Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Health Sciences


Rehabilitation Sciences

First Advisor

Dr. Robert Marshall


This cross sectional study examined problem solving by typically developing children on the Rapid Assessment of Problem Solving test (RAPS). The RAPS, a modification of Mosher and Hornsby’s 20Q task, requires the examinee to solve three problems. Each problem involves asking yes/no questions to identify a target picture from a 32-picture array with as few questions as possible. Participants were 73 young (ages 7-9), 79 early adolescent (ages 10-13) and 77 adolescent (ages 14-17) children residing in Kentucky. Children were seen in the summer months and administered the RAPS on a single occasion, with 22 of the children being testing twice. All children passed screening tasks and completed RAPS testing without difficulty. Test-retest stability for the RAPS was adequate for clinical purposes and no learning effects were seen on the test. Results were examined to identify group differences in components of executive functioning (planning, strategy selection, strategy execution, and strategy shifting) that impact problem solving efficiency. To determine how children went about solving problems, questions were classified by type and in terms of when they were asked in the sequence of questions leading to solving of a problem. Results revealed that the young group differed from the early adolescent and adolescent groups on several objective measures: number of questions to problem solve, use of constraint questions, problem solving efficiency, mean integration planning score, and overall RAPS efficiency. The young group also differed from the two older groups in terms of the types of questions asked and when certain types of questions were asked in solving a problem. Young children were more prone to guess on early questions whereas older children asked effective constraint questions. Many of the differences suggest young and older children and young and older adults differ in their ability to integrate information needed to solve RAPS problems effectively. Findings of this study suggest there are age-related differences in solving fixed-alternative 20Q problems and provide a normative data base for using the RAPS to assess problem solving of both normal and disabled children in the age range studied.