Research on the impact of virtual reality (VR) on learning and the integrated model of processing evidence and its uncertainty values (Glaser, Lengyel, Toulouse, & Schwan, 2022) indicates that learners are in danger of mistakenly confusing the representational fidelity of information with scientific certainty and neglect uncertain information during the learning process. In order to find a solution to this problem, we have examined how high-fidelity VR presentations may be sup-plemented by markers of scientific uncertainty in a way that provides learners with an under-standing of the evidential nature of the VR's content and, at the same time, have them elaborate on particular elements of the VR, irrespective of their status of scientific uncertainty. We assumed that certain information is better retained than uncertain information, and additionally inferred that reception goals might influence the retention of uncertain information. Participants (n = 75) were assigned to one of two experimental conditions: learning goal and research goal (between -subjects factor). In both conditions, participants walked through the virtual reconstruction of a historic building with architectural elements including uncertain, medium, and certain archi-tectural elements (within-subject factor). Uncertainty was expressed visually by stop light colors and verbally in accompanying audio explanations. Retention of the appearance of architectural elements, uncertainty values, their verbal scientific justifications, and also exploratorily historical experience were measured in both reception goal conditions. Results showed that with a learning goal but not a research goal the appearance of certain architectural elements was better remembered than that of uncertain architectural elements. The memory of uncertainty values and their verbal scientific justifications was equal for uncertain, medium, or certain architectural elements in both reception goal conditions. Affective engagement during the historical experience was higher with a research than with a learning goal. The study extends existing theories on processing uncertain information and shows that orchestrating basic characteristics of VR with carefully chosen instructional strategies for successful knowledge acquisition is especially important for VRs containing uncertain information.
