Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Learning new words through social and self-perspective interaction in virtual reality
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Ana Zappa1, Silvia Rondini1, Antoni Rodriguez-Fornells1,2, Mel Slater3; 1Department of Cognition, Development and Educational Psychology & Neuroscience Institute, University of Barcelona, Barcelona, Spain, 2Institució Catalana de Recerca i Estudis Avançats, Barcelona, Spain, 3Department of Clinical Psychology and Psychobiology, Institute of Neurosciences, University of Barcelona, Barcelona
Growing evidence points to face-to-face human interaction as one of the most powerful pathways to second language learning. Yet most laboratory research on adult language learning has relied on artificial, non-social tasks that fail to capture how second language learners acquire word meanings in interactive contexts. This study addresses this gap by investigating how different social and non-social learning perspectives influence the learning of new words when embedded in an immersive virtual reality (VR) environment. VR offers a unique opportunity to simulate social interaction while maintaining experimental control. It allows learners to engage with realistic virtual agents and experience learning perspective in ways that are not possible in traditional settings. The current study compares word learning across three perspectives: an Other condition, in which participants learn from a social virtual agent; a Self condition, in which participants learn from a realistic self-avatar; and an Object-only condition, which serves as a non-social control. The social agent condition is motivated by research showing that linguistic input from a social partner is processed differently than input from a non-social source, often resulting in deeper encoding and higher engagement. The self-avatar condition introduces a more exploratory manipulation. Learning from one's own avatar may engage self-referential processing in ways that benefit encoding, but this has not yet been empirically tested in second language word learning in immersive VR, making this an open question. Linguistic input will be held constant across conditions, allowing us to isolate the contribution of social perspective to learning outcomes. Eye-tracking during VR will provide real-time indices of attentional engagement - including gaze distribution between agents and objects - allowing us to explore whether social perspective modulates attention during learning and whether attentional patterns predict learning outcomes. In a within-subject design, forty participants will learn novel German words (e.g., Halskette [necklace]) in an interactive VR environment across two learning phases. In the first phase, participants will see four objects, select one by pointing, hear a label produced by the agent, and repeat it. In the second phase, participants will attempt to produce each label independently, rate their confidence, and receive the correct label from the agent. Behavioural measures will include response times, confidence ratings, and learning curves. Post-training tests will measure recognition and production immediately after learning and at a ten-day follow-up to assess retention. We predict that semantic learning will follow the pattern Other > Self > Object-only, reflecting the advantage of social interaction for word learning. We further predict that greater attentional engagement with the agent - as indexed by gaze distribution - will be associated with better learning outcomes in the social conditions. These findings will advance our understanding of how social context and perspective shape second language learning, with implications for language education and rehabilitation.
Topic Areas: Meaning: Lexical Semantics, Multilingualism