Processing Instruction (PI) (e.g., VanPatten, 1996) is a type of form-focused instruction that draws on the processing model of second language acquisition. It is designed to alter learnersŐ default processing strategies in order to direct them to attend to and make better form-meaning connections of the targeted form in the input they receive. The theory claims that PI, which is input-based, is more effective than an output-based instructional approach in enabling learners interpret and produce the targeted form. A series of PI studies conducted by VanPatten and his collaborators have been challenged by various experimental studies (e.g., Dekeyser and Sokalski, 2001; Allen, 2000). The discrepancy among the research results is due to the variation among such factors as the type of targeted linguistic form, the type of production assessment task, instructional variables, and adherence to the PI design. This prevents us from reaching a conclusion about the relative effectiveness of two practice modes: input in PI and output in an alternative output-based instruction.
The present study compares the effects of PI and namely output-based PI in enabling learners to interpret and produce two Japanese forms on four different assessment tasks. The study attempts to investigate how the effects differ depending on the degree of cognitive processing load in producing the target form, and the degree of monitoring required in a task. The tasks include interpretation, written sentence-completion, written sentence-level open-ended, and conversational task. All the variables but practice mode are identical across the two instructions.
The data is collected from the college-level second-year Japanese students. The students are divided into one control group and two experimental groups, and the latter group receives two hours of instructional treatment. The results of immediate posttests with four dependent variables will be described in this presentation and pedagogical implication will be suggested.