Abstract: Over the past decade, there has been an explosion of research on false memory. Although linguistic stimuli have always been used in the Deese-Roediger-McDermott paradigm in research on false memory, surprisingly few of these studies have involved bilingual participants and cross-language effects in false memory. In the present experiment, language was utilized as a new variable to explore whether false memory of less proficient Chinese-English bilinguals can cross language boundaries just like in the case of proficient bilinguals, as demonstrated by recent studies. Furthermore, Chinese and English belong to different language systems; therefore, the modality effect reported by Smith and Hunt (1998), Maylor and Mo (1999), and Gallo et al. (2001) deserves more careful scrutiny. In this experiment, we also tested whether false memory with respect to Chinese characters has the same modality effect as it does on English words in order to enrich the explanations for this effect.
The research materials included 16 lists from Stadler, Roediger, and McDermott’s (1999) article (8 English and 8 Chinese lists). Each list consisted of the 12 most common associates to a critical lure. The recognition list comprised of 96 randomly arranged items, including 48 studied words, 16 nonstudied critical lures from each list, and 32 unrelated words, which were all presented in the same language or a different language used at study.
A mixed design was used, with two between-subject variables (study and test modalities) and one within-subject variable (the same and different languages between study and test). The participants were 112 undergraduate students who were randomly assigned to 4 experimental conditions (visual-visual, visual-auditory, auditory-visual, and auditory-auditory). The study lists and recognition lists were presented visually (English list: lowercase letters; Chinese list: Song font) in the center of the computer screen or auditorily (by a male voice) over headphones.
The results are reported in terms of the mean of true and false recognition proportions as well as the corrected recognition scores Pr for each language condition and modality condition. A 2 (language: English, Chinese) × 2 (study modality: visual, auditory) × 2 (test modality: visual, auditory) repeated measures ANOVA on different conditions indicated that there were main effects of language, study modality, test modality, and interactions among language, study modality, and test modality.
False memory of less proficient Chinese-English bilinguals can cross language boundaries, but false memory is higher in the same language than in a different language, and in the Chinese lists than in the English ones. Remarkably, in the present experiment, visual study led to significantly more false memory than auditory study, which was different from the findings reported in the existing Western studies. These findings are important to understand the mechanism of false memory

Key words: alse memory, DRM paradigm, cross-language, modality effect, bilingual memory representation