Cognitive ability has been identified as a factor which is associated with accounting expertise, yet little is known about its effects on accounting task performance. The widely-accepted model of accounting performance proposed by Libby and Luft (1993) predicts that cognitive ability influences accounting performance directly and indirectly by affecting task-specific knowledge. Empirical inconsistencies found in these relationships are largely unexplored in the accounting literature, and so a complete understanding of the ability~performance relation remains elusive. The psychology literature suggests that cognitive style, as well as cognitive ability, contributes to the cognitive processes that underlie accounting performance. Research has shown that cognitive style can affect accounting performance, particularly under conditions of cognitive misfit (Chan 1996). Cognitive misfit occurs when an accountant's cognitive style interacts with incongruent demands of the accounting task. Theory suggests the condition of cognitive misfit may explain empirical inconsistencies found in prior ability~performance research. The existing accounting literature has not investigated how cognitive misfit impacts the ability~performance relationship in accounting tasks. The current study explores how the condition of cognitive misfit affects direct and indirect relationships between cognitive ability and accounting performance proposed in the Libby and Luft (1993) performance model. The new knowledge generated in this research will benefit the accounting profession in its quest to identify the determinants of expert performance.