In the present study, a computerized Chinese character matching task was designed to elaborate the cognitive process of character recognition in Chinese. Due to the capability of gathering a great amount of measures in t his task, the analysis of individual differences becomes possible. Some earlier conceptions about internal representation of Chinese characters were also re-examined with a data-analytic strategy which did not assume that different subjects follow the same psychological process. Two experiments were conducted and the results showed that (1) using dummy variable coding, linear model could be adopted to fit the process model with reaction time data, and (2) the morphological code seemed to play a more important role than phonological code in this kind of Chinese character matching task.