Answers are grounded in this paper and cite their sources.
This scoping review maps execution-trace techniques relevant to connecting game behavior with source code, considering live operation, representation, runtime cost, and evaluation. Searches of IEEE Xplore, ACM Digital Library, ScienceDirect, and Scopus in July 2026 yielded 289 records. English-language, obtainable reports meeting general and topical eligibility criteria were screened and charted initially by one reviewer. Claude Opus 5 (Anthropic) assisted preliminary screening; all decisions were manually verified, and all coauthors verified results and resolved disagreements by consensus. Of 240 screened records, 97 reports representing 92 studies were included; three sought reports were not retrieved. Nineteen studies (20.7%) address live or mixed operation, 19 (20.7%) involve human participants, and 25 (27.2%) address instrumentation. Contextual sources include industrial profilers and JOVE, a Java execution visualizer demonstrated with pinball. Existing work supports live observation, trace reduction, and links to code, but the examined evidence does not establish comparative effectiveness in contemporary game-engine comprehension tasks. Heterogeneous overhead measures prevent a common frame-time estimate. We distinguish capture granularity, temporal resolution, aggregation, refresh, and observation latency, and propose task-based evaluations linking correctness and time to execution perturbation and information retention. Search vocabulary, retrospective criteria changes, and heterogeneous evaluations limit the evidence map.
No takes yet. Share an insight, caveat, or question.
Silva et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: