The aim of this study was to examine the effect of contextual interference using bandwidth and self-control feedbacks on learning of parameter and generalized motor program in pursuit tasks. 180 novice, right hand and volunteer subjects (18-25 years) were participated in two separate experiments. Each experiment included six groups of blocked, serial and random of bandwidth and self-control (6 groups of 15 subjects in each experiment). The task of subjects in first experiment was to pursuit light target on the rotary pursuit system with stable motor program (quadrangle) and variable parameters (with speeds of 20, 30 and 40 cycle pre minute). In the second experiment, the light target was pursuit with stable parameter (20 cycle pre minute) and variable motor programs (patterns of quadrangle, circle and triangle). Both experiments were conducted with, different arrangements and feedbacks. The time on target (TOT) was recorded by computer as performance score of subjects. After participating in the pre-test (including one blocked of 9 trials), all subjects, practiced for eight sessions and in each session two blocks of nine trials were performed and then they participated in the acquisition, retention and transfer (after 72 hour) tests. The data were analyzed with using one way analysis of variance (ANOVA), factorial ANOVA with Repeated measures and Bonferroni post hoc tests. The finding showed that in both experiments, the blocked groups performance of bandwidth and self-control was significantly better than other groups in the acquisition stage (P<0.05). However, there was no significant difference in the performance of groups in the retention and transfer stages in first experiment (P>0.05). However, the performance of bandwidth and self-control random groups in second experiment was significantly better than other groups in retention and transfer stages (P<0.05). On the other hand, the performance of self-control feedback groups in both experiments was better than bandwidth feedback groups (P<0.05). Finally, the results of the present research show the useful effects of bandwidth and self-control feedback on learning of generalized motor program relative to the parameter and provided other support for the notion of dissociation between these processes, and Magill & Hall hypothesis.