A virtual reality food court to study meal choices in youth: Design and assessment of usability

Background Regular consumption of take-out and fast foods with sugary drinks is associated with poor quality diets and higher prevalence of obesity. Among the settings where such food is consumed is the food court typically found in shopping malls prominent in many countries. Objective The objective of this research was to develop a virtual reality food court that could be used to test food environmental interventions, such as taxation, and ultimately to facilitate the selection of healthier food choices. Methods Fourteen food courts in Sydney, Australia were selected to include those in the city center and suburbs of high and low socioeconomic status. Researchers visited the courts to collect information on number and type of food outlets, all menu items for sale, cost of foods and beverages and sales promotions. This information was used to assemble 14 food outlets typically found in food courts, and representative menus were compiled. The UNITY gaming platform was used to design a virtual reality food court that could be used with HTC VIVE goggles. Participants navigated the virtual reality food court using the head-mounted display, keyboard, and mouse and selected a lunch meal, including food and beverage. A validated questionnaire on presence within the virtual reality food court and system usability was completed at the end of the session. The constructs for presence included a sense of control, sensory fidelity, realism, distraction, and involvement. Questions were rated on a scale from 1 (worst) through 7 (best) for each of 28 questions giving a maximum total score of 196. The systems usability scale (SUS) that gives a final score out of 100 was also assessed. Results One hundred and sixty-two participants with a mean age of 22.5 (SD 3.1) years completed the survey. The mean score for total presence was 144 (SE 1.4) consisting of control: 62.1 (SE 0.8), realism: 17.5 (SE 0.2), involvement: 9.6 (SE 0.2), sensory fidelity: 34.9 (SE 0.4), and distraction: 24.0 (SE 0.3). The mean SUS was 69 (SE 1.1). Conclusions Virtual reality shows promise as a tool to study food choice for test interventions to inform practice and policy.