Unveiling Self-Related Deficits in Neurosurgical Patients: Insights from the Self-Other Voice Discrimination Task as a Neuropsychological Tool

conducted pre-operatively to those that are conducted intra-operatively. For example, covert object naming (ON) looks at testing the expressive component of language in task-based fMRI. This is completely different from intra-operative testing where naming errors are used to localise the expressive speech areas during DCS. The aim of this study was to develop feasible and reproducible language specific task-based fMRI paradigms for assessment providing an alternative to intraoperative DCS in event of failure to perform or complete an awake craniotomy. Methods: 15 English speaking healthy volunteers with no known intracranial pathology or surgery were recruited for this study. Along with other demographics, their handedness and education level were collected as part of the study. A modified pre-operative task-based fMRI protocol was designed for this study. In addition to T1 weighted structural image, single echo resting state and diffusion weighted imaging, each participant was asked to do the following tasks covert object naming, overt objecting naming, sentence completion, and semantics association. Task based fMRI images were analysed using SPM12®. Resting state data was analysed using CONN22a toolbox. The regions of interest (ROIs) showing activation obtained from this healthy volunteer cohort were then applied to resting state and diffusion weighted imaging in patients with gliomas to understand functional and anatomical connectivity. Results: 7 males and 8 females, right-handed English speaking was recruited for this study. All volunteers completed the structural and functional MRI within 50 minutes. This was important to ensure that patients were able to tolerate the entire duration of the scan without resulting in fatigability and errors during tasks. All tasks were completed in 15 volunteers allowing 2nd level analysis generating ROIs of interest. An example of individual level analysis for overt naming has been shown in Figure 1. Both 1st and 2nd level analysis allowed us to understand language representation for various tasks. These ROIs were then applied to resting state analysis in patients with gliomas to understand functional connectivity based on ROIs. This provided further insight into how gliomas can distort functional connectivity in presence of the tumour. Conclusions: The newly devised protocol to perform the same intra-operative tasks in the pre-operative fMRI is feasible providing robust results which can be replicated across patients. Our data also provides information that can be applied to resting state analysis for those patients who may have severe deficits which prevents them from performing any tasks in the scanner.