The purposes of this study were to identify the cognitive skills and strategies that research experts use in experimental research and to evaluate the extent to which experts and non-experts can accurately report their problem-solving processes. Analysis identified three strategies used across levels of expertise: (1) begin with a strong prediction about a theory and interpret results in relation to it; (2) begin with a highly complex exploratory design to maximize the likelihood of patterns suggestive of a hypothesis; or (3) use the least complex designs possible and interpret the cumulative results in a linear fashion. Despite clear articulation of these strategies, participants omitted at least 48% of the steps they took to generate research designs and evaluate results.