https://orcid.org/0000-0003-4384-9277
Figures
Abstract
Objective
To understand the impact of fidelity and perceived realism on virtual reality food choices, and task motivation, engagement, and interest.
Intervention
Participants were randomly assigned to either a high- (n = 43) or a low- (n = 41) visual fidelity environment and were asked to select foods to have a meal with a friend.
Analysis
Simple linear regressions between visual fidelity and perceived realism, and log-linear regressions for visual fidelity or perceived realism on either motivation, interest, or engagement. Poisson models between visual fidelity or perceived realism, and food selections.
Results
Manipulating visual fidelity was not associated with perceived realism, motivation, interest, or engagement in the food selection task. Perceived realism increased motivation by 0.3% (SE 0.056; p = 0.022), interest by 1.4% (SE 0.002; p<0.001), and engagement by 0.9% (SE 0.001; p<0.001) in the food selection task. High visual fidelity decreased the total number of foods selected (B = 0.216; CI (-0.384; -0.047); p = 0.012).
Citation: Curi Braga B, Sajjadi P, Bagher M, Klippel A, Menold J, Masterson T (2025) The impact of visual fidelity on screen-based virtual reality food choices: A randomized pilot study. PLoS ONE 20(1): e0312772. https://doi.org/10.1371/journal.pone.0312772
Editor: Ziyu Qi, University of Marburg: Philipps-Universitat Marburg, GERMANY
Received: November 6, 2023; Accepted: October 13, 2024; Published: January 30, 2025
Copyright: © 2025 Curi Braga et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data file is available in OSF: https://osf.io/mjcrv/.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The selection of foods for consumption relies on both intrinsic (i.e., individual motivations) and extrinsic (i.e., environmental contexts) factors [1–6]. Consumers often must balance the rewarding properties of a food against a variety of perceived short and long-term consequences of consuming them [7]. For example, it has been demonstrated that food selection is based on costs, sensory quality (e.g., taste, texture, smell), convenience, nutrition value, environmental impact, health benefits, and social issues [8–10].
When investigating food choice, most studies rely on hypothetical scenarios and often employ visual representations of foods. Novel technologies such as virtual reality (VR) have the potential to enhance our comprehension of food choices and the motivations driving them. Food selection in a VR food buffet, for example, have been shown to reflect food selection made in real food buffets [11–13]. VR makes the gap between the virtual and the real experience smaller by providing ecologically valid scenarios that reflect many aspects of real-life experiences. VR and real-life experiences can be perceived similarly [14]; but perception of realism dependent on the illusion of presence (degree of technical immersion) and the plausibility of the hypothetical situation (degree of credibility of scenario) [15].
Through experiences such as presence and embodiment, VR has the potential to afford users natural and realistic means of interaction with foods and environments [16]. Visual fidelity, or the extent to which an image looks like a photo rather than generated by a computer [17], may influence users’ perception of virtual foods, their engagement in virtual experiences, and virtual food choices [18]. For example, a recent study demonstrated that the visual quality of digital food items can influence participants desire to consume them [19].
Visual fidelity depends on geometric realism (how much virtual objects seem like their real-world counterparts) and illumination realism (lighting fidelity) [20]. In relation to digital imagery, high visual fidelity means realistic, while low visual fidelity is simpler and more cartoonish [21]. High visual fidelity can potentially make VR-based scientific experiments more realistic and engaging [21], but the extent to which manipulating visual fidelity predicts user experience and behavior in relation to virtual food selection is unclear. To the best of our knowledge, only one study has systematically manipulated the visual fidelity of foods [19] although they did not manipulate the visual fidelity of the surrounding environment where food was displayed in the simulation.
Therefore, this pilot study sought to understand if manipulating visual fidelity affects users’ perception of task realism and if either visual fidelity or perception of realism are related to motivation, interest and enjoyment, or engagement within a food selection task. Our goal was to test the effect of different levels of visual fidelity and the perception of realism of the virtual foods and environment where they were displayed on virtual food choices. To accomplish this, we created two versions of a virtual food selection task that differed in visual fidelity. Difference in visual fidelity of foods and the environment were achieved by manipulating lighting, geometry [22], rendering quality [23], and texture quality [24] of food items and the surrounding environment. In this study we hypothesized that higher visual fidelity would be associated with higher levels of perceived realism and increased motivation, interest and engagement in the food choice task. We also hypothesized that perceived realism, regardless of visual fidelity, would be related to motivation, engagement, and interest in the food choice task. Finally, we hypothesized that interacting with high fidelity models leads to a larger number of foods being selected.
This is an important topic of study, given the costs and effort for producing high visual fidelity models. If low visual fidelity has similar or better user performance to high visual fidelity, it will help researchers save money in application development and produce training content more quickly [21]. This is particularly pertinent to academia where both financial and human resources are scarce.
Materials and methods
In this study we measured the impact of visual fidelity and perceived realism on food choice, food choice motivation, interest, and engagement by randomly assigning 88 participants to either a high or a low visual fidelity condition on a web-based VR experience. Participants were recruited from Amazon Mechanical Turk (MTurk). Remote studies ran on MTurk have shown to be an effective way in obtaining a heterogenous sample with adequate diversity of participants, capturing reliable data, and saving costs [29–32]. Our excluding criteria were (1) repeating the survey, (2) leaving the surveys incomplete, (3) spending 2.2 standard deviations from the mean time to finish the study [25]. No one was excluded based on the first two criteria, but four participants were excluded for taking too long to complete the study. We analyzed the data from 84 participants, 43 in the high visual fidelity and 41 in the low visual fidelity conditions. The Human Research Protection Program Institutional Review Board of The Pennsylvania State University determined that this study (#00019284) did not require formal IRB review because it met the criteria for exemption according to the policies of The Pennsylvania State University and the federal regulations. The study started on January and ended in June of 2022. The participants provided informed written consent. We followed the principles of the Declaration of Helsinki. The data collected are available in the S1 Data.
Intervention
Two versions of the same virtual experience were created, one with high visual fidelity models and one with low visual fidelity models. The virtual experience represented a kitchen with 21 different food items grouped as fruits & vegetables, main course or dessert, and placed on a table at the center of the scene. The users could change the angle of the camera view (rotate) and zoom in or out to explore and examine the food items from different perspectives. The virtual experience was developed in Unity3D and was made accessible through conventional web browsers.
Manipulating fidelity
The visual fidelity of the virtual experience was manipulated through a process in which we purchased high fidelity photogrammetry models, and systematically reduced their visual quality through removal of details and stylization. A specific collection of models with a wide variety of high-quality food objects based on real photographs was used [26]. It was our aim to maintain a good level of similarity between the two levels of visual fidelity in terms of recognizability. As such, we utilized a top-down approach where we removed detail from the base high-fidelity models to synthesize low fidelity ones. This process included reducing polygon counts and modifying textures. Polygon reduction was automatically processed, while reducing texture quality involved applying the cutout filter in Adobe Photoshop [27]. Furthermore, a toon shader was applied to the synthesized lower fidelity models to intensify the reduction effect The toon shader gave the “flat” look to the objects in the low visual fidelity condition. Toon shaders also include an outline around the object to make it stand out distinctly. Toon shaders have been applied previously to give cartoonish appearance to 3D characters in a medical virtual reality program [28]. Research has shown that intentionally making images more cartoonish avoids the uncanny valley, or the aversion to almost but not quite realistic avatars, for example [29]. When the toon shader we used was paired with the polygon and texture reduction, the result was a good balance of reduced quality while retaining enough fidelity to enable easy recognition. Colors of the foods were not manipulated beyond this procedure. Fig 1 shows the high and low visual fidelity environments. Fig 2 shows the difference between a high- and low-visual fidelity banana (polygon counts of 4860 and 940, respectively).
The top image shows the high-visual fidelity and bottom shows the low-visual fidelity condition.
The left image shows the high-visual banana and the right image shows the low-visual fidelity banana.
Study procedure
The study began with the participants providing consent and answering a set of pre-stimuli questions. Upon completion, they received a code to start the VR experience and then another code to continue with the post-stimuli questions once the VR experience was over. The instructions explained that they should explore the 21 food items on the table of the virtual kitchen and make food choices to have a meal with a friend. They explored the virtual foods for three to five minutes from different angles by rotating around and zooming in and out using a mouse. After three minutes, they were allowed to end the VR experience and start the post-stimuli questionnaire. If they reached the end of the five-minute threshold, the experience would automatically end. In the post-stimuli survey, participants selected foods for a hypothetical meal with a friend and answered questions about the VR experience. After finishing the study, they received a code to enter into MTurk to be compensated for participation.
Pre- and post-VR surveys
All questions from the pre- and post-VR surveys had either Likert scale or multiple-choice answers or asked to enter a value. Appendix 1 shows the pre- and post-VR surveys. We used the Eating Motivation Survey (TEMS), that encompasses a broad spectrum of motives for normal eating behavior (i.e., food choice) in a validated, psychometric-based, comprehensive systematic survey [30]. Interest and enjoyment (herein referred as interest) were measured using five questions from the modified version of the interest-enjoyment subscale [31]. Engagement was measured using three questions from a 32-question scale related to realism, sensory fidelity, and distraction factors [32]. We measured individuals’ external food cue reactivity as a measure of eating behavior using an adapted version of the scale developed by [33]. Fullness rating was measured pre-survey using a visual analogue scale (VAS) to control for the hunger level of participants as a potential confound. We measured perceived realism of the virtual experience in light of the perception of typicality, plausibility, and perceptual quality. Typicality, plausibility, and perceptual quality measured how typical, plausible and realistic participants thought the virtual environment and scenario were. We measured perceived realism as part of the post-VR survey using an adapted version of the scale proposed by [18].
Data analysis
To investigate the relationship between visual fidelity and perceived realism we ran a log-linear regression with visual fidelity as the independent variable and perceived realism as the dependent variable. We followed this with a sensitivity analysis correcting for participants fullness ratings, external food cue reactivity ratings, and ethnicity. We also ran log-linear regressions using visual fidelity and perceived realism as independent variables and either task motivation, interest, or engagement as the dependent variables. In all models we used a logarithmic transformation to make the betas interpretable since food choice motivation, interest, and engagement were the sums of Likert scale scores from each individual survey.
To investigate the relationship between visual fidelity, perceived realism, and food selections we ran separate Poisson regressions for visual fidelity and perceived realism as the independent variable on foods chosen including the number of low-energy dense foods, the number of high-energy dense foods, and the total number of overall foods chosen. We also ran a combined model that included both visual fidelity and perceived realism in the same model while also correcting for external food cue reactivity, fullness rating, and ethnicity. We conducted similar follow up analyses using linear regressions with estimated kilocalories (kcal) of foods selected rather than the number of foods. We also stratified this by low-energy dense foods, high-energy dense foods, and total foods. As an exploratory analysis we analyzed how differences in visual fidelity and perceived realism affected the selection of any of the 21 individual foods displayed. To do this we used Probit regressions. We again used a follow up analysis examining estimated kcal of foods selected using linear regression. The significance level for all tests conducted was established a priori as a p<0.05.
Results
Participant demographics
The average age of our sample was 39.8 (11.9) years. White/Caucasians were 75% and black were 16% of the analyzed participants. Half of the sample were female, 39% were college educated or more, and 20% had an income lower than the median income of the United States [34]. High- and low-visual fidelity groups were balanced at baseline for all characteristics but being Hispanic or Latino (p = 0.053). Participant groups differed in fullness rating at baseline (p = 0.034). The demographics are described in Table 1.
Relationships between visual fidelity, perceived realism, and motivation, interest, and engagement
The log-linear regression between visual fidelity and perceived realism showed that there was no significant relationship between visual fidelity and perceived realism (B = 0.04, SE = 0.07, p = 0.570). Log-linear regression models are a type of generalized linear model that are used for modeling categorical data. The model uses a log-link function to relate the linear predictors to the response variable. In other words, it’s possible to interpret the impact of the independent variables in the dependent variable as a percentage. The results of the log-linear regression models investigating the relationship between visual fidelity, perceived realism, motivation, interest, and engagement are summarized in Table 2. In short, visual fidelity was not directly associated with motivation, interest, or engagement in the food selection task. This was true even in the combined model and when controlling for external food cue reactivity, fullness rating, and ethnicity. However, perceived realism was associated with all three independent variables. Specifically, our models showed that a one-point change in perceived realism independently increased overall motivation in the task by 0.3% (SE 0.05; p = 0.022), interest in the task by 1.4% (SE 0.002; p<0.001), and engagement 0.9% (SE 0.001; p<0.001) in the food selection task. As perceived realism could range from 15 to 105 points, a 30 points difference between subjects’ perception of realism, for example, would suggest a 9% difference in motivation between the two participants. This was true even in our combined models which controlled for external food cue reactivity, fulness rating, and ethnicity.
Relationships between visual fidelity, perceived realism, and food choice
The results of the Poisson regressions for visual fidelity and perceived realism as independent variables on food choice using the total number of foods chosen are summarized in Table 3. The main difference between a Poisson regression and a log-linear regression is that the Poisson regression is used when the primary goal is to model count data where the dependent variable represents the count of occurrences of an event. Perceived realism was not associated with any of the food selection variables. However, visual fidelity was associated with food selections with significantly less foods being selected in the high-fidelity condition (B = 0.216; CI (-0.38; -0.05); p = 0.012). This was driven by the number of high energy dense foods selected (B = 0.322 (CI (-0.55; -0.09); p = 0.006) in the high-fidelity condition but not the number of low-energy foods. These results remained significant even in our combined model when controlling for perceived realism, external food cue reactivity, fullness rating and ethnicity. The results of the linear regressions, when food choice was modeled as the sum of estimated kilocalories (kcal) rather than just number of foods, are summarized in Table 4. These regressions also showed an effect of visual fidelity for all foods (B = -385.49; CI (-688.02; -82.97); p = 0.013) and for high energy dense foods (B = -400.10; CI (-712.47; -87.72); p = 0.013). Food selection results remained consistent even when controlling for perceived realism, external food cue reactivity, fullness rating and ethnicity.
Results of fidelity when considering the selection of each individual food item are summarized in S1 and S2 Tables. In short, four food items were selected significantly more in the low- than in the high-visual fidelity condition: eggs (B = -0.745; SE = 0.29; p = 0.01); bacon (B = -0.533; SE = 0.30; p = 0.05); cheese (B = -0.600; SE = 0.31;
p = 0.05); croissant (B = -1.000; SE = 0.35; p = 0.005). Perceived realism was associated with choosing more corn (B = 0.014; SE = 0.01; p = 0.10); more pears (B = 0.036; SE = 0.01; p = 0.006) and less eggs (B = -0.015; SE = 0.01; p = 0.05). Only the significance of pear (B = 0.035; SE = 0.01; p = 0.012), eggs (B = -0.016; SE = 0.01; p = 0.05) and bacon (B = 0.013; SE = 0.01; p = 0.10) were maintained in the combined models, that adjusted visual fidelity by perceived realism, fullness rating, external food cue reactivity, and ethnicity.
Discussion
As described previously, VR models can exhibit different degrees of visual fidelity, which may influence VR-based study outcomes. This study tested the impact of high compared to low visual fidelity environment and objects and how this may modify perceived realism and relate to task related motivation, interest, and engagement during a food selection task in a web VR environment. Our findings suggest that visual fidelity was not associated with motivation, interest, and engagement, nor with perceived realism. In contrast, perceived realism was associated with motivation, interest, and engagement. This suggests that the perception of realism by users does not necessarily depend on the technological visual fidelity of the experience, but perhaps does more on the experiential typicality and plausibility it elicits in users. This observation has rather important implications for the design and operationalization of virtual experiences, particularly with serious applications. The perceived realism of an experience is shown to be a determining factor in users’ motivation, interest, and engagement, and should therefore be a key indicator of a good user experience. Nevertheless, when designing VR experiences, developers should focus on the plausibility and typicality of the experience rather than the merely the visual fidelity of it.
While visual fidelity was not related to task engagement measure it did appear to be weakly related to food choice. This appears to partially support previous work [19] in which higher visual fidelity of a food stimuli was related to a higher desire to eat that food. The response to visual fidelity in their study reflected a minimum required quality to trigger a significant desire to eat. More specifically, going from level one to two within a one (low) to seven (high) scale of visual fidelity led to no effect on desire to eat, but the increase in visual fidelity led to higher desire to eat from level two and on, with diminishing returns at top levels [19]. Our work suggests that some but not all foods may be impacted by changes in visual quality, however future studies are required to tease apart why this may be the case.
In contrast with [19], our study manipulated both foods and the surrounding virtual environment. Additionally, our study was designed to understand the effect of the complete context, considering both the foods presented and the virtual kitchen where it was displayed. We did this because both food and the environmental context can influence food behavior and desire to eat [19]. For example, the context in which food is experienced can alter its appeal [35]. Altering both the foods and the environment may be important for the congruity of the virtual experience [35]. Moreover, we rarely consume foods isolated from the environment in natural experiences, such as within sensory booths [35]. This design, however, limited our ability to parse the effects of the environment versus the effect of the food as a focal object. We could not reconcile the effect of the kitchen’s fidelity on the sense of presence of the participant, which could influence the evaluation of the foods themselves. We plan to address this problem in a future study through placing the high and low fidelity foods in different types of environments and see if that influences the reactions of participants. For example, we can place the high and low fidelity foods in a 3D environment vs an environment with a photo as backdrop vs a gray-box environment as control.
Overall, our findings show that perception of realism, that is, the believability of the visual stimuli and the task, but not the visual fidelity at which some foods are modeled, could be an important consideration. These results have implications for specific study designs and for testing in specific populations. Specifically, it may be important to assess and control for perceived realism within study designs. Additionally, if a study relies on the selection of specific foods, it may be important to pilot the stimuli prior to use within a virtual experiment to assess its fidelity and perception to ensure that the food model itself is not impacting study results. It is also important to consider that other studies have shown that visual fidelity can alter the effect of training paradigms [36]. However, this should be balanced against constraints such as processing power of available equipment and the effort, cost, and time needed to model objects precisely. Our data suggest that the perception of visual fidelity, rather than hyper realistic modeling, may be more important. Therefore, VR may be able to provide more ecologically valid conditions for learning and performing tasks [13, 32, 37, 37–41] without the need for hyper realistic food models. However, our findings are preliminary and future work is needed to understand what aspects (i.e., lighting, context, colors) improve the perception of lower-visual fidelity models. This would lead to effective model designs that could be used with a variety of technologies at a lower cost.
Strengths and limitations
A strength of remote experiments like this one is that they offer a practical solution to simulate realistic and validated interactions with food at scale, while having more nuanced results at lower cost and no food waste [25, 42, 43]. Most experiments on food selection for commensality and most human-computer interaction studies have been conducted in laboratory (lab) settings [44–48]. Lab studies have homogeneous participants, usually college students, leading to limited external validity and contradictory findings in the literature. In-person VR studies require a dedicated open space in the lab, besides the burden of scheduling and managing participant’s recruitment [49–52].
Our sample had about the gender/sex and race proportion of the general American population [53]. The groups were not balanced in terms of fullness rating at the start of the study, but we controlled the regressions for the fullness rating variable. A limitation of at-home VR studies is that it is difficult to collect precise data about the hardware and screen setup that participants used, since most people do not know this information. We did not know the screen size of each participant, but we pre-screened out those who did not have a laptop or a desktop and they could not proceed to the experiment if they did not maximize their screen. An important limitation was that the FPS was higher for the high- than for the low-fidelity VR, but both high- and low-fidelity had high FPS. The app was lightweight, so even the high-fidelity version could run in an old machine. This was pre-tested. Another technical consideration is that VR studies lack smell and touch, future studies should consider this using open-source devices for smell experiences which have been proposed as a solution [54].
In this study participants were given five minutes for the VR experience to standardize the exposure period between participants, which is a strength. However, some participants may have forgotten the foods while doing the questionnaire. Future studies should consider longer exposure periods or a back button to review the scene.
It is possible that the appearance of the food and environment from our study did not seem realistic for the participants, despite being created by a high-quality photogrammetry process. Future work may need to consider more accurate processes to capture lighting, lighting reflection, and coloring contrasts [55], and test models for higher fidelity of foods to test for the impact of realism on food choices. Moreover, additional studies are needed to test the individual effects of lighting, geometry and rendering quality separately. Another study could also test for different levels of food fidelity in the same stable environment to assess context. An immersive VR experiment, as opposed to a screen-based VR experiment such as this one, may also be needed to improve the realism and feeling of “being there”. It is crucial to understand to what extent virtual reality is perceived, in fact, as real by participants.
Conclusion
Our results indicated that perceived realism, rather than visual fidelity, was associated with motivation, interest, and engagement in a food choice task. Overall, studies should measure and control for participants’ perception of realism in virtual and immersive virtual reality tasks. Additionally, manipulating visual fidelity appears to influence the selection of some foods, therefore caution should be used in studies that rely on specific food selections as part of the study outcomes. However, further studies are needed to determine why some foods may be affected by visual fidelity while others may not.
Supporting information
S1 Table. Probit regressions for the impact of visual fidelity and perceived realism on the number of times each food was chosen.
https://doi.org/10.1371/journal.pone.0312772.s001
(DOCX)
S2 Table. Linear regressions for the impact of visual fidelity and perceived realism on the number of kilocalories chosen for each food separately.
https://doi.org/10.1371/journal.pone.0312772.s002
(DOCX)
References
- 1. Higgs S. Social norms and their influence on eating behaviours. Appetite. 2015 Mar 1;86:38–44. pmid:25451578
- 2. Köster EP. Diversity in the determinants of food choice: A psychological perspective. Food Quality and Preference. 2009 Mar 1;20(2):70–82.
- 3. Prinsen S, de Ridder DTD, de Vet E. Eating by example. Effects of environmental cues on dietary decisions. Appetite. 2013 Nov 1;70:1–5. pmid:23791633
- 4. Rozin P. The Meaning of Food in Our Lives: A Cross-Cultural Perspective on Eating and Well-Being. Journal of Nutrition Education and Behavior. 2005 Nov 1;37:S107–12. pmid:16246277
- 5. Sobal J, Bisogni CA, Jastran M. Food Choice Is Multifaceted, Contextual, Dynamic, Multilevel, Integrated, and Diverse. Mind, Brain, and Education. 2014;8(1):6–12.
- 6. Stok FM, Hoffmann S, Volkert D, Boeing H, Ensenauer R, Stelmach-Mardas M, et al. The DONE framework: Creation, evaluation, and updating of an interdisciplinary, dynamic framework 2.0 of determinants of nutrition and eating. PLOS ONE. 2017 Feb 2;12(2):e0171077. pmid:28152005
- 7. de Boer J, Hoogland CT, Boersema JJ. Towards more sustainable food choices: Value priorities and motivational orientations. Food Quality and Preference. 2007 Oct 1;18(7):985–96.
- 8.
Barjolle D, Gorton M, Milošević Đorđević J, Stojanović Ž, editors. Food Consumer Science: Theories, Methods and Application to the Western Balkans [Internet]. Dordrecht: Springer Netherlands; 2013 [cited 2023 Oct 3]. Available from: https://link.springer.com/10.1007/978-94-007-5946-6
- 9. Lindeman M, Väänänen M. Measurement of ethical food choice motives. Appetite. 2000 Feb 1;34(1):55–9. pmid:10744892
- 10. Steptoe A, Pollard TM, Wardle J. Development of a Measure of the Motives Underlying the Selection of Food: the Food Choice Questionnaire. Appetite. 1995 Dec 1;25(3):267–84. pmid:8746966
- 11. Persky S, Goldring MR, Turner SA, Cohen RW, Kistler WD. Validity of assessing child feeding with virtual reality. Appetite. 2018 Apr 1;123:201–7. pmid:29277518
- 12. Ung CY, Menozzi M, Hartmann C, Siegrist M. Innovations in consumer research: The virtual food buffet. Food Quality and Preference. 2018 Jan 1;63:12–7.
- 13. Cheah CSL, Barman S, Vu KTT, Jung SE, Mandalapu V, Masterson TD, et al. Validation of a Virtual Reality Buffet environment to assess food selection processes among emerging adults. Appetite. 2020 Oct 1;153:104741. pmid:32445771
- 14.
Bailenson J. Experience on demand: What virtual reality is, how it works, and what it can do. New York, NY, US: W. W. Norton & Company; 2018. 290 p. (Experience on demand: What virtual reality is, how it works, and what it can do).
- 15. Slater M, Spanlang B, Corominas D. Simulating virtual environments within virtual environments as the basis for a psychophysics of presence. ACM Trans Graph. 2010 Jul 26;29(4):1–9.
- 16. Parsons TD, Rizzo AA. Affective outcomes of virtual reality exposure therapy for anxiety and specific phobias: A meta-analysis. Journal of Behavior Therapy and Experimental Psychiatry. 2008 Sep 1;39(3):250–61. pmid:17720136
- 17. Fan S, Ng TT, Koenig BL, Herberg JS, Jiang M, Shen Z, et al. Image Visual Realism: From Human Perception to Machine Computation. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2018 Sep;40(9):2180–93. pmid:28866484
- 18. Cho H, Shen L, Wilson K. Perceived Realism: Dimensions and Roles in Narrative Persuasion. Communication Research. 2014 Aug 1;41(6):828–51.
- 19. Ramousse F, Raimbaud P, Baert P, Helfenstein-Didier C, Gay A, Massoubre C, et al. Does this virtual food make me hungry? effects of visual quality and food type in virtual reality. Frontiers in Virtual Reality [Internet]. 2023 [cited 2023 Oct 3];4. Available from: https://www.frontiersin.org/articles/10.3389/frvir.2023.1221651
- 20. Slater M, Khanna P, Mortensen J, Yu I. Visual Realism Enhances Realistic Response in an Immersive Virtual Environment. IEEE Computer Graphics and Applications. 2009;
- 21. Hvass J, Larsen O, Vendelbo K, Nilsson N, Nordahl R, Serafin S. Visual realism and presence in a virtual reality game. In: 2017 3DTV Conference: The True Vision—Capture, Transmission and Display of 3D Video (3DTV-CON) [Internet]. 2017 [cited 2023 Oct 3]. p. 1–4. Available from: https://ieeexplore.ieee.org/document/8280421
- 22.
Rademacher P, Lengyel J, Cutrell E, Whitted T. Measuring the Perception of Visual Realism in Images. In: Gortler SJ, Myszkowski K, editors. Rendering Techniques 2001. Vienna: Springer; 2001. p. 235–47. (Eurographics).
- 23. Mania K, Wooldridge D, Coxon M, Robinson A. The effect of visual and interaction fidelity on spatial cognition in immersive virtual environments. IEEE Transactions on Visualization and Computer Graphics. 2006 May;12(3):396–404. pmid:16640253
- 24.
Huang J, Klippel A. The Effects of Visual Realism on Spatial Memory and Exploration Patterns in Virtual Reality. In: Proceedings of the 26th ACM Symposium on Virtual Reality Software and Technology [Internet]. New York, NY, USA: Association for Computing Machinery; 2020 [cited 2023 Oct 3]. p. 1–11. (VRST ‘20). Available from: https://dl.acm.org/doi/10.1145/3385956.3418945
- 25. Sajjadi P, Edwards CG, Zhao J, Fatemi A, Long JW, Klippel A, et al. Remote iVR for Nutrition Education: From Design to Evaluation. Frontiers in Computer Science [Internet]. 2022 [cited 2023 Oct 3];4. Available from: https://www.frontiersin.org/articles/10.3389/fcomp.2022.927161
- 26.
3D Models for Professionals:: TurboSquid [Internet]. [cited 2023 Oct 3]. Available from: https://www.turbosquid.com/
- 27. Sobal J, Bisogni CA, Devine CM, Jastran M. A conceptual model of the food choice process over the life course. The psychology of food choice. 2006 Jan;1–18.
- 28. Volante M, Babu SV, Chaturvedi H, Newsome N, Ebrahimi E, Roy T, et al. Effects of Virtual Human Appearance Fidelity on Emotion Contagion in Affective Inter-Personal Simulations. IEEE Transactions on Visualization and Computer Graphics. 2016 Apr;22(4):1326–35. pmid:26780808
- 29. Tinwell A, Grimshaw M, Nabi DA, Williams A. Facial expression of emotion and perception of the Uncanny Valley in virtual characters. Computers in Human Behavior. 2011 Mar 1;27(2):741–9.
- 30. Renner B, Sproesser G, Strohbach S, Schupp HT. Why we eat what we eat. The Eating Motivation Survey (TEMS). Appetite. 2012 Aug 1;59(1):117–28. pmid:22521515
- 31. McAuley E, Duncan T, Tammen VV. Psychometric Properties of the Intrinsic Motivation Inventory in a Competitive Sport Setting: A Confirmatory Factor Analysis. Research Quarterly for Exercise and Sport. 1989 Mar 1;60(1):48–58. pmid:2489825
- 32. Allman-Farinelli M, Ijaz K, Tran H, Pallotta H, Ramos S, Liu J, et al. A Virtual Reality Food Court to Study Meal Choices in Youth: Design and Assessment of Usability. JMIR Formative Research. 2019 Jan 9;3(1):e12456. pmid:30684440
- 33. Masterson TD, Gilbert-Diamond D, Lansigan RK, Kim SJ, Schiffelbein JE, Emond JA. Measurement of external food cue responsiveness in preschool-age children: Preliminary evidence for the use of the external food cue responsiveness scale. Appetite. 2019 Aug 1;139:119–26. pmid:31047939
- 34.
Bureau UC. Census.gov. [cited 2023 Oct 3]. Income in the United States: 2021. Available from: https://www.census.gov/library/publications/2022/demo/p60-276.html
- 35. Picket B, Dando R. Environmental Immersion’s Influence on Hedonics, Perceived Appropriateness, and Willingness to Pay in Alcoholic Beverages. Foods. 2019 Feb;8(2):42. pmid:30691117
- 36.
Ragan ED, Bowman DA, Kopper R, Stinson C, Scerbo S, McMahan RP. Effects of field of view and visual realism on virtual reality training effectiveness for a visual scanning task. IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS.
- 37. Siegrist M, Ung CY, Zank M, Marinello M, Kunz A, Hartmann C, et al. Consumers’ food selection behaviors in three-dimensional (3D) virtual reality. Food Research International. 2019 Mar 1;117:50–9. pmid:30736923
- 38. Hoenink JC, Mackenbach JD, Waterlander W, Lakerveld J, van der Laan N, Beulens JWJ. The effects of nudging and pricing on healthy food purchasing behavior in a virtual supermarket setting: a randomized experiment. International Journal of Behavioral Nutrition and Physical Activity. 2020 Aug 3;17(1):98. pmid:32746928
- 39. Xu C, Demir-Kaymaz Y, Hartmann C, Menozzi M, Siegrist M. The comparability of consumers’ behavior in virtual reality and real life: A validation study of virtual reality based on a ranking task. Food Quality and Preference. 2021 Jan 1;87:104071.
- 40. Yaremych HE, Kistler WD, Trivedi N, Persky S. Path Tortuosity in Virtual Reality: A Novel Approach for Quantifying Behavioral Process in a Food Choice Context. Cyberpsychology, Behavior, and Social Networking. 2019 Jul;22(7):486–93. pmid:31241349
- 41. van der Waal NE, Janssen L, Antheunis M, Culleton E, van der Laan LN. The appeal of virtual chocolate: A systematic comparison of psychological and physiological food cue responses to virtual and real food. Food Quality and Preference. 2021 Jun 1;90:104167.
- 42. Blascovich J, Loomis J, Beall AC, Swinth KR, Hoyt CL, Bailenson JN. Immersive Virtual Environment Technology as a Methodological Tool for Social Psychology. Psychological Inquiry. 2002 Apr 1;13(2):103–24.
- 43. Cheah CSL, Kaputsos SP, Mandalapu V, Tran T, Barman S, Jung SE, et al. Neurophysiological Variations in Food Decision-Making within Virtual and Real Environments. In: 2019 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI) [Internet]. 2019 [cited 2023 Oct 3]. p. 1–4. Available from: https://ieeexplore.ieee.org/document/8834497
- 44.
Gustarini M, Ickin S, Wac K. Evaluation of challenges in human subject studies “in-the-wild” using subjects’ personal smartphones. In: Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publication [Internet]. New York, NY, USA: Association for Computing Machinery; 2013 [cited 2023 Oct 3]. p. 1447–56. (UbiComp ‘13 Adjunct). Available from: https://dl.acm.org/doi/10.1145/2494091.2496041
- 45. Henze N, Pielot M, Poppinga B, Schinke T, Boll S. My App is an Experiment: Experience from User Studies in Mobile App Stores. International Journal of Mobile Human Computer Interaction (IJMHCI). 2011;3(4):71–91.
- 46.
Mottelson A, Hornbæk K. An affect detection technique using mobile commodity sensors in the wild. In: Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing [Internet]. New York, NY, USA: Association for Computing Machinery; 2016 [cited 2023 Oct 3]. p. 781–92. (UbiComp ‘16). Available from: https://dl.acm.org/doi/10.1145/2971648.2971654
- 47.
Reinecke K, Gajos KZ. LabintheWild: Conducting Large-Scale Online Experiments With Uncompensated Samples. In: Proceedings of the 18th ACM Conference on Computer Supported Cooperative Work & Social Computing [Internet]. New York, NY, USA: Association for Computing Machinery; 2015 [cited 2023 Oct 3]. p. 1364–78. (CSCW ‘15). Available from: https://dl.acm.org/doi/10.1145/2675133.2675246
- 48. Schulte EM, Avena NM, Gearhardt AN. Which Foods May Be Addictive? The Roles of Processing, Fat Content, and Glycemic Load. PLOS ONE. 2015 Feb 18;10(2):e0117959. pmid:25692302
- 49. Kourtesis P, Korre D, Collina S, Doumas LAA, MacPherson SE. Guidelines for the Development of Immersive Virtual Reality Software for Cognitive Neuroscience and Neuropsychology: The Development of Virtual Reality Everyday Assessment Lab (VR-EAL), a Neuropsychological Test Battery in Immersive Virtual Reality. Frontiers in Computer Science [Internet]. 2020 [cited 2023 Oct 3];1. Available from: https://www.frontiersin.org/articles/10.3389/fcomp.2019.00012
- 50. Ma X, Cackett M, Park L, Chien E, Naaman M. Web-Based VR Experiments Powered by the Crowd. In: Proceedings of the 2018 World Wide Web Conference [Internet]. Republic and Canton of Geneva, CHE: International World Wide Web Conferences Steering Committee; 2018 [cited 2023 Oct 3]. p. 33–43. (WWW ‘18). Available from: https://dl.acm.org/doi/10.1145/3178876.3186034
- 51.
Mottelson A, Hornbæk K. Virtual reality studies outside the laboratory. In: Proceedings of the 23rd ACM Symposium on Virtual Reality Software and Technology [Internet]. New York, NY, USA: Association for Computing Machinery; 2017 [cited 2023 Oct 3]. p. 1–10. (VRST ‘17). Available from: https://dl.acm.org/doi/10.1145/3139131.3139141
- 52.
Saffo D, Yildirim C, Di Bartolomeo S, Dunne C. Crowdsourcing Virtual Reality Experiments using VRChat. In: Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems [Internet]. New York, NY, USA: Association for Computing Machinery; 2020 [cited 2023 Oct 3]. p. 1–8. (CHI EA ‘20). Available from: https://dl.acm.org/doi/10.1145/3334480.3382829
- 53.
U.S. Census Bureau QuickFacts: United States [Internet]. [cited 2023 Oct 3]. Available from: https://www.census.gov/quickfacts/fact/table/US/RHI125222
- 54.
Javerliat C, Elst PP, Saive AL, Baert P, Lavoué G. Nebula: An Affordable Open-Source and Autonomous Olfactory Display for VR Headsets. In: Proceedings of the 28th ACM Symposium on Virtual Reality Software and Technology [Internet]. New York, NY, USA: Association for Computing Machinery; 2022 [cited 2023 Oct 3]. p. 1–8. (VRST ‘22). Available from: https://dl.acm.org/doi/10.1145/3562939.3565617
- 55. Paakki M, Sandell M, Hopia A. Visual attractiveness depends on colorfulness and color contrasts in mixed salads. Food Quality and Preference. 2019 Sep 1;76:81–90.