The effect of a foreign language on moral judgment: Boundary conditions and replicability in Japan and China

This project aims to clarify the boundary conditions of the Moral Foreign-Language Effect (MFLE) in moral judgment using a combined approach of survey-based measurement and laboratory experimentation. Specifically, we examine the replicability of MFLE and its boundary conditions under a second language (L2) condition, potential differences in effect magnitude among third language (L3) learners (L2–L3 differences), and the influence of cultural context (Japan vs. China) on MFLE in an integrated framework. Prior work has suggested that using a foreign language may attenuate affective responses and facilitate more deliberative processing, thereby increasing utilitarian judgments. At the same time, accumulating evidence indicates that findings can be inconsistent across studies, potentially due to moderators such as language proficiency, task characteristics, and presentation conditions. Accordingly, rather than treating MFLE as a simple “present vs. absent” phenomenon, this project focuses on specifying when, under what conditions, and to what extent the effect emerges. Concretely, we explore whether and to what degree MFLE varies as a function of factors such as foreign-language proficiency, different types of moral dilemmas, stimulus presentation modality (reading vs listening), and the source of decision recommendations (human vs AI), thereby improving the testability and cross-situational generalizability of conclusions in the MFLE literature. This Study 1 examines whether moral judgments differ when moral dilemmas are processed in a native language (L1) versus a foreign language (L2: English) and clarifies which decision components drive any observed language effects. Using the CNI (Consequences–Norms–Inaction) model and a standardized set of 24 CNI dilemmas, we test whether L2 reduces sensitivity to consequences (C) and moral norms (N) and whether it changes generalized inaction tendencies (I) (Qian et al., 2024). We collect parallel samples from Japan and China to evaluate replicability across cultural contexts (Białek et al., 2019; Nadarevic et al., 2021), and we explore whether individual differences in death anxiety and empathy (Strupp-Levitsky et al., 2020) moderate the magnitude of the L1–L2 differences