In three experiments, native Chinese speakers were asked to use their native and non-native languages to read and translate Chinese words and to name pictures. In Experiment 1, four groups of subjects with various degrees of proficiency in their second language, English, participated. In Experiments 2 and 3, subjects were first asked to learn a list of words in a new language, French, using either Chinese words or pictures as media; then they performed the reading, naming, and translation tasks. All subjects performed better in reading words than in naming pictures, when responding in Chinese. When the response was in the non-native language (English or French), high-learning subjects were equally efficient in translation and picture-naming tasks. Low-learning subjects, however, performed better in either the translation or the picture-naming task, depending on their learning strategies. These results are consistent with the idea that both proficiency in a non-native language and the strategy for acquiring the language are main determinants for the pattern of lexical processing in that language.One important question in the study of bilingual or multilingual processing concerns how people process words in their non-native language. This question has been investigated by many researchers, using a variety of semantic tasks (see, e.g., Snodgrass, 1984, for a relevant review). These have included, for example, bilingual naming and translation (e.g., Chen & Leung, 1989;Kroll & Curley, 1987;Potter, So, Von Eckardt, & Feldman, 1984), the semantic/sentence priming paradigm with a lexical decision task (e.g., Chen & Ng, 1989; Jin & Fischler, 1987; Kirsner, Smith, Lockhart, King, & Jain, 1984;Kroll & Borning, 1987;Schwanenflugel & Rey, 1986), and the Stroop type of interference task (e.g., Biederman & Tsao, 1979;Chen & Ho, 1986). The results of these studies with proficient, bilingual subjects are generally consistent with the concept-mediation hypothesis that the native and non-native languages of a proficient bilingual are operated independently, so that lexical items in the This research was supported by a Faculty Research Grant from The Chinese University of Hong Kong. This article was prepared while the author was a Research Fellow at the Institute for Perception Research (IPO), and the recipient of a research fellowship from the Eindhoven University of Technology, the Netherlands. I am grateful to the principal, Chun Ping Ng, and the students of the Po Leung Kuk Siu HonSum Primary Schoolfor their kind assistance and participation; to YuenSum Leung, Siu-Yee Mak, Kin-TongChan, Siu-Yee Wong, and FungYing Siu for assistancein conductingthe experiments; and to MaryPotter, Annette de Groot, James Chumbley, and an anonymous reviewer for helpful comments and suggestions on an earlier draft. Correspondence should be addressed to Hsuan-Chih Chen, Department of Psychology, The Chinese University of Hong Kong, Shatin, NT, Hong Kong. two languages are not directly associated but are connected through an amodal conceptual system ...