Dissertation Information for Peter L. PirolliNAME:
DEGREE:
DISCIPLINE:
SCHOOL: ADVISORS: COMMITTEE MEMBERS: MPACT Status: Incomplete - Not_Inspected Title: PROBLEM SOLVING BY ANALOGY AND SKILL ACQUISITION IN THE DOMAIN OF PROGRAMMING Abstract: Previous research (Anderson, J. R., Pirolli, P. L., and Farrell, R. in press . Learning to program recursive functions. In M. Chi, R. Glaser, & M. Farr (Eds.) The nature of expertise.; Pirolli, P. L. & Anderson, J. R. 1985 . The role of learning from examples in the acquisition of recursive programming skills. Canadian Journal of Psychology, 39, 240-272.) on skill acquisition in programming recursion is extended by the development of a computer simulation model called X that addresses how people achieve solutions by analogy to examples and how learning from such analogies takes place. The X model is a production system model of cognition, based on the ACT* theory (Anderson, J. R. 1983 . The architecture of cognition, Cambridge, MA: Harvard University Press.), that invokes problem-solving by analogy to examples when domain-specific skill is lacking. The model uses knowledge of the structure, functionality, and preconditions of components of an example solution to guide problem-solving. Skill acquisition mechanisms create new production rules that summarize problem solutions and facilitate future performance. Verbal protocols of novices and experts were used to develop simulations of novice performance, expert performance, and skill acquisition in programming recursion. Two experiments relevant to problem-solving by analogy and learning recursion are also reported. In Experiment 1, subjects given examples discussed in the context of instruction that elaborated how recursive functions are written in a manner consistent with expert knowledge learned to program recursion faster than subjects given examples in the context of instruction that elaborated how recursive functions operate. In Experiment 2, it was found that subjects given examples that were highly similar to training programs showed more intrusions of the example solution in transfer programs than subjects given low-similarity examples or no examples. The implications of the X model and these results for fundamental issues of knowledge acquisition and teaching are discussed. |
MPACT Scores for Peter L. PirolliA = 0 Advisors and Advisees Graphgenerating graph, please reload |
Students under Peter L. Pirolli
ADVISEES:
- None
COMMITTEESHIPS:
- Michael George Berger - University of California, Berkeley (1994)