Prior General Problem-Solving Ability and Transfer in Unfamiliar Computer Science Domains

Abstract Computer Science (CS) students often encounter novel topics that demand learning new concepts and skills. A central question is whether students’ prior general problem-solving ability provides a transferable advantage when they learn an unfamiliar CS domain, above and beyond any advantage of greater prior domain knowledge. This integrative review critically examines existing theory and evidence regarding transfer of general problem-solving skills into new CS contexts. I review transfer theory (e.g. near vs. far transfer; low-road/high-road; common-elements) and distinguish key constructs such as general problem-solving ability, computational thinking, algorithmic thinking, programming skill, domain knowledge, intelligence, and others. I organize evidence from CS education and cognitive science at several levels of relevance (direct, related, CS-to-CS transfer, computational thinking transfer, programming transfer, indirect theory, and contradictory findings). The strongest related evidence comes from meta-analyses and studies of programming instruction and computational thinking (e.g. moderate cognitive transfer from learning programming, positive cross-domain CT transfer. However, no prior study has explicitly isolated general problem-solving skill while controlling target-domain knowledge. I identify this empirical gap precisely, and develop a proposed conceptual framework (“Problem-Solving Transfer Framework”) that outlines how inputs (problem-solving ability, domain knowledge, computational thinking skills, etc.), instructional conditions, and cognitive mechanisms (abstraction, analogy, decomposition, metacognition) jointly influence outcomes (learning, near transfer, far/novel transfer, retention). Finally, I describe how this research question could be tested via a rigorous design: e.g. assessing CS students’ domain knowledge and general problem-solving ability at pretest, giving standardized instruction, then measuring learning gains and transfer tasks (near and novel/far). This review contributes clearer conceptual distinctions (problem solving vs. CT vs. programming skill, etc.), a synthesis of transfer evidence in CS, a proposed theory of conditions enabling (or limiting) transfer, and a detailed empirical plan to examine the relationship.