Investigating Language and Domain-General Processing in Neurotypicals and Individuals With Aphasia — A Functional Near-Infrared Spectroscopy Pilot Study

Abstract

Brain reorganization patterns associated with language recovery after stroke have long been debated. Studying mechanisms of spontaneous and treatment-induced language recovery in post-stroke aphasia requires a network-based approach given the potential for recruitment of perilesional left hemisphere language regions, homologous right hemisphere language regions, and/or spared bilateral domain-general regions. Recent hardware, software, and methodological advances in functional near-infrared spectroscopy (fNIRS) make it well-suited to examine this question. fNIRS is cost-effective with minimal contraindications, making it a robust option to monitor treatment-related brain activation changes over time. Establishing clear activation patterns in neurotypical adults during language and domain-general cognitive processes via fNIRS is an important first step. Some fNIRS studies have investigated key language processes in healthy adults, yet findings are challenging to interpret in the context of methodological limitations. This pilot study used fNIRS to capture brain activation during language and domain-general processing in neurotypicals and individuals with aphasia. These findings will serve as a reference when interpreting treatment-related changes in brain activation patterns in post-stroke aphasia in the future. Twenty-four young healthy controls, seventeen older healthy controls, and six individuals with left hemisphere stroke-induced aphasia completed two language tasks (i.e., semantic feature, picture naming) and one domain-general cognitive task (i.e., arithmetic) twice during fNIRS. The probe covered bilateral frontal, parietal, and temporal lobes and included short-separation detectors for scalp signal nuisance regression. Younger and older healthy controls activated core language regions during semantic feature processing (e.g., left inferior frontal gyrus pars opercularis) and lexical retrieval (e.g., left inferior frontal gyrus pars triangularis) and domain-general regions (e.g., bilateral middle frontal gyri) during hard versus easy arithmetic as expected. Consistent with theories of post-stroke language recovery, individuals with aphasia activated areas outside the traditional networks: left superior frontal gyrus and left supramarginal gyrus during semantic feature judgment; left superior frontal gyrus and right precentral gyrus during picture naming; and left inferior frontal gyrus pars opercularis during arithmetic processing. The preliminary findings in the stroke group highlight the utility of using fNIRS to study language and domain-general processing in aphasia.

Language reorganization patterns in global aphasia–evidence from fNIRS

“…Moreover, we found the latency of the left SMG was negatively correlated with the ability to name, similar to the results from a previous fMRI study (41). In a recent fNIRS study performed by Gilmore et al, the left SMG was also reported to be involved in semantic features and picture-naming tasks (42), while differences should be noted between the two studies. Gilmore et al enrolled patients with an average AQ of 67.17 (42.20-93.20), while in our present study, patients with an average AQ of 13.34 (2.6-21.7) were included.…”

“…A disadvantage of fNIRS that is not unique to this study is that it measures signals ∼1.5-cm deep into the cortex, meaning that the contribution of subcortical structures in naming tasks in global aphasia could not be captured in this study. Nevertheless, the cortical coverage in this study was greater than in other aphasia studies to date [e.g., 64 channels in this study vs. 56 channels in Gilmore et al (42)]. Similarly, methods for managing the lesion in fNIRS studies in aphasia are still emerging.…”

Concurrent Structural and Functional Patterns in Patients With Amnestic Mild Cognitive Impairment

“…In this study, we found that the IFG_L of patients with aMCI had concurrent structural and functional changes, which suggested that IFG_L might be a better indicator for predicting cognitive deficits in aMCI ( Gilmore et al, 2021 ). IFG_L was related to language/semantic processing.…”

Neural mechanisms underlying different aspects of pseudostuttering versus fluent speech during spontaneous speech production.

“…The activation increases in the IFG, MTG, PMC and SMG were expected due to their connection to the language network (44,45). While activations in the IFG and SMG are a typical neural correlate for picture naming (68), the PMC is associated with speech motor control in general (for a review, see Tremblay, et al (32)) but also with the prolongation and repetition of syllables (69). Notably, our additional finding in the MTG can be explained by its associated functional role in sentence generation (70,71), which was more relevant in our task (complex panorama picture description) than in the tasks implemented by other studies (reading sentences), the activation in the DLPFC can be explained by increasing working memory demands (33)(34)(35) and/or the inhibition of fluent speech (72).…”