The effects of different approaches to learning Chinese characters were investigated. Ninety-two high-school students were randomly assigned to one of five treatment groups: translation, verbal mnemonics, visual mnemonics, dual coding mnemonics, or self-generated mnemonics. All groups received instruction and completed posttests in a computer-based environment. The results indicate that participants who generated their own mnemonics demonstrated higher post-test performance than those in visual coding, verbal coding, and translation groups; subjects in the dual coding group scored higher than those in the translation group. Those who generated their own mnemonics spent more time on task than any other group, and those in the verbal coding group took more time than those in the translation group. Survey and qualitative data suggest that learners' interpretations of the Chinese characters were rooted in their cultural backgrounds and personal experiences.