Individuals with different L1s rely differently on suprasegmental cues when recognizing spoken words in L1 and in L2. The present study investigates the use of stress information in L2 word recognition by intermediate-level learners of Spanish with different L1s: German (i.e., a language with distinctive word stress; DE), French and Korean (i.e., languages without distinctive word stress; FR, KR). Contrary to DE, the absence of word stress in French and Korean was expected to hinder FR and KR ability to use stress in L2 Spanish word recognition. In a cross-modal word-identification task, participants listened to semantically ambiguous auditory sentences ending with incomplete two-syllable word fragments and had to choose the word that matched the heard fragment, which either contained (stressed condition) or lacked a stressed syllable (unstressed condition). While the results revealed no difference between three language groups in the unstressed condition (62%), the groups differed in the stressed condition. KR accuracy (77%) was unexpectedly higher than FR (63%) and as good as DE (76%), suggesting that, contrary to FR, KR were able to use stress to access L2 words, although word stress does not exist in their L1.
