When designing a new user interaction (UI) technology or applying UI modalities to a product or system, the designer can select from many methods and tools to assist them with evaluating the UI’s function and appeal with end users. Testing early in a design process is highly desirable since any issues found can be resolved more easily and often at less expense. However, for lighting solutions, these methods and tools are less suitable due to the qualities of light as a medium. Light is often detached from the UI itself and the light output is generally experienced throughout an environment which is often encompassing the users. For example, testing a new UI to control a yet to be installed media façade is not a simple system to mock up in advance, due to scale and cost. There is a need therefore, within the lighting industry, to have tools or methods with which design teams can test lighting UI, in conjunction with the light output, early in the design process. A potential solution is to use virtual environments. These would provide designers with a space in which they can show virtual light output that can be controlled using any developmental UI; this would enable them to evaluate lighting UI much earlier and potentially in more detail than is currently possible. In this paper we report on a user study that compares three different environments (physical, virtual CAVE and screen) in a bid to determine whether the virtual environments could provide reliable evaluations of UI for lighting versus a real setup. Our findings show that virtual environments indeed have the potential to elicit similar evaluative feedback from end users as a real environment when considering the functional utilitarian elements..