This study investigates compound verb processing in second language speakers (L2) of Persian. Forty-six near-native L2 speakers of Persian were tested to examine the processing of transparent (non-idiomatic) and opaque (idiomatic) compound verbs, under masked priming paradigm. The results revealed a significant nominal priming effect in the opaque condition, and a numerically stronger nominal priming effect in the transparent condition. There was also an increase in the processing load on the parser when the target was an opaque compound. The results of this study seem to be compatible with the dual access or dual route hypothesis, yet with the version that assumes the two routes are activated in parallel rather than the version that assumes high frequency words are represented lexically but low frequency words are decomposed.
