Sections 1 and 2 reports on two research studies carried out by the authors, section 3 is a series of abstracts of other research studies stimulated by the authors and section 4 is an appendix containing the instructional materials, tests, raw data, and other items of interest. In the first study the concept "trochaic meter" was taught to 180 college students by means of eight treatment conditions. The independent variables involved presenting a definition or instances or both combined with attribute definition and/or attribute prompting. Dependent variables were errors in correct classification and specified classification. Hypotheses consisted of prediction of particular errors for each treatment and six of the eight hypotheses were supported. In the second study, 100 education psychology students were used in an investigation of four instructional strategies for promoting the acquisition of an infinite concept class. The independent variables were 1) probability level of concept instances; 2) matching of positive instances to a negative instance so that the irrelevant attributes are similar; and 3) divergency of positive instances with one another so that all of their irrelevant attributes differ. The four predicted dependent variables were all statistically significant. (MBM)