With growing technological advancements driving the progress of virtual reality and augmented reality, a recent study observed behavioral impacts in the applications of AR in a manufacturing setting.
Professor of operations and information management Enno Siemsen co-authored the paper. He said the goal of the study was to observe how AR changes the nature of work for people in manufacturing.
“I knew that there were clear benefits to using the technology,” Siemsen said. “But every time you go on a project like this you also want to understand what are the possible drawbacks, right?”
The study was conducted in a German manufacturing company, where they tested workers’ speed in performing new tasks with and without the aid of AR instructions. Siemsen said they found that workers who used AR support were more productive than those using traditional printed instruction. But, when the researchers removed the instructions those who used AR struggled to continue with the task compared to others.
Another important aspect of the study was the concept of process improvement. People using AR instructions could not provide beneficial input for improving manufacturing processes, Siemsen said, while the printed instructions group came up with constructive feedback immediately.
“This experiment suggests that, in terms of sort of innovation and continuous improvement to the process, you’ll suffer if you don’t implement augmented reality in a smart way,” Siemsen said.
The study was only conducted over a short period of time, but it hints at the potential long-term behavioral effects of AR. Siemens said possible ways of overcoming the innovation and learning drawbacks of AR in manufacturing could be limited use of the tech within a team or programming AR as learning devices rather than a means of step-by-step instructions.
Siemsen pointed to a phenomenon called the productivity dilemma, in which learning about a process decreases as that process becomes more standardized for immediate efficiency. One example of this can be seen in GPS navigation, where step-by-step instructions for a journey don’t allow the user to build the mental model they would create when using a map to navigate themselves, according to Siemsen.
“Industry 4.0 is all about digitization and bringing in technology and so on,” Siemsen said. “But ultimately, it will never entirely replace human beings in the workforce, and we really want to make sure that the ingenuity of a frontline worker’s creativity and their potential is still there and still a key element of every sort of manufacturing process.”
Siemsen said the study is not trying to say not to use AR technology, but rather to be cautious and aware of how this technology interacts with the people that use it.
Using print instructions reduces the time it takes to complete a task the more it’s done — this is called the power law of learning. But using AR instructions causes people to get worse at the task when the instructions are removed, Siemsen said, which shows rare evidence of breaking the learning curve.
“It demonstrates how strong the effect is,” Siemsen said. “That this is not a trivial small, kind of effect that would just wash out with a learning curve. It’s a pretty strong effect, so I don’t celebrate this as a good thing. I celebrate this as an unusual empirical phenomenon — you usually don’t break the learning curve.”
AR technology can be applied to other fields as well, Siemsen said, and this study serves as a cautionary tale. In settings like the medical field, it could be dangerous to rely on step-by-step instructions.
Graduate student Jingjie Li works in the Wisconsin Embedded Systems and Computing Lab with a focus on privacy and security. A current project at WISEST is a medical application of AR called readyVue.
Li said readyVue is looking to expand on current products that can be used in surgery training to give information on human organs of interest.
One implication to consider in VR and AR technology is privacy, Li said, as these devices can collect biometric data of the user as well as the environment and surroundings. This leaves room for sensitive information to be stored and sold to unwanted third parties for advertisement — especially in medical settings.
Li is working to give users a means of control over what data collected by AR and VR technology is secure. With advanced technology, biometric and environmental data can be used to infer personality traits or emotions.
“The malicious party could also use these techniques to exploit private attributes that weren’t revealed in the previous old days,” Li said. “This is something important for our research as well. To nudge the user — to inform them of this privacy risk before addressing them.”
Li said his angle of research is aiming to give a generalizable framework for privacy in AR for all sorts of scenarios, like gaming or healthcare.
While VR and AR have been around for a while, the advancement of this technology is leading to a variety of applications. No matter the setting, researchers are working to see potential tradeoffs as more and more people use this technology. Whether its privacy or behavioral implications, studying the drawbacks of this technology will be important before the widespread adoption of mixed reality both researchers’ work shows.