hide
Free keywords:
-
Abstract:
Left-hemisphere stroke patients often suffer a profound loss of spontaneous speech — known as non-fluent aphasia. Yet, many patients are still able to sing entire pieces of text fluently. This striking finding has inspired mainly two research questions. If the experimental design focuses on one point in time (cross section), one may ask whether or not singing facilitates speech production in aphasic patients. If the design focuses on changes over several points in time (longitudinal section), one may ask whether or not singing qualifies as a therapy to aid recovery from aphasia. The present work addresses both of these questions based on two separate experiments.
A cross-sectional experiment investigated the relative effects of melody, rhythm, and lyric type on speech production in seventeen patients with non-fluent aphasia. The experiment controlled for vocal frequency variability, pitch accuracy, rhythmicity, syllable duration, phonetic complexity and other influences, such as learning effects and the acoustic setting. Contrary to earlier reports, the cross-sectional results suggest that singing may not benefit speech production in non-fluent aphasic patients over and above rhythmic speech. Previous divergent findings could very likely be due to affects from the acoustic setting, insufficient control for syllable duration, and language-specific stress patterns. However, the data reported here indicate that rhythmic pacing may be crucial, particularly for patients with lesions including the basal ganglia. Overall, basal ganglia lesions accounted for more than fifty percent of the variance related to rhythmicity. The findings suggest that benefits typically attributed to singing in the past may actually have their roots in rhythm. Moreover, the results demonstrate that lyric type may have a profound impact on speech production in non-fluent aphasic patients. Among the studied patients, lyric familiarity and formulaic language appeared to strongly mediate speech production, regardless of whether patients were singing or speaking rhythmically. Lyric familiarity and formulaic language may therefore help to explain effects that have, up until now, been presumed to result from singing.
A longitudinal experiment investigated the relative long-term effects of melody and rhythm on the recovery of formulaic and non-formulaic speech. Fifteen patients with chronic non-fluent aphasia underwent either singing therapy, rhythmic therapy, or standard speech therapy. The experiment controlled for vocal frequency variability, phonatory quality, pitch accuracy, syllable duration, phonetic complexity and other influences, such as the acoustic setting and learning effects induced by the testing itself. The longitudinal results suggest that singing and rhythmic speech may be similarly effective in the treatment of non-fluent aphasia. Both singing and rhythmic therapy patients made good progress in the production of common, formulaic phrases — known to be supported by right corticostriatal brain areas. This progress occurred at an early stage of both therapies and was stable over time. Moreover, relatives of the patients reported that they were using a fixed number of formulaic phrases successfully in communicative contexts. Independent of whether patients had received singing or rhythmic therapy, they were able to easily switch between singing and rhythmic speech at any time. Conversely, patients receiving standard speech therapy made less progress in the production of formulaic phrases. They did, however, improve their production of unrehearsed, non-formulaic utterances, in contrast to singing and rhythmic therapy patients, who did not. In light of these results, it may be worth considering the combined use of standard speech therapy and the training of formulaic phrases, whether sung or rhythmically spoken. This combination may yield better results for speech recovery than either therapy alone. Overall, treatment and lyric type accounted for about ninety percent of the variance related to speech recovery in the data reported here.
The present work delivers three main results. First, it may not be singing itself that aids speech production and speech recovery in non-fluent aphasic patients, but rhythm and lyric type. Second, the findings may challenge the view that singing causes a transfer of language function from the left to the right hemisphere. Moving beyond this left-right hemisphere dichotomy, the current results are consistent with the idea that rhythmic pacing may partly bypass corticostriatal damage. Third, the data support the claim that non-formulaic utterances and formulaic phrases rely on different neural mechanisms, suggesting a two-path model of speech recovery. Standard speech therapy focusing on non-formulaic, propositional utterances may engage, in particular, left perilesional brain regions, while training of formulaic phrases may open new ways of tapping into right-hemisphere language resources — even without singing.
