In this paper, the study model, which presents the patterns of how the changing mechanisms are adopted to the changing stages of smoking cessation, was developed by modifying the integrative model of Prochaska et al. (1983) with including seven changing mechanisms which were identified by Oh and Kim(1996). Then the developed study model was exploratively tested with 155 University student between 20 and 29 years of age as subjects. According to the study results. among the five mechanisms. which are significant in explaining the differences between stages. DUNCAN post-comparison showed that the least applied ones were Stimulus Control, Self Determinism, Cognitive Restructuring in the precomplation stage, and Reinforcement and Dramatic Relief. in the relapsed stage. In the comtemplation stage. it was observed that Dramatic Relief is used most frequently and that the other two mechanisms, Information Management and Cognitive Restructuring, showed different results from those inferred in the study hypothesis. In the case of Information Management, it was excluded from the analysis it was not included in explaining significant difference among changing stages, but Cognitive Reconstruction turns out to be a more frequently used mechanism in the action stage rather than in the comtemplation stage. Helping Relationship was also excluded in the post-comparison analysis since it was not included in explaining significant difference among changing stages and Reinforcement was a more frequently used mechanisms in the comtemplation stage. Stimulus Control turns out to be the connecting mechanism which was most frequently used in both the contemplation and action stages. Self Determination was most frequently used in the action stage rather than in both the comtemplation stage and action stage, differing from the presumption of the model. Lastly, subjects in the relapsed stage were utilizing Stimulus Control and Self Determination at the same level as subjects in the precontemplation stage. and the utilization of both Reinformcement and Dramatic Relief was lower than that of the precomtemplation stage, that is, at the lowest level. Only Cognitive Restructuring was used of the same level as the comtemplation stage. The relapsed stage in this study did not represent the preparation for action stage as presumed in the model of Prochaska et. al. (1983) but did show a pattern similar to the initial stage of smoking cessation. However, since this interpretation about the relapser was based on only a small number of relapsers(n=5), this conclusion may not be reliable.