Speech Motor Stability in IPD Effects of Rate and Loudness Manipulations Research Article
Open Access
Research Article  |   October 01, 2001
Speech Motor Stability in IPD
 
Author Affiliations & Notes
  • Jennifer Kleinow
    Department of Audiology and Speech Sciences Purdue University West Lafayette, IN
  • Anne Smith
    Department of Audiology and Speech Sciences Purdue University West Lafayette, IN
  • Lorraine Olson Ramig
    Department of Communication Disorders and Speech Science University of Colorado-Boulder and The Wilbur James Gould Voice Center The Denver Center for the Performing Arts Denver, CO
  • Contact author: Jennifer Kleinow, PhD, Purdue University, Department of Audiology and Speech Science, Heavilon Hall, West Lafayette, IN, 47907.
    Contact author: Jennifer Kleinow, PhD, Purdue University, Department of Audiology and Speech Science, Heavilon Hall, West Lafayette, IN, 47907.×
  • Corresponding author: E-mail: jkleinow@expert.cc.purdue.edu
Article Information
Hearing & Speech Perception / Acoustics / Hearing Disorders / Attention, Memory & Executive Functions / Speech, Voice & Prosody / Speech / Research Articles
Research Article   |   October 01, 2001
Speech Motor Stability in IPD
Journal of Speech, Language, and Hearing Research, October 2001, Vol. 44, 1041-1051. doi:10.1044/1092-4388(2001/082)
History: Received October 3, 2000
 
Journal of Speech, Language, and Hearing Research, October 2001, Vol. 44, 1041-1051. doi:10.1044/1092-4388(2001/082)
History: Received October 3, 2000
Web of Science® Times Cited: 68

Increasing phonatory effort, an integral component of the Lee Silverman Voice Treatment, LSVT, has been identified as an effective management strategy for adults with hypokinetic dysarthria associated with Parkinsonism. The present study compares the effects of increased loudness on lower lip movements to those of changes in speaking rate, another approach to the treatmentm of hypokinetic dysarthria. Movements of the lower lip/jaw during speech were recorded from 8 adults with IPD, 8 healthy aged adults, and 8 young adults. The spatiotemporal index (STI), a measure of spatial and temporal variability, revealed that for all speaker groups slow rate was associated with the most variability. Compared to the other conditions, STI values from the loud condition were closest to those from habitual speech. Also, the normalized movement pattern for the loud condition resembled that of habitual speech. It is hypothesized that speaking loudly is associated with a spatial and temporal organization that closely resembles that used in habitual speech, which may contribute to the success of the LSVT.

Idiopathic Parkinson’s disease (IPD) affects as many as 10% of Americans over the age of 60 (Schoenberg, 1987). In addition to the cardinal symptoms of rigidity, tremor, and hypokinesia, an estimated 75%– 100% of adults diagnosed with IPD develop speech and voice problems (Canter, 1965; Logemann, Fisher, Boshes, & Blonsky, 1978; Oxtoby, 1982; Streifler & Hofman, 1984). Such speech impairments are typical of hypokinetic dysarthria, a motor speech disorder characterized by short rushes of speech, imprecise consonants, hoarse and breathy voice, reduced prosody, and reduced loudness (Aronson, 1990; Boshes, 1966; Critchley, 1981). Although pharmacological and surgical treatments may alleviate motor disturbances of the limbs and, to some extent, the speech and voice impairments in some patients with IPD (see Schulz & Grant, 2000  for review), behavioral speech therapy remains the most effective treatment for the speech and voice symptoms of optimally medicated patients (Schulz & Grant, 2000).
One management strategy that may be employed in the behavioral treatment of hypokinetic dysarthria is rate reduction. Yorkston, Beukelman, Strand, and Bell (1999)  review a variety of techniques that have been shown to be effective in reducing speaking rate for many individuals with dysarthria. In addition to rate control techniques, clinical studies have also documented the efficacy of increasing vocal loudness as an intervention for hypokinetic dysarthria (Adams & Lang, 1992; Hanson & Metter, 1983; Johnson & Pring, 1990; Rubow & Swift, 1985; Scott & Caird, 1983). Recently, Ramig and her colleagues (Ramig, 1995) have developed the Lee Silverman Voice Treatment (LSVT), which focuses on increasing phonatory effort to facilitate vocal fold adduction in individuals with Parkinson’s disease. It is hypothesized that increasing loudness (a fundamental element of LSVT) generates positive alterations in speech motor output across the respiratory, laryngeal, and orofacial systems.
Clinical efficacy studies of LSVT have documented improvements in speech and voice parameters of adults with IPD. Respiratory functioning has been examined using the aerodynamic measures of maximum flow declination rate and subglottal pressure, both of which have shown significant improvements following LSVT (Ramig & Dromey, 1996). Lung volume excursion during speech increases significantly following therapy (Dromey & Ramig, 1998). At the phonatory/laryngeal level, significant improvement in pre-and post-treatment variables such as duration of sustained vowel phonation, maximum fundamental frequency range, and fundamental frequency variability have been reported (Ramig, Countryman, Thompson, & Horii, 1995). In addition, improvements have been consistently documented in vocal fold adduction (Dromey, Ramig, & Johnson, 1995; Ramig & Dromey, 1996; Smith, Ramig, Dromey, Perez, & Samandari, 1995) and sound pressure level following treatment (Countryman, Ramig, & Pawlas, 1994; Dromey et al., 1995; Ramig, 1995; Ramig, Countryman, O’Brien, Hoehn, & Thompson, 1996; Ramig, Countryman, Thompson, & Horii, 1995). Though studies supporting the efficacy of LSVT appear promising, few studies compare this treatment with alternative therapies for hypokinetic dysarthria, most notably rate reduction techniques.
Given these issues, direct examination of orofacial speech movement parameters associated with rate and loudness manipulations in people with IPD may yield interesting clinical and theoretical observations. In the present study, a composite measure of spatial and temporal variability in lower lip movement sequences—the spatiotemporal index (STI; Smith, Goffman, Zelaznik, Ying, & McGillem, 1995)—was employed to assess the stability of speech movements across multiple repetitions of a phrase during rate and loudness manipulations. The STI reflects the degree to which a set of movement trajectories produced for multiple repetitions of a motor behavior converge onto a single movement template after linear normalization (Smith et al., 1995). For typical adult speakers, habitual rate is most stable, with movement trajectories for each trial converging onto a single pattern (Kleinow & Smith, 2000; Smith et al., 1995; Wohlert & Smith, 1998). Thus, typical adults generate consistent speech movement patterns when producing habitual, highly practiced speech behaviors. When speakers move away from their preferred rate (in this case, after being verbally instructed to speak half or twice as fast as normal), they become less consistent at producing stable motor behaviors. It may be hypothesized that adults with IPD are not as consistent at producing repetitive speech motor behaviors in their habitual mode and that manipulating speech parameters such as rate and loudness may actually improve their performance. In other words, rate and loudness manipulations may affect speech movements in adults with IPD in ways quite distinct from those in normal adult speakers.
Method
Subjects
Three groups of 8 adults each participated in this experiment. Three men and 5 women diagnosed with idiopathic Parkinson disease made up Group IPD. Subjects with IPD were recruited from a local Parkinson disease support group and by referral from community neurologists. These participants ranged in age from 57 to 78 years (M = 70, SD = 7 years), and all were being treated pharmacologically at the time of the study. Perceptual evaluations of conversational speech revealed mild-moderate hoarseness, monotonicity, and reduced vocal loudness for all participants with IPD. These perceptual characteristics are consistent with the diagnosis of hypokinetic dysarthria (Logemann, Fisher, Boshes, & Blonsky, 1978). Additional patient information is recorded in Table 1.
Table 1 Participant descriptions.
Participant descriptions.×
Case Sex Age Years post-onset Stage (Hoehn & Yahr, 1967) Medications
1 M 62 20 3.0 Levodopa-carbidopa controlled-release Bromocriptine
2 F 65 3 3.5 Levodopa-carbidopa controlled-release Pramipexole
3 F 57 5 3.5 Levodopa-carbidopa controlled-release Selegiline
4 F 77 2 2.0 Levodopa-carbidopa Amantadine
5 F 73 1 2.0 Levodopa-carbidopa controlled-release
6 M 67 6 * Levodopa-carbidopa controlled-release Selegiline, amantadine
7 M 78 12 * Levodopa-carbidopa Pergolide
8 F 72 3 1 Levodopa-carbidopa controlled-release
* No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.
No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.×
Table 1 Participant descriptions.
Participant descriptions.×
Case Sex Age Years post-onset Stage (Hoehn & Yahr, 1967) Medications
1 M 62 20 3.0 Levodopa-carbidopa controlled-release Bromocriptine
2 F 65 3 3.5 Levodopa-carbidopa controlled-release Pramipexole
3 F 57 5 3.5 Levodopa-carbidopa controlled-release Selegiline
4 F 77 2 2.0 Levodopa-carbidopa Amantadine
5 F 73 1 2.0 Levodopa-carbidopa controlled-release
6 M 67 6 * Levodopa-carbidopa controlled-release Selegiline, amantadine
7 M 78 12 * Levodopa-carbidopa Pergolide
8 F 72 3 1 Levodopa-carbidopa controlled-release
* No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.
No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.×
×
Three male and 5 female healthy aged controls made up Group CONT-A. Participants in this group ranged in age from 67 to 78 years (M = 73, SD = 5 years). A third experimental group, CONT-Y, consisted of 4 male and 4 female young adults, who ranged in age from 21 to 28 years (M = 26, SD = 2 years). All subjects in the CONTA and CONT-Y groups reported good health and indicated no history of speech, language, hearing, or other neurological problems.
Experimental Task
During five different experimental conditions, each participant repeated the phrase “Buy Bobby a puppy” (Smith et al., 1995), modulating his or her vocal loudness or speech rate as directed. This phrase was selected because it contains bilabial consonants, which constrain the motion of the lips to achieve movement targets. First, the participants were instructed to produce the phrase at their habitual rate and at a comfortable loudness level. Participants repeated the phrase, pausing for several seconds between utterances, for approximately one minute. After a brief rest period, the participants repeated the trial for an additional minute, ensuring that at least 20 accurate and fluent productions of the phrase were recorded. The next two conditions involved modulations in vocal intensity. During the Loud condition, the participants were instructed to speak at “twice” their normal vocal intensity while continuously repeating the phrase as before. The participants were then allowed to re-habituate to their normal rate and intensity by producing the phrase at habitual rate and intensity. For the Soft condition, they were instructed to perform the task again at “half” their normal vocal intensity. The final two experimental conditions involved rate modulations. Again, the participants repeated the phrase “Buy Bobby a puppy,” with the instructions to speak at twice their normal rate for the Fast condition and half their normal rate for the Slow condition. Participants were instructed to slow rate by “stretching out” words rather than by increasing the pause time between words or syllables. Though the subjects were encouraged to self-scale rate and loudness during these manipulations, their task compliance was monitored with a sound pressure level meter as well as by viewing the acoustic signal on a storage oscilloscope. All participants appeared to adjust their verbal behavior as directed.
Data Collection
Participants were seated comfortably and fitted with a lightweight, head-mounted strain gauge to transduce the movements of the lower lip and jaw. A small bead threaded through the cantilever beam was attached to the vermilion border of each participant’s lower lip at midline, providing a signal of superior-inferior lip and jaw movements (Barlow, Cole, & Abbs, 1983). This kinematic signal was digitized at 500 samples/s and then digitally low-pass filtered in forward and backward directions using a fourth order Butterworth filter (15 Hz cutoff). Instantaneous velocity was digitally derived using the three-point difference method. The audio signal, transduced by a head-mounted microphone, was collected simultaneously and digitized at 20 kHz. The audio and kinematic signals were synchronized, which provided the opportunity to listen to each audio record to ensure that all corresponding kinematic data entered into analysis were obtained from fluent productions.
To measure the vocal intensity of each repetition of the utterance, a sound-level meter was placed 30 cm in front of the participant’s lips. The SLM distance was monitored continuously as the experiment progressed. The experimenter manually recorded the peak vocal SPL measures, which were continuously displayed at 1-s intervals from the digital output of the SLM during all speaking tasks. Output from the SLM, as well as a pure tone calibration signal, was recorded onto digital audiotapes (Fox & Ramig, 1997).
Data Analysis
For each condition, 15 individual tokens of the phrase “Buy Bobby a puppy” were extracted from the long kinematic records containing multiple repetitions of the target phrase. As in a previous study (Smith et al., 1995), these individual tokens consisted of the data points between the peak velocities of the first and last opening movements (from the initial release of the /b/ in “buy” to the final release of /p/ in the second syllable of “puppy”). Utterances containing disfluencies, false starts, misarticulations, or other overt errors were excluded from analysis.
The 15 displacement waveforms from each condition were then amplitude- and time-normalized. Amplitude-normalized signals were derived by subtracting the mean and dividing by the standard deviation of each displacement record. Because records corresponding to single tokens of the target phrase varied in duration, time-normalization was required to fix each record to a constant length. In this case, a cubic spline procedure was used to interpolate each displacement signal to a length of 1000 points (Smith, Johnson, McGillem, & Goffman, 2000). For each set of 15 time- and amplitude-normalized waveforms, the standard deviation was computed at successive 2% intervals. The 50 resultant SDs were then summed to produce the spatiotemporal index (STI).
Pattern Recognition
A custom-designed pattern-recognition program introduced by Smith et al. (1995)  demonstrated that, for normal speakers, utterances spoken at habitual, fast, and slow rates could be reliably classified into groups corresponding to distinct movement templates. This analysis was expanded in the present study to include loud and soft speech movement templates as well. For each participant, the time- and amplitude-normalized waveforms for each production in the five conditions were zero-lag cross-correlated with five individualized movement templates. The templates were derived by averaging 14 waveforms (produced by the individual) from each condition. Though each condition originally contained 15 waveforms, the “test” waveform was not included in the average of the condition from which it came. To balance the number of utterances included in each average, one waveform was randomly excluded from the average of the other four conditions. The pattern-recognition program automatically sorted the “test” waveform into the class of the template producing the highest correlation.
Vocal Intensity and Utterance Duration
Peak vocal intensity for each repetition, measured in dB SPL, was displayed at 1-s intervals on the digital output screen of the sound level meter. For every speaker, the average peak vocal intensity value was computed for each of the five conditions. For every speaker, the average duration of the displacement signal was recorded for each condition.
Results
Utterance Duration
All participants modified their speaking rate in the expected direction during the Fast and Slow conditions. On average, participants spoke at 80% of their habitual rate for the Fast condition and at 202% of their habitual rate for the Slow condition. Mean utterance duration and standard deviation (in seconds) for each group and all conditions are presented in Table 2. A repeatedmeasures ANOVA with one between factor (Group) and one within factor (Condition) with five levels (Habitual, Fast, Slow, Loud, Soft) was performed on the average duration of the phrase “Buy Bobby a puppy.” Not surprisingly, a significant effect of condition was observed [F(4, 84) = 91.64, p < 0.001]. Though the group of adults with IPD exhibited the smallest range of speaking rate modulation, no significant group differences or Group × Condition interactions were observed. Thus, all groups responded to the experimental manipulations in a similar manner. It should be noted that overall duration was not affected by the loudness conditions; participants spoke at 99% of Habitual rate during the Loud condition and at 102% of habitual rate during the Soft condition. Condition contrasts revealed that only the Fast and Slow conditions were associated with utterance durations that were statistically different from the Habitual rate (Tukey, HSD, p < .01).
Table 2 Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)
Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)×
Condition Young adult controls Aged adult controls Adults with IPD



M (SD) % of Habitual M (SD) % of Habitual M (SD) % of Habitual
Habitual 1.07 (.14) 100 1.13 (.09) 100 1.12 (.15) 100
Fast .83 (.10) 78 .90 (.08) 79 .92 (.10) 83
Slow 2.16 (.51) 203 2.32 (.58) 207 2.18 (.66) 194
Loud 1.02 (.11) 96 1.13 (.18) 100 1.12 (.13) 100
Soft 1.01 (.15) 95 1.23 (.32) 108 1.15 (.13) 104
Table 2 Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)
Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)×
Condition Young adult controls Aged adult controls Adults with IPD



M (SD) % of Habitual M (SD) % of Habitual M (SD) % of Habitual
Habitual 1.07 (.14) 100 1.13 (.09) 100 1.12 (.15) 100
Fast .83 (.10) 78 .90 (.08) 79 .92 (.10) 83
Slow 2.16 (.51) 203 2.32 (.58) 207 2.18 (.66) 194
Loud 1.02 (.11) 96 1.13 (.18) 100 1.12 (.13) 100
Soft 1.01 (.15) 95 1.23 (.32) 108 1.15 (.13) 104
×
Utterance Intensity
Participants with IPD scaled loudness similarly to their healthy aged peers as well as young adults. The means and standard deviations, in dB SPL, for each group and all conditions are presented in Table 3. A repeated-measures ANOVA analogous to the test presented above revealed a significant condition effect on vocal intensity [F(4, 84) = 172.55, p < 0.01]. No significant group or group × condition effects were observed. Condition contrasts revealed that only the Loud and Soft conditions were statistically different from Habitual intensity (Tukey, HSD, p < .01), ensuring that speech rate and utterance intensity were not confounded.
Table 3 Average utterance intensity (dB SPL) for each speaker group across conditions.
Average utterance intensity (dB SPL) for each speaker group across conditions.×
Condition Young adult controls (dB SPL) Change (dB SPL) Healthy aged controls (dB SPL) Change (dB SPL) Adults with IPD (dB SPL) Change (dB SPL)
Habitual 72.39 (3.7) 72.53 (4.8) 72.43 (4.9)
Fast 72.51 (2.4) .12 74.27 (3.6) 1.76 72.77 (3.5) .35
Slow 70.15 (2.9) −2.24 70.51 (4.0) −2.02 70.63 (5.8) −1.79
Loud 81.65 (4.3) 9.26 81.58 (4.3) 9.05 79.69 (3.8) 7.27
Soft 65.10 (2.4) −7.29 65.90 (2.6) −6.63 65.85 (4.0) −6.57
Table 3 Average utterance intensity (dB SPL) for each speaker group across conditions.
Average utterance intensity (dB SPL) for each speaker group across conditions.×
Condition Young adult controls (dB SPL) Change (dB SPL) Healthy aged controls (dB SPL) Change (dB SPL) Adults with IPD (dB SPL) Change (dB SPL)
Habitual 72.39 (3.7) 72.53 (4.8) 72.43 (4.9)
Fast 72.51 (2.4) .12 74.27 (3.6) 1.76 72.77 (3.5) .35
Slow 70.15 (2.9) −2.24 70.51 (4.0) −2.02 70.63 (5.8) −1.79
Loud 81.65 (4.3) 9.26 81.58 (4.3) 9.05 79.69 (3.8) 7.27
Soft 65.10 (2.4) −7.29 65.90 (2.6) −6.63 65.85 (4.0) −6.57
×
To address the concern that participants with IPD might experience vocal fatigue over the 15 consecutive utterances, mean vocal intensity of the first eight repetitions was compared to mean vocal intensity of the last seven repetitions for all conditions. Paired t tests revealed no significant differences in SPL for utterances spoken by the IPD patients in the first half of the trial compared to the final utterances for any condition.
Spatiotemporal Stability
Figure 1 displays typical results for a young adult speaker performing at habitual rate and intensity. The top panel shows 15 extracted lower lip/jaw movement trajectories corresponding to the phrase “Buy Bobby a puppy.” These trajectories are time- and amplitude-normalized in the middle panel, and the standard deviations are summed to compute the STI in the third panel. The STI from this participant can be compared to the STI values obtained from an adult with Stage 2 IPD, which is shown in the left panel of Figure 2, and a healthy aged control, shown in the right panel of Figure 2. Figure 3 compares data from the same IPD participant from the Loud (left) and Slow (right) conditions. For this particular participant, speaking loudly is associated with a decrease in STI, whereas speaking slowly is associated with an increase in STI.
Figure 1

Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase, “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase,  “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 1

Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase, “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 2

Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 2

Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 3

Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 3

Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 4 presents the mean STI values as a function of experimental condition for the three groups of speakers. Data from this plot show that, for young adults, habitual rate and loudness yields the most stable speech (as indicated by the lowest STI value). For this group of speakers, manipulating speech rate and loudness is associated with an increase in STI, though the effects of speaking loudly appear minimal. As in earlier studies (Smith et al., 1995; Smith & Kleinow, 2000) the greatest effect is seen in the Slow condition. For the aged adults and adults with IPD, however, speaking loudly was associated with the lowest STI values. Speaking slowly was associated with the highest STI values for all groups.
Figure 4

Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.

 Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.
Figure 4

Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.

×
A repeated-measures ANOVA was performed on the STI to examine possible group and condition differences in motor stability across a repeated speech behavior. Results indicated a significant group effect [F(2,21) = 3.47, p = .05] and condition effect [F(4,84) = 18.47, p < 0.01] and no group × condition interaction. Clearly, the young adults had mean STI values that were lower than the aged controls and IPD patients for all conditions (see Figure 4). An additional ANOVA performed to compare only the mean STI values for healthy aged adults with those obtained from IPD speakers revealed no significant differences. For all groups, condition contrasts revealed that the Habitual condition was significantly different from only the Slow condition (Tukey, HSD, p < .05).
Pattern Recognition
In the pattern-recognition procedure, each individual waveform from each speaker was sorted automatically into the group for which it best matched the average movement template (as determined by cross-correlations between the individual token and each of the five average waveforms for each condition). Table 4 summarizes the classification accuracy for each group of subjects. Pattern-matching was most reliable for young adults (80.7% correct) but failed considerably more often for healthy aged adults (65.3% correct). Pattern-matching was least reliable for individuals with IPD, whose movement waveforms were classified correctly for only 58.5% of the trials. A χ2 test showed significant differences among young adults, aged adults, and adults with IPD in pattern-matching accuracy ( χ2 = 71.26, p < .001). A similar test performed for the healthy aged adults and adults with IPD revealed significant differences in pattern classification accuracy between these two groups ( χ2 = 5.94, p = .01).
Table 4 Pattern recognition classification results.
Pattern recognition classification results.×
Group Total classified correctly Total classified incorrectly
Young Adults Aged Adults Adults with IPD
484 (80.7%) 392 (65.3%) 351 (58.5%)
116 (19.3%) 208 (34.7%) 249 (41.5%)
Table 4 Pattern recognition classification results.
Pattern recognition classification results.×
Group Total classified correctly Total classified incorrectly
Young Adults Aged Adults Adults with IPD
484 (80.7%) 392 (65.3%) 351 (58.5%)
116 (19.3%) 208 (34.7%) 249 (41.5%)
×
Table 5 presents results of this sorting task for three conditions (Habitual, Loud, and Soft) in the form of confusion matrices. In the top panel, figures indicate the number of habitual productions for each speaker group that were classified as Habitual, Loud, Soft, Fast, and Slow, respectively. The middle panel reflects the same organization for the loud productions, whereas the bottom panel displays sorting results for the slow productions for all speaker groups.
Table 5 Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.
Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.×
Condition Classified as:

Habitual Loud Soft Fast Slow
Habitual
  Young Adults 95 16 4 5 0
  Aged Adults 71 16 22 8 3
  Adults with IPD 73 24 15 8 0
Loud
  Young Adults 20 88 10 2 0
  Aged Adults 20 68 12 20 0
  Adults with IPD 28 74 8 9 1
Slow
  Young Adults 3 0 0 1 116
  Aged Adults 3 0 9 0 108
  Adults with IPD 11 8 1 8 92
Table 5 Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.
Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.×
Condition Classified as:

Habitual Loud Soft Fast Slow
Habitual
  Young Adults 95 16 4 5 0
  Aged Adults 71 16 22 8 3
  Adults with IPD 73 24 15 8 0
Loud
  Young Adults 20 88 10 2 0
  Aged Adults 20 68 12 20 0
  Adults with IPD 28 74 8 9 1
Slow
  Young Adults 3 0 0 1 116
  Aged Adults 3 0 9 0 108
  Adults with IPD 11 8 1 8 92
×
Discussion
To produce speech, the central nervous system must generate sets of command signals to control muscle activation. This muscle activity interacts with the biomechanical properties of the structures to be moved, and patterned movement output is produced. The STI is a composite measure of spatial and temporal variability over repeated motor behaviors. As people repeat a target utterance, variable movement trajectories are produced from trial to trial. A higher STI value indicates greater variability in generating a movement trajectory for a single target sentence. Thus, greater variability can be the result of more variable neural command signals to muscles, as well as more variable biomechanical properties at the periphery. In the case of healthy aged participants and patients with Parkinson’s disease, both factors are likely to contribute. For example, the basal ganglia play a role in motor planning, execution, and learning (Graybiel, Aosaki, Flaherty, & Kimura, 1994), and basal ganglia disease could contribute to the generation of more variable neural command signals. Also, patients with Parkinson’s disease may show atypical muscle tone, indicating differences in tonic levels of muscle activity, which could alter muscle stiffness (Glendinning & Enoka, 1994).
Speech Rate
In the present study slow rate was associated with the highest STI for all subject groups. This finding is consistent with previous studies of normal speakers (Smith et al., 1995) and adults who stutter (Smith & Kleinow, 2000). Clearly, individuals’ ability to produce consistent speech motor patterns over repeated utterances does not improve with increased movement duration, a result that may have significant implications for speech therapy. It should also be noted that STI values do not appear to be an epiphenomenon of rate. If the STI were positively correlated with utterance duration, one would expect decreased STI values during the Fast condition. This, however, was not observed.
Verbally instructing a patient with hypokinetic dysarthria to speak half as fast as normal may move him or her away from a “stable” mode of speech. There are a variety of techniques appropriate for use in rate control, of which this study examined only one (simple instructions to slow down). It is possible that other strategies to reduce rate (e.g., paced speech) may lead to more stable movement patterns. Additionally, investigating the effects of alternate rate control techniques could provide further insight as to whether group STI differences arise from motor planning or execution processes.
Speech Intensity
For all speaker groups, the Loud condition was associated with STI values that were similar to or less than those observed for the habitual rate and loudness condition. In addition, speaking loudly was associated with less temporal and spatial variability (lower STI) than speaking slowly. These results are corroborated by a recent kinematic study focusing on adults with IPD (Dromey, 2000). Dromey reported that speaking loudly was associated with lower STI values than exaggerated speech, an alternative therapy technique for treating dysarthria. Dromey hypothesized that speaking loudly may be motorically simpler than using exaggerated articulation. Further, individuals with IPD, 40%–60% of whom may experience decreased cognitive functioning (Mahler & Cummings, 1990), may benefit from instructions to change only one speech parameter (vocal effort) rather than attempt to control the force and velocity of multiple articulators.
It is also possible that speaking loudly is associated with a temporal and spatial organization that more closely resembles the organization for habitual speech than speaking slowly. In support of this hypothesis, early studies of rate change (Kozhevnikov & Chistovich, 1965) suggested that the relative duration of the sounds in a syllable was not preserved across changes in speech rate. Likewise, relative timing of submovement sequences changes nonlinearly with speech rate (Smith et al., 1995). Changes in vocal intensity, however, have been associated with relatively linear transformations in timing and amplitude between habitual and loud speech output (Schulman, 1989). Thus, it may be simpler, especially for disordered populations, to maintain speech motor stability while performing at a loud intensity, because they can use a preexisting, well-practiced movement organization.
Pattern Recognition
Differences between adults with IPD and typical speakers emerged in the pattern-recognition procedure. In this procedure, normalized displacement waveforms for the entire utterance are used to determine if distinct movement templates exist for each rate and loudness condition (Smith et al., 1995). If separate templates did not exist, individual waveforms would not have been classified into the correct template categories. This algorithm successfully sorted the majority of waveforms into correct condition classes. Utterances produced by adults with IPD were sorted least reliably, however, indicating that adults with IPD are less reliable at generating the same movement template from trial to trial. The loud condition was often confused with habitual for adults with IPD (26% of loud utterances were classified as habitual). For all speakers, the slow condition was associated with the most distinctive pattern. For IPD speakers, only 9% of slow utterances were classified as habitual. Thus, slow-rate templates least resemble templates produced at habitual rate and loudness. Loud-rate templates, on the other hand, most resemble templates produced during habitual speaking mode.
Some Caveats
The majority of patients with IPD (5/8) participating in this study demonstrated mild symptoms based on the Hoehn and Yahr (1967)  scale. Further studies should include patients exhibiting more severe symptoms to adequately describe the effects of rate and loudness manipulations on speech stability across a broader range of impairment. Additionally, the Hoehn and Yahr scale (1967)  measures the severity of limb symptoms. Limb motor systems and speech motor systems may deteriorate at different rates (Metter & Hanson, 1986), indicating that for future studies it may be more appropriate to use measures of dysarthria severity for patient classification. Further, though current data and previous studies in adults who stutter (Smith & Kleinow, 2000) demonstrate that slow speech is associated with increased variability, this finding should not be generalized to other clinical groups. Individuals with dysarthria secondary to traumatic brain injury, for example, have demonstrated more consistent movement patterns for slow speech (McHenry, 1999).
It should also be emphasized that the STI is a composite measure of spatial and temporal variability. Thus, the specific contributions of speaking modulations to the spatial and/or temporal properties of kinematic waveforms cannot be directly determined with the STI. With the goal of increasing intelligibility while maintaining fluency and naturalness, the effects of speech modulations on independent measures of spatial and temporal characteristics should continue to be studied. For example, adults with IPD may show both decreased amplitude and velocity of mandibular movements (Forrest, Weismer, & Turner, 1989; Svensson, Henningson, & Karlsson, 1993) and lip movements (Caligiuri, 1987) during syllable repetition tasks when compared to controls. Forrest, Weismer, and Turner (1989)  found lower lip movement amplitude and velocity to be decreased in adults with IPD for opening gestures, but lower lip movement velocity was increased for speakers with IPD for closing gestures. Loud phonation has been associated with increased articulatory displacement as well as with increased velocity for the upper and lower lip in normal adults and adults with IPD (Dromey & Ramig, 1998). Slow speaking rates are associated with lower movement velocities (Smith & Kleinow, 2000), and faster than normal rates produce variable effects on velocity in normal adults (Kuehn & Moll, 1976; McClean, 2000). Thus, in general, although asking a patient with IPD to speak loudly would seem to place more demands on the motor system in terms of movement amplitude and velocity, it does not appear to negatively affect performance. In contrast, slow speech rates result in more variable output patterns that may or may not be adaptive for IPD. It would be useful for future studies to combine measures of intelligibility and kinematics to address the question of the best treatment strategies for patients with IPD.
Conclusion
The STI was designed to reflect motor stability across repeated productions of an entire phrase. Low STI values across multiple repetitions of a motor behavior reflect the presence of stable underlying processes involved in movement planning and execution, and low values are typical of normal adults repeating words or phrases (Goffman & Smith, 1999; Smith & Goffman, 1998). High STI values, in contrast, suggest that the performer is using multiple solutions to reach task goals. In other words the speaker may be exploring more of the available movement space. This may indicate a disordered state (e.g., dysarthria or stuttering), or it may indicate normal behavior, as in the case of young children (Goffman & Smith, 1999; Smith & Goffman, 1998). In the developing speech motor system, it may be adaptive to continue to explore a variety of movement solutions to achieve phonetic goals.
The present data lead to the conclusion that changes in vocal loudness (an integral component of LSVT) and rate manipulations may operate differently to affect speech motor stability. Slow speech rates were associated with greater movement trajectory variability. In contrast to rate changes, increasing vocal intensity did not produce a change in motor stability. This means that, as vocal intensity was increased, the lower lip/jaw movements of aged adults and adults with Parkinson’s disease maintained the stability associated with their preferred production mode. Changes in speech motor performance resulting from rate and intensity manipulations may document neuromotor correlates of behavioral management for motor speech disorders. For example, one factor contributing to the success of LSVT for adults with IPD may be that speaking loudly activates a speech production mode that is a relatively simple scaling of the habitual mode in terms of spatiotemporal organization. Thus, individuals with IPD are recruiting an existing coordinative organization that is highly stable despite the increase in amplitude of movement associated with loud productions. This may be an ideal treatment strategy for individuals whose dysarthria is characterized by a reduction in overall output (Ramig, 1995).
Acknowledgments
This work was supported by NIH grants DC01150 and DC00559 from NIDCD.
References
Adams, S. G., & Lang, A. E. (1992). Can the Lombard effect be used to improve voice intensity in Parkinson’s disease?. European Journal of Disorders of Communication, 27, 121–127. [Article] [PubMed]
Adams, S. G., & Lang, A. E. (1992). Can the Lombard effect be used to improve voice intensity in Parkinson’s disease?. European Journal of Disorders of Communication, 27, 121–127. [Article] [PubMed]×
Aronson, A. E. (1990). Clinical voice disorders. New York: Thieme.
Aronson, A. E. (1990). Clinical voice disorders. New York: Thieme.×
Barlow, S. M., Cole, K. J., & Abbs, J. H. (1983). A new head-mounted lip-jaw movement transduction system for the study of motor speech disorders. Journal of Speech and Hearing Research, 26, 283–288. [Article] [PubMed]
Barlow, S. M., Cole, K. J., & Abbs, J. H. (1983). A new head-mounted lip-jaw movement transduction system for the study of motor speech disorders. Journal of Speech and Hearing Research, 26, 283–288. [Article] [PubMed]×
Boshes, B. (1966). Voice changes in Parkinsonism. Journal of Neurosurgery, 24, 286–288.
Boshes, B. (1966). Voice changes in Parkinsonism. Journal of Neurosurgery, 24, 286–288.×
Caligiuri, M. (1987). Labial kinematics during speech in patients with parkinsonian rigidity. Brain, 110, 1033–1044. [Article] [PubMed]
Caligiuri, M. (1987). Labial kinematics during speech in patients with parkinsonian rigidity. Brain, 110, 1033–1044. [Article] [PubMed]×
Canter, G. J. (1965). Speech characteristics of patients with Parkinson’s disease: II. Physiological support for speech. Journal of Speech and Hearing Disorders, 30, 44–49. [Article]
Canter, G. J. (1965). Speech characteristics of patients with Parkinson’s disease: II. Physiological support for speech. Journal of Speech and Hearing Disorders, 30, 44–49. [Article] ×
Countryman, S., Ramig, L. O., & Pawlas, A. A. (1994). Speech and voice deficits in Parkinsonian plus syndromes: Can they be treated?. Journal of Medical Speech-Language Pathology, 2, 211–225.
Countryman, S., Ramig, L. O., & Pawlas, A. A. (1994). Speech and voice deficits in Parkinsonian plus syndromes: Can they be treated?. Journal of Medical Speech-Language Pathology, 2, 211–225.×
Critchley, E. M. R. (1981). Speech disorders of Parkinsonism: A review. Journal of Neurology, Neurosurgeryand Psychiatry, 44, 751–758. [Article]
Critchley, E. M. R. (1981). Speech disorders of Parkinsonism: A review. Journal of Neurology, Neurosurgeryand Psychiatry, 44, 751–758. [Article] ×
Dromey, C. (2000). Articulatory kinematics in patients with Parkinson disease using different speech treatment approaches. Journal of Medical Speech-Language Pathology, 8, 155–161.
Dromey, C. (2000). Articulatory kinematics in patients with Parkinson disease using different speech treatment approaches. Journal of Medical Speech-Language Pathology, 8, 155–161.×
Dromey, C., & Ramig, L. O. (1998). Intentional changes in sound pressure level and rate: Their impact on measures of respiration, phonation, and articulation. Journal of Speech, Language, and Hearing Research, 41, 1003–1018.
Dromey, C., & Ramig, L. O. (1998). Intentional changes in sound pressure level and rate: Their impact on measures of respiration, phonation, and articulation. Journal of Speech, Language, and Hearing Research, 41, 1003–1018.×
Dromey, C., Ramig, L. O., & Johnson, A. B. (1995). Phonatory and articulatory changes associated with increased vocal intensity in Parkinson disease: A case study. Journal of Speech and Hearing Research, 38, 751–764. [Article] [PubMed]
Dromey, C., Ramig, L. O., & Johnson, A. B. (1995). Phonatory and articulatory changes associated with increased vocal intensity in Parkinson disease: A case study. Journal of Speech and Hearing Research, 38, 751–764. [Article] [PubMed]×
Forrest, K., Weismer, G., & Turner, G. S. (1989). Kinematic, acoustic, and perceptual analyses of connected speech produced by parkinsonian and normal geriatric adults. Journal of the Acoustical Society of America, 85, 2608–2622. [Article] [PubMed]
Forrest, K., Weismer, G., & Turner, G. S. (1989). Kinematic, acoustic, and perceptual analyses of connected speech produced by parkinsonian and normal geriatric adults. Journal of the Acoustical Society of America, 85, 2608–2622. [Article] [PubMed]×
Fox, C., & Ramig, L. O. (1997). Vocal sound pressure level and self-perception of speech and voice in men and women with idiopathic Parkinson disease. American Journal of Speech-Language Pathology, 6, 85–94. [Article]
Fox, C., & Ramig, L. O. (1997). Vocal sound pressure level and self-perception of speech and voice in men and women with idiopathic Parkinson disease. American Journal of Speech-Language Pathology, 6, 85–94. [Article] ×
Glendinning, D. S., & Enoka, R. M. (1994). Motor unit behavior in Parkinson’s disease. Physical Therapy, 74, 61–70. [Article] [PubMed]
Glendinning, D. S., & Enoka, R. M. (1994). Motor unit behavior in Parkinson’s disease. Physical Therapy, 74, 61–70. [Article] [PubMed]×
Goffman, L., & Smith, A. (1999). Development and phonetic differentiation of speech movement patterns. Journal of Experimental Psychology: Human Perception and Performance, 25, 649–660. [Article] [PubMed]
Goffman, L., & Smith, A. (1999). Development and phonetic differentiation of speech movement patterns. Journal of Experimental Psychology: Human Perception and Performance, 25, 649–660. [Article] [PubMed]×
Graybiel, A. M., Aosaki, T., Flaherty, A. W., &Kimura, M. (1994). The basal ganglia and adaptive motor control. Science, 265, 1826–1831. [Article] [PubMed]
Graybiel, A. M., Aosaki, T., Flaherty, A. W., &Kimura, M. (1994). The basal ganglia and adaptive motor control. Science, 265, 1826–1831. [Article] [PubMed]×
Hanson, W. R., & Metter, E. J. (1983). DAF speech rate modification in Parkinson’s disease: A report of two cases. In Berry, W. R. (Ed.), Clinical dysarthria (pp. 231–251). San Diego, CA: College-Hill Press.
Hanson, W. R., & Metter, E. J. (1983). DAF speech rate modification in Parkinson’s disease: A report of two cases. In Berry, W. R. (Ed.), Clinical dysarthria (pp. 231–251). San Diego, CA: College-Hill Press.×
Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: Onset, progression, and mortality. Neurology, 17, 427–442. [Article] [PubMed]
Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: Onset, progression, and mortality. Neurology, 17, 427–442. [Article] [PubMed]×
Johnson, J. A., Pring, T. R. (1990). Speech therapy and Parkinson’s disease: A review and further data. British Journal of Disorders of Communication, 25, 182–194. [Article]
Johnson, J. A., Pring, T. R. (1990). Speech therapy and Parkinson’s disease: A review and further data. British Journal of Disorders of Communication, 25, 182–194. [Article] ×
Kleinow, J., & Smith, A. (2000). Influences of length and syntactic complexity on the speech motor stability of the fluent speech of adults who stutter. Journal of Speech, Language, and Hearing Research, 43, 548–559. [Article]
Kleinow, J., & Smith, A. (2000). Influences of length and syntactic complexity on the speech motor stability of the fluent speech of adults who stutter. Journal of Speech, Language, and Hearing Research, 43, 548–559. [Article] ×
Kozhevnikov, V. A., & Chistovich, L. A. (1965). Speech: Articulation and perception (Publication nos. JPRS: 30, 543; TT: 65-31233). Washington, DC : U.S. Department of Commerce.
Kozhevnikov, V. A., & Chistovich, L. A. (1965). Speech: Articulation and perception (Publication nos. JPRS: 30, 543; TT: 65-31233). Washington, DC : U.S. Department of Commerce.×
Kuehn, D. P., & Moll, K. (1976). A cineradiographic study of VC and CV articulatory velocities. Journal of Phonetics, 4, 303–320.
Kuehn, D. P., & Moll, K. (1976). A cineradiographic study of VC and CV articulatory velocities. Journal of Phonetics, 4, 303–320.×
Logemann, J. A., Fisher, H. B., Boshes, B., & Blonsky, E. (1978). Vocal tract control in Parkinson’s disease: Phonetic feature analyses of misarticulations. Journal of Speech and Hearing Disorders, 43, 47-57. [Article] [PubMed]
Logemann, J. A., Fisher, H. B., Boshes, B., & Blonsky, E. (1978). Vocal tract control in Parkinson’s disease: Phonetic feature analyses of misarticulations. Journal of Speech and Hearing Disorders, 43, 47-57. [Article] [PubMed]×
Mahler, M. E., & Cummings, J. L. (1990). Alzheimer’s disease and the dementia of Parkinson’s disease: Comparative investigations. Alzheimer Disease and Associated Disorders, 4, 133–149. [Article] [PubMed]
Mahler, M. E., & Cummings, J. L. (1990). Alzheimer’s disease and the dementia of Parkinson’s disease: Comparative investigations. Alzheimer Disease and Associated Disorders, 4, 133–149. [Article] [PubMed]×
McClean, M. D. (2000). Patterns of orofacial movement velocity across variations in speech rate. Journal of Speech Language & Hearing Research, 43, 205–216. [Article]
McClean, M. D. (2000). Patterns of orofacial movement velocity across variations in speech rate. Journal of Speech Language & Hearing Research, 43, 205–216. [Article] ×
McHenry, M. A. (1999, November). Kinematic correlates of pacing strategies in dysarthric speech. Poster session presented at the ASHA Convention, San Francisco.
McHenry, M. A. (1999, November). Kinematic correlates of pacing strategies in dysarthric speech. Poster session presented at the ASHA Convention, San Francisco.×
Metter, E. J., Hanson, W. R. (1986). Clinical and acoustical variability in hypokinetic dysarthria. Journal of Communication Disorders, 19, 347–366. [Article] [PubMed]
Metter, E. J., Hanson, W. R. (1986). Clinical and acoustical variability in hypokinetic dysarthria. Journal of Communication Disorders, 19, 347–366. [Article] [PubMed]×
Oxtoby, M. (1982). Parkinson’s disease patients and their social needs. London: Parkinson’s Disease Society.
Oxtoby, M. (1982). Parkinson’s disease patients and their social needs. London: Parkinson’s Disease Society.×
Ramig, L. (1995). Speech therapy for Parkinson’s disease. In Koller, W. & Paulson, G. (Eds.), Therapy of Parkinson’s disease (pp. 539–550). New York: Marcel Dekker.
Ramig, L. (1995). Speech therapy for Parkinson’s disease. In Koller, W. & Paulson, G. (Eds.), Therapy of Parkinson’s disease (pp. 539–550). New York: Marcel Dekker.×
Ramig, L. O., Countryman, S., O’Brien, C., Hoehn, M., & Thompson, L. (1996). Intensive speech treatment for patients with Parkinson disease: Short-term and longterm comparison of two techniques. Neurology, 47, 1496–1504. [Article] [PubMed]
Ramig, L. O., Countryman, S., O’Brien, C., Hoehn, M., & Thompson, L. (1996). Intensive speech treatment for patients with Parkinson disease: Short-term and longterm comparison of two techniques. Neurology, 47, 1496–1504. [Article] [PubMed]×
Ramig, L. O., Countryman, S., Thompson, L., & Horii, L. (1995). A comparison of two forms of intensive speech treatment for Parkinson disease. Journal of Speech and Hearing Research, 38, 1232–1251. [Article] [PubMed]
Ramig, L. O., Countryman, S., Thompson, L., & Horii, L. (1995). A comparison of two forms of intensive speech treatment for Parkinson disease. Journal of Speech and Hearing Research, 38, 1232–1251. [Article] [PubMed]×
Ramig, L. O., & Dromey, C. (1996). Aerodynamic mechanisms underlying treatment related changes in SPL in patients with Parkinson disease. Journal of Speech and Hearing Research, 39, 798–807. [Article] [PubMed]
Ramig, L. O., & Dromey, C. (1996). Aerodynamic mechanisms underlying treatment related changes in SPL in patients with Parkinson disease. Journal of Speech and Hearing Research, 39, 798–807. [Article] [PubMed]×
Rubow, R., & Swift, E. (1985). A microcomputer-based wearable biofeedback device to improve transfer of treatment in Parkinsonian dysarthria. Journal of Speech and Hearing Disorders, 50, 178–185. [Article] [PubMed]
Rubow, R., & Swift, E. (1985). A microcomputer-based wearable biofeedback device to improve transfer of treatment in Parkinsonian dysarthria. Journal of Speech and Hearing Disorders, 50, 178–185. [Article] [PubMed]×
Schoenberg, B. S. (1987). Descriptive epidemiology of Parkinson’s disease: Disease distribution and hypothesis formulation. Advances in Neurology, 45, 277–283. [PubMed]
Schoenberg, B. S. (1987). Descriptive epidemiology of Parkinson’s disease: Disease distribution and hypothesis formulation. Advances in Neurology, 45, 277–283. [PubMed]×
Schulman, R. (1989). Articulatory dynamics of loud and normal speech. Journal of the Acoustical Society of America, 85, 295–312. [Article] [PubMed]
Schulman, R. (1989). Articulatory dynamics of loud and normal speech. Journal of the Acoustical Society of America, 85, 295–312. [Article] [PubMed]×
Schulz, G. M., & Grant, M. K. (2000). Effects of speech therapy and pharmacologic and surgical treatments on voice and speech in Parkinson’s disease: A review of the literature. Journal of Communication Disorders, 33, 59–88. [Article] [PubMed]
Schulz, G. M., & Grant, M. K. (2000). Effects of speech therapy and pharmacologic and surgical treatments on voice and speech in Parkinson’s disease: A review of the literature. Journal of Communication Disorders, 33, 59–88. [Article] [PubMed]×
Scott, S., & Caird, F. I. (1983). Speech therapy for Parkinson’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 46, 140–144.
Scott, S., & Caird, F. I. (1983). Speech therapy for Parkinson’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 46, 140–144.×
Smith, A., & Goffman, L. (1998). Stability and patterning of speech movement sequences in children and adults. Journal of Speech, Language, and Hearing Research, 41, 18–30. [Article]
Smith, A., & Goffman, L. (1998). Stability and patterning of speech movement sequences in children and adults. Journal of Speech, Language, and Hearing Research, 41, 18–30. [Article] ×
Smith, A., Goffman, L., Zelaznik, H. N., Ying, G., & McGillem, C. (1995). Spatiotemporal stability and patterning of speech movement sequences. Experimental Brain Research, 104, 493–501. [Article] [PubMed]
Smith, A., Goffman, L., Zelaznik, H. N., Ying, G., & McGillem, C. (1995). Spatiotemporal stability and patterning of speech movement sequences. Experimental Brain Research, 104, 493–501. [Article] [PubMed]×
Smith, A., Johnson, M., McGillem, C., & Goffman, L. (2000). On the assessment of stability and patterning of speech movements. Journal of Speech, Language, and Hearing Research, 43, 277–286. [Article]
Smith, A., Johnson, M., McGillem, C., & Goffman, L. (2000). On the assessment of stability and patterning of speech movements. Journal of Speech, Language, and Hearing Research, 43, 277–286. [Article] ×
Smith, A., & Kleinow, J. (2000). Kinematic correlates of speaking rate changes in stuttering and normally fluent adults. Journal of Speech, Language, and Hearing Research, 43, 521–536. [Article]
Smith, A., & Kleinow, J. (2000). Kinematic correlates of speaking rate changes in stuttering and normally fluent adults. Journal of Speech, Language, and Hearing Research, 43, 521–536. [Article] ×
Smith, M. E., Ramig, L. O., Dromey, C., Perez, & K. S., Samandari, R. (1995). Intensive voice treatment in Parkinson’s disease: Laryngostroboscopic findings. Journal of Voice, 9, 453–349. [Article] [PubMed]
Smith, M. E., Ramig, L. O., Dromey, C., Perez, & K. S., Samandari, R. (1995). Intensive voice treatment in Parkinson’s disease: Laryngostroboscopic findings. Journal of Voice, 9, 453–349. [Article] [PubMed]×
Streifler, M., & Hofman, S. (1984). Disorders of verbal expression in Parkinsonism. In Hassler, R. G. & Christ, J. F. (Eds.), Advances in neurology (pp. 385–393). New York: Raven Press.
Streifler, M., & Hofman, S. (1984). Disorders of verbal expression in Parkinsonism. In Hassler, R. G. & Christ, J. F. (Eds.), Advances in neurology (pp. 385–393). New York: Raven Press.×
Svensson, P., Henningson, C., & Karlsson, S. (1993). Speech motor control in Parkinson’s disease: A comparison between a clinical assessment protocol and a quantitative analysis of mandibular movements. Folia Phoniatrica, 45, 157–164. [Article] [PubMed]
Svensson, P., Henningson, C., & Karlsson, S. (1993). Speech motor control in Parkinson’s disease: A comparison between a clinical assessment protocol and a quantitative analysis of mandibular movements. Folia Phoniatrica, 45, 157–164. [Article] [PubMed]×
Wohlert, A. B., & Smith, A. (1998). Spatiotemporal stability of lip movements in older adult speakers. Journal of Speech, Language, and Hearing Research, 41, 41–50. [Article]
Wohlert, A. B., & Smith, A. (1998). Spatiotemporal stability of lip movements in older adult speakers. Journal of Speech, Language, and Hearing Research, 41, 41–50. [Article] ×
Yorkston, K. M., Beukelman, D. R., Strand, E. A., & Bell, K. R. (1999). Management of motor speech disorders in children and adults. Austin, TX: Pro-Ed.
Yorkston, K. M., Beukelman, D. R., Strand, E. A., & Bell, K. R. (1999). Management of motor speech disorders in children and adults. Austin, TX: Pro-Ed.×
Figure 1

Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase, “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase,  “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 1

Data from one young adult from the habitual rate and intensity condition. The top plot includes the original displacement waveforms for the target phrase, “Buy Bobby a puppy.” The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 2

Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 2

Data from an adult with IPD from the habitual rate and intensity condition (right panel) and a healthy aged control (left panel). The center panel illustrates the time and amplitude normalized wave-forms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 3

Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

 Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.
Figure 3

Data from the same adult with IPD for the Loud condition (left panel) and the Slow condition (right panel). These values reflect the range of values obtained for this participant. The center panel illustrates the time and amplitude normalized waveforms. The lower panel shows the standard deviation computed at 2% intervals across the 15 normalized waveforms. The sum of these standard deviations, the STI, is shown in the upper right corner of the bottom plot.

×
Figure 4

Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.

 Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.
Figure 4

Mean STI values for the phrase “Buy Bobby a puppy” for each speaker group as a function of condition.

×
Table 1 Participant descriptions.
Participant descriptions.×
Case Sex Age Years post-onset Stage (Hoehn & Yahr, 1967) Medications
1 M 62 20 3.0 Levodopa-carbidopa controlled-release Bromocriptine
2 F 65 3 3.5 Levodopa-carbidopa controlled-release Pramipexole
3 F 57 5 3.5 Levodopa-carbidopa controlled-release Selegiline
4 F 77 2 2.0 Levodopa-carbidopa Amantadine
5 F 73 1 2.0 Levodopa-carbidopa controlled-release
6 M 67 6 * Levodopa-carbidopa controlled-release Selegiline, amantadine
7 M 78 12 * Levodopa-carbidopa Pergolide
8 F 72 3 1 Levodopa-carbidopa controlled-release
* No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.
No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.×
Table 1 Participant descriptions.
Participant descriptions.×
Case Sex Age Years post-onset Stage (Hoehn & Yahr, 1967) Medications
1 M 62 20 3.0 Levodopa-carbidopa controlled-release Bromocriptine
2 F 65 3 3.5 Levodopa-carbidopa controlled-release Pramipexole
3 F 57 5 3.5 Levodopa-carbidopa controlled-release Selegiline
4 F 77 2 2.0 Levodopa-carbidopa Amantadine
5 F 73 1 2.0 Levodopa-carbidopa controlled-release
6 M 67 6 * Levodopa-carbidopa controlled-release Selegiline, amantadine
7 M 78 12 * Levodopa-carbidopa Pergolide
8 F 72 3 1 Levodopa-carbidopa controlled-release
* No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.
No Hoehn & Yahr, (1967)  or UPDRS available. Both individuals were still ambulatory, placing them in the mild to moderate range of IPD. Participant 6’s primary complaint was shuffled gait and tremor in the upper extremities. Participant 7 showed overall reduced movement while walking that was well controlled with medication.×
×
Table 2 Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)
Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)×
Condition Young adult controls Aged adult controls Adults with IPD



M (SD) % of Habitual M (SD) % of Habitual M (SD) % of Habitual
Habitual 1.07 (.14) 100 1.13 (.09) 100 1.12 (.15) 100
Fast .83 (.10) 78 .90 (.08) 79 .92 (.10) 83
Slow 2.16 (.51) 203 2.32 (.58) 207 2.18 (.66) 194
Loud 1.02 (.11) 96 1.13 (.18) 100 1.12 (.13) 100
Soft 1.01 (.15) 95 1.23 (.32) 108 1.15 (.13) 104
Table 2 Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)
Average group mean (and SD) for duration for each condition. (Both mean and standard deviation are given in seconds.)×
Condition Young adult controls Aged adult controls Adults with IPD



M (SD) % of Habitual M (SD) % of Habitual M (SD) % of Habitual
Habitual 1.07 (.14) 100 1.13 (.09) 100 1.12 (.15) 100
Fast .83 (.10) 78 .90 (.08) 79 .92 (.10) 83
Slow 2.16 (.51) 203 2.32 (.58) 207 2.18 (.66) 194
Loud 1.02 (.11) 96 1.13 (.18) 100 1.12 (.13) 100
Soft 1.01 (.15) 95 1.23 (.32) 108 1.15 (.13) 104
×
Table 3 Average utterance intensity (dB SPL) for each speaker group across conditions.
Average utterance intensity (dB SPL) for each speaker group across conditions.×
Condition Young adult controls (dB SPL) Change (dB SPL) Healthy aged controls (dB SPL) Change (dB SPL) Adults with IPD (dB SPL) Change (dB SPL)
Habitual 72.39 (3.7) 72.53 (4.8) 72.43 (4.9)
Fast 72.51 (2.4) .12 74.27 (3.6) 1.76 72.77 (3.5) .35
Slow 70.15 (2.9) −2.24 70.51 (4.0) −2.02 70.63 (5.8) −1.79
Loud 81.65 (4.3) 9.26 81.58 (4.3) 9.05 79.69 (3.8) 7.27
Soft 65.10 (2.4) −7.29 65.90 (2.6) −6.63 65.85 (4.0) −6.57
Table 3 Average utterance intensity (dB SPL) for each speaker group across conditions.
Average utterance intensity (dB SPL) for each speaker group across conditions.×
Condition Young adult controls (dB SPL) Change (dB SPL) Healthy aged controls (dB SPL) Change (dB SPL) Adults with IPD (dB SPL) Change (dB SPL)
Habitual 72.39 (3.7) 72.53 (4.8) 72.43 (4.9)
Fast 72.51 (2.4) .12 74.27 (3.6) 1.76 72.77 (3.5) .35
Slow 70.15 (2.9) −2.24 70.51 (4.0) −2.02 70.63 (5.8) −1.79
Loud 81.65 (4.3) 9.26 81.58 (4.3) 9.05 79.69 (3.8) 7.27
Soft 65.10 (2.4) −7.29 65.90 (2.6) −6.63 65.85 (4.0) −6.57
×
Table 4 Pattern recognition classification results.
Pattern recognition classification results.×
Group Total classified correctly Total classified incorrectly
Young Adults Aged Adults Adults with IPD
484 (80.7%) 392 (65.3%) 351 (58.5%)
116 (19.3%) 208 (34.7%) 249 (41.5%)
Table 4 Pattern recognition classification results.
Pattern recognition classification results.×
Group Total classified correctly Total classified incorrectly
Young Adults Aged Adults Adults with IPD
484 (80.7%) 392 (65.3%) 351 (58.5%)
116 (19.3%) 208 (34.7%) 249 (41.5%)
×
Table 5 Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.
Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.×
Condition Classified as:

Habitual Loud Soft Fast Slow
Habitual
  Young Adults 95 16 4 5 0
  Aged Adults 71 16 22 8 3
  Adults with IPD 73 24 15 8 0
Loud
  Young Adults 20 88 10 2 0
  Aged Adults 20 68 12 20 0
  Adults with IPD 28 74 8 9 1
Slow
  Young Adults 3 0 0 1 116
  Aged Adults 3 0 9 0 108
  Adults with IPD 11 8 1 8 92
Table 5 Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.
Confusion matrices for the pattern recognition task. Results are shown for Habitual, Loud, and Slow conditions. Figures indicate the number of utterances classified as normal, loud, soft, fast, and slow for each condition and speaker group.×
Condition Classified as:

Habitual Loud Soft Fast Slow
Habitual
  Young Adults 95 16 4 5 0
  Aged Adults 71 16 22 8 3
  Adults with IPD 73 24 15 8 0
Loud
  Young Adults 20 88 10 2 0
  Aged Adults 20 68 12 20 0
  Adults with IPD 28 74 8 9 1
Slow
  Young Adults 3 0 0 1 116
  Aged Adults 3 0 9 0 108
  Adults with IPD 11 8 1 8 92
×