Rendaku (Japanese sequential voicing) as rule application: An ERP study

Japanese sequential voicing (rendaku) is a process of voicing the initial obstruent of the second member of a compound word in Japanese (e.g. hon 'book'+tana 'shelf'→hon-dana). We conducted an event-related potential measurement experiment to investigate whether rendaku is a regular process of rule application or an analogical process based on lexical memory. When rendaku was applied wrongly to words lexically specified to resist rendaku, a left anterior negativity component, followed by a P600 was observed, whereas applying rendaku against a phonological constraint known as Lyman's law elicited a P600 component alone. Failure of rendaku application where it should apply yielded an N400. These results suggest that rendaku is a process involving rule application. NeuroReport 25:1296-1301