The processing and production of L2 tense-aspect by Chinese and Arabic learners of English

The acquisition of L2 morphosyntax is a challenge for L2 learners both in comprehension and production, particularly when grammatical structures differ between a learner’s L1 and the target language (DeKeyser, 2005). Studies on L2 acquisition of grammar have found varying performance depending on modality and also on what type of knowledge is being tapped into. For instance, learners may display metalinguistic/explicit knowledge but are less able to apply it in real-time comprehension (Roberts & Liszka, 2013). Some learners might be sensitive to grammatical distinctions during on-line comprehension, but persistent difficulties are found in their production (Trenkic et al., 2014). This study investigates the explicit knowledge (comprehension: AJT; production: gap-filling), online comprehension (visual world eye-tracking) and production (sentence imitation), focusing on tense-aspect in L2 English, of upper-intermediate L1 Chinese (24) and Arabic (24) learners, compared with 24 native English controls.

Like in the other studies, all the participants patterned together in the judgment task, demonstrating similar explicit knowledge of the present progressive, present perfect and past simple items. The results of the eye-tracking task from cluster-based permutation analyses found that both L2 groups predicted the correct construction (looked to the pictures containing ongoing events when hearing present progressive sentences, and to the pictures depicting completed events hearing present perfect sentences, e.g., The girl is planting/has planted flowers in the garden), although this prediction emerged later than the English group. Cross-linguistic influences were suggested: the L1 Chinese group failed to show prediction when hearing simple past sentences, and their divergency points of looking at the target picture appeared significantly later than the Arabic group (95% CI = [-202, -750]ms, p =.003). In contrast, L1 Chinese showed predication 100ms earlier than L1 Arabic group upon hearing present perfect sentences, but the difference was not significant.

Similar to the above, all participants scored highly in the gap-filling task (Chinese: 88%; Arabic: 86%; English: 92%), in contrast to the task tapping more implicit processes. The results of the online production task (elicited imitation) found that both L2 groups performed less well in comparison to the AJT and the comprehension task. The two learner groups performed similarly numerically (Chinese - corrected 34%; Arabic - 37%), and much worse than the controls (81%). The Chinese group tended to omit verb inflections and the Arabic learners mainly used simple past structures in present perfect contexts, which could be attributed to L1 influence.

The initial findings provided positive evidence for successful L2 predictive processing in tense-aspect and the potential influence of L1. It also suggested that while learners may be able to utilise morphosyntactic information during comprehension, their production knowledge often falls behind.

References:
DeKeyser, R. M. (2005). What makes learning second-language grammar difficult?. Language Learning, 55, 1–25.
Roberts, L., & Liszka, S. A. (2013). Processing tense/aspect-agreement violations on-line in the second language. Second Language Research, 29(4), 413–439.
Trenkic, D., Mirkovic, J., & Altmann, G. T. M. (2014). Real-time grammar processing by native and non-native speakers. Bilingualism, 17(2), 237–257.