Background to the research
Language education remains an active area of research, and inquiry on a global level, and numerous studies conducted by language educators are sensitive to many issues faced by learners, communities, and national education systems concerning language learning and education. It is believed that conversational interaction can facilitate inter-language development, but compared to laboratory settings realistic research to identify its efficacy in classroom settings is few. The interaction hypothesis, within the context of second language acquisition (SLA), suggests that receiving comprehensible input and interactional feedback and negotiating for meaning are helpful for second language (L2) learning. Literature on conversational interaction studies reveal that much of the empirical research has been carried out in laboratory settings, hence some researchers opine that patterns may differ if the research is done in L2 classroom settings The research by Gass et al (2005) is an attempt to compare interaction in classrooms and laboratories to provide insight into the nature of interactions in each setting.
It is assumed that interaction, when it involves negotiations for meaning and feedback, facilitates second language learning (SLA), which includes receiving comprehensible input and interactional feedback, and may provide opportunities for learners to test target language to identify the difference between their “interlanguage and target language” (Gass, 2005, p. 576). According to Pica (1994a), negotiation for meaning is an activity ‘in which learners seek clarification, confirmation, and repetition of L2 utterances they do not understand” (cited by Gass). The important benefit of interaction is its capacity for negotiation to connect input and output.
Research paradigm
Earlier research by Long (1981, 1983, 1985), Pica (1985, 1988, 1989, 1992), and others, in the domain of interaction research, was mainly investigations of various aspects of the role of negotiation in comprehensions, which advanced to interaction-comprehension links. Recent empirical studies by Ayoun (2001), Braidi (2002), Long et al (1998), and Philp (2003), to name a few, convincingly demonstrated a relationship between various types of interaction and L2 learning. These claims that there are benefits from conversational interaction by a number of researchers are supported with empirical data collected in laboratory settings. The research findings of Foster (1998), on the basis of classroom observation, prompts her that “learners appear to choose not to negotiate for meaning,” because the classroom environment is less controlled than that in the laboratory. In her data, dyads carrying out information exchange tasks were the most successful in producing negotiation for meaning, suggesting that it may be worthwhile to investigate task type further. In the opinion of Nicholas et al (2001, p.721 as cited by Gass) in many English as the second language (ESL) classrooms “response to learner utterances are even more likely to be interpreted as reactions to meaning.” Thus, the relevance of laboratory research to L2 classroom settings is important in interaction research, and the study by Gass is “a first step toward investigating the difference between classrooms and laboratories by examining the extent to which interaction is present in two highly comparable settings” (p.581). They compared task-based interactions, using materials that have been used as instructional tools and research instruments, in everyday classes and in a laboratory setting. In a controlled laboratory setting, learners worked in dyads to carry out tasks, whereas in a foreign language classroom, learners from the same population carried out the same tasks as in the laboratory and the results were compared. The research questions posited by Gass et al were (p.581-582):
- How does task-based interaction in the classroom compare to task-based interaction in a laboratory setting?
- How do different tasks influence interaction in classrooms and laboratories?
Sample and methodology
For their research, students enrolled in 3rd-semester university-level Spanish courses, in the age range from 17 to 25 were selected. There were 55 female and 19 male students, generally native speakers of English, who had been taking the Spanish class for 11 weeks which met 4 days a week for 50 minutes. The three tasks, selected from material appearing in popular textbooks that were typically used in the student’s language classrooms, were: one optional information exchange task (consensus); and two required information exchange tasks (picture difference and map). The tasks were adapted and translated by an experienced language instructor and the appropriateness of the task was checked by a second native-speaking Spanish instructor. The contextualized and communicative pictures of the picture difference task describe real-life scenes, and the participants were instructed to find 10 differences between their pictures. This allowed students to use and learn the authentic and meaningful language and value of which is well documented in the literature. The consensus task consisted of descriptions of several Spanish universities, in Spanish, requiring the participants to help prospective students to rank the universities and decide which school to attend. Under the Map task, the participants had to share their knowledge of the street conditions and cooperate with each other to successfully locate different street properties from the map, as well as to draw a driving route from starting point to a waiting friend. Considering the usefulness and interactive features of (a) negotiations for meaning, (b) language-related episodes, and (c) recasts, for analyzing interaction studies, Gass et al adopted these three coding methods in their data analysis. To compare the settings and tasks and to explore whether there were any significant interactions between the settings and tasks, repeated-measures analyses of variance were performed.
Research findings, suggestions, and conclusions
Repeated-measures analysis of variance revealed that there were neither significant interactions between setting and task nor significant differences between classroom and the laboratory on the total amount of negotiation and confirmation checks, except significant differences in the amount of the tasks for the total amount of negotiation. “Post hoc Bonferroni” analyses also revealed that there was more negotiation and confirmation on the picture differences task (p.588). In the summary of results, Gass et al suggest that “in terms of amount and type or negotiation for meaning Language Related Episodes (LREs), and recasts, whether learners interacted in classroom or laboratory settings had very little impact on the interactional patterns” (p.596-597). On examining the interactional features of the study sample it was found that there were fewer differences in the classroom and laboratory contexts. The incidences of LREs and recasts in the classroom interactions were highly comparable to those in the laboratory. They found some interesting differences in interactional processes according to the task, but not the setting. The claims of Foster (1998) that there is a lower incidence of negotiation of meaning in a classroom setting than in a laboratory setting was not supported by the data of Gass et al.
The research of Gass et al (2007) points to the need to carefully examine task types in any investigation of interaction and considering a number of variables when looking at negotiation in a classroom context. The primary variables are one-way versus the two-way flow of information, the role of the interlocutors, task familiarity, and gender. They conclude that when considering generalization researchers and teachers need to consider what makes sense in their context as findings will be different in different settings; “classrooms can vary tremendously and not all laboratories are equivalent either” (p.601). Thus, interaction may be more task-dependent, varying in each context from task to task. It is suggested that in future research, it might be interesting to consider gender differences and examine this in the light of individual initiative in interactional modifications, and length and linguistic target of interactional feedback, as well as other qualitative information.
Critical observations about the study
Though the research findings were aimed to reflect comparative relevance of classroom setting over laboratory setting, considering the wide range of classrooms that can be examined, the generalization of the findings is clearly limited. This study was carried out in a foreign, rather than an L2, classroom, which may affect the test results. The number of social and contextual features that a classroom has not been focused on in this study. The researchers have not taken into account the various complexities of the classroom context, such as the social relationships of the participants. As the evaluation was carried out in a typical foreign language classroom and a typical interaction laboratory experiment, the findings cannot be taken as representative of a comparable population.
It is pertinent to note that impediments to research may include access to the kinds of literature, the degree to which the quantitative and qualitative research methods are valued, the ways in which research methodology and reporting are taught, guidance in data collection and analysis, and other constraints posed by limited access to assistance or other physical resources. Choices of research topics and methodology may also reflect different value systems of priorities which may not be popular outside the researcher’s contexts. As such, original data-driven research in language education will influence the type of research questions, how these questions are investigated, and how the results and interpretations are reported or not reported. Research topics may include language acquisition in formal and informal settings, effects of innovative teaching methodologies on language learning, language testing, technology in language learning, task-based learning, influences on instructional changes, developing locally appropriate textbooks and educational materials, language teacher problems solving, and program evaluation. Hence, for the future of TESL, Gass et al suggest that “replication studies are obviously advisable in order to permit greater confidence in the results” (p. 601).
Works cited
Ayoun, D. (2001). The role of negative and positive feedback in the second language acquisition of the passe´ compose´ and imparfait. Modern Language Journal, 85, 226–243.
Braidi, S. M. (2002). Reexamining the role of recasts in native-speaker/nonnative-speaker interactions. Language Learning, 52, 1–42.
Foster, P. A Classroom perspective on the negotiation of meaning. Applied Linguistics, 19, 1-23. 1998.
Gass, Susan; Mackey, Alison and Ross-Feldman, Lauren. Task-based Interactions in Classroom and Laboratory Settings. Language Learning 55:4, 2005, pp.575-611.
Long, M. H. (1983). Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4, 126–141.
Long, M. H., Inagaki, S., & Ortega, L. (1998). The role of implicit negative feedback in SLA: Models and recasts in Japanese and Spanish. Modern Language Journal, 82, 357–371.
Nicholas, H., Lightbown, P. M., & Spada, N. (2001). Recasts as feedback to language learners. Language Learning, 51, 719–758.
Philp, J. (2003). Constraints on ‘‘noticing the gap’’: Nonnative speakers’ noticing of recasts in NS-NNS interaction. Studies in Second Language Acquisition, 25, 99–126.
Pica, T. (1994a). Questions from the language classroom: Research perspectives. TESOL Quarterly, 28, 49–79.