We’ve been talking a lot about VR lately, but the world of XR content and experiences is wide and inclusive of mixed reality (MR), augmented reality (AR), and more, each with its own set of strengths and applications. MR is a broad term that describes content situated "anywhere between the extrema of the virtuality continuum". AR, on the other hand, specifically describes the addition of a new layer of visual information onto a real-world experience, whether that information is in the form of a HUD, character models, animations, or some other visual cue. One application we’re particularly excited about is the intersection of AR and job training - the ability to layer information onto a task in real-time has huge implications for training the working world. Fortunately, research is starting to emerge that provides evidence for this potential.
Accenture Labs and Meta (an AR company) recently released a study of AR that used another playful medium - that is, Lego. Respondents in this study were split into three groups tasked with completing a Lego lighthouse set, and each group worked with differentiated instructional models. One group used standard paper instructions, one group used a static (non-moving) 3D AR model of the instructions, and one group used a dynamic (moving) 3D AR model of the instructions, which also demonstrated the movements the respondents would have to make in order to assemble the lighthouse.
The study generated four key findings:
Respondents using the dynamic AR instructions were able to more quickly complete each step. This suggests that modeling the motions of assembly was critical to increasing their velocity.
Participants using static AR instructions were the slowest. This was surprising to the researchers; they had assumed that the instructions would increase in usefulness as they gained fidelity and complexity, but this result suggested that there are still advantages for a 2D representation over a 3D representation, in some cases.
Regardless of the instruction types participants used, there was no statistically significant difference across perceived speed, helpfulness, and effectiveness. This suggested that the benefits respondents experienced were unconscious; that is, the respondents did not perceive that they were going faster than normal.
Participants overwhelmingly (86%) preferred AR instructions over paper instructions. In spite of the last bullet that suggests that respondents are unaware of the AR benefits as they experience them, participants nevertheless reported a strong preference for AR.
The ability to co-locate training information with the targeted task has many benefits not listed here; the demand on working memory is decreased, ambiguity around proper assembly motions is reduced, focus and velocity correspondingly increase, and so on. Taken as a whole, these findings suggest that the working world should pay close attention to AR as an instructional medium for job training.