O feedback implícito-negativo na aquisição de segunda língua: modelos e recasts na aquisição da língua inglesa por aprendizes brasileiros

• Ano de defesa: 2011
• Autor(a) principal: Flavia Cristina de Azeredo Silva
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Foco de Atenção; Memória de Trabalho; Noticing; Modelo; Interação; Recast; Língua inglesa Erros; Língua inglesa Estudo e ensino Falantes de português Brasil; Aquisição da segunda linguagem; Cognição; Língua inglesa Métodos de ensino Falantes estrangeiros
• Link de acesso: http://hdl.handle.net/1843/DAJR-8MLJLP

The role of implicit negative feedback in the form of recasts has been extensively investigated in the SLA field; however, the benefit to learners from this form of feedback is still an issue of debate. This experimental research aims to examine the role of focus attention, noticing and working memory capacity in the effectiveness of two corrective feedbacks, recasts and models on Brazilian learners of English acquiring two language structures - adjective word order and indirect (embedded) question formation. The present research followed the pre-test, pos-test, delayed pos-test, and control design. Thirty-three students were randomly selected from a local language course in Brazil to perform two communicative activities in three different sessions. During the activities, the participants received corrective feedback or no feedback according to the experimental condition they were assigned. Before starting the study, the participants were pre-tested by the means of three versions of two instruments: a grammaticality judgment test (GJ) and an oral picture description; they were pos-tested immediately after the last day of treatment; five days later, the participants were administered a delayed pos-test. All of the instruments and the conditions were counterbalanced. Along with the language activities, the participants also performedattention control tests, working memory tests and a stimulated recall test.The data analyses suggest that recasts and models were more effective than no feedback. Both forms were, as well, more effective for adjective word order than no corrective feedback - a tendency that is evidencedin both GJ test and oral picture description. Data of the indirect questions from the GJ and from the oral picture description task showed that participants who were in the recast condition had greater improvementthan those who were in the model and control conditions. As for the cognitive factors, the results showed a complex scenario. Overall, the data give partial support to the research hypotheses. In general, attentionseems to be the most relevant component of the cognitive variables contemplated on this research.