A Survey of Robot-Assisted Language Learning (RALL)

Using social robots for language learning: are we there yet?

Along with the development of speech and language technologies and growing market interest, social robots have attracted more academic and commercial attention in recent decades. Their multimodal embodiment offers a broad range of possibilities, which have gained importance in the education sector. It has also led to a new technology-based field of language education: robot-assisted language learning (RALL). RALL has developed rapidly in second language learning, especially driven by the need to compensate for the shortage of first-language tutors. There are many implementation cases and studies of social robots, from early government-led attempts in Japan and South Korea to increasing research interests in Europe and worldwide. Compared with RALL used for English as a foreign language (EFL), however, there are fewer studies on applying RALL for teaching Chinese as a foreign language (CFL). One potential reason is that RALL is not well-known in the CFL field. This scope review paper attempts to fill this gap by addressing the balance between classroom implementation and research frontiers of social robots. The review first introduces the technical tool used in RALL, namely the social robot, at a high level. It then presents a historical overview of the real-life implementation of social robots in language classrooms in East Asia and Europe. It then provides a summary of the evaluation of RALL from the perspectives of L2 learners, teachers and technology developers. The overall goal of this paper is to gain insights into RALL's potential and challenges and identify a rich set of open research questions for applying RALL to CFL. It is hoped that the review may inform interdisciplinary analysis and practice for scientific research and front-line teaching in future.

Social robots in a translanguaging pedagogy: A review to identify opportunities for robot-assisted (language) learning

This mini review discusses the use of social robots in a translanguaging pedagogy: the use of robots to enable students to use their full linguistic repertoire within schools, so any language that they speak at home or in another aspect of their lives. Current research on robot-assisted second-language learning is reviewed with the aim of finding out whether students’ languages have been employed strategically to support learning of another language. A total of 83 articles has been analyzed on the use of first and second languages in student-robot interactions. Most interactions were either exclusively in the second language, or exclusively in the first language, with only target words in the second language. Few studies strategically mixed the two languages to bootstrap learning, and only one study used the first language of students with migrant backgrounds to learn the second language. The review concludes with recommendations for future use of social robots in a translanguaging pedagogy.

Robot-Assisted Language Learning: Integrating Artificial Intelligence and Virtual Reality into English Tour Guide Practice

This action research created an application system using robots as a tool for training English-language tour guides. It combined artificial intelligence (AI) and virtual reality (VR) technologies to develop content for tours and a 3D VR environment using the AI Unity plug-in for programming. Students learned to orally interact with the robot and act as a guide to various destinations. The qualitative methods included observation, interviews, and self-reporting of learning outcomes. Two students voluntarily participated in the study. The intervention lasted for ten weeks. The results indicated the teaching effectiveness of robot-assisted language learning (RALL). The students acknowledged the value of RALL and had positive attitudes toward it. The contextualized VR learning environment increased their motivation and engagement in learning, and students perceived that RALL could help develop autonomy, enhance interaction, and provide an active learning experience. The implications of the study are that RALL has potential and that it provides an alternative learning opportunity for students.

Pitch It Right: Using Prosodic Entrainment to Improve Robot-Assisted Foreign Language Learning in School-Aged Children

Robot-assisted language learning (RALL) is a promising application when employing social robots to help both children and adults acquire a language and is an increasingly widely studied area of child–robot interaction. By introducing prosodic entrainment, i.e., converging the robot’s pitch with that of the learner, the present study aimed to provide new insights into RALL as a facilitative method for interactive tutoring. It is hypothesized that pitch-level entrainment by a Nao robot during a word learning task in a foreign language will result in increased learning in school-aged children. The results indicate that entrainment has no significant effect on participants’ learning, contra the hypothesis. Research on the implementation of entrainment in the context of RALL is new. This study highlights constraints in currently available technologies for voice generation and methodological limitations that should be taken into account in future research.

Robot-Enhanced Language Learning for Children in Norwegian Day-Care Centers

In a case study, we transformed the existing learning program Language Shower, which is used in some Norwegian day-care centers in the Grorud district of Oslo municipality, into a digital solution using an app for smartphones or tablets with the option for further enhancement of the presentation by a NAO robot. The solution was tested in several iterations and multiple day-care centers over several weeks. Measurements of the children’s progress across learning sessions indicated a positive impact of the program using a robot as compared to the program without a robot. In situ observations and interviews with day-care center staff confirmed the solution’s many advantages, but also revealed some important areas for improvement. In particular, the speech recognition needs to be more flexible and robust, and special measures have to be in place to handle children speaking simultaneously.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children

Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Engaging English Language Learners as Cultural Informants in the Design of a Social Robot for Education

Background: There are 4.9 million English Language Learners (ELLs) in the United States. Only 2% of educators are trained to support these vulnerable students. Social robots show promise for language acquisition and may provide valuable support for students, especially as we return to needing smaller classes due to COVID-19. While cultural responsiveness increases gains for ELLs, little is known about the design of culturally responsive child–robot interactions. Method: Therefore, using a participatory design approach, we conducted an exploratory study with 24 Spanish-speaking ELLs at a Pacific Northwest elementary school. As cultural informants, students participated in a 15-min, robot-led, small group story discussion followed by a post-interaction feedback session. We then conducted reflexive critiques with six ELL teachers who reviewed the group interactions to provide further interpretation on design feature possibilities and potential interactions with the robot. Results: Students found the social robot engaging, but many were hesitant to converse with the robot. During post-interaction dialogue students articulated the specific ways in which the social robot appearance and behavior could be modified to help them feel more comfortable. Teachers postulated that the social robot could be designed to engage students in peer-to-peer conversations. Teachers also recognized the ELLs verbosity when discussing their experiences with the robot and suggested such interactions could stimulate responsiveness from students. Conclusion: Cultural responsiveness is a key component to successful education in ELLs. However, integrating appropriate, cultural responsiveness into robot interactions may require participants as cultural informants to ensure the robot behaviors and interactions are situated in that educational community. Utilizing a participatory approach to engage ELLs in design decisions for social robots is a promising way to gather culturally responsive requirements to inform successful child–robot interactions.