Dutch listeners ’ perception of Korean fortis, lenis, and aspirated stops: First exposure

This study addresses Dutch listeners ’ perception of Korean fortis, lenis, and aspirated stop triplets. The Korean stop triplets were expected to be extremely difficult for Dutch listeners, as all Korean stops are voiceless (in initial position), while Dutch only distinguishes voiced and voiceless stops. Dutch listeners were not familiar with Korean and received no training; they only heard six examples of each target sound before performing a phonetic categorization task. The Korean stops were lenis, fortis, and aspirated bilabial (/p/-/p*/-/ph/), denti-alveolar (/t/-/t*/-/th/), and velar (/k/-/k*/-/kh/) stops. Stimuli were blocked by type and place of articulation; thus one block concerned either a fortis-lenis, a lenis-aspirated, or a fortis-aspirated contrast, and either bilabial, denti-alveolar, or velar stops. Both Korean and Dutch listeners performed best on fortis-aspirated, intermediate on lenis-fortis, and poorest on aspirated-lenis contrasts. Not surprisingly, Dutch listeners performed less accurately than Korean listeners. Remarkably, however, the Dutch listeners performed significantly above chance level (if only just above it numerically) for each contrast type at each place of articulation. Thus, even at first exposure and without any training, Dutch listeners managed to identify the extremely difficult L2 sounds relatively successfully