In this research on individual differences in thought processes, the primary interest was to study the process followed by the subject in order to reach the solution of the problem. For this purpose, the performance of 38 experimental subjects was studied throughout 24 problems of the training sessions. The methods used were group norms, length of plateaux calculated from the schemata norms, and performance curves based on group norms and on schemata norms. The second aim of the research was to study the individual differences in the process of solving a problem between subjects with training and subjects without training. The performances of 38 experimental subjects were compared with the performances of 38 control subjects individually matched before the experiment. These comparisons were made on the basis of 12 problems, 3 pretesting and 9 posttesting. The methods used to measure their performance were schemata norms, length of plateaux, individual performance curves based on schemata norms, and convex sets based on schemata norms. The third aim of the research was to study whether the educational level had an influence on the process followed by a subject in order to solve this type of problem. Results of the first study showed that the main effect schemata and the main effect content are statistically significant, as is the interaction between schemata and content. Schemata norms give more useful information about the problem-solving process followed. The second study results were that experimental subjects show a "better" performance than control subjects. The third study showed that the college student's process is always better than that of the high school student. The length of plateaux is shown to be a good measure in the characterization of process. (Author/DB)