Rule Differences, Practice, and Response Label in Complete Learning

Rule differences in a complete learning (CL) task were investigated over practice and conditions of neutral and directional response labels. The Ss were forty 13- to 14-year-old paid volunteers, 20 males and 20 females. Ss were given 20 problems, in four blocks of five problems each. Within each block, problems representing the Affirmative (Af), Conjunctive (Cj), Conditional (Cd), Inclusive Disjunctive (Id), and Biconditional (Bd) rules were presented. The order of rule difficulty was found to be Af < Cj = Cd = Id < Bd. Performance improved after the first block of five rules with no further change over the next three blocks of problems. No significant effect was obtained for Label, and there were no significant interactions. These results for rule difficulty and practice were concluded to be generally consistent with both Neisser and Weene's and Hunt, Marin, and Stone's models of CL, and to clarify previous results concerning Rule by Practice interactions. The absence of a Label effect is inconsistent with previous findings in attribute indentification and rule learning, and may be unique to the CL selection paradigm.