Training Via Augmented Reality
August 23, 2017 by Bill Rosenthal

A worldwide survey this year of 26,000 consumers found that in 2016, 48% of respondents were driven by access to virtual reality (VR) or augmented reality (AR) in making smartphone purchase decisions. In 2017, the proportion rose to 54%. The survey, which was performed by Accenture, is summarized by an infographic on Accenture’s site, and it’s worth your time to go have a look at it.

But I am particularly interested in this rapidly growing phenomenon of people buying smartphones to get access to VR or AR, because AR, it turns out, has the potential to influence training. Last week, an article on the Training Industry website reported on a study showing that AR can beat both traditional paper-based instructions and static 3D representations in a training application.

Unlike VR, which requires a headset to sort of seal the user into a virtual world, AR is the overlay of digital information on the real world. It might be done with a smart lens (think fighter pilot’s heads-up display or Google Glass) or through a smartphone, like the Pokemon Go game. 

The Training Industry article reported on a study that was presented at the 2016 IEEE International Symposium on Mixed and Augmented Reality: a group of men and women, aged 16 to 65, were divided into three groups and given the task to set up a Lego™ lighthouse. One group received conventional written instructions, one group received a static 3D representation of the task, and one group received dynamic 3D AR instructions that included the motions needed to perform the assembly.

The group that received instructions via AR performed the fastest. Second place went to conventional written instructions, and static 3D representation came in last. It’s interesting to me, and somewhat unexpected, that conventional written instructions beat static 3D representation.

Why is AR so effective? For one thing, AR users were able to overlay the tasks with the instructions. That is, they were able to work without having to divide their attention between the instructions and what they were doing. In other words, they spent less effort on translating the instructions into actions. Second, they got the instructions as they were needed, saving them time. How many times do you refer back to the instructions when you are assembling a toy or a piece of furniture? Those trips back to the instructions apparently add up fairly quickly.

I think the retail sector should take note of these findings, at least those businesses that sell products requiring assembly, such as furniture and small appliances. There has been a trend lately of supplying assembly instructions that substitute static 3D representation for written directions. I suppose they do this to save the expense of translation in an age of global commerce, but they should note they are putting the assembly process (and with it, customer satisfaction) at risk. Customers on the whole might assemble the products more efficiently and more easily with written instructions.

Beyond retail considerations, however, any time you talk about instructions, you are talking about training. While it’s true that there aren’t many businesses based on building with Legos, all manufacture involves assembly at some point, and AR is clearly an effective technology for training assemblers. But I think it could also apply to any activity that involves manipulation of real-world objects, and perhaps one day someone will figure out how to apply it to mental work as well. Training developers, it’s time to start learning about AR.