This study explores the effect of self-regulated learning support on learners’ cognitive load and problem-solving performance, considering cases from well-structured to ill-structured tasks in a computer programing course. Sixty-seven undergraduate students in a computer programing fundamentals course were randomly assigned into one of two groups: (1) experimental group ( N = 34; received self-regulated learning support) and (2) control group ( N = 33). Participants in both groups were asked to solve programing problems using Python over 3 weeks, and their cognitive load levels were measured after each week. The results showed that novice learners’ self-regulated learning skills can be influenced by learning support, which contributes to solving complex problems. The results also suggest that, given the sequence of tasks (i.e., basic and advanced well-structured and ill-structured problems), self-regulated learning support may require time to be effective. The germane load of the experimental group was higher for well-structured problems than that of the control group. Using logistic regression, the two groups were differentiated through different variables measured in this study, which showed a high predictive explanatory power. The results provide implications for designing self-regulated learning support for programing courses, using problem-based learning in association with problem-solving skills, self-regulated learning skills, and cognitive load management.