The Resolution of Relative Clause Attachment Ambiguity in L1-English L2-Italian Instructed and Immersed Bilinguals

Research on relative clause (RC) attachment ambiguity has shown that parsing preferences vary cross-linguistically, and that bilingual exposure can influence both the L2 and the L1. However, few studies have examined parsing preferences in both languages of the same bilingual populations, and the English–Italian pairing remains underexplored. This study addressed these gaps by investigating RC attachment resolution in 112 participants across four groups: British university students learning Italian in UK classrooms (n = 27; age of L2 acquisition: M = 17.6, SD = 1.3), British long-term residents in Italy (n = 27; age of L2 acquisition: M = 24.0, SD = 7.7; length of residence: M = 20.4, SD = 14.1), and English (n = 31) and Italian (n = 27) monolingual controls. Using self-paced reading tasks, we measured attachment preferences and reading times in both languages. In Italian ambiguous trials, controls showed a significant high-attachment preference while neither bilingual group showed a significant preference in the same direction. The model revealed a significant overall group difference driven by a contrast between Italian controls and students, though not specific to ambiguous trials. Both bilingual groups were also overall slower to parse RCs than Italian controls after controlling for proficiency or age differences. In English, all groups performed at chance level with no significant differences in preferences or reading times. These exploratory findings suggest that L1 influence on L2 RC attachment may persist even in proficient immersed bilinguals. No clear evidence of L1 attrition emerged, contrasting with previous Spanish–English findings and with anaphora resolution results in the same sample. Methodological limitations constrain interpretation and should be addressed in future research.