[Verbal memory disorders provoked by a variety of stokes].

In recent years a consensus has been reached about the neuroanatomical substrate of verbal memory, but this state of knowledge has not yet been implemented in clinical practice. One reason for this may be that most of the neuroscientific studies on verbal memory used different neuropsychological instruments and that only a small set of patient groups with the same etiology but different lesion sites were analysed. Returning to three earlier studies, we analyse the possibility to make differential judgements on the verbal memory impairments of four different patient groups by using the California Verbal Learning Test (CVLT). We compare patients with left-sided (n = 16) and right-sided (n = 10) posterior cerebral artery infarcts and patients with infarcts of the left (n = 10) or right (n = 21) frontal lobe, and we integrated data about their retention errors that had not been analysed so far. Our findings reveal significant differences between these patient groups, concerning the quantitative aspects of impairments, and also the profiles of memory errors (recall, interference and perseverations). Our study documents a relation between the site of the lesion and the type of verbal memory impairment, agreeing with some of the most recent neuroscientific findings. Starting from this observation we try to define a neuropsychological pattern of memory impairment which enables differential clinical diagnoses using the CVLT as a memory test.