Locus of morphological processing during visual L2 compound recognition: An ERP investigation

The extent to which second language (L2) speakers process morphologically complex words through morphem-based processes remains debated, with conflicting evidence from behavioral studies suggesting that L2 learners may rely primarily on orthographic overlap rather than morphological structure. The present study employed morpho-semantic priming paradigm and event-related potentials (ERPs) to investigate whether and how Chinese-English bilinguals engage in morphological decomposition and morpho-semantic integration during real-time English compound word recognition. We manipulated two factors: Relatedness (primes semantically related vs. unrelated to the first constituent morpheme of the compound word) and Semantic Transparency (transparent vs. opaque compound words). In the N400 time window, a significant main effect of Relatedness emerged, with both transparent and opaque compound words showing comparable N400 attenuation, indicating early semantically blind morphological decomposition. In the late negativity window (800–1200 ms), a significant main effect of Semantic Transparency emerged, with opaque compounds eliciting greater negativity than transparent compounds, reflecting the late morpho-semantic integration process. These findings demonstrate a biphasic pattern of L2 morphological processing encompassing early obligatory morphological decomposition followed by late morpho-semantic integration. Overall, our findings suggest that Chinese-English bilinguals can engage in native-like morphological processing during recognition of English compound words.