Most people know that Parkinson’s disease affects movement. What is less widely understood is that Parkinson’s disease also impacts two of the most essential tools of human communication: voice and facial expression.
Hypophonia is a soft or reduced loudness voice. Hypomimia is often described as ‘masked’ or reduced facial expression’. Combined, these can create significant social barriers that result in isolation, yet, these symptoms of Parkinson’s are not as frequently discussed.
The Person Behind the Masked Face
Research on hypomimia has found that observers watching silent video recordings of people with Parkinson’s consistently rate them as more anxious, hostile, suspicious, and unhappy, despite the fact they are feeling none of those things. Families report that reduced expressiveness is misread as anger, and clinicians can mistake it for depression. People living with Parkinson’s describe feeling stigmatized, excluded from conversations, and less confident because the face they present to the world no longer reflects who they are inside.
What is LSVT LOUD® and Why Would Anyone Expect It to Help Facial Expression?
LSVT LOUD (Lee Silverman Voice Treatment) was developed to address the soft, breathy, monotone voice and mumbled speech, called hypokinetic dysarthria (a movement disorder of speech), that affects roughly 90% of people with Parkinson’s. The treatment is intensive by design: 16 individual one-hour sessions over four weeks, built around a deceptively simple instruction: think loud. Sessions focus on sustained vowels, pitch range exercises, functional phrases, and personalized speech hierarchy exercises, all with the goal of recalibrating how much vocal effort feels “normal.”
The treatment works by targeting a core deficit in Parkinson’s motor control: the brain’s difficulty regulating the amplitude (e.g., loudness in speech; bigness in movement). When amplitude is driven through the vocal system with intensity and effort, improvements have been documented across the speech mechanism: articulation, respiratory function, speech intelligibility, and even select aspects of swallowing. This “spread of effects” makes the training target very efficient; there is no need to work on each speech subsystem separately.
Given those broad system effects and years of clinical anecdotes of improved facial expression, researchers asked a natural question: what about the face?
Can LSVT LOUD Help Hypomimia Too?
A study by Spielman, Borod, and Ramig (Cognitive and Behavioral Neurology, 2003) examined 44 people with typical Parkinson’s who had received either LSVT LOUD or a high-effort, intensive respiratory treatment. Blinded raters judged facial mobility and engagement from silent video clips recorded before and after treatment and consistently rated LSVT LOUD participants as more facially mobile after treatment. This study provided a foundation for examining this question with additional scientific rigor.
The follow-up study by Dumer and colleagues (Journal of the International Neuropsychological Society, 2014) did just that. It enrolled 56 participants: 16 received LSVT LOUD, 12 an intensity-matched articulation treatment (ARTIC), 17 were untreated people with PD, and 11 were controls without PD. PD participants were stratified based on age, sex, disease stage, time since diagnosis, cognitive status, depression scores, and severity of voice, speech and swallowing symptoms.
This study used the Facial Action Coding System (FACS), the gold-standard objective method for coding facial movement. FACS-certified coders scored one-minute silent video clips where they coded each small facial movement and counted how often people changed expression (Frequency) and how many different expressions they used (Variability). They were blind to treatment assignment and time point (i.e., before or after treatment). The total number of facial expressions (Frequency) and the number of distinct expressions (Variability) were calculated. At baseline, all three PD groups showed significantly lower Frequency and Variability than controls, confirming hypomimia pre-treatment.
These graphs demonstrate the mean Frequency and Variability for the total sample of each participant group. Source: Dumer et al., 2014, Journal of the International Neuropsychological Society.
The post-treatment results revealed:
- LSVT LOUD was the only group whose mean Frequency and Variability scores increased from before to after treatment.
- Improvements in Frequency and Variability were significantly greater in the LSVT LOUD group than in either ARTIC or the untreated group.
- After treatment, the facial expression Frequency and Variability of LSVT LOUD participants were no longer significantly different from non-Parkinson controls.
- By contrast, the ARTIC group still scored significantly below non-Parkinson controls on both measures after treatment.
Participant happiness ratings were measured before and after each speech sample to control for the possibility that a boost in mood, rather than the treatment itself, was driving facial expression changes. The LSVT LOUD group did not differ significantly from the ARTIC and untreated group on happiness ratings at the post-treatment time point, suggesting facial expressivity gains were attributable to the treatment, not simply to feeling better.
Both LSVT LOUD and ARTIC are intensive, high-effort, neuroplasticity-based treatments, matched on dosage, effort, frequency, and duration. The only difference is the target: voice (respiratory-laryngeal system) versus articulation (orofacial-articulatory movements). Yet the outcomes on facial expression are strikingly different.
Intuitively, articulation might seem the better bet for facial expression, but it was not. The explanation likely lies in the deep neural overlap between voice, emotion, and face. Voice and facial movement are tightly coupled neurologically, and the brain networks involved in vocal amplitude control overlap with those regulating emotionally driven facial movement. LSVT LOUD’s high-effort focus on the voice appears to engage neural circuits that articulation training, working at the level of lips, tongue, and jaw, does not reach in the same way.
Important Caveats
The evidence is promising but not definitive. Most participants had moderate Parkinson’s (Hoehn–Yahr stages 2.5–3.5), so generalizability to earlier or more advanced disease stages is unknown. The post-treatment measurement window was approximately one month, meaning durability of facial expressivity gains specifically was not assessed. In other words, we still need larger, longer studies to know how long the facial changes last and whether they improve everyday social life.
Finding LSVT LOUD Treatment
The face is central to how we connect with others. When Parkinson’s reduces facial expression, it can create a disconnect between the emotions you feel and what people around you perceive. These studies offer preliminary evidence that voice treatment can help.
LSVT LOUD is delivered by speech-language pathologists who are specifically trained and certified in the treatment. Not every SLP specializes in Parkinson’s, and not every voice therapy is LSVT LOUD; the protocol’s intensity, single-target approach, sensory recalibration component, and homework structure are what the research supports, and they matter.
LSVT Global maintains a Find a Clinician directory at lsvtglobal.com. Search by location to find trained and certified LSVT LOUD providers near you. Telehealth options (online treatment) are also available.
A referral to a speech-language pathologist, preferably made early, is one of the most impactful things a neurologist or primary care physician can do for a patient with Parkinson’s disease. Not waiting until the voice is barely audible.
LSVT LOUD was designed to give people their voices back. There’s growing evidence it may give them something else too: a face that reflects the person they know themselves to be.
This post draws on three peer-reviewed studies. An initial draft was generated by Claude Sonnet 4.6 (Anthropic). Significant editing and fact checking were completed by human contributors (C Fox & A Halpern).
Spielman, J.L., Borod, J.C., & Ramig, L.O. (2003). The effects of intensive voice treatment on facial expressiveness in Parkinson disease: preliminary data. Cognitive and Behavioral Neurology, 16(3):177–188.
Dumer, A.I., Oster, H., McCabe, D., Rabin, L.A., Spielman, J.L., Ramig, L.O., & Borod, J.C. (2014). Effects of the Lee Silverman Voice Treatment (LSVT LOUD) on hypomimia in Parkinson’s disease. Journal of the International Neuropsychological Society, 20:1–11.
Rasing, N.B., van de Geest-Buit, W.A., Chan, O.Y.A., et al. (2024). Treatment approaches for altered facial expression: A systematic review in facioscapulohumeral muscular dystrophy and other neurological diseases. Journal of Neuromuscular Diseases. DOI: 10.3233/JND-230213.