The purpose of the study was to identify differences in cognitive skills associated with differing rates of predictive reasoning success for high school biology students, to determine possible correlations between predictive success and cognitive operational level, and to assess whether directed practice facilitated problem-solving success. The Group Test of Logical Thinking (GALT) was used to assess cognitive development, while written prediction sheets and oral interview were used to identify skill use patterns and measure predictive success. Treatment-group subjects then received 8 hours of directed practice in prediction using interactive computer simulations, and all subjects were retested. Predictive success showed a significant correlation both to subjects' operational level and to five specific cognitive skills. ANCOVA indicated a significant treatment effect, with marked increases in predictive success following practice in the identified problem-solving skills. (Author)