The appeal of virtual chocolate: A systematic comparison of psychological and physiological food cue responses to virtual and real food

The Difference between PC-Based and Immersive Virtual Reality Food Purchase Environments on Useability, Presence, and Physiological Responses

Computer simulations used to study food purchasing behavior can be separated into low immersion virtual environments (LIVE), which use personal computers and standard monitors to display a scene, and high immersion virtual environments (HIVE) which use virtual reality technology such as head-mounted displays to display a scene. These methods may differ in their ability to create feelings of presence or cybersickness that would influence the usefulness of these approaches. In this present study, thirty-one adults experienced a virtual supermarket or fast-food restaurant using a LIVE system or a HIVE system. Feelings of presence and cybersickness were measured using questionnaires or physiological responses (heart rate and electrodermal activity). The participants were also asked to rate their ability to complete the set task. The results of this study indicate that participants reported a higher sense of presence in the HIVE scenes as compared to the LIVE scenes (p < 0.05). The participant's heart rate and electrodermal activity were significantly higher in the HIVE scene treatment when compared to the LIVE scene (p < 0.05). There was no difference in the participant's ability to complete tasks in the different scenes. In addition, feelings of cybersickness were not different between the HIVE and LIVE scenes.

Does virtual reality enhance the effects of inhibitory control training for loss-of-control eating? A pilot factorial experiment

Inhibitory control, one's ability to inhibit automatic responses to desirable stimuli, may be inadequately targeted in interventions for loss-of-control eating (LOC). Promising evidence has identified inhibitory control trainings (ICTs) as an avenue to target inhibitory control directly; however, effects of ICTs on real-world behavior are limited. Compared to typical computerized trainings, virtual reality (VR) presents several potential advantages that may address key shortcomings of traditional ICTs, i.e. poor approximation to everyday life. The present study utilized a 2 × 2 factorial design of treatment type (ICT vs sham) by treatment modality (VR vs standard computer), which allows for increased statistical power by collapsing across conditions. Our primary aim was to examine the feasibility and acceptability of six weeks of daily training among groups. A secondary aim was to preliminarily assess main and interactive effects of treatment type and modality on target engagement and efficacy (i.e., training compliance, change in LOC episodes, inhibitory control, and implicit liking of foods). Participants (N = 35) with ≥1×/weekly LOC were assigned to one of four conditions and completed ICTs daily for six weeks. The trainings were feasible and acceptable, evinced by high retention and compliance across time and condition. Although completing daily trainings across treatment types and modalities was associated with large decreases in LOC, there were no meaningful effects of either treatment type or modality, nor a significant interaction effect, on LOC or mechanistic variables. Future research should aim to increase the efficacy of ICT (both standard and VR-based) and test in fully-powered clinical trials.

Exploring the effects of habituation and scent in first-person 360-degree videos on consumption behavior

Although immersive virtual environments can influence food-related thoughts, emotions and behavior, the influence of repeated exposure to food cues in such environments has rarely been explored. This study seeks to understand if habituation, a decrease in one's physiological and behavioral response that results from repeated simulation, can take place while repeatedly watching 360-degrees of food being consumed. The influence of scent as an olfactory cue is further explored, based on past research on embodied cognition. In Study One (n = 42), participants who viewed 30 repetitions of someone eating an M&M ate significantly fewer M&Ms than those who viewed three repetitions. Study Two (n = 114) used a 2 (behavior: eating M&M/inserting a coin) × 2 (repetitions: 3/30) between-subjects experiment to confirm that results from Study One were due to habituation of the consumption video, finding that there were only significant differences between repetitions in the M&M condition. Finally, Study Three (n = 161) comprised a 2 (repetition: 3/30) × 2 (scent: present/absent) between-subjects experiment. Participants in the 30-repetition condition and those in the scent-present condition ate significantly fewer M&Ms respectively, but no interaction effects were found. The theoretical and practical implications of these findings are discussed.

Food cue reactivity: Neurobiological and behavioral underpinnings

The modern obesogenic environment contains an abundance of food cues (e.g., sight, smell of food) as well cues that are associated with food through learning and memory processes. Food cue exposure can lead to food seeking and excessive consumption in otherwise food-sated individuals, and a high level of food cue responsivity is a risk factor for overweight and obesity. Similar food cue responses are observed in experimental rodent models, and these models are therefore useful for mechanistically identifying the neural circuits mediating food cue responsivity. This review draws from both experimental rodent models and human data to characterize the behavioral and biological processes through which food-associated stimuli contribute to overeating and weight gain. Two rodent models are emphasized - cue-potentiated feeding and Pavlovian-instrumental transfer - that provide insight in the neural circuits and peptide systems underlying food cue responsivity. Data from humans are highlighted that reveal physiological, psychological, and neural mechanisms that connect food cue responsivity with overeating and weight gain. The collective literature identifies connections between heightened food cue responsivity and obesity in both rodents and humans, and identifies underlying brain regions (nucleus accumbens, amygdala, orbitofrontal cortex, hippocampus) and endocrine systems (ghrelin) that regulate food cue responsivity in both species. These species similarities are encouraging for the possibility of mechanistic rodent model research and further human research leading to novel treatments for excessive food cue responsivity in humans.

Remote iVR for Nutrition Education: From Design to Evaluation

While different crowdsourcing platforms promote remote data collection, experiments in the immersive Virtual Reality (iVR) research community are predominantly performed in person. The COVID-19 pandemic, however, has forced researchers in different disciplines, including iVR, to seriously consider remote studies. In this paper, we present a remote study using the Immersive Virtual Alimentation and Nutrition (IVAN) application, designed to educate users about food-energy density and portion size control. We report on the results of a remote experiment with 45 users using the IVAN app. In IVAN, users actively construct knowledge about energy density by manipulating virtual food items, and explore the concept of portion size control through hypothesis testing and assembling virtual meals in iVR. To explore the feasibility of conducting remote iVR studies using an interactive health-related application for nutrition education, two conditions were devised (interactive vs. passive). The results demonstrate the feasibility of conducting remote iVR studies using health-related applications. Furthermore, the results also indicate that regardless of level of interactivity learners significantly improved their knowledge about portion size control after using the IVAN (p &lt; 0.0001). Adding interactivity, however, suggests that the perceived learning experience of users could be partially affected. Learners reported significantly higher scores for immediacy of control in the interactive condition compared to those in the passive condition (p &lt; 0.05). This study demonstrates the feasibility of conducting an unsupervised remote iVR experiment using a complex and interactive health-related iVR app.

Development and Evaluation of a Virtual Reality Puzzle Game to Decrease Food Intake: Randomized Controlled Trial

Background

Virtual reality (VR) has gained popularity in daily life, and VR food cues seem to elicit food cravings, similar to real food cues. However, little is known about the impact of VR food cues on actual food intake.

Objective

In real life (RL), exposure to food cues in a situation in which the desire to eat food interferes with the completion of a food-related task reduces the subsequent food intake (ie, the pre-exposure effect). In this study, we examine, on the one hand, whether the pre-exposure effect could be replicated in RL and, on the other hand, whether this effect could be extended to VR contexts.

Methods

The study used a 2 (stimulus type: food vs nonfood) × 2 (mode: VR vs RL) between-subject design (n=175). Participants were randomly assigned to 1 of the 4 conditions.

Results

We found the main effect of mode on food intake, with a higher food intake after both VR conditions than after RL conditions (P=.02). In addition, among female participants, we found that exposure to both food cues (ie, VR and RL) resulted in lower food intake than exposure to both nonfood cues (P=.05). In contrast, this effect was not observed among male participants (P=.34). Additionally, VR and RL cues generated similar emotional and behavioral responses (eg, arousal and game difficulty).

Conclusions

We were unable to replicate the exposure effect in our complete sample. Subgroup analyses, however, showed that for women, exposure to food cues (either in VR or in RL) reduces food intake, indicating that a VR pre-exposure procedure may effectively be applied exclusively for women.

Trial Registration

ClinicalTrials.gov NCT05169996; https://clinicaltrials.gov/ct2/show/NCT05169996

Examining the neural correlates of goal priming with the NeuroShop, a novel virtual reality fMRI paradigm

OBJECTIVE

Health goal priming has been shown to stimulate healthy food choices by activating an individual's weight-control goal. The present study combined fMRI with a novel virtual reality food choice task to elucidate the underlying neural mechanisms of health goal priming. Previous research has suggested that the ventromedial prefrontal cortex (vmPFC) and dorsolateral prefrontal cortex (dlPFC) play a role in the incorporation of health considerations into the food choice process. Responses may be more representative for those found in real life when assessed in an environment similar to the actual choice environment. Therefore, the first aim of the study was to explore if a novel virtual reality food choice task is sufficiently sensitive to detect basic valuation processes in food choice. The second aim was to examine whether increased activation in the dlPFC drives the effects of health goal priming.

METHODS

Fifty-six female participants performed an fMRI food choice task embedded in a virtual supermarket environment. They chose between perceived healthy and unhealthy products in a health prime, hedonic prime, and non-food control condition, while activation in brain areas involved in self-control and valuation (vmPFC, dlPFC) was assessed.

RESULTS

There were no differences in relative preference for perceived healthy products over unhealthy products between the conditions. There were also no main effects of prime condition on brain activation in the vmPFC and dPFC during food choice. Across conditions, activation in the vmPFC correlated with the tastiness of the chosen product during food choice.

CONCLUSIONS

Although the study does not provide support for health goal priming triggering neural self-control mechanisms, results did show that virtual reality has potential for a more realistic fMRI food choice paradigm.

Investigating the relative merits of using a mixed reality context for measuring affective response and predicting tea break snack choice

Sensory evaluation for the investigation of food consumption is often conducted in a controlled laboratory environment, which does not reflect consumption behaviour in real world. Here, we compared the effect of consumption setting (traditional sensory booth, mixed reality projection café, and a café) on consumer affective responses, and to investigate the effectiveness of using Microsoft HoloLens technology, an Augmented Mixed Reality device, as an ecologically valid alternative to natural consumption eating for sensory evaluation. Participant [(n = 120): 86 females/34 males, aged 18-65 years] affective response (overall liking, attribute liking, emotional response, and snack choice) towards two commercially available tea break snacks (caramel slice and chocolate digestive biscuit) was assessed in three different consumption settings using a balanced crossover design. There were no significant differences for most affective ratings between data obtained from the HoloLens evoked café and real café (p ≥ 0.10), suggesting that mixed reality could provide an ecologically valid context for consumer research. However, response differences were observed between these two contexts and the sensory booths. For example, interested, joy, enthusiastic emotion terms were rated slightly higher in the evoked café in comparison to the booth context and slightly higher emotional engagement was observed for joy in the café compared to the booths (all p < .10). This study highlights key considerations for deciding where consumer testing should be conducted and the importance of using a combination of overall liking, attribute liking and emotional response to obtain data representative of real-world environments in consumer studies.