Delayed Auditory Feedback Research Paper
Journal of Fluency Disorders 27 (2002) 187–201
The effect of fast speech rate on stuttering frequency during delayed auditory feedback
Garen Sparks a,∗ , Dorothy E. Grant a , Kathleen Millay a , Delaina Walker-Batson a , Linda S. Hynan b b Department of Communication Sciences and Disorders, Texas Woman’s University, Dallas, TX, USA Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, TX, USA Received 26 June 2000; received in revised form 1 October 2001; accepted 14 March 2002
a
Abstract Delayed auditory feedback (DAF) has been documented to improve fluency in those who stutter. The increased fluency has been attributed to the slowed speech rate induced by DAF, but recent experiments have suggested …show more content…
that increasing the speech rate may also decrease stuttering under DAF. This investigation described the effect of combining a fast speech rate and DAF on the fluency of four people who stutter. Fluency of the two mildly dysfluent subjects was the same for both no DAF and DAF conditions at normal and at fast oral reading rates. In contrast, the two severely dysfluent subjects improved in fluency from the no DAF to the DAF conditions. They were found to be dysfluent at both normal and fast oral reading rates without DAF. The results of the study point to the need for further research on the relationship between speech rate and stuttering frequency under conditions of DAF and no DAF. Educational objectives: Readers will learn about and be able to describe how the frequency of stuttering is affected by: (1) speech rates; (2) DAF; and (3) how stuttering severity influences such effects. © 2002 Elsevier Science Inc. All rights reserved.
Keywords: Stuttering frequency; Normal and fast speech rates; Delayed auditory feedback; Non-altered auditory feedback
Corresponding author. Present address: Department of Communication Sciences and Disorders, University of Texas at Dallas, 5349 Amesbury #504, Dallas, TX 75206, USA. E-mail address: garen@utdallas.edu (G. Sparks). 0094-730X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 9 4 - 7 3 0 X ( 0 2 ) 0 0 1 2 8 - 6
∗
188
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
1. Introduction In his book The Structure of Scientific Revolutions, Kuhn (1970) explained that during the early development of a study paradigm, many researchers investigate the same phenomenon, yet form different interpretations of the data obtained and come to different conclusions regarding it. He further explained that ongoing investigation of any scientific question eventually exposes an irregularity, or an unexpected finding, which violates the investigator’s expected outcome of the experiment, causing more questions about the reigning model of thought on the issue. The exposure of an anomaly in the investigator’s paradigm creates a “crisis” which calls for a scientific revolution. Research investigating the phenomenon of stuttering therapy and the importance of the fluency enhancing effects of using the technique of delayed auditory feedback (DAF) has followed the path set forth by Kuhn and has recently created a “crisis.” The anomaly that has recently created the “crisis” in stuttering research can be found in reports about the influence of auditory feedback and its relationship to fluency enhancement. Initially, researchers were driven by the notion that auditory feedback deficits were causal to stuttering (Butler & Stanley, 1966; Mysak, 1960; Webster & Lubker, 1968; Wolf & Wolf, 1959; Yates, 1963). Many investigators attempted to explain that stuttering behavior could be reversed through the fluency enhancing effects of DAF. Despite the many suggestions that the cause of stuttering might be related to the individual’s auditory feedback abilities, interest in the study of the fluency enhancing effects of DAF has declined. Several reasons can be offered in explanation of the decline of interest related to the fluency enhancing effects of DAF. Criticism developed regarding the explanations for the success of DAF therapy. Borden (1979) argued that auditory feedback during speech production was too slow for continuous correction to be useful to speech sound production. Starkweather (1987) pointed out that the effects of DAF were variable, unpredictable, and even unreplicable among some stutterers. Another reason for the decline in interest in the fluency enhancing effects of DAF pertains to Goldiamond’s (1962, 1965) widespread clinical application of DAF as a tool to produce prolonged speech. Following Goldiamond’s lead, some researchers began to emphasize factors other than DAF that could be responsible for producing stutter free speech and the interest in studying the effects of DAF through research declined. For example, Perkins (1979) insisted that DAF was only effective as a tool to produce syllable prolongation, which he believed was the factor responsible for fluency enhancement. He supported this by pointing out that the same effect could be obtained without the use of DAF as long as the technique of syllable prolongation was used. One of the most powerful causes of the decline in research interest in fluency enhancement when using DAF can be attributed to Wingate (1970, 1976). He proposed that the fluency enhancing effect of DAF was due to a reduced speech rate and was a motoric by-product. The Wingate hypothesis was accepted as true
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
189
and, consequently, delivered a death blow to continued research on the fluency enhancing effects of DAF. Wingate’s (1970, 1976) notion that all of the fluency enhancement effects of DAF could be attributed to speech rate reduction is presently being questioned (Kalinowski, Armson, Roland-Meiszkowski, Stuart, & Gracco, 1993; Kalinowski, Stuart, Sark, & Armson, 1996; MacLeod, Kalinowski, Armson, & Stuart, 1995). Several points of conflict have arisen. Wingate’s position was based on the assumption that all speakers under DAF slow their speech rate. However a review of studies by Soderberg (1969) shows that this is not always true. Additionally, the method that was used to measure speech rate in many of the earlier studies is currently being questioned. Duration of total speaking time was the common method of measuring speech rate. Any time-consuming disruption in speech flow, such as with syllable and whole word repetition, increased the duration measurement of speech. Recent experiments however (Kalinowski et al., 1993, 1996; MacLeod et al., 1995) have shown evidence that a slow speech rate is not necessary for the amelioration of stuttering under various fluency evoking conditions. This implies that auditory feedback plays a definite role in fluency enhancement. However, the relatively small number of subjects on whom this effect has been shown heightens the need for further investigation. The current study used procedures similar to those of Kalinowski et al. (1993, 1996) and MacLeod et al. (1995) to investigate the role of speech rate on stuttering frequency under conditions of DAF. The finding of decreased stuttering behavior while a stutterer reads rapidly under DAF conditions would suggest that the fluency enhancing effect of DAF is not solely contingent on decreasing speech rate. This study’s research questions ask whether (a) stutterers improve in fluency, over the no DAF conditions, while speaking at normal oral reading rates under conditions of 55, 80, and 105 ms auditory feedback delays, and (b) whether stutterers improve in fluency, over the no DAF conditions, while speaking at fast oral reading rates under conditions of 55, 80, and 105 ms auditory feedback delays.
2. Method 2.1. Participants One 13-year-old (Subject A), two 19-year-olds (Subjects C and D), and one 21-year-old (Subject B) participated in the study. All the participants were male. The first two subjects (Subjects A and B) were identified as having mild dysfluency, and the last two subjects (Subjects C and D) were identified as having severe dysfluency according to the Stuttering Severity Instrument-3 (Riley, 1994). All subjects were recruited from the Dallas-Ft. Worth area through the primary investigator’s contact with their previous speech–language pathologist. All had a history of receiving speech therapy for remediation of dysfluency through their school systems.
190
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Only Subject C had been enrolled in a therapy program for at least 1 year prior to data collection.
None of the subjects reported having had any experience with DAF during speech therapy. All of the subjects passed a bilateral hearing screening test and showed sensitivity equal to or better than 20 dB HL for the frequencies of 250, 500, 1000, 2000, 4000, and 8000 Hz (ASHA, 1990). They also had normal middle ear function bilaterally according to impedance screening (ANSI, 1989). Standard American English was the first language of each participant. Subject B reported that he was initially diagnosed with attention deficit hyperactive disorder at 7 years of age and was rediagnosed with attention deficit disorder at 18 years of age. Subject C’s conversational speech included characteristics of Black English. No other subject had any additional speech, language, or neurological differences. 2.2. Materials All testing and data collection were conducted in one side of a double-walled, sound-treated, audiometric test suite. The DAF conditions (55, 80, and 105 ms delays) were created by a Phonic Mirror mini DAF (Model PM 505) and the gain was arbitrarily set at 20 dB for each subject. A microphone for the DAF (Phonic Ear AT164-l omni-directional lapel microphone) was attached to the subjects’ clothing approximately 4–6 in. below the chin. All the subjects’ speech samples were video and audio recorded with a camera (Panasonic VHS Reporter AG-185 on a Bogen tripod Model 3126) and audio …show more content…
recorded with a cassette recorder (Sony Model TCM-929). An additional microphone for the audio recorder (Labtec AM-232 lapel microphone) was also attached to the subjects’ clothing approximately 4–6 in. below the chin. 2.3. Procedures Before testing began in the audiometric suite, all subjects were asked to complete a short questionnaire (Guitar, 1998) concerning the characteristics of their fluency disorder and history of their speech therapy treatment. In the control condition, the subjects read aloud four different 200–210 syllable passages taken from Read All About It! (Bisset & Fino, 1988) at both a normal and fast reading rate without DAF. Normal oral reading rate was explained to the subjects as the individual’s usual, or normal, oral reading rate while speaking naturally. Fast oral reading rate was explained to the subject as his fastest reading aloud rate that would maintain intelligibility. The primary investigator demonstrated a normal and a fast oral reading rate using a reading passage from the Riley Stuttering Severity Instrument for Children and Adults-3. The subjects were instructed not to use any strategies that they might have learned in therapy, such as gentle onset or slowed speech, to control or reduce their stuttering. Then, the subjects were asked to read each of the four passages at both a normal and fast oral reading rate for each of the three conditions of auditory feedback delay (55, 80, and 105 ms). Each passage read differed from the previous passage read in terms of the content of the material, rate,
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
191
and delay used. The delays were always presented in a 55, 80, and 105 ms delay sequence, and the reading rates always alternated from normal to fast. Before and after each passage was read, the subjects were instructed to keep silent while the next DAF delay was selected for the following reading condition so as to decrease the possibility of creating a practice and improved fluency effect through extended use of the DAF. Stuttering frequency was based on the first 150 syllables read. Any part-word repetition, prolongation, or block was considered a stuttering episode. A sample of the first 50 fluent syllables, representative of the normal or fast oral reading rate, was taken from each paragraph read under each of the conditions to be analyzed to determine speech rate (i.e., 50 syllables/s). Sample duration was determined by the primary investigator who timed the beginning of the first syllable through the last syllable of the sample using a stopwatch, rounding to the nearest tenth of a second. In a few cases, samples containing 50 continuous fluent syllables could not be obtained; therefore speech rate was not determined for these situations. After a period of 3 weeks, the same no DAF and DAF procedures described above were conducted using four different 200–210 syllable reading passages, also taken from Read All About It! (Bisset & Fino, 1988). A second procedure was conducted in part to confirm or challenge the patterns of performance of the participants from the first procedure. Means for the data gathered in both the first and second recording sessions were then obtained. In total, eight different passages were read under eight different reading conditions by each subject during both of the recording sessions. The eight reading conditions included (1) No DAF: Normal Rate, (2) No DAF: Fast Rate, (3) 55 ms DAF: Normal Rate, (4) 55 ms DAF: Fast Rate, (5) 80 ms DAF: Normal Rate, (6) 80 ms DAF: Fast Rate, (7) 105 ms DAF: Normal Rate, and (8) 105 ms DAF: Fast Rate.
3. Results The results of an experiment on the fluency enhancing characteristics of DAF under conditions of normal and fast oral reading rates in four stutterers are presented in this section. The primary purpose of this investigation was to determine whether the use of a fast oral reading rate would increase stuttering under the fluency enhancing conditions of DAF. Each subject’s number of syllables spoken fluently per second and number of dysfluent episodes during normal and fast oral reading rates with no DAF and the three DAF delays are displayed separately in the four tables in Appendix A. The average performances of the mild and severe stuttering pairs will be compared below. 3.1. Mild stutterers compared to severe stutterers Data were grouped according to stuttering severity to determine whether mild stutterers (Subjects A and B) differed from severe stutterers (Subjects C and D)
192
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Fig. 1. Mild stutterers compared to severe stutterers: mean rate of utterance in syllables/s obtained from 16 passages read at normal oral reading rates with and without DAF.
in terms of the effect of DAF on evoking fluency at either a normal or a fast oral reading rate. Figs. 1–4 present the data obtained for the mild and severe stutterers of this investigation. Normal oral reading rate decreased for the mild stutterers from 4.8 syllables/s under no DAF to 4.6 syllables/s for the 55 and 80 ms delays in auditory feedback. The mild stutterers obtained an average rate of 4.4 syllables/s with the 105 ms delay in feedback. The severe stutterers did not produce any 50 continuously fluent syllable samples during the no DAF reading conditions, thus normal and fast reading rates were not calculated. Normal oral reading rate decreased for the severe stutterers from 4.1 syllables/s in the 55 and 80 ms delay conditions to 3.9 syllables/s in the 105 ms delay conditions. Fast oral reading rates decreased for the mild stutterers from 5.5 syllables/s under no DAF to 5.3 syllables/s for 55 and 80 ms DAF. The mild stutterers continued to decrease in speech rate producing 5.2 syllables/s in the 105 ms delay condition.
Fig. 2. Mild stutterers compared to severe stutterers: mean rate of utterance in syllables/s obtained from 16 passages read at fast oral reading rates with and without DAF.
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
193
Fig. 3. Mild stutterers compared to severe stutterers: mean number of stuttering episodes occurring in 16 passages read at a normal oral reading rate with and without DAF.
Mean fast oral reading rate could not be calculated for the severe stutterers from the no delay conditions because 50 continuously fluent syllables could not be obtained.
Fast oral reading rate decreased for the severe stutterers from 5.5 syllables/s in the 55 ms DAF condition to 5.1 syllables/s in the 80 and 105 ms DAF conditions. Stuttering episodes remained fairly low for the mild stutterers under no DAF while reading at their normal rates. A mean of 1.3 stuttering episodes was found for the no DAF conditions and means of 0.5, 0.9, and 0.9 stuttering episodes were found for 55, 80, and 105 ms DAF, respectively. Stuttering episodes during normal reading rates decreased for the severe stutterers from a mean of 15.2 stuttering episodes in the no DAF condition to means of 1.2, 1.0, and 1.0 stuttering episodes in the 55, 80, and 105 ms DAF conditions, respectively. The number of stuttering episodes decreased slightly for the mild stutterers reading at their fast rates from the no DAF to the DAF conditions. An average
of
Fig. 4. Mild stutterers compared to severe stutterers: mean number of stuttering episodes occurring in 16 passages read at a fast oral reading rate with and without DAF.
194
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
1.3 stuttering episodes were produced during the no DAF conditions and an average of 0.7 stuttering episodes were produced during both the 55 and 80 ms delays in auditory feedback. In the 105 ms delay condition, the mild stutterers showed an average of 1.1 stuttering episodes. Stuttering episodes decreased considerably from the no DAF conditions to the DAF conditions when severe stutterers used a fast oral reading rate. An average of 27.6 stuttering episodes was found for the no DAF conditions and an average of 1.3 stuttering episodes was found for the 55 ms delay condition. Averages of 1.5 and 1.4 stuttering episodes were found for the 80 and 105 ms delays, respectively. 3.2. Reliability measures At the time of the study, the rater (first author) was enrolled in an intensive graduate fluency course in which calculation of fluency rate and dysfluency for people who stutter using both video- and audio-tapes was monitored for accuracy on a consistent basis. In addition, the professor of the course, who is an experienced speech–language pathologist with fluency disorders, was a member of the research committee and supervised and reviewed the results of this research study. To assess intra-reliability, each session for all subjects was rated two times. After each experimental session, the audio- and video-tapes were reviewed by the rater and analyzed (i.e., stuttering frequency and oral reading rate). One week after the first ratings were performed, the audio- and video-tapes from each experimental session were again reviewed and analyzed independently from the first ratings. Intra-rate reliability was assessed for each participant separately using test–retest reliability using a Pearson Product Moment Correlation Coefficient and intra-class correlation coefficient (ICC) (Bartko, 1994; Bartko & Carpenter, 1976). The test–retest reliability coefficient provides a measure of the association between two measurements, while the ICC provides a measure of agreement
Table 1 Subjects A–D: overall test–retest reliability and intra-class correlation coefficients Intra-reliability measures Test–retest correlation coefficient P < 0.0001 Intra-class correlation coefficient P < 0.0001 N Rate (syllables/s) A 0.972 – >0.999 – 64 B 0.981 – 0.980 – 64 C 0.976 – 0.975 – 54∗ D 0.983 – 0.983 – 48∗ Stuttering episodes A 0.919 – 0.917 – 64 B 0.864 – 0.860 – 64 C 0.998 – 0.998 – 64 D 0.996 – 0.995 – 64
Asterisk (∗) indicates that 50 continuously fluent syllables were not always produced (i.e., speech rate was not calculated), thereby affecting the number of observations that could be statistically compared.
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
195
comparing the between-subject variance to the within-subjects variance from an appropriate analysis of variance model. The overall results for each subject by session for each type of measure are provided in Table 1. Both the test–retest reliability and ICCs were found to be very high and very similar to each other (P < 0.0001). These overall results were found to be similar to those for subject by session for reading rate (normal versus fast), auditory delay (0, 55, 80, and 105 ms), and passage (paragraph #s 1–4).
4. Discussion Research question 1 of this investigation asked whether stutterers reading at a normal rate would improve in fluency under conditions of 55, 80, and 105 ms delays in auditory feedback as compared to their performances under conditions of no DAF. The two mildly dysfluent participants of this study showed minimal improvements in fluency with the application of delays in auditory feedback. The two severely dysfluent participants, however, showed an obvious decrease in stuttering frequency under the DAF conditions for the normal reading rate conditions. Research question 2 of this investigation asked whether stutterers would improve in fluency while using a fast oral reading rate under conditions of no delay and 55, 80, and 105 ms delays in auditory feedback. The two mildly dysfluent subjects showed little improvement in fluency while using a fast oral reading rate under conditions of no DAF and under each of the DAF conditions. The two severely dysfluent participants, however, markedly improved fluency while reading at a fast rate under the influence of each of the DAF times. Thus, the findings of the present study provide support for the conclusions drawn in the studies conducted by Hargrave, Kalinowski, Stuart, Armson, and Jones (1994); Kalinowski et al. (1993, 1996); Kalinowski, Armson, and Stuart (1995); and MacLeod et al. (1995) that a reduced rate of speech is not necessary for improvement in fluency during conditions of DAF. Also, the effect of speech rate on stuttering frequency under conditions of no DAF was observed in the data collected for this investigation. It was found that changing from the normal to the fast reading speech rate conditions under no DAF did not affect the stuttering frequency of the two participants who were identified as mildly dysfluent. The performances of the two severely dysfluent participants, however, were found to be affected negatively in terms of an increase in stuttering frequency when asked to read at a fast rate with no DAF. This finding supports the work of Vanryckehem, Glessing, Brutten, and McAlindon (1999) who reported that a significant increase in stuttering frequency occurred under no DAF among the slow, normal, and fast rate conditions for their most dysfluent stutterers. Vanryckehem et al. (1999) reported that stuttering frequency did not increase for their least dysfluent stutterers under these same conditions. The current findings of the study also support Kalinowski et al. (1995) who found a numeric increase in stuttering frequency between normal and fast reading rates under no
196
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
DAF. These studies collectively imply that there may be a relationship between stuttering frequency and speech rate depending on severity of dysfluency (i.e., a group effect). The finding that fluency can improve under DAF without slowing speech rate implies that the improvement may be influenced not only by a motor component but also, perhaps, a sensory (auditory) component. The effect of fluency improvement under DAF at fast speech rates was more prominent in the severely dysfluent subjects of this study than the mildly dysfluent subjects. The reason for this is presently unknown, and points to the need for further research. The findings of this study, that mild stutterers did not increase stuttering frequency when they used a fast oral reading rate under no DAF, but the severe stutterers did, implies that there may be considerable variability among subjects with regard to the effects of rapid speaking rates. Some stutterers, for example, may rely more on auditory feedback than others to achieve fluency (Yates, 1963). Findings that fast speech rate with and without DAF may affect fluency as a function of the degree of stuttering severity warrants consideration in clinical management procedures. The use of DAF in stuttering therapy should be revisited. Many stuttering treatment programs support the idea that stuttering is due to a type of neuromotor timing disorder (Kent, 1984; Perkins, 1979). These programs emphasize slowing the speaking rate of those who stutter. It assumes stutterers can achieve fluency without the use of DAF and other alterations in auditory feedback because they can be taught specific motoric strategies such as a slowed speech rate during therapy. DAF is thought to improve the fluency of stutterers by simplifying their speech coordination through slowed transitional movements from sound to sound and/or through longer planning time for speech movement coordination. However, this study and Vanryckehem et al. (1999) have found that people who are more dysfluent tend to be more affected by the demand or pressure associated with increased speech rate than the people who are less dysfluent. Therefore, speech rate reduction as a therapy treatment may only be effective for those who stutter more frequently. DAF may be the key that allows severely dysfluent participants to improve in fluency when using faster (i.e., more natural) speaking rates (Hargrave et al., 1994; Kalinowski et al., 1993, 1996; MacLeod et al., 1995). Further, the need for more research into the use of DAF in therapy has been recently emphasized by other findings. It has been reported that motorically-based therapy programs have an adverse effect on speech naturalness. In other words, the speech produced by conscious alterations to the speech motor plan is typically perceived to sound unnatural (Franken, Boves, Peters, & Webster, 1992; Kalinowski, Noble, Armson, & Stuart, 1994). In contrast, the fluent speech produced using DAF in experiments has been evaluated by both the subjects and the experimenters to sound natural. These studies imply that the use of DAF therapy may be most beneficial to those who have difficulty producing natural sounding speech and/or those clients who have difficulty in maintaining “motoric” targets. Finally, Hargrave et al. (1994), Kalinowski et al. (1993, 1995, 1996), and MacLeod et al. (1995) supported the idea that auditory alterations such as DAF
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
197
produce an immediate amelioration of stuttering behavior whereas speech motor targets have to be taught over time. They also pointed out that most subjects have reported a feeling of invulnerability to stuttering under conditions of altered auditory feedback at both normal and fast speech rates. This feeling was rarely reported, however, when stutterers relied solely on motoric strategies, even when their speech production sounded relatively fluent. From this observation they concluded that a sense of invulnerability to stuttering results from changes in neural organization which are precipitated by alterations in auditory feedback. One research avenue that has been proposed is to explore the effects of an altered auditory prosthetic device as an addition to or an alternative to current stuttering therapy. Other relevant research options include examining the effectiveness of fluency enhancing conditions (alone or combined) using normal and fast rates of speech. For example, the effectiveness of DAF alone compared to DAF in combination with frequency altered feedback could be explored. The current study reported the stuttering severity of each subject determined by the Riley Stuttering Severity Instrument-3 (1994), which is in contrast to past investigations which have reported data in terms of percentages of dysfluency without reporting severity of the stutterers. However, a few aspects of the study raise concern about the representativeness and generalizability of the data. The investigators of the study used 200–210 syllable passages, which meets the requirement of the Riley Stuttering Severity Instrument for Children and Adults-3 (1994) to determine stuttering severity, basing stuttering frequency on the first 150 syllables and calculating speech rate on 50 continuously fluent syllables. Other investigations have used passages of 300–500 syllables in length, basing stuttering frequency on only the first 300 syllables (Kalinowski et al., 1996). However, it is important to note that despite the shortened criteria used by the study compared to earlier investigations, a similar pattern of results was obtained between them. In addition, investigators have argued for the importance of articulatory rate measurements in determining the relationship between stuttering frequency and measurement of speaking rate (Kalinowski, Armson, & Stuart, 1995). This study did not eliminate all possible pause time from the speech samples, such as with the use of a computer aided waveform analysis program, so thus measured speech rate instead of articulatory rate. However, the investigators eliminated all time-consuming stuttering interruptions from speech rate analysis. The investigators did not report normal and fast speech rates under no DAF based on shorter samples than 50 continuously fluent syllables for the severely dysfluent subjects (Subjects C and D) because they felt it was more important to emphasize that the severely dysfluent subjects met the 50 fluent syllable criterion only when reading under DAF and at both normal and fast reading rates. Finally, each passage read during the study differed from the previous passage read in terms of the content of the material, rate, and delay in an attempt to decrease the possibility of task familiarization. The DAF delays in the study proceeded in a 55, 80, and 105 ms order, and the reading rates alternated from normal to fast. The lack of pure randomized delay and reading rates for each subject restricts interpretation of data. For example, increased dysfluency for Subject A from the 55 ms delay to the 105 ms DAF
198
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
delay when using a fast reading rate may be due to longer delay effects, negative adaptation effects, or to some other potential factors (see Table A.1 in Appendix A). Clearly, further research incorporating more subjects with differing fluency levels, varied DAF times, passages of varying length, measures of both articulatory and speech rate, and better randomization procedures is indicated with the ultimate goal of improved therapy practices and/or more options for the remediation of stuttering. Acknowledgments This study was conducted in fulfillment of the thesis requirement for the degree of Master of Science with a major in Speech–Language Pathology at Texas Woman’s University. The authors would like to acknowledge the participants and their families without whom this project would not have been possible. Appendix A Averaged results for Subjects A–D as shown in Tables A.1–A.4.
Table A.1 Subject A: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 4.1 3.8 3.8 3.3 Fast 4.4 4.2 3.9 3.9 Stuttering episodes (first 150 syllables/sample) Normal 1.8 0.6 0.8 0.9 Fast 1.4 0.4 0.4 1.3
Table A.2 Subject B: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 5.4 5.4 5.4 5.4 Fast 6.6 6.4 6.6 6.4 Stuttering episodes (first 150 syllables/sample) Normal 0.8 0.4 0.9 0.9 Fast 1.1 1.0 1.0 0.8
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
199
Table A.3 Subject C: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 –∗ 4.0 4.1 3.8 Fast –∗ 5.3 4.8 4.9 Stuttering episodes (first 150 syllables/sample) Normal 5.1 0.9 1.4 1.4 Fast 27.8 2.0 2.3 2.3
Asterisk (∗) indicates that 50 continuously fluent syllables were not produced; therefore, speech rate was not calculated. Table A.4 Subject D: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 –∗ 4.1 4.0 3.9 Fast –∗ 5.7 5.4 5.3 Stuttering episodes (first 150 syllables/sample) Normal 25.3 1.5 0.5 0.6 Fast 27.4 0.5 0.6 0.4
Asterisk (∗) indicates that 50 continuously fluent syllables were not produced; therefore, speech rate was not calculated.
References
American National Standards Institute. (1989). Specification for audiometers (ANSI S3.6 -1969). New York, NY: American National Standards Institute. American Speech-Language-Hearing Association. (1990). Guidelines for screening for hearing impairments and middle ear disorders. ASHA 32 (Suppl. 2), 17–24. Bartko, J. J. (1994). General methodology II measures of agreement: A single procedure. Statistics in Medicine, 13, 737–745. Bartko, J. J., & Carpenter, W. T., Jr. (1976). On the methods and theory of reliability. The Journal of Nervous and Mental Disorders, 163, 307–317. Bisset, J. D., & Fino, M. S. (1988). Read all about it! Tucson, AZ: Communication Skill Builders. Borden, G. J. (1979). An interpretation of research on feedback interruption in speech. Brain and Language, 7, 307–319. Butler, R. R., & Stanley, P. E. (1966). The stuttering problem considered from an automatic control point of view. Folia Phoniatricia, 18, 33–44. Franken, M. C., Boves, L., Peters, H. F. M., & Webster, R. L. (1992). Perceptual evaluation of the speech before and after fluency shaping stuttering therapy. Journal of Fluency Disorders, 17, 223–241. Goldiamond, I. (1962). The maintenance of ongoing fluent behavior and stuttering. Journal of Mathematics, 1, 57–95. Goldiamond, I. (1965). Stuttering and fluency as manipulatable operant response classes. In L. Krasner & L. P. Ullmann (Eds.), Research in behavior modification (pp. 106–156). New York: Holt, Rinehart & Winston.
200
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Guitar, B. (1998). Stuttering: An integrated approach to its nature and treatment (2nd ed.). Baltimore: Williams & Wilkins. Hargrave, S., Kalinowski, J., Stuart, A., Armson, J., & Jones, K. (1994). Effect of frequency altered feedback on stutterers’ fluency at two speech rates. Journal of Speech and Hearing Research, 37, 1313–1319. Kalinowski, J. S., Armson, J., Roland-Meiszkowski, M., Stuart, A., & Gracco, V. L. (1993). Effect of alterations in auditory feedback and speech rate on stuttering frequency. Language and Speech, 36(1), 1–16. Kalinowski, J., Noble, S., Armson, J., & Stuart, A. (1994). Pretreatment and posttreatment speech naturalness ratings of adults with mild and severe stuttering. American Journal of Speech Language-Pathology, 3, 65–70. Kalinowski, J., Armson, J., & Stuart, A. (1995). Effect of normal and fast articulatory rates on stuttering frequency. Journal of Fluency Disorders, 20, 293–302. Kalinowski, J., Stuart, A., Sark, S., & Armson, J. (1996). Stuttering amelioration at various auditory feedback delays and speech rates. European Journal of Disorders of Communication, 31, 259–269. Kent, R. D. (1984). Stuttering as a temporal programming disorder. In R. F. Curlee & W. H. Perkins (Eds.), Nature and treatment of stuttering: New directions (pp. 283–301). San Diego: College Hill Press. Kuhn, T. S. (1970). The structure of scientific revolutions (2nd ed.). Chicago, IL: University of Chicago Press. MacLeod, J., Kalinowski, J., Armson, J., & Stuart, A. (1995). Effect of single and combined altered auditory feedback on stuttering frequency at two speech rates. Journal of Communication Disorders, 28, 218–226. Mysak, E. D. (1960). Servo-theory and stuttering. Journal of Speech and Hearing Disorders, 25, 188–189. Perkins, W. H. (1979). From psychoanalysis to discoordination. In H. H. Gregory (Ed.), Controversies about stuttering therapy (pp. 97–127). Baltimore, MD: University Park Press. Riley, G. (1994). Stuttering severity instrument for children and adults-3. Austin, TX: Pro-Ed Publications. Soderberg, G. A. (1969). Delayed auditory feedback and the speech of stutterers: A review of studies. Journal of Speech and Hearing Disorders, 34, 20–29. Starkweather, C. W. (1987). Fluency and stuttering. Englewood Cliffs, NJ: Prentice Hall. Vanryckehem, M., Glessing, J. J., Brutten, G. J., & McAlindon, P. (1999). The main and interactive effect of oral reading rate on the frequency of stuttering. American Journal of Speech-Language Pathology, 8, 164–170. Webster, R. L., & Lubker, B. B. (1968). Interrelationships among fluency producing variables in stuttered speech. Journal of Speech and Hearing Research, 11, 754–766. Wingate, M. E. (1970). Effect on stuttering of changes in audition. Journal of Speech and Hearing Research, 13, 861–873. Wingate, M. E. (1976). Stuttering: Theory and treatment. New York: Irvington Publishers. Wolf, A. A., & Wolf, E. G. (1959). Feedback processes in the theory of certain speech disorders. Speech Pathology and Therapy, 2, 48–55. Yates, A. J. (1963). Recent empirical and theoretical approaches to the experimental manipulation of speech in normal subjects and in stammerers. Behavior Research and Therapy, 1, 95–119.
CONTINUING EDUCATION The effect of fast speech rate on stuttering frequency during delayed auditory feedback QUESTIONS 1. It has been hypothesized that stuttering speakers’ fluency is enhanced during delayed auditory feedback (DAF) because it:
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
201
2.
3.
4.
5.
a. corrects such speakers’ auditory feedback deficits b. produces syllable prolongation which reduces stuttering c. is a by-product of a reduction in speaking rate d. all of the above e. none of the above The oral reading rates of these stuttering speakers during 0 ms DAF conditions were: a. similar for both pairs of speakers b. similar to those during DAF for the severe pair c. similar to those during DAF for the mild pair d. much slower in the fast condition for the severe pair than that of the mild pair e. both (a) and (b) This study found that stuttering frequency: a. varied little between normal and fast readings for either pair of speakers b. increased substantially during fast readings for both pairs of speakers c. increased substantially during fast readings for the severe pair d. varied little between normal and fast readings only for the mild pair e. both (c) and (d) This study found that DAF reduced the frequency of stuttering: a. during both normal and fast readings b. similarly in both pairs c. of both pairs the most at the longest delay d. of the mild pair the most at fast speech rates e. both (b) and (c) This study’s findings imply that how much stuttering speakers’ fluency improves under DAF conditions is influenced most strongly by: a. speakers’ reading rates b. speakers’ severity of stuttering c. DAF delay levels d. all of the above e. none of the above
The effect of fast speech rate on stuttering frequency during delayed auditory feedback
Garen Sparks a,∗ , Dorothy E. Grant a , Kathleen Millay a , Delaina Walker-Batson a , Linda S. Hynan b b Department of Communication Sciences and Disorders, Texas Woman’s University, Dallas, TX, USA Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, TX, USA Received 26 June 2000; received in revised form 1 October 2001; accepted 14 March 2002
a
Abstract Delayed auditory feedback (DAF) has been documented to improve fluency in those who stutter. The increased fluency has been attributed to the slowed speech rate induced by DAF, but recent experiments have suggested …show more content…
that increasing the speech rate may also decrease stuttering under DAF. This investigation described the effect of combining a fast speech rate and DAF on the fluency of four people who stutter. Fluency of the two mildly dysfluent subjects was the same for both no DAF and DAF conditions at normal and at fast oral reading rates. In contrast, the two severely dysfluent subjects improved in fluency from the no DAF to the DAF conditions. They were found to be dysfluent at both normal and fast oral reading rates without DAF. The results of the study point to the need for further research on the relationship between speech rate and stuttering frequency under conditions of DAF and no DAF. Educational objectives: Readers will learn about and be able to describe how the frequency of stuttering is affected by: (1) speech rates; (2) DAF; and (3) how stuttering severity influences such effects. © 2002 Elsevier Science Inc. All rights reserved.
Keywords: Stuttering frequency; Normal and fast speech rates; Delayed auditory feedback; Non-altered auditory feedback
Corresponding author. Present address: Department of Communication Sciences and Disorders, University of Texas at Dallas, 5349 Amesbury #504, Dallas, TX 75206, USA. E-mail address: garen@utdallas.edu (G. Sparks). 0094-730X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 9 4 - 7 3 0 X ( 0 2 ) 0 0 1 2 8 - 6
∗
188
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
1. Introduction In his book The Structure of Scientific Revolutions, Kuhn (1970) explained that during the early development of a study paradigm, many researchers investigate the same phenomenon, yet form different interpretations of the data obtained and come to different conclusions regarding it. He further explained that ongoing investigation of any scientific question eventually exposes an irregularity, or an unexpected finding, which violates the investigator’s expected outcome of the experiment, causing more questions about the reigning model of thought on the issue. The exposure of an anomaly in the investigator’s paradigm creates a “crisis” which calls for a scientific revolution. Research investigating the phenomenon of stuttering therapy and the importance of the fluency enhancing effects of using the technique of delayed auditory feedback (DAF) has followed the path set forth by Kuhn and has recently created a “crisis.” The anomaly that has recently created the “crisis” in stuttering research can be found in reports about the influence of auditory feedback and its relationship to fluency enhancement. Initially, researchers were driven by the notion that auditory feedback deficits were causal to stuttering (Butler & Stanley, 1966; Mysak, 1960; Webster & Lubker, 1968; Wolf & Wolf, 1959; Yates, 1963). Many investigators attempted to explain that stuttering behavior could be reversed through the fluency enhancing effects of DAF. Despite the many suggestions that the cause of stuttering might be related to the individual’s auditory feedback abilities, interest in the study of the fluency enhancing effects of DAF has declined. Several reasons can be offered in explanation of the decline of interest related to the fluency enhancing effects of DAF. Criticism developed regarding the explanations for the success of DAF therapy. Borden (1979) argued that auditory feedback during speech production was too slow for continuous correction to be useful to speech sound production. Starkweather (1987) pointed out that the effects of DAF were variable, unpredictable, and even unreplicable among some stutterers. Another reason for the decline in interest in the fluency enhancing effects of DAF pertains to Goldiamond’s (1962, 1965) widespread clinical application of DAF as a tool to produce prolonged speech. Following Goldiamond’s lead, some researchers began to emphasize factors other than DAF that could be responsible for producing stutter free speech and the interest in studying the effects of DAF through research declined. For example, Perkins (1979) insisted that DAF was only effective as a tool to produce syllable prolongation, which he believed was the factor responsible for fluency enhancement. He supported this by pointing out that the same effect could be obtained without the use of DAF as long as the technique of syllable prolongation was used. One of the most powerful causes of the decline in research interest in fluency enhancement when using DAF can be attributed to Wingate (1970, 1976). He proposed that the fluency enhancing effect of DAF was due to a reduced speech rate and was a motoric by-product. The Wingate hypothesis was accepted as true
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
189
and, consequently, delivered a death blow to continued research on the fluency enhancing effects of DAF. Wingate’s (1970, 1976) notion that all of the fluency enhancement effects of DAF could be attributed to speech rate reduction is presently being questioned (Kalinowski, Armson, Roland-Meiszkowski, Stuart, & Gracco, 1993; Kalinowski, Stuart, Sark, & Armson, 1996; MacLeod, Kalinowski, Armson, & Stuart, 1995). Several points of conflict have arisen. Wingate’s position was based on the assumption that all speakers under DAF slow their speech rate. However a review of studies by Soderberg (1969) shows that this is not always true. Additionally, the method that was used to measure speech rate in many of the earlier studies is currently being questioned. Duration of total speaking time was the common method of measuring speech rate. Any time-consuming disruption in speech flow, such as with syllable and whole word repetition, increased the duration measurement of speech. Recent experiments however (Kalinowski et al., 1993, 1996; MacLeod et al., 1995) have shown evidence that a slow speech rate is not necessary for the amelioration of stuttering under various fluency evoking conditions. This implies that auditory feedback plays a definite role in fluency enhancement. However, the relatively small number of subjects on whom this effect has been shown heightens the need for further investigation. The current study used procedures similar to those of Kalinowski et al. (1993, 1996) and MacLeod et al. (1995) to investigate the role of speech rate on stuttering frequency under conditions of DAF. The finding of decreased stuttering behavior while a stutterer reads rapidly under DAF conditions would suggest that the fluency enhancing effect of DAF is not solely contingent on decreasing speech rate. This study’s research questions ask whether (a) stutterers improve in fluency, over the no DAF conditions, while speaking at normal oral reading rates under conditions of 55, 80, and 105 ms auditory feedback delays, and (b) whether stutterers improve in fluency, over the no DAF conditions, while speaking at fast oral reading rates under conditions of 55, 80, and 105 ms auditory feedback delays.
2. Method 2.1. Participants One 13-year-old (Subject A), two 19-year-olds (Subjects C and D), and one 21-year-old (Subject B) participated in the study. All the participants were male. The first two subjects (Subjects A and B) were identified as having mild dysfluency, and the last two subjects (Subjects C and D) were identified as having severe dysfluency according to the Stuttering Severity Instrument-3 (Riley, 1994). All subjects were recruited from the Dallas-Ft. Worth area through the primary investigator’s contact with their previous speech–language pathologist. All had a history of receiving speech therapy for remediation of dysfluency through their school systems.
190
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Only Subject C had been enrolled in a therapy program for at least 1 year prior to data collection.
None of the subjects reported having had any experience with DAF during speech therapy. All of the subjects passed a bilateral hearing screening test and showed sensitivity equal to or better than 20 dB HL for the frequencies of 250, 500, 1000, 2000, 4000, and 8000 Hz (ASHA, 1990). They also had normal middle ear function bilaterally according to impedance screening (ANSI, 1989). Standard American English was the first language of each participant. Subject B reported that he was initially diagnosed with attention deficit hyperactive disorder at 7 years of age and was rediagnosed with attention deficit disorder at 18 years of age. Subject C’s conversational speech included characteristics of Black English. No other subject had any additional speech, language, or neurological differences. 2.2. Materials All testing and data collection were conducted in one side of a double-walled, sound-treated, audiometric test suite. The DAF conditions (55, 80, and 105 ms delays) were created by a Phonic Mirror mini DAF (Model PM 505) and the gain was arbitrarily set at 20 dB for each subject. A microphone for the DAF (Phonic Ear AT164-l omni-directional lapel microphone) was attached to the subjects’ clothing approximately 4–6 in. below the chin. All the subjects’ speech samples were video and audio recorded with a camera (Panasonic VHS Reporter AG-185 on a Bogen tripod Model 3126) and audio …show more content…
recorded with a cassette recorder (Sony Model TCM-929). An additional microphone for the audio recorder (Labtec AM-232 lapel microphone) was also attached to the subjects’ clothing approximately 4–6 in. below the chin. 2.3. Procedures Before testing began in the audiometric suite, all subjects were asked to complete a short questionnaire (Guitar, 1998) concerning the characteristics of their fluency disorder and history of their speech therapy treatment. In the control condition, the subjects read aloud four different 200–210 syllable passages taken from Read All About It! (Bisset & Fino, 1988) at both a normal and fast reading rate without DAF. Normal oral reading rate was explained to the subjects as the individual’s usual, or normal, oral reading rate while speaking naturally. Fast oral reading rate was explained to the subject as his fastest reading aloud rate that would maintain intelligibility. The primary investigator demonstrated a normal and a fast oral reading rate using a reading passage from the Riley Stuttering Severity Instrument for Children and Adults-3. The subjects were instructed not to use any strategies that they might have learned in therapy, such as gentle onset or slowed speech, to control or reduce their stuttering. Then, the subjects were asked to read each of the four passages at both a normal and fast oral reading rate for each of the three conditions of auditory feedback delay (55, 80, and 105 ms). Each passage read differed from the previous passage read in terms of the content of the material, rate,
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
191
and delay used. The delays were always presented in a 55, 80, and 105 ms delay sequence, and the reading rates always alternated from normal to fast. Before and after each passage was read, the subjects were instructed to keep silent while the next DAF delay was selected for the following reading condition so as to decrease the possibility of creating a practice and improved fluency effect through extended use of the DAF. Stuttering frequency was based on the first 150 syllables read. Any part-word repetition, prolongation, or block was considered a stuttering episode. A sample of the first 50 fluent syllables, representative of the normal or fast oral reading rate, was taken from each paragraph read under each of the conditions to be analyzed to determine speech rate (i.e., 50 syllables/s). Sample duration was determined by the primary investigator who timed the beginning of the first syllable through the last syllable of the sample using a stopwatch, rounding to the nearest tenth of a second. In a few cases, samples containing 50 continuous fluent syllables could not be obtained; therefore speech rate was not determined for these situations. After a period of 3 weeks, the same no DAF and DAF procedures described above were conducted using four different 200–210 syllable reading passages, also taken from Read All About It! (Bisset & Fino, 1988). A second procedure was conducted in part to confirm or challenge the patterns of performance of the participants from the first procedure. Means for the data gathered in both the first and second recording sessions were then obtained. In total, eight different passages were read under eight different reading conditions by each subject during both of the recording sessions. The eight reading conditions included (1) No DAF: Normal Rate, (2) No DAF: Fast Rate, (3) 55 ms DAF: Normal Rate, (4) 55 ms DAF: Fast Rate, (5) 80 ms DAF: Normal Rate, (6) 80 ms DAF: Fast Rate, (7) 105 ms DAF: Normal Rate, and (8) 105 ms DAF: Fast Rate.
3. Results The results of an experiment on the fluency enhancing characteristics of DAF under conditions of normal and fast oral reading rates in four stutterers are presented in this section. The primary purpose of this investigation was to determine whether the use of a fast oral reading rate would increase stuttering under the fluency enhancing conditions of DAF. Each subject’s number of syllables spoken fluently per second and number of dysfluent episodes during normal and fast oral reading rates with no DAF and the three DAF delays are displayed separately in the four tables in Appendix A. The average performances of the mild and severe stuttering pairs will be compared below. 3.1. Mild stutterers compared to severe stutterers Data were grouped according to stuttering severity to determine whether mild stutterers (Subjects A and B) differed from severe stutterers (Subjects C and D)
192
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Fig. 1. Mild stutterers compared to severe stutterers: mean rate of utterance in syllables/s obtained from 16 passages read at normal oral reading rates with and without DAF.
in terms of the effect of DAF on evoking fluency at either a normal or a fast oral reading rate. Figs. 1–4 present the data obtained for the mild and severe stutterers of this investigation. Normal oral reading rate decreased for the mild stutterers from 4.8 syllables/s under no DAF to 4.6 syllables/s for the 55 and 80 ms delays in auditory feedback. The mild stutterers obtained an average rate of 4.4 syllables/s with the 105 ms delay in feedback. The severe stutterers did not produce any 50 continuously fluent syllable samples during the no DAF reading conditions, thus normal and fast reading rates were not calculated. Normal oral reading rate decreased for the severe stutterers from 4.1 syllables/s in the 55 and 80 ms delay conditions to 3.9 syllables/s in the 105 ms delay conditions. Fast oral reading rates decreased for the mild stutterers from 5.5 syllables/s under no DAF to 5.3 syllables/s for 55 and 80 ms DAF. The mild stutterers continued to decrease in speech rate producing 5.2 syllables/s in the 105 ms delay condition.
Fig. 2. Mild stutterers compared to severe stutterers: mean rate of utterance in syllables/s obtained from 16 passages read at fast oral reading rates with and without DAF.
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
193
Fig. 3. Mild stutterers compared to severe stutterers: mean number of stuttering episodes occurring in 16 passages read at a normal oral reading rate with and without DAF.
Mean fast oral reading rate could not be calculated for the severe stutterers from the no delay conditions because 50 continuously fluent syllables could not be obtained.
Fast oral reading rate decreased for the severe stutterers from 5.5 syllables/s in the 55 ms DAF condition to 5.1 syllables/s in the 80 and 105 ms DAF conditions. Stuttering episodes remained fairly low for the mild stutterers under no DAF while reading at their normal rates. A mean of 1.3 stuttering episodes was found for the no DAF conditions and means of 0.5, 0.9, and 0.9 stuttering episodes were found for 55, 80, and 105 ms DAF, respectively. Stuttering episodes during normal reading rates decreased for the severe stutterers from a mean of 15.2 stuttering episodes in the no DAF condition to means of 1.2, 1.0, and 1.0 stuttering episodes in the 55, 80, and 105 ms DAF conditions, respectively. The number of stuttering episodes decreased slightly for the mild stutterers reading at their fast rates from the no DAF to the DAF conditions. An average
of
Fig. 4. Mild stutterers compared to severe stutterers: mean number of stuttering episodes occurring in 16 passages read at a fast oral reading rate with and without DAF.
194
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
1.3 stuttering episodes were produced during the no DAF conditions and an average of 0.7 stuttering episodes were produced during both the 55 and 80 ms delays in auditory feedback. In the 105 ms delay condition, the mild stutterers showed an average of 1.1 stuttering episodes. Stuttering episodes decreased considerably from the no DAF conditions to the DAF conditions when severe stutterers used a fast oral reading rate. An average of 27.6 stuttering episodes was found for the no DAF conditions and an average of 1.3 stuttering episodes was found for the 55 ms delay condition. Averages of 1.5 and 1.4 stuttering episodes were found for the 80 and 105 ms delays, respectively. 3.2. Reliability measures At the time of the study, the rater (first author) was enrolled in an intensive graduate fluency course in which calculation of fluency rate and dysfluency for people who stutter using both video- and audio-tapes was monitored for accuracy on a consistent basis. In addition, the professor of the course, who is an experienced speech–language pathologist with fluency disorders, was a member of the research committee and supervised and reviewed the results of this research study. To assess intra-reliability, each session for all subjects was rated two times. After each experimental session, the audio- and video-tapes were reviewed by the rater and analyzed (i.e., stuttering frequency and oral reading rate). One week after the first ratings were performed, the audio- and video-tapes from each experimental session were again reviewed and analyzed independently from the first ratings. Intra-rate reliability was assessed for each participant separately using test–retest reliability using a Pearson Product Moment Correlation Coefficient and intra-class correlation coefficient (ICC) (Bartko, 1994; Bartko & Carpenter, 1976). The test–retest reliability coefficient provides a measure of the association between two measurements, while the ICC provides a measure of agreement
Table 1 Subjects A–D: overall test–retest reliability and intra-class correlation coefficients Intra-reliability measures Test–retest correlation coefficient P < 0.0001 Intra-class correlation coefficient P < 0.0001 N Rate (syllables/s) A 0.972 – >0.999 – 64 B 0.981 – 0.980 – 64 C 0.976 – 0.975 – 54∗ D 0.983 – 0.983 – 48∗ Stuttering episodes A 0.919 – 0.917 – 64 B 0.864 – 0.860 – 64 C 0.998 – 0.998 – 64 D 0.996 – 0.995 – 64
Asterisk (∗) indicates that 50 continuously fluent syllables were not always produced (i.e., speech rate was not calculated), thereby affecting the number of observations that could be statistically compared.
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
195
comparing the between-subject variance to the within-subjects variance from an appropriate analysis of variance model. The overall results for each subject by session for each type of measure are provided in Table 1. Both the test–retest reliability and ICCs were found to be very high and very similar to each other (P < 0.0001). These overall results were found to be similar to those for subject by session for reading rate (normal versus fast), auditory delay (0, 55, 80, and 105 ms), and passage (paragraph #s 1–4).
4. Discussion Research question 1 of this investigation asked whether stutterers reading at a normal rate would improve in fluency under conditions of 55, 80, and 105 ms delays in auditory feedback as compared to their performances under conditions of no DAF. The two mildly dysfluent participants of this study showed minimal improvements in fluency with the application of delays in auditory feedback. The two severely dysfluent participants, however, showed an obvious decrease in stuttering frequency under the DAF conditions for the normal reading rate conditions. Research question 2 of this investigation asked whether stutterers would improve in fluency while using a fast oral reading rate under conditions of no delay and 55, 80, and 105 ms delays in auditory feedback. The two mildly dysfluent subjects showed little improvement in fluency while using a fast oral reading rate under conditions of no DAF and under each of the DAF conditions. The two severely dysfluent participants, however, markedly improved fluency while reading at a fast rate under the influence of each of the DAF times. Thus, the findings of the present study provide support for the conclusions drawn in the studies conducted by Hargrave, Kalinowski, Stuart, Armson, and Jones (1994); Kalinowski et al. (1993, 1996); Kalinowski, Armson, and Stuart (1995); and MacLeod et al. (1995) that a reduced rate of speech is not necessary for improvement in fluency during conditions of DAF. Also, the effect of speech rate on stuttering frequency under conditions of no DAF was observed in the data collected for this investigation. It was found that changing from the normal to the fast reading speech rate conditions under no DAF did not affect the stuttering frequency of the two participants who were identified as mildly dysfluent. The performances of the two severely dysfluent participants, however, were found to be affected negatively in terms of an increase in stuttering frequency when asked to read at a fast rate with no DAF. This finding supports the work of Vanryckehem, Glessing, Brutten, and McAlindon (1999) who reported that a significant increase in stuttering frequency occurred under no DAF among the slow, normal, and fast rate conditions for their most dysfluent stutterers. Vanryckehem et al. (1999) reported that stuttering frequency did not increase for their least dysfluent stutterers under these same conditions. The current findings of the study also support Kalinowski et al. (1995) who found a numeric increase in stuttering frequency between normal and fast reading rates under no
196
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
DAF. These studies collectively imply that there may be a relationship between stuttering frequency and speech rate depending on severity of dysfluency (i.e., a group effect). The finding that fluency can improve under DAF without slowing speech rate implies that the improvement may be influenced not only by a motor component but also, perhaps, a sensory (auditory) component. The effect of fluency improvement under DAF at fast speech rates was more prominent in the severely dysfluent subjects of this study than the mildly dysfluent subjects. The reason for this is presently unknown, and points to the need for further research. The findings of this study, that mild stutterers did not increase stuttering frequency when they used a fast oral reading rate under no DAF, but the severe stutterers did, implies that there may be considerable variability among subjects with regard to the effects of rapid speaking rates. Some stutterers, for example, may rely more on auditory feedback than others to achieve fluency (Yates, 1963). Findings that fast speech rate with and without DAF may affect fluency as a function of the degree of stuttering severity warrants consideration in clinical management procedures. The use of DAF in stuttering therapy should be revisited. Many stuttering treatment programs support the idea that stuttering is due to a type of neuromotor timing disorder (Kent, 1984; Perkins, 1979). These programs emphasize slowing the speaking rate of those who stutter. It assumes stutterers can achieve fluency without the use of DAF and other alterations in auditory feedback because they can be taught specific motoric strategies such as a slowed speech rate during therapy. DAF is thought to improve the fluency of stutterers by simplifying their speech coordination through slowed transitional movements from sound to sound and/or through longer planning time for speech movement coordination. However, this study and Vanryckehem et al. (1999) have found that people who are more dysfluent tend to be more affected by the demand or pressure associated with increased speech rate than the people who are less dysfluent. Therefore, speech rate reduction as a therapy treatment may only be effective for those who stutter more frequently. DAF may be the key that allows severely dysfluent participants to improve in fluency when using faster (i.e., more natural) speaking rates (Hargrave et al., 1994; Kalinowski et al., 1993, 1996; MacLeod et al., 1995). Further, the need for more research into the use of DAF in therapy has been recently emphasized by other findings. It has been reported that motorically-based therapy programs have an adverse effect on speech naturalness. In other words, the speech produced by conscious alterations to the speech motor plan is typically perceived to sound unnatural (Franken, Boves, Peters, & Webster, 1992; Kalinowski, Noble, Armson, & Stuart, 1994). In contrast, the fluent speech produced using DAF in experiments has been evaluated by both the subjects and the experimenters to sound natural. These studies imply that the use of DAF therapy may be most beneficial to those who have difficulty producing natural sounding speech and/or those clients who have difficulty in maintaining “motoric” targets. Finally, Hargrave et al. (1994), Kalinowski et al. (1993, 1995, 1996), and MacLeod et al. (1995) supported the idea that auditory alterations such as DAF
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
197
produce an immediate amelioration of stuttering behavior whereas speech motor targets have to be taught over time. They also pointed out that most subjects have reported a feeling of invulnerability to stuttering under conditions of altered auditory feedback at both normal and fast speech rates. This feeling was rarely reported, however, when stutterers relied solely on motoric strategies, even when their speech production sounded relatively fluent. From this observation they concluded that a sense of invulnerability to stuttering results from changes in neural organization which are precipitated by alterations in auditory feedback. One research avenue that has been proposed is to explore the effects of an altered auditory prosthetic device as an addition to or an alternative to current stuttering therapy. Other relevant research options include examining the effectiveness of fluency enhancing conditions (alone or combined) using normal and fast rates of speech. For example, the effectiveness of DAF alone compared to DAF in combination with frequency altered feedback could be explored. The current study reported the stuttering severity of each subject determined by the Riley Stuttering Severity Instrument-3 (1994), which is in contrast to past investigations which have reported data in terms of percentages of dysfluency without reporting severity of the stutterers. However, a few aspects of the study raise concern about the representativeness and generalizability of the data. The investigators of the study used 200–210 syllable passages, which meets the requirement of the Riley Stuttering Severity Instrument for Children and Adults-3 (1994) to determine stuttering severity, basing stuttering frequency on the first 150 syllables and calculating speech rate on 50 continuously fluent syllables. Other investigations have used passages of 300–500 syllables in length, basing stuttering frequency on only the first 300 syllables (Kalinowski et al., 1996). However, it is important to note that despite the shortened criteria used by the study compared to earlier investigations, a similar pattern of results was obtained between them. In addition, investigators have argued for the importance of articulatory rate measurements in determining the relationship between stuttering frequency and measurement of speaking rate (Kalinowski, Armson, & Stuart, 1995). This study did not eliminate all possible pause time from the speech samples, such as with the use of a computer aided waveform analysis program, so thus measured speech rate instead of articulatory rate. However, the investigators eliminated all time-consuming stuttering interruptions from speech rate analysis. The investigators did not report normal and fast speech rates under no DAF based on shorter samples than 50 continuously fluent syllables for the severely dysfluent subjects (Subjects C and D) because they felt it was more important to emphasize that the severely dysfluent subjects met the 50 fluent syllable criterion only when reading under DAF and at both normal and fast reading rates. Finally, each passage read during the study differed from the previous passage read in terms of the content of the material, rate, and delay in an attempt to decrease the possibility of task familiarization. The DAF delays in the study proceeded in a 55, 80, and 105 ms order, and the reading rates alternated from normal to fast. The lack of pure randomized delay and reading rates for each subject restricts interpretation of data. For example, increased dysfluency for Subject A from the 55 ms delay to the 105 ms DAF
198
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
delay when using a fast reading rate may be due to longer delay effects, negative adaptation effects, or to some other potential factors (see Table A.1 in Appendix A). Clearly, further research incorporating more subjects with differing fluency levels, varied DAF times, passages of varying length, measures of both articulatory and speech rate, and better randomization procedures is indicated with the ultimate goal of improved therapy practices and/or more options for the remediation of stuttering. Acknowledgments This study was conducted in fulfillment of the thesis requirement for the degree of Master of Science with a major in Speech–Language Pathology at Texas Woman’s University. The authors would like to acknowledge the participants and their families without whom this project would not have been possible. Appendix A Averaged results for Subjects A–D as shown in Tables A.1–A.4.
Table A.1 Subject A: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 4.1 3.8 3.8 3.3 Fast 4.4 4.2 3.9 3.9 Stuttering episodes (first 150 syllables/sample) Normal 1.8 0.6 0.8 0.9 Fast 1.4 0.4 0.4 1.3
Table A.2 Subject B: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 5.4 5.4 5.4 5.4 Fast 6.6 6.4 6.6 6.4 Stuttering episodes (first 150 syllables/sample) Normal 0.8 0.4 0.9 0.9 Fast 1.1 1.0 1.0 0.8
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
199
Table A.3 Subject C: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 –∗ 4.0 4.1 3.8 Fast –∗ 5.3 4.8 4.9 Stuttering episodes (first 150 syllables/sample) Normal 5.1 0.9 1.4 1.4 Fast 27.8 2.0 2.3 2.3
Asterisk (∗) indicates that 50 continuously fluent syllables were not produced; therefore, speech rate was not calculated. Table A.4 Subject D: mean rates of utterance in syllables/s and mean number of stuttering episodes for normal and fast oral reading rates and several DAF times Delay (ms) Rate (syllables/s) Normal 0 55 80 105 –∗ 4.1 4.0 3.9 Fast –∗ 5.7 5.4 5.3 Stuttering episodes (first 150 syllables/sample) Normal 25.3 1.5 0.5 0.6 Fast 27.4 0.5 0.6 0.4
Asterisk (∗) indicates that 50 continuously fluent syllables were not produced; therefore, speech rate was not calculated.
References
American National Standards Institute. (1989). Specification for audiometers (ANSI S3.6 -1969). New York, NY: American National Standards Institute. American Speech-Language-Hearing Association. (1990). Guidelines for screening for hearing impairments and middle ear disorders. ASHA 32 (Suppl. 2), 17–24. Bartko, J. J. (1994). General methodology II measures of agreement: A single procedure. Statistics in Medicine, 13, 737–745. Bartko, J. J., & Carpenter, W. T., Jr. (1976). On the methods and theory of reliability. The Journal of Nervous and Mental Disorders, 163, 307–317. Bisset, J. D., & Fino, M. S. (1988). Read all about it! Tucson, AZ: Communication Skill Builders. Borden, G. J. (1979). An interpretation of research on feedback interruption in speech. Brain and Language, 7, 307–319. Butler, R. R., & Stanley, P. E. (1966). The stuttering problem considered from an automatic control point of view. Folia Phoniatricia, 18, 33–44. Franken, M. C., Boves, L., Peters, H. F. M., & Webster, R. L. (1992). Perceptual evaluation of the speech before and after fluency shaping stuttering therapy. Journal of Fluency Disorders, 17, 223–241. Goldiamond, I. (1962). The maintenance of ongoing fluent behavior and stuttering. Journal of Mathematics, 1, 57–95. Goldiamond, I. (1965). Stuttering and fluency as manipulatable operant response classes. In L. Krasner & L. P. Ullmann (Eds.), Research in behavior modification (pp. 106–156). New York: Holt, Rinehart & Winston.
200
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
Guitar, B. (1998). Stuttering: An integrated approach to its nature and treatment (2nd ed.). Baltimore: Williams & Wilkins. Hargrave, S., Kalinowski, J., Stuart, A., Armson, J., & Jones, K. (1994). Effect of frequency altered feedback on stutterers’ fluency at two speech rates. Journal of Speech and Hearing Research, 37, 1313–1319. Kalinowski, J. S., Armson, J., Roland-Meiszkowski, M., Stuart, A., & Gracco, V. L. (1993). Effect of alterations in auditory feedback and speech rate on stuttering frequency. Language and Speech, 36(1), 1–16. Kalinowski, J., Noble, S., Armson, J., & Stuart, A. (1994). Pretreatment and posttreatment speech naturalness ratings of adults with mild and severe stuttering. American Journal of Speech Language-Pathology, 3, 65–70. Kalinowski, J., Armson, J., & Stuart, A. (1995). Effect of normal and fast articulatory rates on stuttering frequency. Journal of Fluency Disorders, 20, 293–302. Kalinowski, J., Stuart, A., Sark, S., & Armson, J. (1996). Stuttering amelioration at various auditory feedback delays and speech rates. European Journal of Disorders of Communication, 31, 259–269. Kent, R. D. (1984). Stuttering as a temporal programming disorder. In R. F. Curlee & W. H. Perkins (Eds.), Nature and treatment of stuttering: New directions (pp. 283–301). San Diego: College Hill Press. Kuhn, T. S. (1970). The structure of scientific revolutions (2nd ed.). Chicago, IL: University of Chicago Press. MacLeod, J., Kalinowski, J., Armson, J., & Stuart, A. (1995). Effect of single and combined altered auditory feedback on stuttering frequency at two speech rates. Journal of Communication Disorders, 28, 218–226. Mysak, E. D. (1960). Servo-theory and stuttering. Journal of Speech and Hearing Disorders, 25, 188–189. Perkins, W. H. (1979). From psychoanalysis to discoordination. In H. H. Gregory (Ed.), Controversies about stuttering therapy (pp. 97–127). Baltimore, MD: University Park Press. Riley, G. (1994). Stuttering severity instrument for children and adults-3. Austin, TX: Pro-Ed Publications. Soderberg, G. A. (1969). Delayed auditory feedback and the speech of stutterers: A review of studies. Journal of Speech and Hearing Disorders, 34, 20–29. Starkweather, C. W. (1987). Fluency and stuttering. Englewood Cliffs, NJ: Prentice Hall. Vanryckehem, M., Glessing, J. J., Brutten, G. J., & McAlindon, P. (1999). The main and interactive effect of oral reading rate on the frequency of stuttering. American Journal of Speech-Language Pathology, 8, 164–170. Webster, R. L., & Lubker, B. B. (1968). Interrelationships among fluency producing variables in stuttered speech. Journal of Speech and Hearing Research, 11, 754–766. Wingate, M. E. (1970). Effect on stuttering of changes in audition. Journal of Speech and Hearing Research, 13, 861–873. Wingate, M. E. (1976). Stuttering: Theory and treatment. New York: Irvington Publishers. Wolf, A. A., & Wolf, E. G. (1959). Feedback processes in the theory of certain speech disorders. Speech Pathology and Therapy, 2, 48–55. Yates, A. J. (1963). Recent empirical and theoretical approaches to the experimental manipulation of speech in normal subjects and in stammerers. Behavior Research and Therapy, 1, 95–119.
CONTINUING EDUCATION The effect of fast speech rate on stuttering frequency during delayed auditory feedback QUESTIONS 1. It has been hypothesized that stuttering speakers’ fluency is enhanced during delayed auditory feedback (DAF) because it:
G. Sparks et al. / Journal of Fluency Disorders 27 (2002) 187–201
201
2.
3.
4.
5.
a. corrects such speakers’ auditory feedback deficits b. produces syllable prolongation which reduces stuttering c. is a by-product of a reduction in speaking rate d. all of the above e. none of the above The oral reading rates of these stuttering speakers during 0 ms DAF conditions were: a. similar for both pairs of speakers b. similar to those during DAF for the severe pair c. similar to those during DAF for the mild pair d. much slower in the fast condition for the severe pair than that of the mild pair e. both (a) and (b) This study found that stuttering frequency: a. varied little between normal and fast readings for either pair of speakers b. increased substantially during fast readings for both pairs of speakers c. increased substantially during fast readings for the severe pair d. varied little between normal and fast readings only for the mild pair e. both (c) and (d) This study found that DAF reduced the frequency of stuttering: a. during both normal and fast readings b. similarly in both pairs c. of both pairs the most at the longest delay d. of the mild pair the most at fast speech rates e. both (b) and (c) This study’s findings imply that how much stuttering speakers’ fluency improves under DAF conditions is influenced most strongly by: a. speakers’ reading rates b. speakers’ severity of stuttering c. DAF delay levels d. all of the above e. none of the above