An experiment was conducted to investigate three topics of current interest in education: critical thinking, programed instruction, and aptitude-treatment interaction. Thirty college students and 31 ninth graders with extreme discrepancies between their verbal and quantitative aptitude scores were taught eight principles of class reasoning by random assignment to one of two versions of a programed instruction unit. One version contained verbal examples and scientific terminology; the other substituted letter symbols for the scientific words. It was hypothesized that if adaptation of instruction to individual differences is an efficacious practice, then students who were highly verbal would learn better from the verbal program version and students with low verbal aptitude would benefit from the symbolic version. Although there was a slight tendency in the hypothesized direction, in no instance was the interaction between program version and discrepancy type significant at the .05 level. However, it was demonstrated that class reasoning principles can be taught by means of a relatively short instructional program and that such instruction will transfer to understanding of conditional reasoning principles. (Author/JY)