| This dissertation investigates cognitive claims about how visual representations affect problem-solving behaviors. The research focuses on the visual programming language LabVIEW. LabVIEW is interesting because of its visual representation and because of prior studies with seemingly conflicting results. The questions arising from the juxtaposition provided the framework for two projects: a survey to collect the opinions of experienced LabVIEW programmers and a controlled experiment to measure the comprehensibility of LabVIEW's visual representation.; The survey was conducted via the World Wide Web and resulted in 227 responses. The survey allowed some distinction between opinions about the visual aspects of LabVIEW and opinions about LabVIEW's many other features. Overall, the respondents were positive in their assessment of LabVIEW. Moreover, respondents rated the value of LabVIEW's visual language significantly higher all other rated features, including LabVIEW's extensive libraries of reusable code. At the same time, respondents' comments suggest other possible sources of LabVIEW's success such as LabVIEW's libraries of reusable code, support for building graphical user interfaces, use of the dataflow paradigm and automatic memory management.; The experiment compared LabVIEW to a textual equivalent. Fifty computer science students participated. Comprehension was measured three ways. First, subjects solved tracing problems; given code and input values, subjects were asked what output the code would produce. Second, subjects solved parallelism problems; given code with several program statements highlighted, subjects were asked about the sequence in which the statements could execute. Third, subjects solved debugging problems; given code and its specifications, subjects were asked to find the error in the code. The textual representation was significantly faster for the tracing problems. The visual representation was significantly faster for the parallelism problems and significantly more accurate for both the parallelism and debugging problems. These results contribute clear evidence in favor of a visual representation, evidence that LabVIEW succeeds in highlighting data dependencies and evidence that LabVIEW helps programmers to maintain an overview perspective of their code. |
