECH DISFLUENCIES AND DELAYED AUDITORY FEEDBACK
861
TABLE 1 PEARSON PRODUCT-MOMENT CORRELATIONS BETWEEN DISFLUENCIES IN SPEECH UNDERNORMAL C O N D I ~ O N AND S DIFFERENCE SCORES SHOWINGREACTION TO DELAYED AUDITORYFEEDBACK Group
D
113 Stutterers 5- to f~-~r.-olds 9- to 13-yr.-olds Non-sruaerers 5- to 8-yr.-olds 9-to 13-yr.-olds All Subjects 'p < .05. t p < .Ol.
25
10 15 25 10 15 50
-.81t -.88t -.89t -.71t -.64*
-.88t -.81t
Delay of Feedback (msec.) 226 306 413 -.84$ -.86t -.87t -.78t -.60* -.86t -.82t
-.77t -.83t -.73t -.60t -.01 -.86t -.72t
-.Got -.65* -.59* -.31 .36 -.76t -.48t
520 -.28 -.83t -.40* .57 -.53t -.40t
Negative correlations were also found for the non-stuttering group. Correlations were statistically significant (p < .01) for 113-, 226-, and 306-msec. delays, while the correlation for 520 msec. was significant ( p < .05). For the 5- to 8-yr.-olds within the non-stuttering group, the correlation for scores at 113-msec. delay was significant ( p < .05 ) . None of the other correlations differed significantly from zero. In the 9- to 13-yr.-old group, correlations at delays of 113, 226, 306, and 413 msec. were significant ( p < .Ol), while the correlation for 520-msec. delay was significant ( p < .05). Age differences in reaction to specific delays were more noticeable for the non-stuttering group than for the stuttering group, with non-stuttering 5- to 8-yr.-olds being significantly affected by only one delay. Age differences in reactions of non-stutterers to delay have been noted in the literature (Chase, Sutton, First, & Zubin, 1961; MacKay, 1968; Buxton, 1969; Timmons & Boudreau, 1976). When both stutterers and non-stutterers were considered together, correlations for scores at all delays were negative and statistically significant ( p < .01). These data confirm the observations that for both stutterers' (Ham & Steer, 1967) and non-stutterers' (Atkinson, 1954) reactions to delayed auditory feedback are negatively related to intelligibility under conditions of normal feedback. However, use of difference scores, as Cronbach and Furby (1970) pointed out, may contribute to the high negative values observed here. Delayed auditory feedback seems to be useful when dealing with severe problems of disfluency, whether or not they are diagnosed as stuttering. Further investigation of this therapeutic tool is necessary to demonstrate its effectiveness with speech difficulties other than stuttering. However, there may be a point (related to degree of disfluency) at which diminishing returns make the delays used in this study inappropriate for therapy. The sample tested is not considered large enough to determine whether, or at what point, these delays become of little use.
B. A. TIMMONS
&
J. P. BOUDREAU
REFERENCES ATKINSON,C. J. Some effects' on intelligibility as the sidetone level and amount of sidetone delay are changed. Iowa Academy of Science, 1954, 61, 334-340. BEAUMONT,J. T., & FOSS, B. M. Individual differences in reacting to delayed auditory feedback. British Journal of Psychology, 1957, 48, 85-89. BUXTON, L. F. An investigation of sex and age differences in speech behaviors under delayed auditory feedback. Unpublished doctoral dissertation, Ohio State Univer.. 1969. CHASE, R. A., SUITON, S., FIRST, D., & ZUBIN, J. A developmental study of changes in behavior under delayed auditory feedback. Journal of Genetic Psychology, 1961, 99, 101-112. CHASE, R. A., S u r r o ~ ,S., & RAPIN, J. Sensory feedback influences on motor performance. Journal of Auditory Research, 1964, 3, 212-223. CHASE, R. A., SUTON, S., RAPIN, J., STANDFAST,S., & K~RVBY, S. Sensory feedback influences on motor performance. Communications Lab., Columbia Universiry and Department of Biometrics Research, State of New York, Research Report, 1959. CRONBACH,L. J., & FURBY, L. How we should measure "changew---or should we? Psychological Bulletin, 1970, 74, 68-80. FAIRBANKS,G., & GU'ITMAN. N. Effects of delayed auditory feedback upon articulation Journal of Speech and Hearing Research, 1958, 1, 12-22. GOLDIAMOND, I. Stuttering and fluency as mani ulatable operant response classes. In L. Krasner & L. P. Ullmann (Eds.), ResearcK in behavior modification. New York: Holt. Rinehart &Winston, 1966. Pp. 106-156. H m ,R., & STEER, M. D. Certain effects of alterations in auditory feedback. Folia Phoniatrica, 1967, 19, 53-62. LO'IZMANN,G. Zur Anwendung variierter Verzogemngszeiten bei Balbuties. Folia Phoniatrica, 1961, 13, 276-310. MACKAY, D. G. Metamorphosis of a critical interval: age-linked changes in delay in auditory feedback that produces maximal disruption of speech. Journal o f the Acoustical Society of America, 1968, 19, 811-821. NEELLEX, J. N. A study of speech behavior of stutterers and non-stutterers under normal and delayed auditory feedback. Journal o f Speech and Hearing Disorders, Monograph Supplement, 1961, 7 , 63-82. PAR=, D. H. SRA Reading, laboratory, elementary edition, teacher's handbook. Chicago: Science Research Assoc., 1958. RYAN, B., & VAN KIRK, B. The establishment, transfer, and maintenance of fluent speech in 50 stutterers using delayed auditory feedback and operant procedures. journal o f Speech and Hearing Disorders, 1974, 39, 3-10. SODERBERG, G. A. Delayed auditory feedback and the speech of stutterers: a review of studies. Journal o f Speech and Hearing Disorders, 1969, 34, 20-29. TIMMONS,B. A., & BOUDRBAU,J. P. Age, sex, and delay time as factors affecting reaction to delayed auditory feedback. Perceptual and Motor Skills, 1976, 43, 27-30: TIMMONS,B. A., & BOUDREAU,J. P. Delayed auditory feedback and the speech of stuttering and non-stuttering children. Perceptual and Motor Skills, 1978, 46, 551555. VAN RIPEX, C. T h e nature of stuttering. Englewood Cliffs, N . J.: Prencice-Hall, 1971. YATES, A. J. Delayed auditory feedback. Psychological Bulletin, 1963, 60, 213-232.
Accepted August 23, 1978.
