Transfer is typically thought of as requiring individuals to "see" what is the same in the deep structure between a new target problem and a previously encountered source problem, even though the surface features may be dissimilar. We propose that experts can "see" the deep structure by considering the first-order interactions of the explicit surface features and the second-order relationships between the first-order cues. Based on this speculative hypothesis, we propose a domain-specific bottom-up instructional approach that teaches students explicitly to focus on deriving the first-order interactions cues and noticing the second-order relationships among the first-order interaction cues. To do so, researchers and instructional designers need to first extract from experienced solvers or experts how they derive such first-order cues. Transfer is assumed to be based on the similarities in the second-order relationships, which are familiar everyday relationships such as equal to, greater than, and so forth.