In the study, young adolescents’ hypothesis development in a computer-supported and problem-based learning environment was examined in terms of two empirical studies. The first study examined the effect of metacognitive scaffolds to strengthening hypothesis development as well as the influence of hypothesis development in the promotion of young adolescents’ problem solving performance in an ill-structured problem solving environment, Animal Investigator. Data was collected from sixth grade students (N = 172). The findings of the study indicated that participants using metacognitive scaffolds attained significantly higher hypothesis-development performance. Results also revealed that the hypothesis-development performance showed the predictive power of the solution development performance. In the second study, the researcher examined three factors, motivation, metacognition, and prior domain knowledge, as a predictor for children’s hypothesisdevelopment performance in the problem-based learning environment. A hypothesized model was evaluated using structural equation modeling, which is a statistical method of causal relationships. Data were collected from sixth grade students (N = 101) in treatment groups. Two significant factors toward children’s hypothesis-development performance in an ill-structured problem solving environment were determined: Prior domain knowledge and metacognition. Implications and limitations of the present study and issues including the experimental design are discussed.