The relationship between the unique human faculties of language and music has inspired debate for over centuries. As their vocal agents, both speech and singing enable us to communicate linguistic information coupled with conveying emotions. Over the time, the debate on the shared and distinct features of language and music has also been informed by aphasia, a language impairment that most commonly results from stroke and affects speech, comprehension, reading and writing at varying degrees. Case reports since 18th century recount that even persons with severe aphasia that left them unable to speak may retain their ability to sing the words of a familiar song 1. Singing is known to engage more bilateral regions in the brain as opposed to speech processing that is considered to rely more heavily on a left-lateralized language network. Consequently, right hemisphere contributions in singing have appeared a plausible explanation for the observed distinction between speech and singing. Whilst this explanation does not expound why singing ability is not preserved in all persons after left-hemisphere stroke, it has served as a source of inspiration for the development of formal singing-based interventions 2. Meanwhile, the specific underlying neural factors behind preserved singing ability in aphasia have been studied to a limited extent.
In our article “Hodological organization of spoken language production and singing in the human brain” recently published by Communications Biology, we sought to determine the neural networks underlying connected spoken language production and singing in aphasia as well as shine light on the neural underpinnings of preserved singing ability in aphasia.
Forty-five subjects with chronic post-stroke aphasia participated in the study. For reasons that are more than understandable, recruitment of patients with severe spoken language deficits is often challenging and thereby, most studies on aphasia are limited to studying patients with mild or moderate language deficits 3. Nevertheless, we were very fortunate to get together a diverse group of subjects whose aphasia varied from mild to very severe. They joined us in the lab to do one of the things most difficult to them: to produce speech. Along with discourse-level speech production tasks, neuropsychological assessments included singing words of a familiar song and repeating identical phrases through both spoken language and singing. It is not given that people are willing to sing publicly (or in the lab to a strange psychologist), but our subjects, however, showed great courage and put all their effort into the tasks. We are truly indebted to them and their family members for their dedication to the study. All subjects also underwent magnetic resonance imaging for multi-shell diffusion scans that were used for white matter connectometry that maps the white matter networks associated with the tasks of interest 4.
We found that connected spoken language production efficacy was associated with both dorsal and ventral stream structures of the left-lateralized language network that are typically linked to processes of speech production and comprehension, respectively 5. Additionally, projection pathways beyond the prevailing dual-stream model of speech processing, often subserving motor functions and cognitive control, were found to be associated with efficacious spoken language production. These results lend support to the view that fluent spoken language production is a demanding task for the brain, involving both tracts linked to speech production and comprehension and more general cognitive-motor control.
As for singing the words of a familiar song, we were surprised to discover a clearly left-lateralized white matter network comprising ventral stream structures as well as projection pathways. Exploration on the repetition of identical phrases through speech and singing brought forward a similar partition. Our findings indicate that ventral stream pathways may play a key role in producing and repeating words through singing, while word production through speech production and spoken repetition appear to necessitate the interaction of the streams. These indications were further supported by a post hoc analysis comparing two small subgroups of patients with severe spoken language difficulties and either ability or inability to sing the words of a familiar song.
Altogether, these results suggest that spoken language production and word production through singing are underpinned by partly shared neuroanatomical networks in the left hemisphere. The study provides a feasible explanation why some, but not all persons with severe aphasia retain the ability to produce words through singing.
One definite thing about research is that getting responses to your questions always gives rise to new queries. As the present study was focused on singing the words of a song that was familiar to the subjects already prior to stroke, we do not yet know which neural networks are related to learning and singing a novel song. We also hope to learn more about the neural regions and networks associated with the musical aspects of singing in aphasia, such as producing the melody and rhythm of the song.
While our study was cross-sectional and does not inform us about the efficacy of singing-based interventions, it provides an intriguing venue for discussing therapeutic indications. If word production through singing is supported by partly shared networks with spoken language production, as suggested by the present study, can we use singing-based exercises to stimulate the perilesional left-hemisphere language network directly? At present, there are no formal guidelines or recommendations about which patients would benefit most of singing-based interventions (let alone what sort of intervention each patient would gain best advantage of). Meanwhile, it is ever so evident that due to the multifaceted nature of neural recovery and regions related to various aspects of language, identical regimes cannot be directed to all patients with aphasia3. Therefore, we hope that the present study represents one step forward in the progress of finding biomarkers and developing singing-based interventions that are optimized for each individual with aphasia.
1. Johnson JK, Graziano AB. Some early cases of aphasia and the capacity to sing. Prog Brain Res. 216, 73-89 (2015).
2. Albert, M. L., Sparks, R. W., Helm, N. A. Melodic intonation therapy for aphasia. Arch. Neurol. 29, 130–131 (1973).
3. Stefaniak JD, Geranmayeh F, Lambon Ralph MA. The multidimensional nature of aphasia recovery post-stroke. Brain 145, 1354-1367 (2022).
4. Yeh FC, Badre D, Verstynen T. Connectometry: A statistical approach harnessing the analytical potential of the local connectome. Neuroimage 125, 162-171 (2016).
5. Saur, D., Kreher, B. W., Schnell, S., Kümmerer, D., Kellmeyer, P., Vry, M. S., Umarova, R., Musso, M., Glauche, V., Abel, S., Huber, W., Rijntjes, M., Hennig, J., Weiller, C. Ventral and dorsal pathways for language. Proc Natl Acad Sci U S A. 105, 18035–18040 (2008).