Investigators recruited 227 undergraduate and graduate students at Toyo University and ran a four-day discrete choice experiment from November 25–28, 2024. Participants made reservations for time slots without knowing which dates corresponded to the immersive virtual reality (IVR) or real-world (RW) conditions; the authors treated this as producing effectively random assignment. On arrival, participants provided written informed consent and completed a single choice experiment in whichever environment corresponded to their reservation. After the choice task they completed a questionnaire on demographics and usual food consumption behavior. The study team observed that the participant pool—students who occasionally consume ready-to-eat cereals—matched the profile used in prior VR versus RW comparison studies.
The discrete choice experiment presented multiple ready-to-eat cereal (RTEC) options and included an outside (no-purchase) alternative. Participants received a JPY 1,100 voucher to cover product purchases; the chosen product price was deducted from the voucher and any remaining balance was paid in cash. If a participant chose the no-purchase option in the randomly selected question used to determine payment, they received the full JPY 1,100. The authors selected ready-to-eat cereals because their key decision cues—packaging, price, nutrition information, and product composition—translate readily between virtual and physical shelves and because the products are shelf-stable and easy to distribute in an incentive-compatible design.
To derive monetary valuations, the investigators estimated random parameter logit models primarily in WTP space, a modeling strategy that directly recovers willingness-to-pay measures from choice data. They also ran preference-space specifications as robustness checks. The source material reports these model choices but does not provide the full set of model coefficients, p-values, or detailed subgroup estimates within the provided excerpt. The authors characterized their experiment as a real choice experiment (RCE) because actual payments and product distribution were integrated into the study design.
Data indicated no statistically significant differences in WTP between IVR and RW environments for any of the cereal products tested. The study team reported that product choices made in the IVR environment reflected monetary valuations similar to those observed in the RW condition. Based on these observations, the authors concluded that IVR, when paired with real economic incentives, can produce WTP estimates comparable to those from conventional real choice experiments. The manuscript states that this is the first study to compare monetary WTP between IVR and RW using an incentive-compatible experimental design. The source did not report additional numerical summaries, subgroup p-values, or effect sizes in the excerpt provided here.
Investigators discussed both benefits and operational constraints of IVR for consumer research. The immersive head-mounted display (HMD) format offers a 360-degree virtual shopping environment and can reproduce visual cues—product packaging, pricing labels, and shelf placement—consistently across sessions. This allows inexpensive, repeatable experimental manipulations without needing cooperation from retail venues. The authors also acknowledged common IVR challenges described in the literature: discomfort from HMDs, initial learning curves with controllers, and potential VR sickness. Prior work cited by the authors indicated that training reduces controller errors and that moderate presence can be achieved without widespread VR sickness, but the present source excerpt did not quantify training time or measured presence levels for this specific experiment.
The source reports key design elements and an overall null difference in WTP but omits several details in the excerpt. The paper did not present full model output, p-values, or subgroup analyses in the provided text. It also did not report whether response heterogeneity by age, gender, income, or prior VR experience affected WTP estimates. The authors limited the sample to university students; the generalizability to broader consumer populations was not established in the excerpt. Finally, the study focused on ready-to-eat cereals—products whose evaluation depends largely on visual and label information—so whether similar comparability holds for foods that require evaluation of aroma, temperature, or tasting remains an open question according to the source. The paper ends by noting these consequential gaps and by positioning IVR with real incentives as a feasible method for eliciting monetary WTP, while leaving additional empirical tests and population extensions to future work documented in the full manuscript.