This corpus was collected in February 2023 at the LeaF research infrastructure, University of Oulu. The data was collected as a part of the postdoctoral research project “Beyond individual experiences towards social VR” (Research Council of Finland, 2022–2025). We invited 11 pairs (altogether 22 participants) to interact in a quasi-experimental setting in the social VR platform Sansar (https://www.sansar.com/). We had built a private collaboration space in Sansar, which resembled an apartment with different furniture and objects laying around. We instructed the pairs to conduct a collaboration task in the space. The task was to play a pair of roommates and furnish the apartment together. The participants used HTC Vive Focus 3 head-mounted displays (HMD) and hand-held controllers to interact in VR. They were physically situated in two adjacent rooms so they could hear and see one another only as avatar characters in the virtual space. Each pair was video recorded from multiple perspectives using screen capture software and video cameras. We recorded 1) both participants’ first-person perspective to the VR, 2) the researcher’s third-person perspective to the participants in VR, and 3) the participants in their physical environments. The corpus includes 9 hours of video recordings as well as detailed written transcripts of the participants’ audible and multimodal conduct captured in the recordings. We use multimodal conversation analysis as the main research method for analyzing the video-recorded interactions. The languages used in the data are English (10 pairs) and Finnish (1 pair). We also collected background information about the participants’ previous experience with VR technology and online gaming. The ethical guidelines and criteria of the Finnish National Board for Research Integrity were followed during the data collection. The participants were provided with written information about the study and data protection, and they signed an informed consent for being video recorded for research purposes. Due to the personal data revealed in the video materials (participants’ faces and voices), this material cannot be stored or made public via data repositories.

2024