Long-term learning of visual functions in patients after brain damage.

PURPOSE

Systematic vision restoration training has been shown to improve the detection performance of brain-damaged patients with visual-field defects. So far, patients have been trained daily up to 6 months. We wished to determine whether intensive long-term training of 12 months further increases visual detection abilities.

METHODS

Retrospective comparison of 17 patients with visual-field defects using vision restoration training for 12 months with a group of patients training for 6 months. Computer-based home training was completed for 6 months (about 195,000 stimuli presentations) or for 12 months (about 390,000 stimuli presentations). Visual fields were measured at baseline with Rodenstock Perimat 206 (monocular) at 90 degrees eccentricity and at 54 degrees eccentricity with high resolution perimetry (HRP) (binocular) after 6 months (post-6) and after 12 months (post-12) of training.

RESULTS

Near-threshold perimetry revealed minor training effects, beyond 6 months, of 3.5% (p=0.099) in the right eye and of 1.5% (p=0.57) in the left eye. No effects of long-term training were evident in above threshold testing (0.8% detection improvement, n.s.).

CONCLUSIONS

Learning to detect above-threshold stimuli in patients with post-retinal lesions is completed after 6 months of practice with only marginal improvements thereafter. Near-threshold testing reveals that peripheral areas of the visual-field benefit from long-term training even if they are not trained.