Native language acts as a filter in speech perception for L2 learners, so that the sounds of L2 are typically perceived in terms of the categories of the L1. However, learners may adapt their perceptual system so as to be better tuned to the contrasts that are functional in the new language. The aim of our research is to better understand (the extent of) the ability for adult L2 learners to adapt their speech perception to a new language. The present research focuses on adaptation not to the native norm for the target language but to a non-native type of pronunciation which is neither the L1 of the learner nor that of the target language. In the experiment we followed ten adult Chinese learners of English (aged 24-31 at arrival) who studied in the Netherlands for more than two years. The language of instruction, and in much of everyday communication, was English. The Dutch pride themselves of being excellent speakers of English and do not encourage foreigners to speak Dutch to them. Inevitably, therefore, foreigners communicate with Dutch nationals in non-native English (disparagingly called Dunglish by native speakers of English). We wished to determine to what extent the Chinese learners of English would become better tuned to Dutch-accented English, and to what extent their perception of (American) native English might still benefit from exposure to Dutch-accented English. The first part of our experiment was done shortly after the students' arrival in the Netherlands; the second part was done after two years' exposure to Dutch accented-English. In both sections we ran phoneme identification tests using the same materials (19 vowels in /hVd/ contexts and 24 consonants in /aCa/ contexts) produced by Chinese, Dutch and American speakers of English. We test the hypothesis that our Chinese learners will identify the English vowels and consonants more successfully in the re-run of the experiment, but only for the Dutchaccented variety of English. The perception of the Chinese-accented variety of English does not benefit from prolonged exposure to Dutch English, nor do the results generalize to (American) native English. Specifically, our results show that the confusion structure in the vowel and consonant identification tests converges toward the pattern we found for Dutch listeners having to identify vowels and consonants in Dutch-accented English.