“Be careful what you wish for.”
When it comes to using virtual reality to run simulations, people often wish for it to mimic reality exactly.
But what if you could make virtual reality better than reality by giving the user more information?
This is the question that researchers Aaron Joseph, James Schreiner, and ISE professor Deborah Thurston, asked in their paper "Impact of Visual Models on Risk Attitude and Decision Trade-offs,” which recently won the Best Paper Award in Manufacturing and Design at the 2015 Industrial and Systems Engineering Research Conference (ISERC).
To see if they could improve virtual reality simulations to help users make better decisions, they ran an experiment involving a small group of engineering students using a virtual reality system that simulated repairing damage on an airplane part.
According to Thurston, repairing airplane damage used to be pretty straightforward when planes were made primarily out of aluminum. Now planes are typically made up of polymer composites, which adds more room for mistakes and more steps in the decision-making process.
Their experiment tested three hypotheses: could adding more aspects to the virtual reality help the engineer comprehend what their tradeoffs are; could it affect their stated willingness to make tradeoffs; and could it affect how willing they are to make a risk?
They let the participating students approach an example scenario given to them in a virtual reality simulation — a part of a tail wing had been damaged by hail — different from the traditional way. This way typically just gave them a description of the damage and data about how much the repair would cost, the size of the damage, etc.
Instead, the students were given even more information, including photos of the damage and a spreadsheet they could use to weigh options mathematically. The students could look at this information while they worked on the problem within the simulation.
“The basic question we were asking was: if we use the virtual reality system and show these things within the system, does this influence the decision?” Thurston said.
They found that it did, proving all three of their hypotheses to be correct. According to Thurston, it strengthened their understanding of tradeoff options and increased their willingness to make tradeoffs and risks.
Thurston said the fact that 82% of the subjects changed their willingness to make tradeoffs is an interesting finding. She explained how in formal decision making, the user typically finds out what is important to them within the choices they are given.
“We traditionally think of those as they are what they are, we can’t change them,” Thurston said. “But we’ve shown with this, that they can be changed or influenced by a simple thing like showing them of a photo of the thing they’re thinking about.”
Therefore, they found that having just a virtual reality scenario on its own is not enough for optimal decision making. These findings could also apply to much more than just the aviation industry, Thurston said.
Thurston said their way is “enhancing virtual reality and giving the engineer more information than they have in the real, physical world.”
Thurston said this research is important because it is typically thought that computerizing or automating a process will make it significantly better. But that is not always the case.
“You might be automating something that doesn’t work, or something that’s really inefficient,” she said.
Instead, enhancing the system could prevent people from making mistakes they used to make, as their research found.
“Our tagline on the proposal was ‘be careful what you wish for,’” Thurston said. “There’s that whole idea, that if you automate all this stuff … what people typically wish for is to make virtual reality exactly mimic physical reality. And doing that might not be the best thing.”