2
AbstractPurpose: The study investigated whether perceptual learning (PL) of a task consisting in detecting a low contrast Gabor patch flanked above and below by high contrast Gabor patches presented monocularly in the preferred retinal locus (PRL) of patients with macular degeneration (MD), improved their residual visual functions.
Method:We measured contrast detection thresholds using both a Yes/No task (three MD patients and three controls) and a temporal two-alternative forced-choice task (2AFC; four MD patients and three controls).Results: Both tasks produced a significant improvement in contrast sensitivity for the trained target.However, only in the case of the temporal-2AFC this improvement depended on the target-toflankers distance. Furthermore, in both tasks PL improved visual acuity but with the temporal-2AFC task we found a higher degree of generalization of the training to untrained stimuli and tasks.In fact, we found a reduction of the crowding effect and an improvement of the contrast sensitivity for untrained spatial frequencies.Although PL is more effective with a temporal-2AFC task, it is also present with a Yes/No task, suggesting that PL reflects sensory enhancement, rather than improvement in decision mechanisms.Most importantly, follow-up tests on MD patients showed that PL effects were retained between four and eight months, suggesting PL induced long-term neural plasticity in the visual cortex.
Conclusion:The results show for the first time that PL with a collinear configuration has strong, non-invasive and long lasting rehabilitative potential to improve vision in the PRL of patients with central vision loss.3