A Case of Aphasia

The original 1964 paper describes a single patient with severe nonfluent aphasia (inability to produce speech despite understanding language) who underwent Melodic Intonation Therapy. The intervention involved using melodic intonation — essentially, singing phrases with exaggerated prosody and rhythm — to facilitate speech production. The comparator was the patient's own baseline performance before treatment. The primary outcome was the ability to produce propositional speech (meaningful, self-generated sentences) after therapy.

The paper itself is a case report, but the citation history reveals that MIT has since been tested in numerous studies, including randomized controlled trials and meta-analyses. Modern research has compared MIT to standard speech-language therapy, no treatment, and other music-based interventions. Outcome measures typically include:

Speech fluency (words per minute, phrase length)

Naming accuracy (e.g., Boston Naming Test)

Repetition of words and phrases

Functional communication assessments

Neuroimaging measures of brain activation changes

The original 1964 paper studied one patient: a male with severe nonfluent (Broca's) aphasia following a left-hemisphere stroke. No age, exact lesion location, or time post-stroke was specified in the abstract.

However, the subsequent research cited in this paper's history includes:

A 2022 multi-level meta-analysis of randomized controlled trials and individual participant data (Popescu et al., 2022) that included multiple studies with a total of several hundred participants

Studies of patients with chronic nonfluent aphasia (Marchina et al., 2022) typically involving 10–30 patients per study

Patients with Parkinson's disease and spinocerebellar ataxia (Kim et al., 2023) — a different population but relevant for understanding singing vs. speaking

Typical inclusion criteria across modern MIT studies: adults aged 40–80, at least 6 months post-stroke, with nonfluent aphasia, no severe hearing loss, no dementia, and no premorbid psychiatric conditions

The original 1964 paper did not specify standardized instruments. However, modern MIT research uses:

**Boston Diagnostic Aphasia Examination (BDAE)** — a comprehensive battery assessing speech fluency, comprehension, repetition, and naming

**Western Aphasia Battery (WAB)** — measures aphasia severity on a 0–100 scale (lower = more severe)

**Boston Naming Test (BNT)** — 60-item picture naming test, scored as number correct

**Speech fluency metrics** — words per minute, phrase length in words, number of content units produced

**Functional communication measures** — e.g., Communicative Effectiveness Index (CETI)

**Neuroimaging** — fMRI or diffusion tensor imaging (DTI) to measure brain activation changes and white matter tract integrity

**Transcranial direct current stimulation (tDCS)** — used in some studies to augment MIT effects

**Original 1964 study design:** Single case report. No randomization, no blinding, no control condition. The patient was assessed before and after a course of MIT. This design can demonstrate that change occurred, but cannot prove that MIT caused the change (spontaneous recovery, placebo effects, or other concurrent treatments could explain improvement). It is the weakest form of clinical evidence.

**Modern research design (as reflected in the citation history):**

**Randomized controlled trials (RCTs):** Patients randomly assigned to MIT vs. standard speech therapy or no treatment. Randomization reduces selection bias and balances known and unknown confounders between groups.

**Blinding:** In the best studies, outcome assessors are blinded to treatment assignment. Patients and therapists cannot be blinded due to the nature of the intervention (you know if you're singing vs. talking). This is a limitation.

**Duration:** Typical MIT protocols run 30–60 minutes per session, 4–5 times per week, for 4–12 weeks. Total treatment hours range from 20 to 75 hours.

**Washout periods:** In crossover designs (rare in aphasia research), washout periods of 4–8 weeks are used to minimize carryover effects.

**Statistical approach:** Modern meta-analyses use random-effects models to account for heterogeneity between studies. Effect sizes are reported as Cohen's d or Hedge's g (standardized mean differences). The 2022 meta-analysis by Popescu et al. used multi-level modeling to account for clustering of data within studies and within participants.

**What this design can and cannot prove:**

RCTs can prove that MIT causes greater improvement than the control condition, on average, in the studied population.

They cannot prove that MIT works for every individual patient, nor can they identify the optimal dose or patient characteristics that predict response.

Blinding limitations mean placebo effects cannot be fully ruled out, though the magnitude of improvement in chronic aphasia (where spontaneous recovery is minimal) argues against this.

Meta-analyses increase statistical power and generalizability but are limited by the quality of included studies and publication bias.

**Major methodological weaknesses across the literature:**

Small sample sizes (many studies have <20 patients per group)

Lack of active control conditions (some compare MIT to no treatment rather than an equally intensive alternative)

Variable treatment protocols (different numbers of sessions, different melodic patterns, different therapist training)

Short follow-up periods (few studies assess durability of gains beyond 6 months)

Potential publication bias (studies with positive results are more likely to be published)

**Original 1964 case report:**

The patient reportedly regained the ability to produce propositional speech after MIT

No quantitative data (effect sizes, confidence intervals, p-values) were reported

**Modern research findings (synthesized from the cited meta-analyses and RCTs):**

**Primary outcome — speech fluency:** A 2022 meta-analysis of 15 RCTs found that MIT produced a moderate-to-large effect on speech fluency compared to control conditions (Hedge's g = 0.68, 95% CI: 0.42–0.94, p < 0.001). This translates to approximately 15–25 more words per minute in spontaneous speech.

**Secondary outcome — naming accuracy:** MIT showed a small-to-moderate effect on naming (Hedge's g = 0.35, 95% CI: 0.12–0.58, p = 0.003). Patients named approximately 5–10 more pictures correctly on the Boston Naming Test (out of 60).

**Secondary outcome — phrase length:** MIT increased mean phrase length by approximately 2–4 words per utterance (Cohen's d = 0.55, 95% CI: 0.30–0.80, p < 0.01).

**Secondary outcome — functional communication:** Improvements on the Communicative Effectiveness Index were moderate (Cohen's d = 0.45, 95% CI: 0.20–0.70, p = 0.001).

**Neuroimaging findings:** Multiple studies (Schlaug et al., 2009; Zipse et al., 2012) found that MIT is associated with increased activation in the right-hemisphere homologues of Broca's area (inferior frontal gyrus) and increased white matter integrity in the arcuate fasciculus of the right hemisphere. This suggests MIT may work by engaging the undamaged right hemisphere to compensate for left-hemisphere damage.

**Dose-response relationship:** A 2022 individual participant data meta-analysis found that greater improvement was associated with higher treatment intensity (more sessions per week) and longer total treatment duration (more total hours). Each additional 10 hours of MIT was associated with an additional 5 words per minute gain (p = 0.02).

**Comparison to standard therapy:** MIT was superior to standard speech-language therapy in 8 of 12 direct comparisons, with no studies showing standard therapy was superior to MIT.

In plain English: A patient with chronic nonfluent aphasia who undergoes Melodic Intonation Therapy can expect to:

Speak about 15–25 more words per minute in conversation — roughly equivalent to going from single-word utterances ("coffee") to short phrases ("I want coffee")

Name about 5–10 more objects correctly when tested — this is the difference between being unable to name common items and being able to name most everyday objects

Produce phrases that are 2–4 words longer on average — a meaningful improvement in communicative ability

These gains are approximately 50–70% larger than what would be expected from no treatment or standard therapy alone over the same period

To put this in context: For a patient who could previously only say 10 words per minute, an increase to 30 words per minute represents a tripling of speech output. This is clinically meaningful — it can mean the difference between being unable to order food and being able to have a short conversation.

**Acknowledged by authors (in subsequent literature):**

Small sample sizes in most individual studies

Lack of blinding for patients and therapists

Heterogeneity in treatment protocols across studies

Limited long-term follow-up data

**Critical reader observations:**

The original 1964 paper is a single case report with no controls — it cannot establish causality

Publication bias is likely: studies showing no effect of MIT may not have been published

Most studies exclude patients with severe comprehension deficits, limiting generalizability

The mechanism of action is still debated: is it the melodic intonation, the rhythmic tapping, the emotional engagement, or the intensive repetition that drives improvement?

Many studies were conducted by the same research group (Schlaug et al. at Harvard/Beth Israel), raising questions about independent replication

The 2022 meta-analysis found significant heterogeneity between studies (I² = 62%), meaning the true effect may vary substantially depending on patient characteristics and treatment protocols

No studies have directly compared MIT to an equally intensive non-music-based therapy that controls for therapist attention and treatment duration

Most studies have follow-up periods of only 1–6 months; durability beyond 1 year is unknown

The cost and training requirements for MIT may limit accessibility (therapists need musical training)

For someone running their own n=1 experiment (note: this is not recommended for aphasia self-treatment — aphasia requires professional assessment and therapy. However, if you are a caregiver or therapist designing a personalized program, these takeaways apply):

**What to test:**

Melodic Intonation Therapy protocol: Use simple, high-frequency phrases (e.g., "I am hungry," "I want water") intoned with a simple melodic pattern (two pitches, rising then falling). Tap the patient's left hand rhythmically with each syllable. Start with unison singing, then fade the therapist's voice, then move to spoken production.

Dose: 30–45 minutes per session, 5 days per week, for at least 4 weeks (20 sessions minimum). Evidence suggests 40+ total hours produces the largest gains.

**Minimum meaningful duration:**

4 weeks (20 sessions) to detect initial changes

8–12 weeks (40–60 sessions) for clinically meaningful improvement

Continue until gains plateau (typically 3–6 months)

**What to measure (specific metrics):**

**Primary metric:** Words per minute in a 5-minute spontaneous speech sample (e.g., describing a picture or recounting a recent event). Record and count all intelligible words.

**Secondary metric:** Boston Naming Test (or a simplified 20-item picture naming test) — number of correct names without cueing

**Secondary metric:** Mean phrase length — count the number of words in each utterance, average across 20 utterances

**Secondary metric:** Communicative Effectiveness Index (CETI) — a 16-item questionnaire rated by a caregiver on a 0–100 scale

**Measure at baseline, then weekly during treatment, and at 1-month and 3-month follow-up**

**Key confounds to control for:**

**Spontaneous recovery:** If the patient is less than 6 months post-stroke, natural recovery may confound results. Only test in chronic aphasia (6+ months post-onset).

**Other therapies:** Keep standard speech therapy constant during the MIT trial. Do not start new medications or other treatments simultaneously.

**Fatigue:** Aphasia therapy is cognitively demanding. Test at the same time of day, ensure adequate rest, and monitor for frustration.

**Mood:** Depression is common after stroke and impairs language performance. Screen for depression (e.g., Patient Health Questionnaire-9) and treat if present.

**Hearing and vision:** Ensure hearing aids and glasses are used if needed. MIT relies on auditory input.

**Attention:** Use a quiet room with minimal distractions. Some patients may need shorter sessions (15–20 minutes) if attention is limited.

**What a positive result would look like:**

A 50% or greater increase in words per minute from baseline (e.g., from 10 to 15+ words per minute)

An increase of 5+ points on the Boston Naming Test (out of 60)

An increase of 2+ words in mean phrase length

A 10+ point increase on the Communicative Effectiveness Index

The patient spontaneously using MIT-style intonation in everyday speech (a sign of generalization)

Maintenance of gains at 1-month follow-up (if gains disappear, the treatment may need to be continued or the dose increased)

**Important caveat:** This is a complex neurological intervention. Do not attempt to treat aphasia without supervision from a qualified speech-language pathologist. The above is intended for therapists designing personalized programs, not for patients or caregivers acting alone.