Objective To evaluate how effective perceptual learning can be in improving the clinical symptoms of patients with early stage presbyopia. Methods Eighty patients with presbyopia 40-45 years old， mean age 42.3±2.5 years， were randomly recruited （80 eyes） for a prospective control study. The subjects′ amplitudes of accommodation were measured with a minus-lens test. The patients were randomly divided into two groups. Forty patients were trained for 12-15 minutes per session， 3 times per week over a 3-month period using perceptual learning software （the treatment groups）. Another 40 patients were allocated to the control groups. The data for subjective symptoms and visual acuity， reading acuity， maximum reading speed， critical print size， accommodation， and lens thickness changes in the amplitude of accommodation were measured for each patient at five time points： before training， and 3， 6， 9 and 12 weeks after perceptual learning treatment. The data were analyzed using repeated measures analysis of variance. Results Before training， and 3， 6， 9 and 12 weeks after perceptual learning treatment， the subjective symptoms improved and the differences were significant （F=283.9， 482.7， 160.9， 198.3， 126.7， all P<0.05）. The differences in the tendency for change between the treatment groups and the control groups were not significant for visual acuity and accommodation. The differences in the tendency for change between the treatment and the control groups for reading acuity， maximum reading speed， critical print size were significant （F=30.3， 4.6， 41.5， all P<0.05）. Conclusion Perceptual learning can improve the near reading ability of patients with early stage presbyopia.