Native speakers of alphabetic languages, which use letters governed by grapheme-phoneme correspondence rules, often find it particularly challenging to learn a logographic language whose writing system employs symbols with no direct sound-to-spelling connection but links to the visual and semantic information. The visuospatial properties of logographic languages, such as Chinese, have inspired the development of promising instructional methods informed by the Dual Coding Theory (Paivio, 1971) which postulates that coding a stimulus in two different ways, verbally and visually, increases retention compared to a single-coded stimulus. Extending this line of research, this study investigated the effects of dual-coded instruction, including an additional image morphing component, on learning Chinese vocabulary. One hundred twenty adult participants without prior knowledge of logographic languages were randomly assigned to one of four treatment groups: a text-only method group, an animation-only method group, an animation-plus-text method group, and an animation-plus-narration method group. All participants learned the same 24 Chinese characters, of which 12 were concrete characters and 12 were abstract characters. Controlling for participants' spatial ability, e.g., mentally rotating pictures of three-dimensional objects, statistical analyses revealed that the dual coded methods worked better than the single coded methods, and participants performed better on concrete characters than on abstract characters. In addition, under the single coded method, the animation-only method resulted in more substantial gains in Chinese vocabulary learning than the text-only method, while under the dual coded method, the animation-plus-narration method was superior to the animation-plus-text method. These results suggest that dual coded instructional methods involving image morphing based computer animations with multi-sensory input have a significant positive influence on logographic language vocabulary learning.