Date of Award
Doctor of Philosophy
Speech and Hearing Science
Harodl J. Luper
H. Alan Lasater, Harold A. Peterson, Carl W. Asp
The mean and intrasubject response variability of voice reaction time to auditory stimuli was investigated for five year old, nine year old and adult stutterers and nonstutterers. The subjects who participated in this study were twenty-seven stutterers and twenty-seven nonstutterers at each of the three age levels. All of the stutterers had been reported to have exhibited the onset of stuttering behavior by no later than five years of age.
Each of the subjects was presented with a total of fifty-five prerecorded 1000 Hz tones bilaterally through stereo earphones at 80dB SPL. The stimuli were divided into five equal sets of eleven tones each. Data analysis was based on responses two through eleven for each set for a total of fifty responses per subject. The duration of each tone was one second and the silent interval between each of the tones varied randomly among three, four, five and six seconds. A one minute rest period was provided between each set. The subjects were instructed to respond to the onset of each tone as quickly as possible by initiating the neutral vowel sound / Λ / with what they considered to be their natural conversational loudness and effort to hold it until the tone ended. A training period was provided for the five year old and the nine year old subjects until the investigation determined that they understood and could consistently perform the task. Each of the subjects was then given ten practice trials in order to become familiar with the test procedures. Voice reaction time (VRT) was measured with an electronic digital counter triggered by the onset of the pure tones. Vocal onset from the subject was transduced by a condenser microphone two inches from the subject's lips. When the voltage level of the acoustic signal exceeded 460 mVolts (approximately 80 dB SPL vocal intensity level) the circuit to the counter was broken and stopped the clock.
Two three-factor analyses of variance with repeated measures on one factor (Set) were utilized to investigate the effects of the three experimental variables of Age (five, nine, and adult), Group (stutterers and nonstutterers), and Set (one through five) on both the mean and intrasubject response variability (calculated from the log of the variance) of voice reaction time. Conclusion drawn from these analyses may be summarized as follows:
1. Both the stutterers and the nonstutterers exhibited a significant decrease in mean voice reaction time with an increase in age. The five year old subjects for both groups exhibited significantly longer VRTs than the nine year olds and the adults. The VRTs for the nine year olds were longer than those of the adults; however, the differences were not significant.
2. Similarly, the intrasubject response variability decreased significantly with an increase in age for both the stutterers and the nonstutterers with significant differences found between each of the age groups. Thus, while the mean VRT approximated that of the adults by nine years of age, the intrasubject variability continued to decrease through adulthood.
3. Between group comparisons revealed that the mean VRTs for the stutterers were significantly longer than those of the matched nonstutterers at each of the three age levels. The largest difference between the means for the two groups of subjects was found for the five year olds (60 msec.), decreasing to 50 msec. for the nine year olds and 30 msec. for the adults. The Group x Age interaction, however, was nonsignificant.
4. The intrasubject response variability for the stutterers was also significantly greater than that for the nonstutterers at each of the three age levels. As with mean VRT, the greatest difference in response variability between the two groups occurred for the five year olds, progressively decreasing as age increased.
5. The Age x Set interaction for the mean voice reaction time was nonsignificant indicating that age had little or no effect on the rate or degree of adaptation for the mean VRT across response sets. The stutterers as a group, however, demonstrated a pattern of adaptation for mean VRT which was dissimilar to that of nonstutterers. The stutterers exhibited a decrease in mean VRT from set one to set two of 19 msec. This decrease was followed by a progressive increase in VRT with each additional set, attaining a maximum value for set five which was 9 msec. longer than set one. The nonstutterers showed little or no change in mean VRT across sets, attaining a minimum value on set one.
6. The Age x Set interaction for intrasubject response variability was nonsignificant indicating that age had little effect on the rate or degree of adaptation with respect to response variability.
7. Similarly, the Group x Set interaction for intrasubject response variability was also nonsignificant. This indicates that the stutterers and the nonstutterers exhibited similar patterns of change in variability across sets.
8. The combined intrasubject response variability for both the stutterers and the nonstutterers decreased from set one to set two. This was subsequently followed by an overall increase in subject variability with sets three, four, and five.
The results of this study suggest that the slower, more variable voice initiation ability for stutterers may not result from factors associated with the development of the stuttering disorder with age. Difficulty in promptly initiating voicing, on the other hand, may contribute to early disruption in the timing relationship between respiratory, phonatory, and articulatory processes needed for fluent speech production. The results of this study as well as those reported from previous investigations might be interpreted to suggest that the slow voice initiation ability, at least in some individuals, may result from disruption in the development of motor programming involved in early stages of speech production. This disruption also appears to be exhibited in nonspeech as well as speech related motor tasks and may reflect factors involved in central processing, such as lack of hemispheric dominance.
Cross, Douglas Edward, "Speed and Variability of Voice Reaction Times of Stuttering and Nonstuttering Children and Adults. " PhD diss., University of Tennessee, 1978.