Evaluation of Selection Techniques on a Mobile Augmented Reality Game

Selection is one of the most fundamental tasks performed by humans. In Virtual Reality (VR), a number of interaction techniques have been proposed to deal with this type of task. They can be based on different metaphors, that define how the interaction is performed. However, it is still not clear how these different metaphors perform under Augmented Reality (AR). We performed two experiments to evaluate two of those interaction techniques. One was based on a pointing metaphor, and the other on a surface interaction metaphor. To achieve this result, a game prototype was developed, where users had to apply the selection techniques to reach a goal and win. We evaluated the impact of two different types of devices and two types of crowd simulation (with and without flocking behavior) in the user interaction. The scenario allowed the user to easily navigate around the scene from an overview perspective. We performed a two-way ANOVA between the variables and found that the technique based on surface interaction metaphor allowed the users to select multiple objects in the shortest amount of time. However, the technique based on pointing metaphor had a higher accuracy. Moreover, both techniques performed better with a crowd with flocking behavior.