View Content #22961

Bryan Smith is Associate Professor of English at Arizona State University's Tempe campus. His research focuses on computer-assisted language learning (CALL) - specifically, the intersection of CALL and second language acquisition. He is Editor (along with Mat Schulze) of the CALICO Journal, which is an international journal devoted to research and discussion on technology and language learning.

For this month’s theme, I’d like to outline the importance of input, interaction, and output as fundamental to language learning. Few theoretical perspectives on instructed second language acquisition (SLA) would contest the facilitative effect each of these has for second language development; however, theoretical perspectives will differ on their operationalization and relative importance of each. While there are theories that argue that SLA is essentially an input-driven phenomenon (Krashen’s Monitor Model and van Patten’s Input Processing framework) as well as those that ascribe a heightened importance for output (Swain’s Output Hypothesis), most of the recent pedagogically-relevant work on learner interaction has been from what is called the Interactionist Approach to SLA, with a growing interest in studying interaction from a more socio-cultural perspective.

The interaction approach (IA) has its origins in the late 1970s and early 1980s. The IA essentially says that through input from and interaction with other speakers, language learners have opportunities to notice differences between their own output and the formulations of the target language they hear from conversational partners. They also receive corrective feedback, which both modifies the linguistic input they receive and “pushes” them to modify their output during conversation. This interaction helps draw learners’ attention to linguistic problems when they occur and may also lead learners to notice linguistic input in the absence of a problem. Interactionally modified input, refers to modifications made as needed during goal-oriented conversation and is argued to be more effective than pre-modified input, which is modified to be target-like before any learner error occurs. An example of pre-modified input might appear in a reading passage that provides simplified or glossed language for vocabulary. Some of the most commonly investigated phenomena that inform these constructs include negotiated interaction, whereby learners first have a breakdown in understanding and then regain understanding through a series of communication moves, such as recasts, which are a type of corrective feedback and learner uptake, whereby a learner appropriates the target language forms of their interlocutor in some productive way. Such constructs have also been investigated in settings where technology is used to mediate the interactions such as during synchronous chat and texting as well as asynchronous interaction, such as during e-mail or discussion board posts. Some argue that text-based interaction may make aspects of the input more salient due to the slower rate of delivery and the permanence of the message, which remains on the screen, allowing learners to re-read the input. Keep in mind that the discussion above has been from a cognitive interactionist theoretical perspective. The nature and role of input, interaction, and output are viewed differently by researchers espousing other cognitive perspectives, such as information processing and emergentist – as well as more sociocultural or Vygotskyan perspectives.