This paper explores the concept of educational psychology as a "linking science" between (1) theory and research in psychology and learning and (2) the educational arena of the schools. A linking science is defined as a special structure that intervenes between scientific theory and practical application. The author's first major point is that a linking science is needed between psychology and education, but that traditional approaches of applying theory to practice are wrong. To substantiate her opinion, the author reviews the work of three experimental psychologists, A. W. Melton, K. W. Spence, and B. J. Underwood. Among their concerns is that classroom settings and individual differences confound the straightforward application of theories derived in controlled laboratory situations. The author's second major point is that researchers must be aware of the curriculum design problems for which their research will ultimately be applied. This will ensure useful relevance between the two fields, psychology and education, and will promote the success of educational psychology as a linking science. The paper concludes with a discussion of important areas for research that will assist the development of the linking science. (AV)