An ERP investigation of masked cross-script translation priming.

The time course of cross-script translation priming and repetition priming was examined in two different scripts using a combination of the masked priming paradigm with the recording of event-related potentials (ERPs). Japanese-English bilinguals performed a semantic categorization task in their second language (L2) English and in their first language (L1) Japanese. Targets were preceded by a visually presented related (translation equivalent/repeated) or unrelated prime. The results showed that the amplitudes of the N250 and N400 ERP components were significantly modulated for L2-L2 repetition priming, L1-L2 translation priming, and L1-L1 repetition priming, but not for L2-L1 translation priming. There was also evidence for priming effects in an earlier 100-200 ms time window for L1-L1 repetition priming and L1-L2 translation priming. We argue that a change in script across primes and targets provides optimal conditions for prime word processing, hence generating very fast-acting translation priming effects when primes are in L1.