An Experimental Study On The Effects Of Structured Input Activities And Text Reconstruction On Grammar Learning

For the past two decades in SLA field, the role of explicit form-focused instruction (FFI) has been recognized, but it remains controversial as how to implement FFI, especially about the relative effect of input-based and output-based activities. Based on his input processing theory, VanPatten (1993) proposed Processing Instruction, highlighting the role of structured input in facilitating form-meaning mapping so as to promote language learning. On the other hand, Merrill Swain (1985) stressed the role of output in her Output Hypothesis, viewing output not just as L2 learning result, but as the process that promotes learning. Over the years, many empirical studies abroad have been conducted about the relative value of input and output-based instruction for L2 learning, but there has been few in China; moreover, the research results were inconsistent, partly due to inadequate control of research conditions.This study attempts to further explore the issue in Chinese EFL context. By using English past counterfactual conditional (PCC) as the target grammar and adopting structured input and text reconstruction to embody input-based and output-based instruction respectively, and by strictly controlling the duration and difficulty of the two activities, the present study intends to re-examine the relative effectiveness of input-based and output-base activities on grammar learning. The specific purpose is to compare the effects of the two types of activities on 1) the students’overall learning of PCC; 2) their learning in comprehension and production of PCC respectively.Two intact classes of Senior Grade one from a high school in Nanjing were selected as the experimental groups (Input group and Output group respectively). Both having received explicit presentation of the target item, the input group was treated with the structured input activities, while the output group was treated with the text reconstruction work. A pretest and posttest were used, including both comprehension tasks and production tasks, in which participants’ ability to comprehend and produce the target feature was assessed respectively. The data collected were then subjected to independent sample-t test and paired sample-t test using SPSS. The results are as follows:1) Both the structured input activities and output-based text reconstruction task were equally effective in facilitating students’ overall learning of PCC; no significant difference was found.2) Both types of activity were effective in promoting students’ learning of PCC at the comprehension level, but compared with output-based text reconstruction, the structured input activities achieved better effects on comprehension and the difference reached the significance level.3) Both types of activity were effective in helping the students’ learning of PCC at the production level, and they achieved similar effects on students’ learning of past counterfactual conditional.The results prove that both input and output activities which are pedagogically optimized are effective in promoting acquisition of L2 formal features, confirming the validity of both VanPatten’s and Swain’s theories. The results also show that SIA and TR are feasible and effective pedagogical techniques for grammar teaching and should be used in combination and with flexibility according to actual needs.