Social robots are a relatively new digital tool in L2 teaching (Randall, 2019; Vogt et al., 2019). Robot-assisted language learning (RALL) offers interesting possibilities for early language learning, because robots can be used even with illiterate children for practicing speaking in L2. However, relatively little is known about child-robot-interaction (CRI) in L2 learning contexts. In this presentation, we explore these interactions to identify what type of reasons there are for possible communication problems in CRI.
The corpus consists of video recordings of primary school children's first encounters with an educational robot in L2 learning. The 32 participants were 7 to 11 years old children from Swedish speaking schools learning English as their L2. They completed one thematic L2 lesson with a NAO robot pairwise. These recordings were analyzed qualitatively to identify (a) what type of linguistic resources are needed and (b) what is likely to lead to communication breakdowns in the interaction with the robot.
The results show that children's output in robot assisted L2 learning is governed to a great extent by the structure of the pre-programmed lessons and that the majority of communication breakdowns in RALL are related to problems in L2 pronunciation and in robot's voice recognition. These results highlight the importance of adjusting RALL for L2 child speakers and the role of teachers in choosing appropriate teaching materials.
Randall, N. (2019). A survey of robot-assisted language learning (RALL). ACM Transactions on Human–Robot Interaction, 9(1), 1–36.
Vogt, P. et al. (2019). Second Language Tutoring using Social Robots. A Large-Scale Study. Proceedings of the 2019 ACM/IEEE International Conference on Human-Robot Interaction (HRI 2019).
