The purpose of the study is to investigate the effects of mind mapping with cooperative learning (MMCL) and cooperative learning (CL) on: (a) programming performance; (b) problem solving skill; and (c) metacognitive knowledge among computer science students in Malaysia. The moderating variable is the students' logical thinking level with two categories: high logical thinking (HLT) ability and low logical thinking (LLT) ability. Past research has shown problems with regard to computer programming. Many discussion and debates among computer science experts agree that there are problems due to the traditional method used to teach computer programming which did not contribute to the development of above dependent variables. There were gaps or deficiencies identified in students' knowledge in computer programming in each phase of the programming processes. The critical problem, which was due to the lacking of skills in analyzing programming problems, was identified at the first phase of the programming processes. This problem consequently will have an effect on the next phase of a programming sequence. Based on the literature, the above independent variables identified can be used to help lessen the burden at the first phase of the programming processes. The study further investigated the effects of MMCL and CL among the students with high logical thinking (HLT) and low logical thinking (LLT) on the three above mentioned dependent variables. A quasi-experimental study with posttest-only control group method that employed a 3 x 2 Factorial Design was applied in the study. The sample consisted of 127 students from 2 classes in three Universiti Teknologi MARA campuses which were all randomly selected and assigned to the treatment groups (MMCL and CL) and control (T) group. The equivalence of all groups was established using the homogeneity test between the three groups based on the students' Malaysian Education Certificate (SPM) grades in Mathematics, Additional Mathematics, Physic, and Chemistry subjects. The results showed that the students in MMCL and CL groups have significant positive overall effects in programming performance, problem solving skill, and metacognitive knowledge. Students taught via the MMCL method significantly outperformed the students taught via the T method in programming performance and metacognitive knowledge. Meanwhile, students taught via the CL method significantly outperformed the students taught via the T method in metacognitive knowledge. Even though there was no significant effect between MMCL and CL on any dependent variables, the descriptive statistics for programming performance and metacognitive knowledge between the three instructional method groups did show some positive effects. For the effects of instructional methods on programming performance mean scores, there were two major findings. First, the results showed that the overall mean scores for programming performance for the students taught via MMCL method were higher than the CL group whose scores were, in turn, higher than the scores of the students taught via the T method. Second, both HLT and LLT students' mean scores in programming performance for MMCL method were higher compared to the mean scores acquired by the CL and T methods. For the effects of instructional methods on metacognitive knowledge mean scores, there were two major findings. First, the results showed that the mean scores for metacognitive knowledge for the LLT students taught via MMCL method was higher than LLT students taught via CL method whose scores, in turn, were higher than the LLT students taught via the T method. Second, the results showed that the HLT students taught via the CL method outperformed the HLT students taught via the T method in metacognitive knowledge. The results of this study also showed that the difference between HLT and LLT students among the three instructional methods were significant, overall in both programming performance and problem solving skills. Further analysis revealed that the HLT students taught via both MMCL and CL methods significantly outperformed the LLT students taught via the same method in problem solving skills. Finally, the results showed that there were no significant interaction effects between the instructional methods and the students' logical thinking levels for programming performance, problem solving skills, and metacognitive knowledge scores. The findings of this study suggest that mind mapping with cooperative learning method (MMCL) is preferred compared to CL and T methods in programming performance, problem solving skills, and metacognitive knowledge for students of all logical thinking levels. (Contains 5 tables.)