Can we affirm that the current nosologies for mental disorders are accurate? In this paper, I present a concise epistemological review related to mental disorders and their conceptualization from three positions within the philosophy of mind. First, I address the naturalistic approach, delving into Szasz’s eliminativist view, Boorse’s biostatistical account, and the Wakefield’s harmful dysfunction, while also highlighting main challenges to this approach. Second, I analyze the constructivist approach and the arguments provided to critique and oppose the naturalistic perspective, as well as pointing out its weaknesses. Third, I outline the pragmatic approach, whose considerations help transcend the dialectic and confrontations between naturalistic reductionism and sociocultural constructivism. Lastly, I summarize the main conclusions and directly respond to the primary question of this paper.

Several recent studies have utilized neuroimaging to delineate the localization and function of brain regions involved in language. However, many uncertainties persist regarding the organization of the linguistic system in the human brain. The aim of the present study was to characterize the structural changes produced in a sample of 9 patients with post-stroke aphasia (4 women; mean age = 60 years, SD = 14.86) and their relationship with performance in the entire Boston Diagnostic Aphasia Examination (BDAE). Magnetic Resonance Imaging was acquired from the brain of each patient and brain lesions were assessed. Disconnection’s severity of each white matter tract by embedding the lesion into the streamline tractography atlas of the Human Connectome Project was analyzed, and grey matter lesion load using a 7-Network Cortical parcellation template was estimated, with additional subcortical, cerebellar and brainstem parcels. Finally, all data obtained was correlated with performance in the BDAE. Somatomotor network correlated with repetition scale. The disconnection of the left acoustic radiation and inferior longitudinal fasciculus correlated with repetition sub-scale. Finally, the left U-fibers correlated with severity (a BDAE sub-scale that assesses the patient’s communicative skills), conversational speech and reading sub-scales. These findings emphasized that the disconnection of these fronto-parieto-temporal structures correlate with deficits in repetition, beyond the classical hypothesis attributing such deficits solely to the impairment of the arcuate fasciculus.