Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Identification of the neural network involved in naming famous people from voices: a multimodal mapping study in patients with low-grade glioma
Poster Session B, Friday, September 12, 4:30 - 6:00 pm, Field House
Eléonor Burkhardt1,2,3, Hugues Duffau2,4, Bruno Rossion3, Guillaume Herbet1,2,5,6; 1Praxiling, UMR5267, CNRS, University of Paul-Valéry, Montpellier, France, 2Gui de Chauliac Hospital, Regional University Hospital of Montpellier, France, 3IMoPA, UMR7365, CNRS, University of Lorraine, Nancy, France, 4Fonctional Genomics Institute, University of Montpellier, CNRS, INSERM, France, 5University of Montpellier, France, 6Institut Universitaire de France
Proper names (PNs), particularly person names, refer to unique entities and allow humans to structure their social environment. From a psycholinguistic standpoint, common names are related to categories and are meaningful while PNs are related to individualities and lack of meaning. This unique cognitive status makes the PN retrieval especially demanding. Accordingly, brain-damaged patients often have impaired lexical retrieval of PNs. As such, many neuropsychological and neuroimaging studies have examined the brain regions implicated in accessing the lexical output of PNs. They showed the involvement of a predominantly fronto-temporal network, centered on the left anterior temporal lobe (ATL)/the temporal pole. However, only a small body of neuropsychological studies have sought to identify the underlying anatomical tracts. The most recent works have shown the central role of the left inferior longitudinal fasciculus (ILF) in naming famous people from faces, consistent with ILF’s connections between the ATL and the visual associative cortex. In contrast, the left middle longitudinal fasciculus (MdLF), which also projects its fibers into the ATL and is topographically close to the auditory associative cortex, may be preferentially involved in naming people from voices. To investigate this hypothesis, sixty patients who underwent awake neurosurgery for a left low-grade glioma, were behaviorally assessed, at least six months postoperatively. The evaluation consisted in the administration of tasks requiring the naming of famous people based either on spoken or sung voices. Importantly, patients’ performance was compared to that of eighty matched healthy subjects and controlled for the influence of various sociodemographic (age and education) and cognitive factors (mainly within-task control factors and common entity naming, depending on the auditory/visual modality), through several statistical models. Then, lesion-symptom mapping methods were applied, using both lesion topography and structural disconnections. At the cortical level, our results revealed that PN retrieval deficits were associated with lesions of the middle/posterior part of the ventro-temporal structures (rather than the ATL), likely due to their implication in visual-auditory integration processes. In terms of white matter connectivity, the MdLF did not appear to play an essential role. Nevertheless, the low lesion distribution in the mid-to-posterior superior temporal structures, consistent with the spatial stereotypy of low-grade gliomas, does not allow definitive conclusions to be drawn on its involvement. Furthermore, the neuroplasticity usually induced by these tumors may have resulted in its functional compensation. Instead, the posterior part of the corpus callosum and the superficial fibers of the occipito-temporal structures were involved in lexical access to voice names in spoken condition while the ventromedial fibers of the ILF, but also the arcuate fasciculus to a lesser extent, were involved in sung condition. Interestingly, the ILF’s implication in voice-based famous people naming seemed also to reflect a role in visual-auditory integration. In sum, our work advances the understanding of the cognitive and neural mechanisms underlying PN anomia, particularly within the framework of the dual pathway model of language. Although these primary results require replication, they open promising avenues for future research. Clinically, they contribute to the refinement of intraoperative mapping strategies in awake neurosurgery.
Topic Areas: Disorders: Acquired, Language Production