Automated writing evaluation: a posthuman perspective on the development of writing in EFL

• Ano de defesa: 2019
• Autor(a) principal: Nunes, Gisele Medina
• Orientador(a): Leffa, Vilson José
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Pelotas
• Programa de Pós-Graduação: Programa de Pós-Graduação em Letras
• Departamento: Centro de Letras e Comunicação
• País: Brasil
• Palavras-chave em Português: Letras; Automated feedback; Foreign language learning; Written communication strategies; Educational software; Posthuman language learning; Feedback automatizado; Aprendizagem de línguas estrangeiras; Estratégias de comunicação escrita; Aplicativo educacional; Aprendizagem de línguas pós-humana
• Área do conhecimento CNPq: CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA
• Link de acesso: http://guaiaca.ufpel.edu.br/handle/prefix/6412

Digital technology has been permeating language learning for nearly three decades. The English Language, in special, has spread worldwide and the attempts to foster its learning through educational software has increased substantially due to the digital information and communication technologies (DICTs). Online applications with automated feedback bring a new perspective in foreign language learning and teaching practices since a different type of dynamics takes place, the human-machine interaction. Learning via digital technologies has been engendering more autonomous practices, in which learners develop language skills assisted by an automated tutoring system. Taking this scenario as a consolidating reality, it is imperative to analyze how this human-machine dynamics in learning takes place, observing how automatic feedback from an automated writing assessor triggers learners’ communicative strategies and thus supports more effective learning. Therefore, this study focus on investigating the free automated writing evaluator called Write and Improve which assists learners of EFL to develop their writing skills in an independent way as well as the learner-program-teacher triad of interaction. Under a complex, constructivist as well as post-human perspective of language learning, this doctoral dissertation aims at analyzing the application using qualitative methods. The data were collected from texts produced by three subjects who used the online tool to develop their writing skill in beginner and intermediate proficiency levels. Field notes also complemented data analysis, as two of the subjects were part of one of the researcher’s class by the time of the data collection. The data showed automated feedback triggers some communication strategies and there was improvement in learners’ writing proficiency after a period of three months using the tool. We conclude the automated writing evaluator analyzed in this study can integrate a hybrid instructional environment, where human and automated feedback can intertwine and excel at offering favorable conditions for EFL writing skills’ development. From a posthuman perspective, artificial intelligence and human subjectivity are not to be thought as opposing entities but rather as complementary ones, a merge that might contribute to delineate a new pedagogical paradigm for technology-mediated language learning.