Visual hallucinations during spontaneous and training-induced visual field recovery.

Visual hallucinations after post-geniculate visual system lesions were shown to be associated with spontaneous recovery of visual functions. We investigated the occurrence of hallucinations during spontaneous recovery and additionally tested whether hallucinations were re-instated in a phase of vision restoration therapy (VRT). Nineteen patients with post-geniculate lesions and homonymous visual loss participated in a prospective study, and 121 patients with various lesions were included in a retrospective study using a questionnaire including verbal descriptions as well as drawings of hallucinations experienced by the patients. In both samples, visual-field size was determined before and after 6 months of VRT. Many patients in both groups experienced post-lesion hallucinations (mostly colors, objects, motion) which subsided after spontaneous recovery of visual functions (increase of visual field size, recovery of more complex visual function) was ended. Hallucinations re-emerged during training. However, the majority of patients reported simple, unformed visual hallucinations (uncolored phosphenes, spots, flashes), especially when visual field recovery was most intense. Hallucinations were mainly found in patients with large shifts of the visual field border. They occurred in blind areas, particularly in areas of residual vision where recovery was predominantly observed. Hallucinations may reflect functional recovery in partially lesioned brain areas. While the colored/formed hallucinations during spontaneous recovery may represent non-specific activation of higher visual areas, the simple, unformed training-related hallucinations may indicate recovery in the primary visual cortex during treatment. Hallucinations should not generally be discarded as pathological or unimportant symptoms, but they may be functional indicators of visual system plasticity.

Restoration of vision II: residual functions and training-induced visual field enlargement in brain-damaged patients.

PURPOSE: Brain damage is often accompanied by visual field defects which have been considered to be non-treatable. In recent years, however, new diagnostic methods have revealed hitherto unknown residual vision, which was found, for instance, in transition zones near the blind visual field sectors and in spared islands of vision within the blind regions ("blindsight"). Furthermore, animal studies revealed a high degree of plasticity in the visual system suggesting the possibility that recovery of vision may be induced by systematic visual training. METHODS: Here we summarize a series of studies with patients suffering from visual field defects after brain lesion using some most recently developed computer-based programs for the diagnosis and treatment of visual field defects. Specifically, high-resolution perimetry (HRP) was applied to first diagnose residual function in or near the "blind" sector of the visual field. Thereafter, visual restitution training (VRT, see Kasten et al., Nature med. 4, 1998, p. 1083) was used daily for 6 months to provide systematic stimulation of these areas of residual vision. RESULTS: In a number of studies, we have observed not only residual visual functions within or near the field defect, but we were also able to follow the course of spontaneous recovery of visual functions within weeks or months after visual system damage. Furthermore, even long after spontaneous recovery is complete, computer-based visual restitution training (VRT) in or near the areas of residual vision results in a significant enlargement of intact areas, both after optic nerve damage and postchiasmatic lesions. Using VRT, we found a border shift of about 5 degrees of visual angle which cannot be explained by eye movements or eccentric fixation. We observed a transfer of this training effects to other tasks such as form and color detection, as well as to tests of visual exploration which were not specifically trained. Moreover, 72 % of the patients reported subjective improvements of vision. Training-induced visual field enlargement persisted for at least one year, even in the absence of training beyond 6 months of treatment. CONCLUSIONS: The visual system possesses a remarkable plasticity which becomes apparent in visual field enlargement during spontaneous recovery and specific visual training. Animal studies indicate that a minimum number of residual neurons surviving the lesion, in the order of 10%, provides a sufficient substrate for recovery of vision. Though the precise mechanisms of training-induced visual field enlargement need to be further explored, VRT can be introduced for routine clinical treatment of patients with visual field defects.