Dysarthria

Abstract

Dysarthria is a group of motor speech disorders in which weakness, slowness, or incoordination of the muscles of speech distorts the sound of talking while language itself is intact. It is the execution of speech that fails, not the choice of words, which sets it apart from aphasia. In a landmark study at the Mayo Clinic, Darley, Aronson, and Brown listened systematically to hundreds of patients and sorted the disorder into perceptual types — flaccid, spastic, ataxic, hypokinetic, hyperkinetic, and mixed — each the acoustic signature of a lesion at a particular level of the motor system. This article surveys that classification, the five speech subsystems it draws on, the neuroanatomy behind each type, standardized assessment and behavioural treatment, with a worked example and three interactive demonstrations.

Keywords: dysarthria, motor speech, intelligibility

Dysarthria occupies the boundary between the mind and the machinery that expresses it. A person with dysarthria knows exactly what to say, assembles it into perfect grammar, and retrieves every word — and still cannot make the muscles produce it clearly. The disorder therefore isolates the motor-execution stage of speaking from everything upstream, and the *way* it breaks is diagnostic: a slur is not just a slur, but a clue to where along the motor pathway the damage lies (Darley, Aronson, & Brown, 1969a).

The account below moves from what dysarthria is, through the six perceptual types the Mayo studies defined and the five subsystems of speech they rest on, to the neuroanatomy that explains why each type sounds as it does, the standardized assessment of severity and intelligibility, and the behavioural treatments that improve it. Throughout, the theme is that dysarthria is not one disorder but a family of motor breakdowns, each type a fingerprint of its lesion.

Figure 1

The Motor Pathway for Speech and the Types It Produces

What Dysarthria Is

Dysarthria is a disturbance of speech caused by disordered control of the muscles that produce it. The muscles may be weak, slow, imprecise, or poorly coordinated, and the result is speech that is slurred, strained, breathy, monotonous, or imprecise, depending on which muscles and which movements are affected. The defining feature is that the fault lies in the *motor* system — the nerves and muscles of speech and the brain circuits that drive them — not in language.

This is the crucial distinction from aphasia. An aphasic patient may articulate every sound cleanly yet produce the wrong words or no words; a dysarthric patient knows precisely which words to say and in what order, but the delivery is distorted because the muscles will not obey. Dysarthria is also distinct from apraxia of speech, in which the muscles are strong but the *programming* of the movement sequence fails; in dysarthria the deficit is in execution, and it is usually consistent from attempt to attempt rather than groping and variable (Darley, Aronson, & Brown, 1969a).

Because speech is driven by a long motor pathway — from the cortex, through the basal ganglia and cerebellum that shape movement, down to the brainstem nuclei and the cranial nerves that reach the muscles — almost any neurological disease that touches that pathway can cause dysarthria. It is common after stroke (Mackenzie, 2011), in Parkinson disease, in cerebellar disease, in amyotrophic lateral sclerosis (Tomik & Guiloff, 2010), in cerebral palsy (Allison & Hustad, 2018), and in multiple sclerosis and traumatic brain injury. The type of dysarthria that results is a clue to *where* in that pathway the disease has struck.

Types of Dysarthria

The modern classification of dysarthria comes from a pair of studies published in 1969 by Frederic Darley, Arnold Aronson, and Joe Brown at the Mayo Clinic. They had trained listeners rate the speech of patients with a range of neurological diseases on dozens of perceptual dimensions — pitch, loudness, voice quality, articulatory precision, rate, and more — and then asked which dimensions clustered together (Darley, Aronson, & Brown, 1969a). The clusters were not arbitrary: each corresponded to a level of the motor system and to a recognizable *type* of dysarthria (Darley, Aronson, & Brown, 1969b). This perceptual method — diagnosing the site of the lesion by ear — is the foundation of the field and remains in clinical use.

The scheme recognizes six primary types, each the signature of a lesion at a particular level (a seventh, unilateral upper motor neuron dysarthria, was added later as a milder, more focal category).

The mixed type is the commonest in practice, because many diseases do not respect the tidy levels of the motor system. Amyotrophic lateral sclerosis, which attacks both upper and lower motor neurons, produces a mixed spastic–flaccid dysarthria that is often the presenting sign of the disease (Tomik & Guiloff, 2010). In children with cerebral palsy the perceptual profile can be sorted into similar data-driven classes, extending the Mayo logic to a developmental population (Allison & Hustad, 2018).

The Five Speech Subsystems

To understand why each type sounds as it does, it helps to see speech as the joint product of five subsystems, each a set of structures that contributes one aspect of the acoustic result. Dysarthria is, in effect, a failure distributed across these subsystems, and the pattern of which are impaired defines the type.

Respiration supplies the airstream and the subglottal pressure that powers phonation; weak respiratory support gives short phrases and fading loudness. Phonation is the vibration of the vocal folds at the larynx; disordered phonation yields breathy, strained, or harsh voice. Resonance is the shaping of the sound by the oral and nasal cavities, governed by the soft palate; when the palate fails to close off the nose, speech becomes hypernasal. Articulation is the rapid, precise movement of the tongue, lips, and jaw that forms consonants and vowels; imprecise articulation is the single most common sign of dysarthria. Prosody is the melody, stress, and timing that carry emphasis and emotion; its loss produces the flat monotone of hypokinetic dysarthria or the disordered stress of the ataxic type (Darley, Aronson, & Brown, 1969b).

The subsystems are partly independent, so a given disease can strike one while sparing others: Parkinson disease attacks phonation and prosody first, while a lower-motor-neuron lesion of the palate strikes resonance. But they also combine multiplicatively in the ear of the listener, because the intelligibility of speech depends on all of them working together — a point the worked example makes precise (Kent, 2000).

Worked Example: Intelligibility as a Product

Intelligibility — the proportion of what a speaker says that a listener correctly understands — is the outcome that matters most to patients, and it depends on every subsystem at once. A useful illustrative model treats each of the five subsystems as contributing a factor between 0 (complete failure) and 1 (normal), and the overall intelligibility as their product:

$$I = R_{\text{resp}} \times R_{\text{phon}} \times R_{\text{reson}} \times R_{\text{artic}} \times R_{\text{pros}}$$

Suppose a patient with early hypokinetic dysarthria has near-normal respiration ($0.90$), moderately reduced phonation ($0.70$), intact resonance ($0.95$), mildly imprecise articulation ($0.85$), and flattened prosody ($0.75$). The predicted intelligibility is

$$I = 0.90 \times 0.70 \times 0.95 \times 0.85 \times 0.75 \approx 0.38$$

so only about 38% of the message gets through, even though no single subsystem is severely impaired. The model is deliberately multiplicative rather than additive to capture a clinical truth: because the factors multiply, intelligibility is dragged down by the *weakest* subsystems, and several mild impairments compound into a serious communication deficit. It also shows why targeting the worst subsystem yields the largest gain — raising phonation from $0.70$ to $0.95$ here lifts intelligibility from 38% to about 52%. This is exactly the logic behind subsystem-targeted therapy, and the model is illustrative, not an empirically fitted equation (Kent, 2000). The demonstration below lets the reader set the five factors and read off the product.

Neuroanatomy of the Types

The perceptual types are legible precisely because each level of the motor system contributes a different kind of control, and its loss sounds different. The lower motor neurons — the cranial nerves and the muscles they innervate — are the final common path; damage here produces weakness and the breathy, hypernasal, flaccid type. The upper motor neurons descending from cortex normally add both drive and inhibition; their bilateral loss releases spasticity, giving the strained voice and slow, effortful spastic type.

The basal ganglia regulate the scaling and initiation of movement. When dopamine is depleted, as in Parkinson disease, movements become small and under-powered, and speech follows suit into the reduced-loudness, monotone hypokinetic type; when the basal ganglia generate excess involuntary movement, as in dystonia or chorea, the hyperkinetic type results, its speech interrupted unpredictably. The cerebellum coordinates the timing and force of movements; its damage produces the ataxic type, with its irregular breakdowns and the drunken, over-measured “scanning” stress pattern. The cerebellum's contribution to the precise temporal control of speech has been a focus of research linking motor coordination to the acoustic details of articulation (Ackermann, 2008).

That each type maps to a level is what makes dysarthria diagnostically valuable: the sound of the speech is evidence about the location and nature of the underlying neurological disease. The perceptual method that Darley and colleagues built rests on this correspondence, and later work on the neurophysiology of speech motor control has largely borne it out while adding detail about how the levels interact (Kent, 2000).

Assessment

Dysarthria is assessed on two complementary levels: the perceptual features that identify the type, and the functional impact on communication. The standard bedside examination in much of the world is the Frenchay Dysarthria Assessment, which rates the movement and function of the speech musculature — reflexes, respiration, lips, palate, larynx, tongue — on structured scales, turning the clinician's perceptual judgments into scorable, comparable numbers (Enderby, 1980). Such structured examination is what allows the type to be identified reliably and change to be tracked over time.

Alongside the examination sits the measurement of intelligibility, usually as the percentage of words or sentences a naive listener transcribes correctly, and of communication efficiency, how much intelligible information is conveyed per unit time. Sentence-level intelligibility measurement, in particular, captures the functional severity that a feature-by-feature examination can miss, and it is sensitive to the compounding across subsystems the worked example described. In stroke, where dysarthria is common and often co-occurs with aphasia, careful description of the speech profile is needed to separate the two and to plan treatment (Mackenzie, 2011). Assessment also has to weigh listener factors, because intelligibility is a property of the speaker–listener pair, not of the speaker alone (McAuliffe & Fletcher, 2017).

Treatment and Recovery

Dysarthria is treated mainly behaviourally, by a speech-language pathologist, and the evidence that treatment works has grown substantially since the field's efficacy was first synthesized (Yorkston, 1996). The approach follows from the subsystem analysis: therapy targets the impaired subsystems directly — breath support for respiration, effortful closure for phonation, palatal training or a prosthesis for resonance, and articulatory drills for precision — or it adopts compensatory strategies such as slowing the rate to give the impaired system time to hit its targets.

The best-validated single treatment is for the hypokinetic dysarthria of Parkinson disease. Lee Silverman Voice Treatment (LSVT LOUD), developed by Lorraine Ramig and colleagues, trains the patient to produce loud phonation with high effort; because Parkinsonian speech is scaled down across the board, recalibrating loudness cascades into better articulation and prosody as well, and the gains persist for years after the intensive course (Ramig et al., 2001). A systematic review and meta-analysis of speech and language therapy for hypokinetic dysarthria in Parkinson disease confirms benefits across several speech outcomes (Muñoz-Vigueras et al., 2021). For dysarthria after stroke and other non-progressive brain injury, the evidence base is thinner but growing, and a Cochrane review found no definitive evidence of benefit from any one approach while underscoring the need for larger trials (Mitchell et al., 2017).

Discussion

The enduring lesson of dysarthria is that a motor act can fail in structured, diagnostic ways. Speech is one of the most complex voluntary movements the body performs, requiring the rapid coordination of some hundred muscles across the five subsystems, and the ways it breaks are not random. Each pattern of breakdown points back to a level of the motor system, so the sound of disordered speech becomes a window on the integrity of the nervous system — the insight that Darley, Aronson, and Brown converted into a working clinical method (Darley, Aronson, & Brown, 1969b).

The tension running through the field is between the *perceptual* tradition — classifying by trained listening — and the drive toward *instrumental* measurement, which quantifies the acoustic and physiological signal directly. The perceptual approach is fast, requires no equipment, and captures the whole gestalt of the speech, but it depends on listener skill and can be unreliable at the margins; instrumental methods are objective but can miss the functional whole. The productive synthesis, as in the study of speech motor control more broadly, is that both are needed: the perceptual type sets the diagnostic frame, and instrumental and intelligibility measures quantify the severity and track change (Kent, 2000).

Cognitive Implications

For cognitive psychology, dysarthria is a clean dissociation between the *planning* of language and its *motor execution*. A dysarthric patient demonstrates that the entire linguistic system — lexical retrieval, syntax, phonological encoding — can be intact while only the final translation into muscle movement fails. This isolates speech-motor control as a distinct stage in the architecture of speaking, downstream of the language operations that fail in aphasia and separate again from the motor *programming* that fails in apraxia of speech.

The subsystem analysis also connects to the general study of skilled motor control: speech is a model system for how the brain coordinates many effectors in real time under tight temporal constraints, and dysarthria is the natural experiment that shows what each level of the motor hierarchy contributes. The compounding of subsystem impairments into intelligibility loss, meanwhile, is a concrete instance of how a behavioural outcome can depend multiplicatively on several partly independent components — a pattern with parallels across cognition.

Current Directions

The active research front is moving from group-level classification toward the individual speaker and toward objective, often automatic, measurement. A central finding is that speakers with the same dysarthria type and severity can respond very differently to the same treatment, and recent work seeks the speaker and listener factors that explain this variation so that therapy can be matched to the patient (McAuliffe & Fletcher, 2017). In parallel, data-driven classification of the perceptual profile — rather than assignment to the six historical types by ear — is being extended to populations the Mayo scheme was not built for, such as children with cerebral palsy (Allison & Hustad, 2018). A further direction asks how the principles of treatment transfer across languages, since most of the evidence base was built in English and the acoustic targets of therapy may differ in other phonological systems (Levy & Moya-Galé, 2024). The unresolved questions are whether automatic acoustic measures can replace or augment perceptual judgment, and how far treatment can be personalized from a speaker's subsystem profile.

Common Misconceptions

Dysarthria means the person cannot understand or think clearly.

No. Dysarthria affects only the motor production of speech; language, comprehension, and thought are intact. The distortion is in the delivery, not in the message, and mistaking one for the other is both wrong and demeaning (Darley, Aronson, & Brown, 1969a).

Dysarthria and aphasia are the same thing.

They are different disorders that can co-occur. Aphasia is a disorder of language — choosing and understanding words; dysarthria is a disorder of the muscles that produce speech. A stroke can cause either or both, and telling them apart is a core part of assessment (Mackenzie, 2011).

A slur is just a slur — the specific sound does not matter.

On the contrary, the specific perceptual pattern is diagnostic. A breathy, hypernasal voice points to lower motor neuron damage, a strained voice to upper motor neuron damage, and irregular breakdowns to the cerebellum; the type of dysarthria localizes the lesion (Darley, Aronson, & Brown, 1969b).

Nothing can be done about dysarthria.

Behavioural treatment produces measurable gains, most strikingly the intensive voice therapy validated for Parkinson disease, whose benefits persist for years (Ramig et al., 2001), and more broadly across the speech and language therapy evidence base (Yorkston, 1996).

Glossary

Apraxia of speech.

A disorder of programming the movement sequences of speech, in which the muscles are strong but the plan for moving them fails; distinct from dysarthria, which is a disorder of execution.

Articulation.

The movement of the tongue, lips, and jaw that shapes consonants and vowels; imprecise articulation is the most common perceptual sign of dysarthria.

Ataxic dysarthria.

The type produced by cerebellar damage, marked by irregular articulatory breakdowns and excess, equalized stress.

Dysarthria.

A group of motor speech disorders caused by disordered muscular control of speech, sparing language itself.

Flaccid dysarthria.

The type produced by lower motor neuron damage, marked by muscle weakness, breathy voice, and hypernasality.

Frenchay Dysarthria Assessment.

A standardized bedside examination that rates the movement and function of the speech musculature on structured scales.

Hyperkinetic dysarthria.

The type produced by basal-ganglia disorders that generate involuntary movement, interrupting speech unpredictably.

Hypokinetic dysarthria.

The type produced by basal-ganglia dopamine depletion, as in Parkinson disease, marked by monotone, reduced loudness, and short rushes of speech.

Intelligibility.

The proportion of what a speaker says that a listener correctly understands; the functional outcome that matters most in dysarthria.

LSVT LOUD.

Lee Silverman Voice Treatment, an intensive therapy that trains loud, high-effort phonation and is the best-validated treatment for the dysarthria of Parkinson disease.

Mixed dysarthria.

Dysarthria arising from damage at more than one level of the motor system, as in amyotrophic lateral sclerosis; the commonest type in practice.

Perceptual classification.

The method of diagnosing dysarthria type, and thereby lesion site, by trained listening to its perceptual features; the basis of the Mayo scheme.

Phonation.

The vibration of the vocal folds at the larynx that produces voice; disordered phonation yields breathy, harsh, or strained voice.

Prosody.

The melody, stress, and timing of speech that carry emphasis and emotion; its loss produces monotone or disordered stress.

Resonance.

The shaping of the voice by the oral and nasal cavities via the soft palate; failure of palatal closure produces hypernasality.

Respiration.

The breath support that supplies the airstream and subglottal pressure for speech; weak support gives short phrases and fading loudness.

Spastic dysarthria.

The type produced by bilateral upper motor neuron damage, marked by a strained-strangled voice and slow, effortful speech.

Key Researchers

Joseph R. Duffy

(living). Speech-language pathologist at the Mayo Clinic and author of the standard clinical text on motor speech disorders, which carries forward the Mayo perceptual approach to the differential diagnosis of the dysarthrias. ORCID

Pamela Enderby

(living). Speech and language therapist at the University of Sheffield who developed the Frenchay Dysarthria Assessment, the standardized examination that made the perceptual features of dysarthria systematically scorable. ORCID

Katherine C. Hustad

(living). Researcher at the University of Wisconsin–Madison whose data-driven profiling of dysarthria in children with cerebral palsy extends the perceptual-classification tradition to developmental populations. ORCID

Raymond D. Kent

(living). Speech scientist, emeritus at the University of Wisconsin–Madison, whose review of speech motor control and its disorders reframed the study of how the nervous system produces speech and how that production breaks down. ORCID

Megan J. McAuliffe

(living). Speech scientist at the University of Canterbury whose work examines the speaker and listener factors that shape how much intelligibility patients gain from dysarthria treatment. ORCID

Kathryn M. Yorkston

(living). Rehabilitation scientist at the University of Washington whose treatment-efficacy synthesis established the evidence base for behavioural dysarthria management and who helped develop sentence-level intelligibility measurement. ORCID

Frequently Asked Questions

What is the difference between dysarthria and aphasia?

Dysarthria is a disorder of the physical production of speech, caused by weakness or incoordination of the speech muscles, while aphasia is a disorder of language itself — finding words, building sentences, and understanding others. A person with dysarthria knows exactly what to say but cannot produce it clearly; a person with aphasia may articulate clearly but produce the wrong words. The two can occur together, especially after stroke.

What causes dysarthria?

Any condition that disrupts the motor pathway for speech can cause it: stroke, Parkinson disease, cerebellar disease, amyotrophic lateral sclerosis, multiple sclerosis, cerebral palsy, and traumatic brain injury are among the most common. The type of dysarthria depends on where in the motor system the damage falls.

How many types of dysarthria are there?

The Mayo perceptual classification recognizes six primary types — flaccid, spastic, ataxic, hypokinetic, hyperkinetic, and mixed — with a seventh, unilateral upper motor neuron dysarthria, added later. Each maps onto a level of the motor system, so the type of dysarthria helps localize the underlying disease.

Can dysarthria be treated?

Yes. Behavioural treatment by a speech-language pathologist produces measurable gains. The best-validated example is intensive voice therapy (LSVT LOUD) for the dysarthria of Parkinson disease, whose benefits persist for years; other approaches target the impaired subsystems directly or use compensatory strategies such as slowing the rate of speech.

Does dysarthria affect intelligence?

No. Dysarthria affects only the motor control of speech. Language, comprehension, memory, and reasoning are intact; the difficulty is in producing clear speech, not in the thoughts behind it.

What is intelligibility and why is it measured?

Intelligibility is the proportion of a speaker's message that a listener correctly understands. It is measured because it is the functional bottom line for the patient and because it captures the combined effect of impairments across the several speech subsystems, which a feature-by-feature examination can miss.

What are the speech subsystems?

Speech is produced by five subsystems: respiration (breath support), phonation (voicing at the larynx), resonance (shaping by the oral and nasal cavities), articulation (movement of the tongue, lips, and jaw), and prosody (melody, stress, and timing). Dysarthria can affect any combination of them, and the pattern helps define the type.

Is dysarthria progressive?

It depends on the cause. Dysarthria from a single stroke or injury is usually stable or improves, whereas dysarthria from a neurodegenerative disease such as Parkinson disease or amyotrophic lateral sclerosis worsens as the underlying disease progresses. In ALS it is often one of the earliest signs.

Support Organizations

References