Systematic literature review of empirical studies on mental representations of programs

摘要

Programmers are frequently tasked with modifying, enhancing, and extending applications. To perform these tasks, programmers must understand existing code by forming mental representations. Empirical research is required to determine the mental representations constructed during program comprehension to inform the development of programming languages, instructional practices, and tools. To make recommendations for future work a systematic literature review was conducted that summarizes the empirical research on mental representations formed during program comprehension, how the methods and tasks have changed over time, and the research contributions. The data items included in the systematic review are empirical studies of programmers that investigated the comprehension and internal representation of code written in a formal programming language. The eligibility criteria used in the review were meant to extract studies with a focus on knowledge representation as opposed to knowledge utilization. The results revealed a lack of incremental research and a dramatic decline in the research meaning that newly developed or popularized languages and paradigms have not been a part of the research reviewed. Accordingly, we argue that there needs to be a resurgence of empirical research on the psychology of programming to inform the design of tools and languages, especially in new and emerging paradigms.