A developmentally sensitive problem-solving task, the 20-Questions Game, was utilized in this study to examine development-and treatment-related variables among 48 learning disabled and 48 normal boys of two age levels (8-10 years old and 11-13 years old). Subjects performed on three different versions of the experimental task before, immediately after, and two weeks after receiving one of two treatment packages, which differed in their emphasis on task-specific and metacognitive strategies. Results indicated that both younger boys and learning disabled boys demonstrated less efficient problem-solving skills than their older and normally achieving peers. In all subject groups, boys' ability to make use of specific problem-solving skills was highly influenced by variations in stimulus characteristics. Brief training was effective in improving the performance of all subject groups on trained and untrained materials at posttest and follow-up trials. However, expected differences in effectiveness between treatment packages were not realized.Research in the field of learning disabilities has increasingly focused on learning disabled children's difficulties in the development and implementation of problem-solving strategies. These children have been described as "inactive" or "maladaptive" learners (Torgesen, 1977(Torgesen, , 1982, who may fail to adapt to task demands through the use of efficient and organized strategies. Compared with same-aged normal children, they may have a limited understanding of their own cognitive activities and of the ways in which means-end thinking contributes to learning (Gavelek & Raphael, 1982).Successful training studies with learning disabled students suggest that their problem-solving difficulties may stem from "production deficiencies" rather than "mediational deficiencies" (Flavelt, 1970): When provided with sufficient environmental support, learning disabled children can effectively implement strategies (Torgesen, 1980). Still unresolved, however, is the issue of how much and what type of support will most likely elicit the use of efficient problem-solving strategies. The present study addressed this issue by varying the level of instructional support in teaching a problem-solving strategy to learning disabled and normal boys of different age levels.In attempting to remediate the problem-solving difficulties of learning disabled students, some investigators have focused on highly task-specific strategies such as verbal rehearsal (e.g., This article is based on the doctoral dissertation of the author submitted in partial fulfillment of the requirements for the doctoral degree at the University of Washington.The author wishes to thank the administrators, school psychologists, and teaching staffs of the Burlington-Edison, Ferndale, Mt. Vemon, and Sedro-Woolley School Districts in Washington State for their assistance and cooperation. The author also wishes to thank David Koch for his help in the preparation of this article.