Encoding second-language (L2) phonological contrasts into lexical representations is known to be challenging above and beyond the perceptual difficulties that these contrasts may entail. In two experiments, this study assessed the effect of form-focused training during word learning on the lexical encoding of the English /ɛ/-/æ/ contrast into novel L2 minimal pairs (e.g., tendek-tandek) by German learners of English. More specifically, we investigated whether the point in time in which form-focused training is administered (i.e., very first presentation vs. after one training session) determines learners’ success at distinguishing the two vowels in the novel words. In Experiment 1, only native English tokens were presented whereas, in Experiment 2, productions by a fellow German-accented learner of English were also included. Results revealed an early benefit of phonologically-focused training on lexical encoding and novel word recognition (Experiment 1) that nonetheless appeared to be constrained by the demands of the task and the properties of the input presented (Experiment 2). Most importantly, however, learners’ ultimate word recognition performance provided evidence that early lexical representations can rapidly be updated to reflect improvements in phonological knowledge.

众所周知，将第二语言 (L2) 语音对比编码为词汇表示具有挑战性，超出了这些对比可能带来的感知困难。在两个实验中，本研究评估了在单词学习过程中以形式为中心的训练对英语 /ɛ/-/æ/ 对比的词汇编码的影响到新的 L2 最小对（例如，tendek - tandek) 由德国英语学习者提供。更具体地说，我们调查了实施以形式为中心的培训的时间点（即第一次演讲与一次培训课程之后）是否决定了学习者区分新词中两个元音的成功率。在实验 1 中，只呈现了母语英语标记，而在实验 2 中，还包括了一位德语口音英语学习者的作品。结果揭示了在词汇编码和新词识别（实验 1）方面以语音为重点的训练的早期好处，但似乎受到任务需求和所呈现输入属性的限制（实验 2）。然而最重要的是，