It is claimed that emerging technologies, in particular Augmented Reality?(AR), offer new perspectives for training and learning. Nevertheless, actual empirical results do not consistently report a benefit of using AR?applications for training. This article presents a literature review of the ergonomics of?AR for learning and training. On the basis of some current empirical results, we discuss the advantages, the real contributions and the problems of this technology in educational and training settings.
The first part of the paper briefly presents the concept of?AR and the potential of this technology to assist learning. The second part presents a discussion on the dominant orientation of current research in the field of?AR, which is essentially technology-driven. In the third part of the paper, we discuss the advantages and disadvantages of another approach to?AR, which we identified in the literature, that is to say the user-centered approach. Within this framework, AR is considered and evaluated as a learning resource and as an aid to learners' and tutors' activities. In this case, the crucial research questions concern the learning and training objectives implemented within current AR?prototypes, the target end-users' profiles, the pedagogical configurations actually available, and the sensory modalities through which users conduct dialogue.
The fourth part of the paper presents a number of current empirical results on the ergonomics of AR?systems used for training and education. Broadly speaking, empirical results on the ergonomics of AR?systems for learning fall into two categories?: empirical data on the usability of the applications, and studies that try to evaluate the effectiveness of?AR as a training aid (i.e.?its effectiveness for comprehension, retention, transfer,?etc.). There is no clear evidence demonstrating the utility of?AR for learning. Nevertheless, AR seems useful for assisting divided attention in multiple tasks, as well as for assisting the recall of information. We suggest that the limited number of conclusive results on the utility of?AR for learning is partly due to the lack of user-centered and learner-centered design and evaluations of AR?systems. In conclusion, we emphasize the importance of such an orientation and discuss some relationships between usability and learnability.