Title: The effect of representation on learning to reason with problems involving computer program-oriented complex logic
Abstract: The effects of logic representation on subject logical reasoning with computer program oriented complex logic, and of subject skills on reasoning with specific representations of complex logic, were studied. A complex logic test and self-instruction manual were designed to examine reasoning differences attributable to: (a) type of complex logic representation (graphical vs. programmatic); (b) indentation effects in computer program representations; (c) micro-language effects in program representations, and; (d) the manner in which a logic test question was specified. Relationships between reasoning with a specific logic representation and Group Embedded Figures Test score, SAT-Mathematics and SAT-Verbal scores, and degree of programming experience, were also examined. The propositional equivalence construct was the unit of logic for all complex logic representations. Six equivalent representations were employed: a logic table; a logic tree; a program written in a micro-language having a branch-to-label conditional construct, a program in a micro-language having a branch-to-position conditional construct (indented and nonindented versions). Students in a college computer literacy course were assigned to treatment groups and instructed and tested in reasoning with one of the complex logic representations. Experimenter involvement, and the instructional and test protocols were controlled; intervening linguistic effects in the experimental materials were mimimized. There was a 50% difference in mean complex logic test performance scores between subjects using the logic tables or logic trees and subjects using the computer programs. No other representational effects were found. There were strong correlations between Group Embedded Figures Test score and complex logic test score for the logic table (r = 0.55, p $<$.01) and logic tree (r = 0.71, p $<$.01) groups. Evidence of relationships between mathematics, verbal, or programming skills and complex logical reasoning in an computer programming context, was mixed. For the logic tree group, mean performance score was 12% higher on questions where logical predicates were color conditions, compared to that on questions where logical predicates were outcomes.
Thus for reasoning with computer program oriented complex logic, visual representations of the logic appeared more effective than verbal representations.
Publication Year: 1991
Publication Date: 1991-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot