According to the Perceptual Assimilation Model (PAM) by Best (1995) and Best & Tyler (2007), learners of L2 phonology will assimilate two non-native sound categories to one native sound category if there is only one native sound category close to two non-native categories. It follows that L2 speakers will encounter difficulties in perceiving and producing the difference between sounds in a second language (L2) that are not contrastive in their native language (L1). As a result, L2 speakers exhibit indeterminacy between L2 words that differ by a nonnative contrast. Previous studies have shown that native Japanese speakers tend not to resolve the difference between English words such as rocket and locker until the second half of the word is heard. Similar effects have been observed in native Dutch speakers using auditory lexical decision task. These speakers, when hearing an English word including /?/ or /æ/ (e.g., cattle), are faster in responding to its minimal-pairs counterpart (e.g., kettle).
In order to investigate whether such cross-lexical effects exist in visual word recognition, we devised a semantic relatedness decision task that builds on findings from visual word recognition research. 35 Chinese-speaking English learners, all pursuing their MAs in English, participated in this experiment. In order to provide a baseline for comparison, 35 English native speakers were also recruited. The critical stimuli were constructed from 20 homophone pairs (e.g., hear-here) and 20 /i-I/ minimal pairs (e.g., heal-hill). A minimally different spelling control was coupled to each pair (e.g., real or bill for heal-hill), with the constraints that the control differed in only a single grapheme from either member of the pair.
Results indicated that similar to native English speakers, Chinese English learners produced more errors and responded more slowly to the homophone pairs in the critical condition than in the corresponding spelling control condition. This provided direct evidence that viewing a visual word automatically activates its phonological representation, which in turn activates its homophone and causes the semantic interference.
More importantly, results showed that Chinese English learners produced more errors and responded more slowly to the /i-I/ items in the critical condition than in their spelling control condition. No such effect was observed for the English native speakers. This lent support to the conclusion that cross-linguistic phonological interference occurs in L2 visual word recognition (L2 word reading). It can be seen that the transfer of L1 phonology can occur not only in the perception and articulation of L2 sounds, but also in the phonological coding of L2 lexical entries.