Non-native speakers' production of modified comprehensible output and second language learning

This study investigated the ability of NNSs to modify their output toward comprehensibility in the contexts of NS-NNS and NNS-NNS interactions and the degree to which such modified comprehensible output (MCO) was other-or self-initiated. Picture-dictation and opinion-exchange tasks were used to collect data from 8 NSs and 24 NNSs of English representing 13 different L1 backgrounds. The 2 tasks were performed in pairs (NS-NNS and NNS-NNS) and were audiotaped. The results showed that most repairs were self-initiated and that NNS-NNS interactions produced more other-initiations and other-initiated MCOs on the picture-dictation task. The frequencies of these MCOs support the importance of modification toward comprehensible output as a process of second language acquisition.