An Empirical Study of Plan-Based Representations of Pascal and Fortran Code.

Abstract

The first step in program modification is comprehension. Several researchers have argued recently that programmers utilize a plan-based representation when composing or comprehending program code. In a series of studies we are testing the psychological validity of this proposal and examining the nature of plan-based program representations. Here we report the results of our first study in which programmers segmented code and sorted programs. The segmenting data showed the programmers agree on the major components of a program and that these components are defined by goals in a plan representation. Pascal and Fortran programs that employ the same plan structures were segmented into similar components. These components were labelled with similar subgoal descriptions. The majority of subgoals described were abstract, with few being task-specific. Program sorting data also shows clustering into plan groups, however some secondary dimensions, like the type of data used (in Pascal programs), may also be important parts of program representation. Keywords: Man computer interface; Software psychology.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1987
Accession Number
ADA183797

Entities

People

  • Chiung-chen Yu
  • Scott P. Robertson

Organizations

  • Rutgers University–New Brunswick

Tags

Communities of Interest

  • Biomedical
  • C4I
  • Engineered Resilient Systems
  • Weapons Technologies

DTIC Thesaurus Topics

  • Cognitive Systems Engineering
  • Command And Control
  • Computer Programming
  • Computer Programs
  • Computer Science
  • Computers
  • Data Analysis
  • Engineering
  • Human Factors Engineering
  • Human-Computer Interaction
  • Language
  • Military Research
  • Programming Languages
  • Psychology
  • Students
  • Three Dimensional
  • Two Dimensional

Readers

  • Computer Vision.
  • Database Systems and Applications
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.