After a stroke, some individuals develop a language disorder that disrupts how their brains handle the sounds of speech. While their hearing itself may remain intact, the neurological processes that allow speech sounds to be interpreted as meaningful language can be compromised. To better understand how stroke alters these processes, a team of researchers examined changes in brain activity associated with speech comprehension, focusing on how the injured brain differs from a healthy one during everyday listening.
The study was led by Laura Gwilliams, a faculty scholar at the Wu Tsai Neuroscience Institute and Stanford Data Science, and an assistant professor at the Stanford School of Humanities and Sciences, together with Maaike Vandermosten, an associate professor in the Department of Neurosciences at KU Leuven. Their work compared the brain activity of 39 people who had experienced a stroke with that of 24 healthy adults of a similar age. By analysing these two groups side by side, the researchers aimed to uncover the neural mechanisms that support language processing and how these mechanisms are altered after stroke.
To capture brain activity in a naturalistic way, participants were asked to listen to a spoken story while their neural responses were recorded. This approach allowed the researchers to observe how the brain processes speech in real time, rather than relying on artificial tasks involving isolated sounds or words. The results revealed a striking pattern. Individuals with stroke-related difficulties in verbal speech processing were not slower than healthy listeners in responding to speech sounds. Instead, their brain responses were noticeably weaker.
This finding suggests that the core problem is not a delay in processing but a reduction in the strength or quality of that processing. In practical terms, people affected by this language disorder appear to detect sounds just as efficiently as those without brain injury. However, they struggle when it comes to combining these sounds into coherent linguistic units that carry meaning. The brain receives the auditory information, but the integration needed to understand spoken language is less effective.
The study also highlighted differences in how the brain deals with uncertainty in speech. When words were difficult to hear or ambiguous, healthy participants showed prolonged processing of speech sound features. This extended neural activity is thought to reflect the brain’s effort to resolve uncertainty and arrive at the correct interpretation. In contrast, people who had experienced a stroke showed less sustained processing in these situations. This shortened engagement with speech sounds may make it harder for them to identify words that are unclear or masked by noise successfully.
Taken together, these findings point to specific patterns of brain activity that are crucial for understanding spoken language. They suggest that successful speech comprehension relies not only on detecting sounds quickly but also on maintaining and strengthening neural processing when interpretation becomes challenging. After a stroke, this supportive processing may be reduced, contributing to persistent language difficulties.
The authors emphasise that these insights could have important clinical implications. First author Jill Kries expressed enthusiasm about continuing this line of research, particularly the use of simple, natural listening tasks such as hearing a story. Such approaches may improve the diagnosis of language processing disorders, which currently often require lengthy and demanding behavioural assessments. By focusing on how the brain responds during everyday listening, future tools could become both more efficient and more closely aligned with real-world communication challenges.
More information: Jill Kries et al, The spatio-temporal dynamics of phoneme encoding in aging and aphasia, JNeurosci. DOI: 10.1523/JNEUROSCI.1001-25.2025
Journal information: JNeurosci Provided by Society for Neuroscience
