The thesis reports the development and trial of an interactive computer-assisted environment program. A rotational dynamics problem of considerable difficulty was chosen for program development involving a selection process using students and lecturers. Six year one undergraduates, 6 graduates, and a lecturer from the Physics department performed think-aloud problem solving. A high school Physics teacher and 7 year one undergraduates took part in pen and paper problem solving. Data collected focussed on: activities performed at commencement; question clarification and assistance sought; examples referred to for help; general mistakes made; conceptions portrayed; major difficulties encountered; and knowledge used. The problem solving has been categorised into 14 groups. These show the differences between expert and novice styles, highlight the strategies used, and depict the complexity of the problem-solving task through individual flow charts. The problem-solving was performed in an idiosyncratic manner. This has informed the program development to represent knowledge so that it can be accessed at various stages of problem solving when needed. This suggests a constructivist environment with the provision of information through transmission as appropriate. Solutions by the lecturer and the 4 high school teachers were used to determine the knowledge and explanations normally presented. A general solution using the fundamental physical quantities was developed to highlight the complexity of the solution and the vast amount of related knowledge required. This knowledge is usually assumed and normally not presented to students. The findings informed the development of a problem-solving program. The program has been used by 18 and 14 students respectively to solve the problem for which it was developed and a related problem. Data collected highlighted the assistance students seek, the extent of student control allowed, program features used and limitations encountered. Data was also examined to ascertain transfer of learning, and the effectiveness of the program in compiling student conceptions, knowledge and problem-solving styles. Analysis indicates students will use a program if it is helpful in serving their needs and is easy to use. Students were not very effective in transforming knowledge to a new situation. Data collected by the program showed its effectiveness in documenting information and features used, but it was less useful in compiling alternative conceptions.