*3.2. The Effect of the Surrounding Environment on Participants' Hedonic Ratings of Chocolate Samples*

In this study, participants rated the smell, flavour, sweetness, texture, smoothness, and overall liking of three identical pieces of chocolate in three different contextual settings. Participants were unaware that the same chocolate was being served in all three conditions. There was a significant difference in perceived liking of flavour (χ 2 (2) = 7.971, *p* = 0.019) and overall liking (χ 2 (2) = 10.750, *p* = 0.005), depending on the environment in which the chocolate was consumed. A post hoc analysis with Wilcoxon signed-rank tests was conducted with a Bonferroni correction applied, resulting in a significance level set at *p* < 0.017. The results showed that participants rated the flavour (Z = −2.950, *p* = 0.003) and overall liking (Z = −2.885, *p* = 0.004) of chocolate significantly higher when consumed in the VR countryside setting, compared to the traditional sensory booth. The surrounding context did not influence hedonic ratings of other chocolate attributes. However, in general, the VR countryside setting produced higher liking scores for chocolate compared to the VR busy city and sensory booth environment. Descriptive statistics for the data is presented in Table 2.


**Table 2.** Means, standard deviations (SD), medians and interquartile ranges (IQR) for participant hedonic rating \* of chocolate in a traditional sensory booth, a VR busy city and a VR Irish countryside environment.

\* Evaluated on a 9-point hedonic scale, where 1 = dislike extremely and 9 = like extremely.

While our research findings agree with previous studies which show that consumers' hedonic ratings of food can change depending on the specific context [2,12,15,25,26], it is also contradictory with others [8,10,11]. For instance, Kong et al. (2020) [11] studied the impact of three contextual settings, including sensory booths and two VR environments (a pleasant sightseeing tour and a live music concert), on consumer liking of three types of chocolate. While their research showed significant differences in consumer liking of the different chocolate types, the surrounding context did not influence the tasting experience. The different immersive environments used to evoke context might explain the inconsistency in findings between this study and ours. Research suggests that certain food products are better suited to specific consumption contexts [3,30], which in turn may alter the sensory perception of food during consumption, through the elicitation of a positive emotional response [26]. Although in this exploratory study we did not measure consumer emotional response to chocolate in the different immersive environments, the more naturalistic and peaceful setting of the Irish countryside might have improved

the mood of the participants, which favourably influenced their liking of the chocolate flavour, when compared to the sensory booths and VR city environment [31]. Based on this assumption, we would have expected the chocolate consumed in the busy VR city environment to have produced overall lower hedonic ratings in comparison to the quiet sensory booth setting, as observed elsewhere for chocolate ice-cream [26]. Picket and Dando [12], used VR to investigate how the surrounding environment (winery vs bar) influenced sensory perception of two alcoholic drinks (wine and beer). They showed that while participants liked the wine more and were willing to pay more for it when it was consumed in a virtual winery context, these effects were not observed for beer.

VR technology is a promising tool for simulating immersive environments for consumer sensory evaluations and could be used in conjunction with non-invasive biometric sensors to enrich our understanding of the emotions responsible for driving consumer liking in the real world [2,13,32]. Nonetheless, as research in this area is in its infancy, further studies are required to identify the specific conditions in which utilizing VR is relevant and is likely to improve the reliability and ecological validity of sensory testing outcomes [2,13].

Beyond the importance of selecting an appropriate immersive environment for a particular food product, another possible explanation for the conflicting results associated with using VR as a context-enhancing technology for sensory assessments could be related to how it is being applied during sensory tests. For instance, some studies, including ours, have created immersive VR environments by displaying custom-recorded 360-degree VR videos and their corresponding sounds through a HMD [11,12,15]. In contrast, computersimulated 3D virtual environments, in which participants can physically walk around and interact with the virtual space using a HMD, have been used to study various aspects of consumer behaviour [33,34]. While both methods are expected to provide a more engaging experience compared to a traditional testing environment [2,5], the use of the HMD fully replaces the participants' view of the real world, restricting their ability to interact visually with the 'real' food product. In an effort to capture consumers' hedonic responses while fully immersed in the VR environment, the participants in this study provided answers to the questions verbally, which were subsequently recorded by the researcher. In similar studies, participants were instructed to remove their VR headsets after tasting and answer questions on a paper ballot [10,11], while othersoverlaid audio instructions and visual scales in the 360-degree VR video [15]. While each approach has clear advantages and disadvantages, future studies could investigate which approach can best improve the ecological validity of consumer sensory data, while continuing to retain participant engagement. In the meantime, this limitation could be overcome by using a mixed-reality device, such as Microsoft HoloLens, which integrates specific elements from the 'real-world' within the surrounding VR space, enabling the user to interact with the real food product while simultaneously answering questions in a controlled environment [13]. Mixed reality was recently utilised to understand consumer response to tea-break snacks and was shown to evoke ecologically valid data comparable to a real-life context [14].
