Use of L1 phonotactics in initial foreign-language speech segmentation

Von Holzen, K., Wulfert, S., & Hopp, H. (2023). Use of L1 phonotactics in initial foreign-language speech segmentation. Presented at the 2023 Conference Architectures and Mechanisms for Language Processing (AMLaP), San Sebastian, Spain. doi: 10.6084/m9.figshare.24041928Phonotactics provides sublexical cues to speech segmentation, but is language-specific (Weber & Cutler, 2006). We investigate the degree to which L1 phonotactics is applied to foreign-language English segmentation among 96 German 1st and 2nd elementary-school students before receiving instruction in English.In a probe recognition task, students listened to a sentence followed by an isolated probe word. Target pseudowords were embedded in English sentences, with the context (Boundary Type) providing a clear (e.g., lv; dal_vouchen) or ambiguous (kv; dack_vouchen) L1-German phonotactic word boundary cue. Probe words were either short (vouchen), or long (kvouchen/lvouchen) pseudoword candidates (Length). If children use L1 phonotactic cues, they should easily detect pseudowords in clear boundary contexts.An analysis of the 45 1st graders tested so far revealed that students detected target words above chance level (d’prime: t(44) = 7.77; p < .001, Figure 1) . Target accuracy was greatest for short target words with clear boundary cues (EMM = 0.40; Figure 2). This pattern demonstrates a tendency for young children to apply L1 phonotactic cues to segment foreign-language speech. In the full sample, we will also examine the role of individual differences in L1 knowledge and phonological awareness.