This study investigated the effects of input-based and output-based practice on the development of accuracy and speed in recognizing and producing request-making forms in L2 Chinese. Fifty American learners of Chinese with intermediate level proficiency were randomly assigned to an input-based training group, an output-based training group, or a control group. The input and output groups practiced the target forms over four consecutive days. The control group did not practice the forms. The effects of practice were measured by a Listening Judgment Test (LJT) and an Oral Discourse Completion Test (ODCT). The results showed that the effects of input-based and output-based practice were shared across task modalities on measures of performance accuracy (i.e., accuracy in the LJT and ODCT) but not on measures of performance speed (i.e., LJT response times, ODCT planning times, and speech rates).Keywords: interlanguage pragmatics; L2 instruction; Chinese; requests; speech acts; input-based and output-based practice AN IMPORTANT TOPIC IN INSTRUCTIONAL ILP (interlanguage pragmatics) research is the role of various types of instruction in enhancing L2 pragmatic development. Research in this area has centered on comparing the effects of explicit and implicit instructional conditions (hereafter explicit and implicit conditions). However, findings have been inconsistent: Although some studies demonstrated an advantage of explicit over implicit conditions, others showed that both were equally effective or ineffective (see Takahashi, 2010aTakahashi, , 2010b, for recent reviews). A review of the literature indicates that modality of instruction (i.e., input-based vs. output-based) may influence the effects of explicit and implicit conditions, yet such modality effects have not been empirically examined. A closely related but under-researched topic is how to assess instructional effects. Researchers have focused on gains in performance accuracy, which is an indicator of underlying pragmatic knowledge. However, gains in performance speed, which can serve as an indicator of the ability to process pragmatic knowledge (i.e., processing ability), have largely been left unexamined. Because knowledge and processing are key components of pragmatic competence (Kasper, 2001;Taguchi, 2012), both should be examined to better understand instructed L2 pragmatic development. This study is an effort to address the above issues by examining the effects of input-based and output-based practices on the development of accuracy and speed in request-making in L2 Chinese.
BACKGROUND