Abstract
The rise of artificial intelligence (AI) in language education has increased the collaboration between humans and machines (e.g., automated technologies, AI-powered digital tools, robots, etc.), leading to more effective, efficient, inclusive, and sustainable language learning, teaching and assessment. In November 2022 when Open AI's ChatGPT 3.5 was released, human–machine collaboration (HMC) gained more attention. HMC is a concept and practice where humans and machines are treated equally, complementing each other's unique strengths to complete a task (Meniado, 2023a). It is a synergy between humans and intelligent machines, with the latter performing mundane and repetitive tasks allowing humans to handle the more complex tasks. In the current literature, HMC can be associated with collaborative intelligence, collective intelligence and augmented intelligence. Within the contexts of Digital Language Teaching 4.0 and 5.0, HMC places humans at the centre of innovation and creativity (Meniado, 2023a). It facilitates the development of human skills and machine operating abilities and allows humans and machines to increase their strengths and compensate for their flaws. Moreover, it also lessens human errors and improves work effectiveness and efficiency. Lastly, it allows humans and machines to perform jobs/tasks that they are unable to do on their own.
Human–Machine Collaboration in Language Education
In language education, there is a great potential for human–machine (human–AI) collaboration to support and enhance language teaching, learning and assessment. Language teachers and AI can work together to plan lessons, develop instructional materials, facilitate individualized instruction, assess learning and give immediate feedback (Meniado, 2023b). Similarly, second language (L2) learners can also collaborate with AI to comprehend texts, understand unfamiliar words, develop meaning-focused outputs, identify and rectify their language errors and practise using the target language to enhance their fluency development (Meniado, 2023b). In my recent study exploring how English language teachers and learners in the Southeast Asian region collaborate with generative AI (GAI) in teaching and learning L2 writing, I found that teachers used GAI to produce sample texts as models (samples that are easy to comprehend) and then let their students analyse the samples regarding content, language and structure using a checklist. I also found that students independently used GAI to paraphrase, summarize and synthesize the information they gathered from different sources. They also used GAI to evaluate their paragraph/essay outlines (or suggest improvements) based on a given set of criteria and improve them (by pair or by group) based on GAI's feedback. While writing, students also used GAI to translate their ideas/sentences from first language to English and to get real-time feedback and support to make the language of their written work more accurate and its tone more academic. After writing, students used GAI to self-assess their outputs before submitting them to their teachers, and teachers utilized GAI to provide detailed and prompt feedback to their students.
There are many evolving patterns of human–machine interactions and collaborations in English language teaching, learning and assessment. Currently, we have the human-in-the-loop (HITL) and machine-in-the-loop (MITL) systems. HITL is a baseline ethical AI collaboration workflow that avoids offloading the entire workload to (generative) AI tools, while MITL is a system where a human collaborator retains a majority of the workload, an ideal AI collaborative language teaching/learning model (Knowles, 2024). These dynamics will change as AI technologies become more sophisticated and as humans become more mature in exploiting these technologies. But no matter how complex the partnership and synergy may be, it is certain that the harmonious and seamless collaboration between language teachers/learners and AI technologies can cultivate a successful learning environment and lead to better learning outcomes (Chen, 2024; Warschauer and Xu, 2024). Given the important role of HMC in the current and future language education landscapes, it must be recognized as a legitimate concept and practice in the language education systems, should be given more time and space in the macro-level language curriculum and should be integrated into the language teacher education and professional development (PD) programmes as a set of skills or competencies to create a supportive ecosystem where HMC is widely and sustainably practised.
HMC Practices in This RELC Journal Issue
Articles included in this issue present innovative ideas and practices addressing critical issues in applied linguistics and Teaching English to Speakers of Other Languages (TESOL). While no article in this issue explicitly examines the collaboration between teachers/learners and machines (AI), quite a few investigate the use of technologies for teaching and assessment. The use of technologies for these purposes somehow reflects interactions between teachers/learners and automated systems. For example, the article by Astrid Jean Morrison and Paulina Sepulveda-Escobar explores language teacher educators’ perceptions and practices of online assessment, while the article by Jun Chen Hsieh investigates the effect of robot-assisted multimodal digital storytelling presentation on English as a Foreign Language (EFL) learners’ emotions, grit and perceptions on language learning. The technology review article by Bing Mei, Wenya Qi, Xiao Huang and Shuo Huang also discusses the potential of Speeko, an AI-assisted personal public speaking coach, in developing EFL students’ public speaking skills, while the Innovation in Practice article by Lucas Kohnke and Frankie Har describes how Perusall can be used to increase students’ engagement with texts and encourage critical discussion about the texts.
Other Potential Areas for HMC in ELT in the Age of AI
Other articles in this issue explore different perceptions, perspectives and practices in understanding language learning, teaching language skills, assessing learning and giving feedback and language teacher development. The article by Yan Huang and Azirah Hashim unveils the contributing factors to Chinese English learners’ perceptions of different English accents, while Mark Feng Teng's article highlights the important roles of L2 proficiency level and working memory in vocabulary learning. The article by Matthew Sung Chit Cheung also unpacks the relationship between emotions, language ideologies, identities and agency in L2 learning, while Huseyin Uysal and Hyunjin Jinna Kim's interview with Bonny Norton discusses the connection between research, theory and practice in TESOL particularly in language teacher identity and investment. On teaching language skills, the article by Brett Milliner and Blagoja Dimoski introduces the practice of pre-task modelling and communication strategies awareness-raising to help learners overcome disconnects in communicative tasks, while Hong Zhang and Rui Yuan's article describes how the infusion approach can be implemented to cultivate critical thinking abilities and language competence of EFL learners. The article by Limei Zhang, Xiaoqin Yu and Christine CM Goh also introduces the effect of the metacognitive awareness listening questionnaire-based strategy on students’ listening comprehension ability and metacognitive awareness, while Heon Jeon's article illustrates how to promote learning transfer in other contexts of writing by designing and implementing pedagogical tools guided by the notion of high road transfer. Moreover, the article by Yusop Boonsuk, Fa-ezah Wasoh and Eric Ambele also critically examines Global English (GE)-oriented activities to raise learners’ GE awareness.
On language assessment, the article by Philip Greenblatt and Peter McDonald describes how a multimodal project can be used to assess communicative language and develop multimodal literacies, while Huan Mei and Huilin Chen's paper discusses the application, optimization and practical and theoretical uses of cognitive diagnostic assessment. On giving feedback, the article by Deliang Man, Beibi Kong and Meng Huat Chau illustrates how to develop feedback literacy through peer review training, while Kareem Sadeghi and Maryam Esmaeeli's empirical article confirms the effectiveness of different corrective feedback types in improving grammatical accuracy in writing. In addition, the article by Yu Zhou, Shulin Yu and Peisha Wu discusses the effects, practices and issues of praise feedback in L2 writing, while Nourollah Zarrinabadi and Masoumeh Soleimani's article uncovers the effect of positive feedback on goal commitment in directed motivational currents.
On language teacher education and PD, the research article by Mark Bedoya Ulla describes how reflective practice can transform language teachers’ preconceived beliefs and professional values toward teaching, while the research article by Nga Huynh Hong Ngo, Sue Cherrington and David Crabbe presents an integrated framework of effective PD for tertiary EFL lecturers which incorporates three main dimensions of content, context and process. The article by John Macalister and Say Phonekeo also discusses how pre-service teachers’ prior language learning experiences shape their perceptions and practices in ELT, exposing a gap between the education policy that promotes communicative language teaching and the actual practice in the classroom context, while the article by Keith Mathew Graham offers a set of competencies that may be used to guide PD for team teachers in bilingual education. Lastly, the article by Susan Gwee and Hwee Leng Toh-Heng reveals the impact of engaging in classroom research on teachers’ habits of mind regarding their pedagogical practices, professional learning, and students’ learning and engagement.
The perceptions, paradigms and practices covered in the above articles can be enhanced with deliberate HMC in the age of (generative) AI. For example, language teachers may collaborate with AI to understand the beliefs and analyse the factors that shape L2 learning in a particular context. Language teachers may also use AI to identify patterns of behaviours of L2 learners throughout the learning process to develop more informed instructional decisions. In terms of teaching language skills and subskills, language teachers may use varied automated systems or digital tools to facilitate L2 learning both inside and outside the classroom following relevant learning theories, principles and technology integration frameworks (e.g., S-A-M-R, P-I-C-R-A-T). In the current technology landscape, a panoply of AI-powered digital tools is readily available to facilitate the teaching of L2 listening, speaking, reading, writing, vocabulary and grammar under different pedagogical approaches, paradigms and orientations. In terms of language assessment and giving feedback, the use of HMC can be beneficial for both teachers and students. Teachers can work with AI to design and develop different forms of assessment tasks (e.g., paper-based, online, multimodal, etc.) that foster justice, equity, diversity and inclusion, while students can collaborate with AI to plan, draft and revise their assessment outputs/projects through contextualized and immediate corrective feedback. Teachers may also leverage AI to develop feedback literacy, facilitate peer feedback and foster praise feedback and other types of feedback under different circumstances. For professional learning and development, teachers may use AI-powered digital tools to analyse and identify their PD needs, explore relevant PD opportunities, track their learning progress and reflect on their PD experiences.
The Future of Human–Machine Collaboration in Language Education
Artificial intelligence has become an integral part of our daily lives. It has disrupted, enhanced and changed our work perspectives, practices and values. It has changed our perspectives on what to teach and our practices on how to teach language skills and assess learning outcomes. Similarly, it has also revolutionized how students learn and achieve their learning goals. As AI continues to evolve, more exciting and innovative forms of HMC are expected to emerge. The requirements and dynamics of work in various sectors and industries will also change, leaving more challenges to schools, universities and training institutions on how to prepare the graduates and existing workforce to be proficient collaborators with AI. In a world of work where automation is a currency, humans need to be skilled in collaborating with machines. An ideal learner or worker of the future is someone who knows how to proficiently and perceptively collaborate with a machine/AI. Given this, in the context of English language teaching, learning and assessment, some questions remain: What kinds of HMC are considered meaningful and productive for L2 learning? What types of HMC are ethical and to what extent are these allowed? Under what teaching–learning conditions can HITL or MITL systems be used? How can language teachers explicitly and seamlessly integrate HMC into their lessons? What skills or competencies are needed to be able to engage in a meaningful HMC? To answer these questions, relevant policies and guidelines should be developed. Necessary resources should also be provided along with capacity building and continuous training for both language teachers and learners. As we live in an ever-evolving AI-driven society, we hope to find more specific answers to these questions. We also hope to see more articles published in the RELC Journal in the future addressing these questions or issues.
