«Item type text; Electronic Dissertation Authors Goertler, Senta Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
In many cases it is impossible to identify whether something is an instance of error uptake or evidence for the students being at the same language development stage.
Explicit feedback is feedback that is marked overtly in some form. Explicit feedback has a tendency to interrupt the flow of the conversation.
Implicit feedback is feedback that is not marked overtly and usually does not interrupt the flow of the conversation.
Teacher move is any turn produced by the teacher in the chat transcript.
Turn is defined in this study as a posting by a participant. Before each turn, the name and the time of posting is indicated. A turn can be one word or symbol or several sentences long. Every time a person hits enter on the keyboard, what he or she has written in the text box is published to the chat window. This language is considered a turn.
Uptake in this study is considered attempted correct use of a structure or word following a correction within the same chat transcript.
1.7 Organization of the Dissertation After this short overview of the study and its context, chapter two contains more information about the theoretical framework and previous research. First, I outline the interactionist framework and its related theories. Then I discuss previous research on error correction. In the third part I summarize the research on SCMC. The last portion of chapter two provides a short summary and critique of the studies most relevant to this dissertation.
In chapter three I discuss my research questions and how they were examined. A detailed description of the methods, the context, the participants, the instruments, the data collection procedures, and the coding systems is presented in chapter three. The scoring of the pre- and post-test are also outlined.
In chapter four I present the results obtained from each instrument. This is followed by a detailed discussion and interpretation of the results in response to the before-mentioned research questions. In the last part of this chapter, I discuss additional findings not related to the research questions.
In chapter five I summarize the most important findings from chapter 4 and offer some implications for the field of SLA, language teaching, language teaching and technology, teacher training, and program administration. Finally, the limitations of the study are discussed and recommendations for improvement for further studies are given.
CHAPTER II: THEORETICAL FOUNDATION AND LITERATURE REVIEW
2.1 Introduction This dissertation operates within a sociocultural, sociolinguistic view of language learning, specifically the interactionist framework, and within the larger contexts of research on computer-mediated communication (CMC) and corrective feedback. This chapter will address the theoretical foundation for this study, the interactionist framework, and will summarize the relevant research on corrective feedback and CMC.
2.2 Theoretical Foundations of the Interactionist Framework on SLA Second language acquisition (SLA) research is an interdisciplinary field drawing from several areas of interest through research findings and theoretical perspectives.
These fields include, but are not limited to: sociolinguistics, psycholinguistics, neurolinguistics, cognitive science, psychology, anthropology, first language acquisition, theoretical, and functional linguistics. The interactionist framework, the foundation for this dissertation, is a sociolinguistic perspective on second language acquisition.
An interactionist framework is a sociolinguistic, sociocultural perspective on second language acquisition which assumes that elements of interaction are essential for second language learning, and encompasses related language learning theories. The
interaction hypothesis as formulated by Long (1996) states:
… negotiation of meaning, and especially negotiation work that triggers interactional adjustments by the NS or more competent interlocutor, facilitates acquisition because it connects input, internal learner capacities, particularly selective attention and output in productive ways (pp. 451This framework considers all aspects of interaction, such as input, the processing of input, output, and interaction, as crucial for second language learning. On one hand, it focuses on learners and what they do with the input and feedback received and the output produced. On the other hand, it addresses the modifications made and feedback given by the interlocutor, especially a native speaker (NS) or more expert non-native speaker (NNS). Hence, the framework is based on social interaction as a central part of learning, and can be seen in the larger framework of learning set forth by Vygotsky (see for example Lantolf, 2000).
2.2.1 Input Input is the language a learner is exposed to and is the central part of language development in Krashen’s (1985) second language learning theory. Accordingly, input must be comprehensible, and is sufficient for language learning. In the past two decades, these theoretical assumptions have been modified by others such as Gass (1997), who believes that not all input leads to language development, but rather that input is processed through several steps which can lead to further language development. Gass’ (1997) model consists of five stages: apperceived input, comprehended input, intake, integration, and output. Here, input is defined as the language a learner is exposed to, while frequency, affect, attention, and prior knowledge influence the learner’s ability in processing the input. “….Apperceived input is that bit of language that is noticed in some way by the learner because of some particular recognizable feature” (p. 4). What is then comprehended by the learner, either immediately or after negotiation or NS modification, is referred to as the comprehended input. The comprehended input interacts with language universals as well as prior linguistic knowledge for further processing, leading to intake. Intake is the process of hypothesis formation and testing which can result in hypothesis rejection, modification, or confirmation. During the integration process, this hypothesis is stored and either immediately or later integrated into the second language (L2) grammar. During output, the learner has the chance to test hypotheses, practice, and receive feedback. In this sense, output interacts with intake as well as the process of moving apperceived input to comprehended input. Due to the cognitive processes involved, Gass argues that effects of interaction may be delayed rather than immediate.
This poses a challenge for research design as the effects of a particular input segment may be impossible to measure directly.
2.2.2 Output Output, as discussed in relation to Gass’ (1997) model, is based on Swain’s (1985) claim that comprehensible input alone is insufficient for language learning, which she confirmed through her research of immersion students. During input processing, the learner can rely on contextual clues and semantic processing. When being pushed to output, however, a learner must utilize syntactic processing. Output in general plays several roles: (1) creating knowledge from semantic to syntactic processing as described by Swain (1985); (2) practicing or applying existing knowledge; (3) creating automaticity; (4) eliciting further input; and (5) testing hypotheses formed and receiving feedback about them in regard to the target language (Gass & Selinker, 2001). This important role of output in SLA has also been confirmed by Swain and Lapkin (1995) and Izumi, Bigelow, Fujimara, and Fearnow (1999).
However, a major concern with research on output is whether it is really the output that attributed to the language learning or the feedback received as a reaction to the output. It is conceivable that the opportunity to recall and reevaluate one’s hypotheses about the target language in the struggle to produce output is sufficient, or at least that such repeated struggle can lead to greater automaticity.
2.2.3 Noticing Noticing, or attention, is the cognitive process necessary in the learner to identify mismatches between the learner’s own hypotheses about and the actual rules of the target language. This conscious awareness of such differences is important, if not necessary, for second language learning to occur (Schmidt, 1990). In order for input to be processed to intake, i.e., to be incorporated into interlanguage, the learner must be aware of the gap between his or her interlanguage rules and the target language rules. Noticing then is similar to what Gass (1997) refers to as apperception. If noticing is necessary for intake, then an attention to form is important for SLA, which goes against the arguments made by Krashen. The difference between Schmidt and Gass is that, in Schmidt’s model, input automatically becomes intake if it is noticed, whereas Gass argues that input that is noticed, or apperceived, does not necessarily become comprehended input. Regardless of the particular interpretation, theorists agree that attention and awareness in some form is important for language learning. However, the concern in conducting research is to find measures to identify whether an element of language was actually noticed. In general, such a theory appears to operate in a circular argumentation: if something is learned, it was noticed, and it was noticed if it was learned. Through such argumentation, it is difficult for empirical research to prove a hypothesis.
2.2.4 Negotiation of Meaning As discussed, both output and input afford opportunities for negotiation. In interactions with another NNS or with a NS, a learner will encounter many moments of misunderstanding. During those moments the opportunity for the negotiation of meaning arises. Negotiation of meaning is the “attempt made in conversation to clarify a lack of understanding” (Gass & Selinker, 2001, p. 457). During this step, learners can receive feedback on the hypotheses they have formed about the target language, and through the negotiation create a target-like utterance of their intended meaning. “Negotiation of the sort that takes place in conversation is a means to focus a learner’s attention on just those areas of language that do not match those of the language being learned” (Gass & Selinker, 2001, p. 291). Negotiation, then, makes it possible for a conversation to continue after misunderstanding.
Negotiation consists of a trigger and a resolution. The trigger is information that is not understood, while resolution is the process of achieving understanding. Resolution is begun by the participant who has trouble understanding, and who indicates such a struggle through confirmation checks or clarification requests. The conversation partner must respond to the indicator by clarifying the misunderstood output. The negotiation of meaning is completed when the first interlocutor responds to the second interlocutor, indicating understanding (Gass & Varonis, 1985). Pica, Holliday, Lewis, and Morgenthaler (1989) refer to the trigger and resolution as the trigger and signal.
Long, in his dissertation (as cited for example in Ko, Schallert, & Walters, 2003), emphasized the conversational modifications, later referred to as negotiation of meaning, when a NS and NNS talk. Such modifications include: slower speed of talking, clearer articulation, longer pauses, less slang and idioms, more high frequency words, fewer pronouns, gestures, pictures, additional semantic information, short simple sentences, movement of topic to the front of the sentence, repetition, restatements, new information at the end of a sentence, recasts, comprehension checks, and clarification requests (summarized and adapted from Gass & Selinker, 2001, and Ko, Schallert, & Walters, 2003). The modified input used when speaking with NNSs has been referred to as foreigner talk, similar to “motherese” for first language acquisition. Conversational modifications have also been referred to as negotiation (Pica, 1994), negotiation of meaning (Nakahama, Tyler & van Lier, 2001), and negotiated interaction (Mackey & Philip, 1998). More importantly, several researchers have claimed that the negotiation of meaning assists SLA (Gass & Varonis, 1985; Pica, 1994; Pica & Doughty, 1985;
Scarcella & Higa, 1981; Varonis & Gass, 1985).
Long (1996) considers implicit negative feedback received during the negotiation of meaning to be a facilitating factor in SLA. It has been argued that since there exists incomprehensible input (White, 1991) during the negotiation of meaning, learners must focus their attention on the language forms being used during the negotiation, in turn leading to a noticing of the mismatch between interlanguage and target language grammar. It has been argued that this noticing facilitates second language learning (Gass, 1997; Long, 1996; Pica, 1994). However, some researchers suggest that there is little explicit feedback in conversations (Chun, Day, Chenoweth, & Luppescu, 1982), and that most errors do not receive corrections by the NS (Day, Chenoweth, Chun, & Luppescu, 1984: 7.3% corrected).
Long (1983) and Pica and Doughty (1985) have identified five strategies in the negotiation of meaning: clarification requests, confirmation checks, comprehension checks, self-repetition, and other-repetition. Clarification requests are made by the listener to clarify spoken information. Confirmation checks are also made by the listener, but to indicate understanding. Comprehension checks are made by the speaker to check understanding by the listener. Self- and other-repetition are both made by the speaker, to partially, exactly, or in an elaborating manner, repeat an utterance from within the last five speaking turns. A self-repetition is a repetition of a speaker’s own turns, while otherrepetition is of another’s turn.
Just as there are different ways of responding to misunderstanding, there are also different triggers for misunderstanding: lexical (verb phrases and noun phrases, word choice), morphosyntactic (verb inflections, partitives, and plural morphemes), global (based on discourse, content, or both) (Nakahama, Tyler, & van Lier, 2001), and pronunciation. Considering these starting points of the negotiation, Chun, Day, Chenoweth, and Luppescu (1982) found that (a) corrective feedback in natural conversation, in contrast to game-like tasks, is avoided, but that (b) the types of triggers in the two tasks were no different.