Distributional analysis and De Jong, Liang, and Lauber's (1994) dual-process model of the Simon effect

R. De Jong, C.-C. Liang, and E. Lauber (1994) proposed a dual-process model to account for the Simon effect and its reversal. Their proposal included a distributional analysis whose results, they claimed, support the time-course assumptions they make for the 2 processes of the model. It is shown that the 2 functional components of the dual-process model, the unconditional and conditional automaticity, are equivalent to earlier accounts of the Simon effect and its reversal, namely automatic response activation of the dimensional overlap model (S. Kornblum, T. Hasbroucq, gr A. Osman, 1990) and logical recoding (A. Hedge & N. W. A. Marsh, 1975), respectively. It is also shown that the distributional analysis is a simple computational procedure that reflects fundamental statistical properties of the underlying reaction time distributions and their interrelationships and that De Jong et al.'s time-course assumptions precluded at least half of these interrelationships. Indeed, experimental results from tasks in which the Simon effect is obtained often violate these assumptions, as is demonstrated in this article. Finally, it is also shown that De Jong et al.'s data are consistent with the hypothesis that the Simon effect and its reversal, irrespective of the task type in which it is obtained, can be accounted for by a common mechanism with 2 independent functional components.