Bilingual Simulations in Nursing: Virtual Reality, Interlinguistic Pairing, and Authentic Reinforcement

In recent years, researchers in computer-assisted language learning have been exploring the potential of immersive technologies to support second language (L2) acquisition. Virtual reality (VR), defined as the computer-generated simulation of an environment in which users experience a sense of presence (Sadler et al., 2013), can immerse L2 learners in interactive scenarios where they practice communication in real-life situations. Such learning environments can reduce L2 anxiety (Gruber & Kaplan-Rakowski, 2022) while increasing willingness to communicate (Papin, 2018, 2022). However, little is known about the impact of participation in immersive interactive simulations on L2 self-efficacy (i.e., learners' belief in their ability to use the L2 to perform linguistic and non-linguistic tasks (Mills, 2014)). Given the positive impact of self-efficacy on L2 learning (Bai & Wang, 2022), the research objective of this exploratory study is to examine the impact of conducting VR simulations on L2 self-efficacy. The participants will be 48 intermediate French L2 learners enrolled in a nursing education program at two post-secondary institutions, one English-speaking and the other French-speaking. Before participating in an in-person internship at a hospital, they will be asked to individually conduct two simulations on ImmerseMe. During these simulations, which mimic daily conversations in a professional setting, learners will be individually immersed in a virtual environment. This environment will place them in a virtual hospital room using 360-degree videos, and they will have to orally interact with pre-programmed virtual agents. The simulations are based on scenarios depicting authentic interactions encountered in a hospital setting: 1) assessing a patient's level of pain; and 2) interacting with a patient as part of a postoperative protocol. Through ImmerseMe's voice recognition capabilities, the utterances of L2 learners will be compared to a pre-established script, allowing the simulations to progress if the produced sentence matches one of the possible responses.