Abstract

Research in EFL has long highlighted the problematic nature of false friends (FF), which are words in different languages that are formally similar but semantically divergent and often lead to misunderstandings. However, studies on how such items are effectively learned remain limited, particularly in the context of English as learned by Italians. Although pedagogical translation and an explicit focus on form are commonly considered effective in addressing lexical areas prone to cross-linguistic interference, recent evidence suggests that variables related to formal instruction alone do not fully account for the acquisition of English–Italian FF, which are also largely absent from curricular materials. This raises the question of how learners encounter and master such items, and whether informal digital exposure plays a role. This study investigates the relationship between the level of knowledge of FF among a sample of young Italian university students, as assessed through a vocabulary test specifically designed for this purpose, and the students’ informal contact with English, as revealed by a questionnaire on their digital media use habits. Results show that media exposure – measured in terms of quantitative variables such as frequency and duration, and also including platform diversity, film viewing and subtitle use – does not emerge as a primary predictor of performance, while competence in this lexical set tends to correlate more significantly with overall English proficiency and general lexical competence. The disaggregation of results by grammatical category of target items also highlights uneven sensitivity across word classes, with adjectives (as opposed to nouns and verbs) showing the only statistically significant correlation. These findings point to the need for further research into qualitative dimensions of learning this hybrid and elusive lexical set, with a refined methodology and greater attention to individual differences and word-class effects.

This work is licensed under a Creative Commons Attribution 4.0 License.