ERIC Number: ED232857
Record Type: RIE
Publication Date: 1979-Nov
Reference Count: 0
Goals and Methodology of Research on Solving Physics Problems. TR-158.
Novak, Gordon S., Jr.
The research project described will identify the component skills (cognitive processes and associated knowledge structures) required for effective problem-solving, and identify the kinds of learning which can improve problem-solving ability. These areas will be investigated by means of a computer program which can solve physics problems stated in English. The program's problem-solving techniques will then be compared with techniques inferred from protocols taken from expert/novice humans solving the same problems. Using a computer program which can solve physics problems has several advantages. First, requiring that the program be able to solve complete physics problems helps to insure that no essential component skills have been left out or assumed to be trivial when they are not. A second benefit is that the program can provide a testbed for different problem-solving strategies. A third benefit is that such a program could form the basis of an intelligent computer-assisted-instruction program which understands all the component skills required to solve problems. Several hypotheses about the nature of problem-solving in physics are discussed and related to the computer program to be developed and to difficulties of humans in learning to solve physics problems. (JN)
Publication Type: Reports - Descriptive
Education Level: N/A
Sponsor: National Inst. of Education (ED), Washington, DC.; National Science Foundation, Washington, DC.
Authoring Institution: Texas Univ., Austin. Dept. of Computer Science.
Identifiers: Computer Models; National Science Foundation; Science Education Research
Note: Paper presented at the NIE/NSF Project Directors' Meeting (Fredericksburg, VA, November 2, 1979). For related documents, see SE 042 630-635.