← All events

Past event

Grand Challenge — Virtual Reality for the Study of Social Interactions and Cognition

BST (UK)

Virtual reality as a transformative tool for social interaction research — enabling highly controlled yet ecologically realistic simulation of complex worlds — with applications to bilingual communication and to modelling human social perception and action.

Host

  • Jonas Nölle — Nicolaus Copernicus University, Toruń & University of Glasgow

Speakers

  • Jonas Nölle — Nicolaus Copernicus University, Toruń & University of Glasgow
    “VR as a tool to study language and cognition”
  • Alex Titus — Radboud University & MPI Nijmegen
    “How Enhancing Linguistic and Non-linguistic Context Influences Bilingual Communication With the Use of VR”
  • Chen Zhou — University of Glasgow
    “Modelling human social perception and action by quantifying human behaviours in VR”

Agenda

  • 14:00–14:05 Welcome
  • 14:05–14:15 Introduction (Jonas Nölle)
  • 14:15–14:30 Bilingual communication with VR (Alex Titus)
  • 14:30–14:45 Modelling human social perception and action in VR (Chen Zhou)
  • 14:45–15:30 Open discussion

Virtual reality is rapidly emerging as a transformative tool for social interaction research, enabling highly controlled yet ecologically realistic simulation of complex worlds. In this session, MOSAIC host Dr Jonas Nölle and invited guests showcased how these techniques can advance our understanding of human social interaction.

Introductory talk — VR as a tool to study language and cognition (Jonas Nölle)

A concise overview of Virtual Reality (VR) and the broader category of Extended Reality (XR) technologies, and their potential to catalyse a paradigm shift in experimental research — merging rigorous experimental control with high ecological validity. This sets the groundwork for the following presentations and discussion.

Talk 1 — Bilingual communication with VR (Alex Titus)

Our current projects move beyond previous paradigms by adding a jittered approach to virtual avatars in a marketplace, comparing visual and auditory background contexts, comparing more distant language pairs, and running the first moral decision-making experiment in VR. Each incorporates the foundational aspects: bilinguals producing full sentences rather than individual words, addressed to a life-size addressee rather than a microphone, with messages relevant to that addressee, in a rich everyday visual environment rather than in front of a screen.

Talk 2 — Modelling human social perception and action in VR (Chen Zhou)

Humans effortlessly identify social groups in complex environments and navigate crowds accordingly, yet the computational rules governing these behaviours remain a gap. We immersed participants in a well-controlled virtual environment, precisely manipulating virtual humans and measuring behavioural judgments and walking trajectories with classic psychophysical methods. From these results we proposed a social interaction field model of social group perception and a social locomotion model of navigation, validated on open-access databases and real-world scenarios. Translating these human-behaviour-based models into algorithms for social robots achieved significant improvements in perceived human-likeness and comfort during human–robot interactions.