The Korean orthography uses both alphabetic Hangul and logographic Hanja. Two experiments investigated semantic and phonological processing of words written in the two scripts. In the experiments, Korean readers had to respond to words either in a pure context with words from one single script or in a mixed context with words from the two scripts. The task was naming in Experiment 1 and semantic categorization in Experiment 2. The results showed that for Hangul words, there were significant word frequency effects in categorization, but not in naming, and that there were reliable script-switching effects in naming, but not in categorization. For Hanja words, however, there were clear and strong effects of words frequency, regardless of the task used, but significant effects of script-switching were only observed in categorization. These results suggest that the strategies adopted in processing Hangul and Hanja words are determined both by task demand and nature of the script.