The purposes of this study were: to compare students' performance on conceptual and algorithmic chemistry problems; to investigate the relationships between learning orientation, formal reasoning, and mental capacity and students' performance on conceptual and algorithmic problems; and to investigate interactions among learning orientation, formal operational reasoning, and mental capacity. Participants were Grade 11 students enrolled in scientific sections of three Lebanese schools. Learning orientation, formal reasoning and mental capacities were measured using the Learning Approach Questionnaire, the Test of Logical Thinking, and the Figural Intersection Test, respectively. Also, students solved conceptual and low M-demand and high M-demand algorithmic chemistry problems. Students' performance on conceptual and algorithmic problems was compared. Regression analyses were used to examine the predictive power of the cognitive variables on each type of chemistry problems. In addition, performance of meaningful and rote learners was compared on all types of problems. Results showed that students performed significantly better on algorithmic than on conceptual problems. Moreover, meaningful learners outperformed rote learners on a test of conceptual problems while no significant differences existed for both levels of algorithmic problems. The three cognitive variables were significant predictors of performance on conceptual chemistry problems but not on algorithmic problems.