Stimulus structures and mental representations in expert comprehension of computer programs

Comprehension of computer programs involves detecting or inferring different kinds of relations between program parts. Different kinds of programming knowledge facilitate detection and representation of the different textual relations. The present research investigates the role of programming knowledge in program comprehension and the nature of mental representations of programs; specifically, whether procedural (control flow) or functional (goal hierarchy) relations dominate programmers' mental representations of programs. In the first study, 80 professional programmers were tested on comprehension and recognition of short computer program texts. The results suggest that procedural rather than functional units form the basis of expert programmers' mental representations, supporting work in other areas of text comprehension showing the importance of text structure knowledge in understanding. In a second study 40 professional programmers studied and modified programs of moderate length. Results support conclusions from the first study that programs are first understood in terms of their procedural episodes. However, results also suggest that a programmer's task goals may influence the relations that dominate mental representations later in comprehension.