The purposes of this thesis were to study the influences of positive and negative instances in conjunctive and disjunctive concepts of learning, and to study the differences of males and females in conjunctive and disjunctive concepts of learning. The research was done with four groups of 30 subject each. In each group there were numbers of male and female subjects that were randomly selected from Mathayom Suksa 3 of Srinakarinthravirote University Secondary Demonstration School Groups I, II, III, and IV were tested by observing the positive and negative instances which were in proportions of 100 : 0, 75 : 25, 50 : 50, and 25 : 75 respectively. In each instance there were five dimensions and two values. There were twelve concepts, 6 conjunctive and 6 disjunctive. Each concept was randomly selected from two different dimensions. The experimenter showed these instances for about 10 seconds to each subject by reception paradigm. For positive instances the experimenter had to say a nonsense syllable and refuse them if there were negative instances. Subjects learned to know each concept from 12 positive and / or negative instances, and to answer on the answer sheet with about 20 seconds for each item. The subjects were said to have learned the concepts when they had answered all the problems. Test of homogeneity of Variance, analysis of Variance, and Newman - Keul's multiple comparison were, performed on the data. The major findings of this study were as follows: 1) There was a significant difference (at .01 level) among all groups, of subjects in conjunctive and disjunctive concepts of learning when the same proportion, of positive and negative instances were used. 2) There was a significant difference (at .01 level) among all groups of subjects in conjunctive concept learning when different proportions of positive and negative instances were used. It was found that the most and the least of conjunctive concept learning were in the proportions of positive and negative instances, 100 : 0 and 25 : 75 respectively. 3) There was a significant difference (at .01 level) among, all groups of subject in disjunctive concept learning when different proportions of positive and negative instances were used. It was found that the most and the least of disjunctive concept learning were in the proportions of positive and negative instances, 50 : 50 and 100 : 0 respectively.
