Objective: The purpose of this research was to determine the efficacy of attributional and metacognitive-attributional strategies training on reading comprehension of fourth grade female elementary students with dyslexia in Isfahan. Method: 30 subjects were randomly selected, and assigned to 3 groups (two experimental groups and one control group). The reading comprehension test was administered as the pre– and post- test to all groups. The two experimental groups received 8 sessions of 45 minutes training. The control group did not receive training. Results: Covariance analysis showed that there were significant differences among mean scores of reading comprehension in the 3 groups. The mean scores of metacognitive-atributional strategies was significantly bigger than the mean scores of atrributional training. Conclusion: The training of attributional and metacognitive strategies could enhance reading comprehension in students with dyslexia.