Speak along without the song: what promotes fluency in people with aphasia?

Abstract

ABSTRACTBackground: Melodic intonation therapy (MIT) is a widely used treatment for nonfluent aphasia that builds upon a number of musical elements in order to ultimately improve generative language. These include intoning syllables on different pitches, using a metrically regular speech rhythm, and the clinician and patient producing phrases in unison. Studies have investigated which of these musical elements might be the most clinically facilitating, but important differences in methodology and results across studies leave some questions unanswered. In particular, the relative roles of intoning and unison production remain unclear.Aims: The present study examined these elements in isolation and in combination.Methods & Procedures: Twelve people with aphasia (PWA) and 10 control participants sang and rhythmically spoke unfamiliar song lyrics in unison with a recording and by themselves. A subset of eight of the PWA returned for post-hoc testing to evaluate singing familiar lyrics from memory, and propositional speech. Across all tasks, productions were evaluated for syllable accuracy.Outcomes & Results: On average, PWA benefited from unison over solo production and from rhythmic speech over singing when repeating unfamiliar lyrics. They did show a benefit of singing when producing familiar lyrics from memory, but this post-hoc task did not control for syllable duration or task order. Within-group correlations indicated that those with more severe apraxia of speech, worse single-word auditory comprehension, and poorer repetition ability benefited the most from unison production. Those with better sentence-level auditory comprehension and poorer performance on a perceptual rhythm test tended to benefit more from singing compared to rhythmically speaking unfamiliar lyrics.Conclusions: Unison production is an important element for promoting fluency in many PWA, and perhaps especially for those for whom MIT is considered an appropriate treatment option. However, some candidates for MIT are hindered by the inclusion of even moderately complex melodic information, indicating that this factor should be considered when customizing or adapting the therapy. Finally, PWA who show a striking improvement in fluency when singing familiar songs will not necessarily benefit from singing when the aim is to produce new lyrics or propositional content.