Subject–verb agreement by Turkish speakers of English: Differential cue-weighting in L1 and L2 and distinct mechanisms for comprehension and production

Employing a self-paced reading experiment, a read-aloud production task and a pen-and-paper questionnaire, we examined if Turkish (L2) and native (L1) speakers of English rely on similar or different mechanisms in comprehending and producing English subject–verb number agreement. The results showed that the syntactic distance of an attractor to the head noun affected the L1 speakers’ parsing decisions, while the L2 speakers were influenced by its linear proximity to the verb. The results also showed singular attraction in comprehension for both groups and (a tendency for) plural attraction in production. We argue that L1 and L2 speakers differ in the sources of attraction and in their cue weighting/encoding while processing and producing S-V agreement. Whereas syntactic proximity is more crucial in the L1 speakers’ agreement computation, L2 speakers weigh linear distance cue more heavily. We also argue that S-V agreement computation is governed through different mechanisms in comprehension and production.