J. COMMUN. DISORD.

19 (1986) 347-366

CLINICAL AND ACOUSTICAL VARIABILITY

IN HYPOKINETIC DYSARTJ3RIA
E. JEFFREY METTER and WAYNE
Veterans

Administration

Medical

Center,

Sepulveda,

R. HANSON
California

Ten male patients with parkinsonism secondary to Parkinsons disease or progressive supranuclear palsy had clinical neurological, speech, and acoustical speech evaluations. In
addition, seven of the patients were evaluated by x-ray computed tomography (CT) and (F18)-fluorodeoxyglucose (FDG) positron emission tomography (PET). Extensive variability
of speech features, both clinical and acoustical, were found and seemed to be independent
of the severity of any parkinsonian sign, CT, or FDG PET. In addition, little relationship
existed between the variability across each measured speech feature. What appeared to be
important for the appearance of abnormal acoustic measures was the degree of overall
severity of the dysarthria. These observations suggest that a better understanding of hypokinetic dysarthria may result from more extensive examination of the variability between
patients. Emphasizing a specific feature such as rapid speaking rate in characterizing hypokinetic dysarthria focuses on a single and inconstant finding in a complex speech pattern.

The clinical syndrome of hypokinetic dysarthria as presented by Darley

et al. (1975) suggested that the speech abnormality represented a relatively
uniform syndrome that occurred with Parkinsons disease and more recently has been noted in progressive supranuclear palsy (PSP) (Hanson
and Metter, 1980). PSP is a chronic progressive neurologic disorder of
unknown etiology that occurs in middle or late life. The clinical features
first described by Steele, Richardson, and Olezewski (1964) included
ophthalmoplegia (mainly of vertical gaze), dystonic rigidity of the neck,
pseudobulbar palsy, mild dementia, and dysarthria. Subsequent reports
have described prominent parkinsonian features (Klawans and Ringel,
1971; Blumenthal and Miller, 1969; Behrman et al. 1969). Pathologic features are different from Parkinsons disease and include nerve cell loss,
gliosis, neurofibrillary tangles, and demyelination in the basal ganglia and
cerebellum. Marked atrophy of the midbrain and pontine tegmentum usually exists (Steele et al., 1964).
Although the hypokinetic dysarthria is distinctive in these disorders, it
has primarily been described from selected subjects with severe dysarAddress correspondence to E. Jeffrey Metter, Department of Neurology, Veterans Administration Medical Center, 16111 Plummer Street, Sepulveda, CA 91343.
0 1986by Elsevier Science Publishing Co., Inc.
52 Vanderbilt Ave., New York, NY 10017

347
002l-9924/86/$03.50

348

E. J. METTER and W. R. HANSON

thria. Some speech features may vary greatly in parkinsonian patients

who have less severe dysarthria, particularly in measures of speaking rate,
fundamental frequency, and vocal intensity. Marked variability is also
present in other clinical features that may occur in patients with hypokinetic dysarthria, including bradykinesia, tremor, rigidity, and overall degree of disability. Additionally, little information has been available concerning the agreement between speech adequacy and the overall degree
of disability associated with the disease in these patients. Likewise, little
emphasis has been placed on how the dysarthria relates to other features
of the basal ganglia disorders.
The purpose of this study was to compare the clinical syndromes, brain
anatomy, and glucose metabolism with speech characteristics in parkinsonian patients with varying degrees of hypokinetic dysarthria. Of interest
was whether or not specific clinical, metabolic, or structural findings relate to some aspect of the dysarthria. Also of interest was the extent of
variability that may exist between patients across the studied features.
METHODS
Seven male Parkinsons disease patients and one PSP patient were selected who had complete neurologic evaluations, x-ray CT, positron emission tomography (PET) and (F 18)-fluorodeoxyglucose
(FDG), and motor
speech evaluations. These eight subjects included one with severe and
one with moderately severe dysarthrias. For this reason, two other subjects were included with severe hypokinetic dysarthrias who had complete
motor speech and neurological evaluation but who were without CT or
FDG-PET. Of the latter two patients, one had progressive supranuclear
palsy and the other Parkinsons disease. The ten subjects had a mean age
of 62 years, and the range was from 54 to 76 years. Age-matched controls
for comparison of FDG scans were taken from the study of Kuhl et al
(1982). Each subject had a neurologic examination that included ratings
from 0 (normal) to 3 (severe) of bradykinesia, tremor, rigidity, and dementia. They were also rated by the staging scale of Hoehn and Yahr
(1958), and by the scale of Webster (1979). All subjects were receiving
medication at the time of testing. The eight subjects with Parkinsons
disease were all taking levodopa. Neither PSP subject had responded to
levodopa. One was receiving methylsergide, the other was taking anticholinergics. A clinical note was made for the degree of dyskinesia in
each case.
Acoustic speech measurements including speaking rate, mean fundamental frequency, relative intensity, and vowel phonation time were made
from tape-recorded speech samples. The recordings were analyzed using
a microprocessor-controlled
speech analyzer (PM 301, Voice Identification, Inc.), which we have described in a previous publication (Hanson

HYPOKINETIC

DYSARTHRIA

349

and Metter, 1983). Each subject was recorded while seated in a soundtreated test room (IAC model 403A) directly in front of a microphone
(Electrovoice model RE-15) that was coupled to an Ampex tape recorder
(AF 600B) located in an adjacent test room. The mouth-to-microphone
distance was 8 inches. The speech samples consisted of reading aloud the
Grandfather Passage, and maximum sustained phonation of ah. Ten
age-matched normal controls were tape recorded for comparison of acoustic speech measures.
The authors agreed on a rating of overall intelligibility, dysphonia, articulation, prosody, and hypernasality using the methods of Darley et al.
(1975) for each connected speech sample. Each characteristic was graded
on a seven-point scale (1 = normal to 7 = most severe deviation from
normal). An index of total dysarthria severity was derived by adding together the scores for each of the five measures, resulting in a dysarthria
scale ranging from 5 to 35.
X-ray computed tomography (CT) was done as part of the clinical evaluation and was evaluated qualitatively. Positron emission tomography
(PET) was done on the ECAT II (Ortec, Oakridge, Tennessee) (Phelps
et al., 1978). (Fl8)-fluorodeoxyglucose
(FDG) was used as the isotope to
examine glucose metabolism as modeled by Phelps et al. (1979). The resulting scans of the brain were projected onto a monitor, regions of interest
were outlined, and corresponding local cerebral metabolic rate for glucose
(LCMRGlc) were determined by the computer. Quantitative comparison
of CT, PET, and clinical symptomatology of the Parkinsons disease subjects have been reported by Kuhl, Metter, and Riege (1984).
RESULTS
A number of measured speech parameters showed extensive variation in
our subjects. To examine whether or not pattern to the variation existed,
we focused on several features that were of particular interest, including:
the clinical status (ratings of physical disability), severity of overall dysarthria, and speaking rate. These features were then compared across a
number of variables.
Clinical Status: General Features
Clinical severity of parkinsonism was evaluated using the Webster scale,
because it showed a strong correlation with the scales of Hoehn and Yahr
(1958) and with the clinical scales from the neurologic examination. The
Webster scale is linear with a range of 0 (normal) to 30 (most severe).
Mild parkinsonism ranges from 0 to 10, moderate from 11 to 20, and severe
21 to 30. Comparing the level of clinical disability to the severity of dysarthria as judged by the dysarthria scale (Figure 1) demonstrated that

350

E. J. METTER and W. R. HANSON

0
0

10
PARKINSON

15
DISABILITY

20

25

30

SCORE

Figure 1. Scattergram showing relations between perceptual index of total dys-

arthria severity and Parkinson disability as judged by the Webster scale (1979).

the most severe dysarthrias occurred with both mild and severe parkinsonian disability ratings. One subject with severe dysarthria but mild clinical disability was a 58-year-old male with a 6-to-g-year history of Parkinsons disease. He initially presented complaining of a change in his
speech, which was interfering with his ability to work as a traveling salesman. He was noted by the examining physician to have a dysarthria and
very mild features of parkinsonism. Over the years, the dysarthria became
progressively worse more rapidly than other features of his disease. He
is now essentially unable to communicate by speech, but he is independent
in activities and is able to drive while being treated with levodopa. In
contrast to this patient were two patients with severe clinical features
who also displayed severe dysarthria. Both of these individuals had PSP.

HYPOKINETIC

DYSARTHRIA

351

Clinical Status: Specific Features

The lack of an absolute relationship between total parkinsonian disability
and the severity of dysarthria raised the question of whether or not specific
features of the parkinsonian syndrome might show association with the
dysarthria. To answer this question, each subcategory of the Webster
disability scale was studied independently. Ratings for three of these independent features, bradykinesia, rigidity, and facial motility, were compared to total dysarthria scores and no relationships were apparent. Likewise, no relationship was observed between tremor, dyskinesia, or
duration of parkinsonism and severity of dysarthria. The duration of disease for the ten subjects ranged from 2 to 18 years. Two subjects with
severe dysarthria had had their illnesses for 2 and 10 years, respectively.
Subjects with mild to moderate dysarthria showed great variation in the
duration of their disease.

Speaking Rate
One of the most interesting and seemingly variable features of hypokinetic
dysarthria is speaking rate. Our subjects, when compared to controls,
form a continuum from much slower (77 wpm) than normal to much faster
(263 wpm) than normal speaking rates. Rate did not relate to either the
severity of clinical symptoms or severity of dysarthria as seen in Figures
2 and 3. Of three subjects with mild parkinsonism (Webster score < 1l),
one had an abnormally fast (245 wpm), one had a slow (96 wpm), and one
a normal rate (168 wpm) as compared to normal controls (range 118 to
186 wpm). Of two subjects with the most severe clinical parkinsonism,
one had a very rapid rate (263 wpm) and the other a slow rate (77 wpm).
Normal speech rates were not observed in subjects with either severe
clinical parkinsonism or with severe dysarthria. Both subjects with progressive supranuclear palsy had severe dysarthrias, one with a rapid rate,
the other with a slower than normal rate. Likewise, of the subjects with
Parkinsons disease, speech rates were found that were slower and faster
than normal. Thus, extremes of rate were not specific for either disorder.
No relationship was found between the degree of rigidity or bradykinesia
and speaking rate.

Fundamental

Frequency

Mean fundamental frequency for the reading passage was examined in

relation to the clinical and dysarthria scales. Figure 4 demonstrates mean
fundamental frequency in relation to the total dysarthria score and Webster
scale for each patient. Data are also shown for normal control subjects.

352

E. J. METTER and W. R. HANSON

2700

250-

0
0

0
0
0
0
0

51-___-5

10

(5

20
DYSARTHRIA

25

30

35

SCORE

Figure 2. Scattergram between speech rate in words per minute and perceptual
total dysarthria score. x = normal subjects, 0 = total dysarthria scores.

Mean fundamental frequency for each patient was within the normal
range, though there was a tendency for fundamental frequency to increase
with increased clinical disability and with increased severity of the dysarthria. Also, six of 10 subjects had a mean fundamental frequency
greater than 130 Hz, whereas only one control was above this frequency.
Variability of Fundamental

Frequency

Clinically, hypokinetic dysarthria is characterized by monopitch and hypoprosody. To examine monopitch, the variation of fundamental frequency for each subject was examined during the reading of the Grandfather Passage. In Figure 5, coefficients of variability for frequency (the
ratio of individual mean fundamental frequency standard deviation by the
mean fundamental frequency) are plotted in relationship to dysarthria
score, speech rate, and parkinsonian score. Some degree of inverse re-

353
270
0

250

x = Normal
o = Parkinsons

200

%
z 150

?
z!

0
0

8 100

50

15
20
IO
PARKINSON DISABILITY SCORE

25

30

Figure 3. Scattergram between speech rate in words per minute and Parkinson
disability score (Webster, 1979). x = normal subjects, 0 = Parkinson disability
score.
Figure 4. Scattergrams between mean fundamental
scores and Parkinson disability scores.
200

O- Dysarthrio score
l = Parkinson
score

160-

.D

150. a
140.

.O

130.
P
120mx

and total dysarthria

x = Normal

190

ClO-l1
I
10
0

frequency

0
D

OD

.
.

10

15

20

25

30

35 TOTAL DYSARTHRIA
SCORE

IO

15

20

25

PARKINSON
3o DISABILITY SCORE

354

E. J. METTER and W. R. HANSON

5
I3 58!z 5

0
0

10

t5

20

TOTAL
z

X
xx

&
z

SCORE

$20a

35
I

30

>

25

DYSARTHRIA

25-

15-

00

x0

X
X

io0

5-

8
E0

0
t

50

too
SPEECH

150
RATE

200
( WPM)

250

300

25

30

i? 25lx
0
0

000
0

25
;

00
o*

10
PARKINSON

45

20

SCORE

Figure 5. Scattergrams

between frequency CV (coefficient of variability) and total

dysarthria score, speech rate, and Parkinson disability scores. x = normal, 0
= clinical group.

lation was apparent between FF-CV and both dysarthria severity and
clinical disability, but not with speaking rate. Subjects with severe dysarthria showed FF-CV less than normal subjects. A low FF-CV was also
seen in one subject with only a mild to moderate dysarthria, so that the
change was not absolutely related to severity. Abnormal FF-CVs were
observed in the presence of mean fundamental frequencies that were
within the range of the normal controls, demonstrating the specificity of
the problem of fundamental frequency variation.

HYPOKINETIC

355

DYSARTHRIA

Variability of Intensity
Mean relative intensity measures for the reading passage showed no apparent differences between the hypokinetic patients and controls. A comparison of variability of intensity (I-CV) to dysarthria scores, however,
demonstrated that only the parkinsonian subjects with severe dysarthria
showed a decreased variation in intensity as compared to controls (see
Figure 6). Comparing I-CV to the rate of speech, loss of normal intensity
Figure 6. Scattergrams

between intensity variability

and total dysarthria score, speech rate, and Parkinson
mal, 0 = clinical group.

10

15

20

TOTAL

25

DYSARTHRIA

0 Ox4

(coefficient of variability)
disability score. x = nor-

XkP

30

x
0

50

100

150

SPEECH
00
E
>Q

i5-

io-

35

SCORE

200

RATE

250

I
300

25

,
30

(WPM)

00

0
00

v,

65
5-

O*,

10

r
20

15
PARKINSON

SCORE

356

E. J. METTER

and W. R. HANSON

variability was associated with both slow and fast rates but did not appear
in the normal range of speaking rates.

Vowel Phonation Time

The ability to sustain phonation of ah was highly variable in our subjects. The range of phonation time varied from 6.12 to 37.89 seconds, as
compared to the normal subjects who ranged from 11.72 to 33.94 seconds.
Four parkinsonian subjects had phonation times less than 10 seconds, but
the longest vowel prolongation for the 20 subjects was a parkinsonian
individual (37.89 seconds). Three of four parkinsonian subjects with abnormally short vowel prolongation had only mild dysarthria (scale scores
of 8, 14, and 14). No relationship was noted between vowel phonation
time and degree of disability or severity of dysarthria.

Pause Time
An observation commonly made in hypokinetic dysarthria is the presence
of abnormal pause time in connected discourse. This abnormality ranges
from a complete lack of pauses, as seen in the rapid, fused form of the
dysarthria, to the extended pauses noted in some speakers with very slow
speaking rate. To illustrate the variations in pause time that may exist in
Figure 7. Fundamental

frequency, relative intensity, and duration for the phrase

a long flowing beard clings to his chin spoken by three subjects with hypokinetic dysarthria.

HYPOKINETIC

357

DYSARTHRIA

different hypokinetic speakers, tracings of a phrase from the Grandfather

Passage are presented in Figure 7. Outputs (fundamental frequency and
relative intensity) from the Pitch Analyzer (PM301) are displayed simultaneously on two separate channels of a Mingograph (Model 805), operated at a paper speed of 50 mm per second. From these tracings the
total duration of each phrase, indicative of the great variability in speaking
rate, can be clearly seen. Also apparent is the variability in the location
and extent of pauses within each phrase. The rapid, fused, aprosodic
pattern without pauses displayed by patient A stands in marked contrast
to the slow, prolonged pause pattern of patient C. However, common to
both tracings is a flat fundamental frequency contour consistent with the
reduced coefftcient of variability noted earlier, which seems to be characteristic of many speakers with hypokinetic dysarthria. To examine
pauses further, we measured the percentage of the speech sample that
represented pause time for each subject (Figure 8). In both normal and
Figure 8. Scattergrams of the percentage of the speech sample that was pause
time and total dysarthria score and speech rate. x = normal, 0 = clinical group.

0
0

0
0

0
0

154
5

10

50
45

15

20

$5

TOTAL

DYSARTHRIA

SCORE

I
30

35

0
OX

40 i
35-

X
0

xx

30-

x&J
x

25-

x
X

20157
0

30

100
SPEECH

(50

200

RATE (WPM)

250

I
300

358

E. J. METTER

and W. R. HANSON

hypokinetic subjects, we found an inverse relationship between the pause

time and speech rate, although the slopes of the two relationships seem
to be different. In general, for a given speech rate, the hypokinetic subjects
had a greater percentage of pausing than did corresponding normal subjects. The degree of abnormal percentage of pause time was independent
of the degree of dysarthria, occurring even in the subjects with the mildest
degrees of dysarthria.
Effect of Treatment: Individual Case Study
One issue in hypokinetic dysarthria is the effect that medication, particularly levodopa, may have on speech. Drug treatment appears to exert
differential influence on various features of the dysarthria. This can be
demonstrated by considering a 58-year-old patient (Patient D) studied on
and off medication. He had a 3-year history of mild Parkinsons disease
based on the Webster scale (total score = 4) with mild rigidity, decreased
arm swing, facial immobility, and dysarthria while on levodopa. In addition, he had dyskinesias and on-off phenomena that was treated with a
drug holiday. On stopping his levodopa he became clinically worse, with
a Webster score of 19. He was able to ambulate slowly and with great
difficulty. He had severe loss of arm swing, moderate bradykinesia, rigidity, facial immobility, and he required moderate assistance in most
activities of daily living.
On medication, the patient spoke with a slow speech rate (96 wpm), a
prominent dysphonia that was hyperphonic with a relatively high fundamental frequency (mean = 142 Hz, SD = 28 Hz) for the reading sample.
Off medication, his intelligibility went from good to moderately impaired,
with a change in his overall dysarthria scale from 14 to 23. A change also
occurred in his speech, which became slower in rate with increases in
both articulation and pause time (Figure 9). This was associated with little
change in his dysphonia but a moderate decrease in articulatory precision,
prosody, and intensity. No changes were found in mean fundamental frequency, variability of intensity, or percentage pause time. Vowel prolongation that was short initially (7.70 seconds) fell by 51% (3.79 seconds).
These observations demonstrate the differential changes that can occur
in speech functions in conjunction with drug treatment with levodopa.
Variability of Severe Dysarthria in PSP
Hypokinetic dysarthria may include some speech features that are distinctly different among subjects. This can be demonstrated in the two
subjects with PSP (see Table 1). Both subjects were 58 years of age and
showed severe disability, with Webster-scale scores of 21 and 22. Neither
patient showed any improvement with L-DOPA. Both showed moderate

HYPOKINETIC

DYSARTHRIA

359

Figure 9. Fundamental
frequency,
relative intensity,
and duration for the phrase
a long flowing beard clings to his chin produced
by a Parkinsons
disease
patient on and off of treatment
for L-DOPA.

to severe rigidity, bradykinesia,

were unable to look down, and had neck
retroflexion. The first subject had been ill for at least 10 years, and for
the first six years carried a diagnosis of Parkinsons disease before the
full clinical picture of PSP appeared. His speech was rapid, with monopitch, monoloudness, decreased excursion of mouth movements, and decreased intensity. His reading passage showed a speaking rate of 263 wpm,
the highest mean fundamental frequency observed in our patient group

Table 1. Comparison
(PSP).
Measure
Speech rate (wpm)
Articulation time (set)
Pause time (set)
% pause time
Mean frequency (Hz)
Frequency variability
Intensity variability
Phonation time (set)
Dysarthria score
Webster scale
Years of disease

of Two Subjects

with Progressive

Patient 1
263
22
8.35

Supranuclear

Palsy

Patient 2

Normal Range

77
57
45.74

118-186
35-47
7.66-25.71

28%

45%

151
.052
131
12
28
22
10

147
.040
133
7
26
21
2

18-42%

116-152
.lll-,247
188-224
12-34

360

E. J. METTER

and W. R. HANSON

(151 Hz), and a very low FF-CV (.052 compared to the normal range of
. 111 to .247), indicative of his monopitch. He also showed limited ability
in modifying his loudness as measured by the small coefficient of variability for intensity measures (. 131 as compared to the normal range of
.188 to .224). Visual inspection of his speech tracing (Patient A, Figure
7) demonstrated almost constant voicing.
The second subject with PSP had his illness for only 2 years when
studied. His dysarthria was similar in some respects to the patient just
described; however, on two measures he differed greatly. First, he had
a speaking rate of 77 wpm (Patient C, Figure 7), and second, he had a
prolonged percentage of pause time (48%) as compared to the above case
(28%) and to normal subjects (range 18-42%). The marked variability in
speech rate and associated pause time in these two subjects appeared to
be independent of other features of their dysarthria.
Effect of Time
Our general belief is that as parkinsonism progresses clinically a corresponding change should occur in speech. We have shown above a case
in which the dysarthria was the presenting problem for a subject, consistent with a dissociation between speech and other clinical features. A
second approach is to examine subjects repeatedly over time. One subject
was evaluated on two occasions over a 4-year period. Clinically, the subject had shown a progression of his illness with a Webster scale increasing
from 7 to 13 (mild to moderate disability). This progression was associated
with increased rigidity and increased difficulty walking. The patient also
began to have some on-off phenomena. At the most recent evaluation he
also began to have marked problems in vertical eye movements suggesting
that he may evolve a diagnosis of PSP rather than of Parkinsons disease.
Over the 4-year period, his dysarthria actually showed some improvement, with the overall dysarthria severity score falling from 14 to 9.
Acoustic speech measurements indicated that his speech rate slowed from
168 to 153 wpm, and mean fundamental frequency increased from 104 Hz
to 124 Hz. His FF-CV showed a marked increase from .173 to .274, the
latter measure being the highest observed in the 20 subjects. No change
was observed in the I-CV. The principal observation in this case is that
progression of clinical features of parkinsonism can be dissociated from
changes occurring in the dysarthria.
Anatomic Correlates: CT
CT scans of the head did not show any distinct changes that could be
related to any clinical feature of the illness or dysarthria in the seven
subjects who underwent CT scans. Most subjects showed some degree
of cortical atrophy, but this was consistently mild.

HYPOKINETIC

DYSARTHRIA

361

Glucose Metabolic Correlates: FDG PET

The metabolic data for these subjects while on treatment showed their
mean metabolism to be in the low .normal range, as were the rates for
each region studied (Kuhl et al., 1984). Global metabolic rates were significantly decreased as compared to controls, but no significant differences were found for language-related cortex, thalamus, or caudate. No
correlations were found between glucose metabolic rates and duration of
illness, stage of illness, dementia, tremor, rigidity, or dystonia. Scaled
scores of bradykinesia were correlated (r > 0.70, p < .OS) with mean
cortical and caudate LCMRGlc. Correlations were calculated between
the individual components of our dysarthria scale and local glucose metabolic rates for left inferior frontal and caudate regions. None were found
to be significant at the .05 level uncorrected for the number of comparisons. In general, no regional explanations were found for the nature or
extent of dysarthria in these subjects based on the comparison of regional
glucose metabolism.
Several observations were of interest in the subject who was described
on and off treatment (see Figure lo), because he had FDG-PET scans in
both states. Metabolic rates for glucose were far higher while he was
treated than when off of medication (a mean glucose metabolic rate of 26
Figure 10. Positron tomographs of FDG that represent the distribution of glucose
utilization. Four brain sections are presented on (upper figures) and off (lower
figures) of L-DOPA therapy for one subject with Parkinsons disease. The darker
the region, the greater is the LCMRGlc.

362

E. J. METTER and W. R. HANSON

Figure 11. Positron tomographs of glucose utilization in subject with progressive

supranuclear palsy (PSP).
brain regions being 3.77 mg / 100 gms tissue / minute off medication and
6.18 on medication). He was the only subject of three studied both on
and off treatment who showed a difference in metabolic rates. The greatest
relative changes were found in both high frontal areas, Wernickes region,
and in both thalamic regions. The meaning of this response is unclear.
The subject with PSP and rapid speech rate (see Figure 11) demonstrated a 10% depression in his left hemisphere compared to right hemisphere metabolism. This unilateral depression was not seen in any other
subject or in 31 controls but has been seen in some stroke patients with
deep lesions in which the mechanism has been assumed to involve cortical
activating systems (Kuhl et al., 1982; Metter et al., 1981, 1983). Whether
or not this observation can be associated with the clinical state in this
subject can only be speculated.
DISCUSSION
Acoustic characterization
of hypokinetic dysarthria has placed emphasis
on the rapid speaking form with reduced syllable durations, monotone
fundamental frequency, and continuous voicing. Such speech has been
referred to as fused (Kent and Rosenbek, 1982). Our data show that
other acoustic patterns exist in hypokinetic dysarthria, including one with
a very slow speaking rate that is characterized by articulated islands of
speech separated by extended pauses (see Figures 7 and 9). These observations are consistent with a number of other studies that have noted
variability in speech rates (e.g. Canter, 1965; Kreul, 1972; Boshe, 1966).
Of particular interest in our study was the observation that other measured
speech features tended to show large variability without a specific relationship to the speech rate. This implies that variation in rate does not
dictate differences in the overall dysarthria but rather reflects only one
feature of a much more complex problem.

HYPOKINETIC

DYSARTHRIA

363

Mild to moderate severity of dysarthria appeared to be associated with

relatively normal rates of speech. With increasing severity of the dysarthria, the speech tended to be either increased or decreased in rate. Other
measures associated with more severe dysarthria seemed to vary in a
similar manner independent of a fast or slow rate. These included a relative increase in mean fundamental frequency with decreased fundamental
frequency variation. What appeared to be important in the appearance of
a variety of abnormal acoustic measures was the degree of severity of
the dysarthria.
The similarity of acoustical findings between Parkinson disease patients
and those with PSP suggests that hypokinetic dysarthria is not disease
specific but rather reflects structural and/or functional abnormalities
within the brain that are common to more than one disease. The appearance and extent of dysarthria in hypokinetic illnesses appears to be
a phenomenon related to but independent of the appearance of any specific parkinsonian features. Further, our results show that severe dysarthria can occur in the presence of only mild clinical symptoms.
The lack of strong relationships between other clinical features and the
dysarthria suggest that basal ganglia control of speech is somewhat different than for other movements. This may involve differences in specific
locations within the basal ganglia, differences in the physiology of basal
ganglia action on speech as compared to other movements, or differences
in the peripheral organization of the motor system. The latter has been
discussed by Hunker, Abbs, and Barlow (1982). Current techniques for
studying the basal ganglia in vivo are not sensitive enough to allow for
this distinction in man. Each basal ganglia nucleus is anatomically uniform
in structure, but specific regions tend to receive greater innervation from
specific cortical regions, and in fact cortical regions having strong cortical
interconnections
will tend to innervate similar regions of the caudate and
putamen (Yeterian and Van Hoesen, 1978). This implies that if local areas
of the basal ganglia are differentially affected, then selective involvement
of speech function could occur. This would explain the findings in one of
our subjects who had a severe hypokinetic dysarthria but few other clinical signs of Parkinsons disease.
From previous FDG PET studies, we have speculated that part of the
function of the basal ganglia is to assist or allow cortical regions to function
together in carrying out a task, acting much as a pacemaker (Metter et
al., 1984). To examine caudate function, two basal ganglia diseases, Parkinsons (PD) and Huntingtons (HD), were compared to a cortical disease, Alzheimers (AD), and to controls. Examining regional correlation
matrices among 13 regions measured in each hemisphere within each disease group, the number of cortical regional reliable correlations (r values
selected for p < .Ol unadjusted) was found to decrease in HD and PD

364

E. J. METTER and W. R. HANSON

compared to age-matched controls. This differed from AD, in which an

increase was found in cortical to cortical correlations. The findings suggested a loss in focusing of cortical activity related to basal ganglia pathology, which may account for some clinical features in HD and PD.
Thus, basal ganglia damage could disrupt the ability of the cortex to integrate the complexities of speech and tend to dissociate the laryngeal
from articulatory functions and to disrupt prosody. Evaluation of CT
and FDG PET in this study did not allow for further refinement of the
earlier speculations.
To best understand hypokinetic dysarthria, a greater emphasis should
be placed on the extent of variability between patients. Emphasizing a
specific feature such as rapid speech rate to characterize hypokinetic
dysarthria may be deceiving. Few variables specifically related to rate
alone. The most meaningful understanding may result from an evaluation
of a large mixed sample of parkinsonian subjects, examined using acoustical measures, with factor and cluster analysis performed in a manner
similar to that used by Darley et al. (1975). It will also be necessary to
study patients displaying a range of severity, whereas evaluating only
subjects with moderate to severe dysarthria may be misleading. A greater
emphasis should be placed on the extent of variability between patients.
In our study, speech rate seemed to be independent of the severity of the
parkinsonian symptoms exhibited by the subjects. In addition, few of the
acoustic variables measured appeared to relate directly to speaking rate
alone. It is clear that patients within the same type of dysarthria may have
markedly dissimilar acoustic profiles. Univariate comparisons that emphasize a single specific feature as characteristic of the whole group may
be misleading. Multivariate approaches more accurately differentiate subcategories of patients and should be of greater utility in identifying degrees
of impairment and the effects of treatment for dysarthria. The acoustic
profile method used to describe the pattern and severity of speech impairment highlights the differences that exist between patients and may
result in improved treatment strategies. For the most propitious management of patients with hypokinetic dysarthria, the extreme variability in
individual performance must be given due consideration. To understand
the problem further, studies that include large mixed samples of parkinsonian subjects, multiple acoustic measurements, and factor and cluster
analyses must be conducted. In addition, the type and severity of the
impaired speech features must be identified and studied in relationship
to other clinical symptoms.
The FDG studies were funded in part by Department
SSOOO12 and U.S. Public Health Service Research
15654.

of Energy Contract
Grants ROl-GM-24839

#DE-AM03-76and POl-NS-

HYPOKINETIC

DYSARTHRIA

365

REFERENCES
Behrman, S., Carroll, J. D., Janota, A., and Matthews, W. B. (1969). Progressive
supranuclear palsy: Clinico-pathological
study of four cases. Brain 92:663-678.
Blumenthal, H., and Miller, C. (1969). Motor nuclear involvement
supranuclear palsy. Arch. Neurol. 20:362-367.
Boshe, B. (1966). Voice changes in parkinsonism.

J. Neurosur.

in progressive
24:286-288.

Canter, G. J. (1963). Speech characteristics of patients with Parkinsons

I. Intensity, pitch, and duration. J. Speech Hear. Dis. 28:221-229.

disease:

Canter, G. J. (1965). Speech characteristics of patients with Parkinsons disease:

III. Articulation
diadochokineisis,
and overall speech adequacy. J. Speech
Hear. Dis. 301217-224.
Darley, F. L., Aronson, A. E., and Brown, J. R. (1975). Motor Speech Disorders.
Philadelphia: Saunders.
Hanson, W. R., and Metter, E. J. (1980). DAF as instrumental treatment for
dysarthria in progressive supranuclear palsy: A case report. J. Speech Hear.
Dis. 451268-275.

Hanson, W. R., and Metter, E. J. (1983). DAF speech rate modification in Parkinsons disease: A report of two cases. In W. Berry (ed.), Clinical Dysarthria.
San Diego: College Hill Press.
Hoehn, M. M., and Yahr, M. D. (1958). Parkinsonism:
mortality. Neurology 17:427-442.

Onset, progression,

and

Hunker, C. J., Abbs, J. H., and Barlow, S. M. (1982). The relationship between
parkinsonism rigidity and hypokinesia in the orofacial system: A quantitative
analysis. Neurology (NY) 32:749-754.
Kent, R. D., and Rosenbek, J. C. (1982). Prosodic
lesion. Brain and Language 15:259-291.

disturbance

and neurologic

Klawans, H. L., and Ringel, S. P. (1971). Observations on the efficacy of LDOPA in progressive supranuclear palsy. Europ. Neurol. 5: 115-129.
Kreul, E. J. (1972). Neuromuscular control examination (NMC) for parkinsonism:
Vowel prolongations and diadochokinetic and reading rates. J. Speech Hear.
Res. 15:72-83.

Kuhl, D. E., Metter, E. J., and Riege, W. H. (1984). Patterns of local cerebral
glucose utilization determined in Parkinsons disease by the (18F) fluorodeoxyglucose method. Annals of Neurology 15:419-424.
Kuhl, D. E., Metter, E. J., Riege, W. H., and Phelps, M. E. (1982). Effects of
human aging on patterns of local cerebral glucose utilization determined by the
(180 fluorodeoxyglucose
method. J. Cerebral Blood Flow and Metabolism
2: 163-171.

Logemann, J. A., Fisher, B., Boshes, B., and Blonsky, E. R. (1978). Frequency
and cooccurrence of vocal tract dysfunctions in the speech of a large sample
of Parkinson patients. J. Speech Hear, Dis. 43~47-57.
Metter,

E. J., Hanson,

W. R., Riege, W. H., Kuhl, D. E., and Phelps, M. E.

366

E. J. METTER

and W. R. HANSON

(1983). The use of (F18)-fluorodeoxyglucose

positron computed tomography in
the study of aphasia: A review. In R. H. Brookshire (ed.), Clinical Aphasiology
Conference,
1983. Minneapolis: BRK Publishers.
Metter, E. J., Riege, W. H., Hanson, W. R., Kuhl, D. E., Phelps, M. E., Squire,
L. R., Wasterlain, C. G., and Benson, D. F. (1983). Comparisons of metabolic
rates, language and memory in subcortical aphasias. Brain and Language
19:33:47.

Metter, E. J., Riege, W. H., Kameyama, M., Phelps, M. E., and .Kuhl, D. E.
(1984). Cerebral metabolic relationships
for selected brain regions in Alzheimers, Huntingtons,
and Parkinsons diseases. Journal of Cerebral Blood
Flow and Metabolism

4:500-506.

Metter, E. J., Wasterlain, C. G., Kuhl, D. E., Hanson, W. R., and Phelps, M.
E. (1981). 18FDG positron emission computed tomography in a study of
aphasia. Ann. Neurol. 10:173-183.
Phelps, M. E., Hoffman, E. J., Huang, S. C., and Kuhl, D. E. (1978). ECAT: A
new computerized tomographic imaging system for positron emitting radiopharmaceuticals.
J. Nucl. Med. 19:635-647.
Phelps, M. E., Huang, S. C., Hoffman, E. J., Selin, C. S., Sokoloff, L., and
Kuhl, D. E. (1979). Tomographic measurement of local cerebral metabolic rate
in humans with (F-18) 2-fluoro-2-deoxyglucose:
Validation of method. Annals
of Neurology

6:371-388.

Steele, J. C., Richardson, J. C., and Olezewski,

nuclear palsy. Arch. Neurol. 10:333-359.

J. (1964). Progressive

D. D. (1979). Webster Rating Scale, Parkinsons

uation. Garden City, N.Y.: Endo Laboratories,
Inc.

Webster,

supra-

Disease Patient Eval-

Yeterian, E. H., and Van Hoesen, G. W. (1978). Cortico-striate projections in

the rhesus monkey: the organization of certain cortico-caudate
connections.
Brain Research

139:43-63.