The performance of 144 pairs of college students on four successive concept attainment problems was assessed. A 3 x 2 x 2 x 4 factorial design with repeated measures on the last factor was used with the following variables: (a) type of interaction format (cooperation with discussion allowed during solution of the problems, cooperation with discussion not allowed, or competition), (b) memory aid (available or not available), (c) sex (male or female pairs), and (d) problems (four for each pair). Major results were: (a) discussion resulted in better performance on all measures (except time to solution) than did non-discussion or competition, which did not differ significantly; (b) on successive problems, discussing pairs increased their use of the more efficient focusing strategy and decreased time to solution, while non-discussing and competitive pairs showed no change; (c) memory aids resulted in fewer card choices to solution and fewer untenable hypotheses; and (d) no sex differences with the exception that females required less time to solution than males. (BW)