Skip to content
Who's in the Video
Jeremy Bailenson is founding director of Stanford University’s Virtual Human Interaction Lab, Thomas More Storke Professor in the Department of Communication, Professor (by courtesy) of Education, Professor (by courtesy) Program in[…]

Virtual reality isn’t just for gamers and tech hobbyists anymore, it’s also for NFL players. In 2015, Stanford professor Jeremy Bailenson and recent Stanford graduate Derek Blech co-founded a VR training company called STRIVR and in the first six months, they had signed five NFL teams on multiyear contracts as well as about a dozen college teams. What did this tech do for the NFL? It made football a game of the mind like never before, providing players a safe virtual space to make mistakes, learn from feedback, do mental repetitions, and practice communication and decision-making skills. Those same lessons apply to many jobs and industries, and where STRIVR went next was Walmart, implementing VR learning in 200 training academies around the U.S., helping over 150,000 employees improve their customer service skills, look for shoplifters, and prepare for the stress-fest that is the Black Friday sales. Whether you’re a quarterback or a cashier, VR training can raise your potential and you can safely expect it to be part of your future job training in the coming years. Jeremy Bailenson is the author of Experience on Demand: What Virtual Reality Is, How It Works, and What It Can Do.

Jeremy Bailenson: If you think about where we get virtual reality from, there’s something called a flight simulator. In 1929, Edwin Link said he didn’t want to learn how to fly from a book, but flying a plane is very expensive in terms of making a mistake; make a mistake in a plane people die and. obviously, planes get lost. So Edwin Link developed the flight simulator so people could learn by doing in a safe environment.

One of the ways the general public that people who aren't just gamers or technologists are seeing VR is in training. So one thing I've done for the last few years is I've used virtual reality to train athletes. The project began as a master's thesis by Derek Belch, who's a student at Stanford.

We use virtual reality to train quarterbacks to look around, recognize a defensive pattern, make a decision by changing the play—they can keep the original play or they can kill, kill, kill and go down to the next play in the queue.

When Derek graduated in 2015 he founded a company called STRIVR, and STRIVR in the first six months signed five NFL teams to multiyear contracts, about a dozen college teams. And what we've seen over the last few years is many teams adapting using VR so that players can get extra mental repetitions. Now, where this goes down to your everyday person, is the lessons that we learned by training athletes it turns out applies to just about every job.

So think about your own job. You have to look around, you have to see stuff—we call this recognition, pattern recognition—then you have to make a decision and then you have to communicate that decision. So, for a quarterback, he looks around, spots the defense, sees a pattern, changes a play. When he changes the play he calls that out to his teammates. That lesson, that general pattern, applies to just about everybody's job, and the exciting thing for me has been to watch Walmart. So Walmart we began training one of their academies. So Walmart has 200 training academies and basically, if you work at Walmart at any time you can get in your car, drive a few hours and you get to go and train for a week or so at one of these academies.

We started out in one of them where we put VR there and what we were training are things like holiday rush, Black Friday, where there are people everywhere running around and yelling at you and it's this really intense experience, giving employees a sense for what that's going to be like, or having them look around the store to spot safety violations or customers who haven't been helped. The same lessons that we use for quarterbacks in that first training academy, qualitatively we were finding that it was a good solution and that people were enjoying it, and the training was working.

We then went up to 30 training academies and what we had was 30 training academies use VR and we paired that with 30 who were not and we could run a nice controlled experiment to see the efficacy, how well VR worked in terms of training, and we had really good data there.

We're now in all 200 of Walmart's training academies and, to date, over 150,000 employees at Walmart have put on the virtual reality goggles to get better at their job. And it's a really nice use case; training to help you get better at what you do.

One of the most useful things about virtual reality is the tracking data. So for a company who's training someone, what we can do is we can figure out how well you're learning as you're doing the behavior.

We've developed algorithms such that we can quantify attention and engagement and how well someone's encoding a message simply by where they're looking and how they're turning their head. To make virtual reality work you have to track all the body movements, and so people when they think of VR most managers they think about the visual experience aspect of it but, in my opinion, the most valuable part of this from an organizational standpoint is actually having all this data, which is the richest data in history.

So if we can go back to football as an example, we have hundreds of thousands of plays watched over the last couple of years by players. Every player, when they play on the field, they get some level of assessment. They actually get a grade, even if they didn't touch the ball.

Later on, when they analyze film, every player gets graded for many teams. We've got the input data to our machine learning algorithms which is nonverbal behavior, where they're looking when they trained in VR, and we've got our output data which is how well they did on the field. And what we can do is build algorithms that can associate body movements when training in VR with the impact they had on the field. And we've actually developed a number of features that can predict, if you give somebody a novel set of body movements during training, it can actually predict how well that person will perform.

So the NFL, in particular Troy Vincent, who's the vice president of operations there, has been a leader in—they want to be known as the best place to work and they want to be a thought leader in this area. And so what we've done for the NFL is we've built an interview trainer. So we talked earlier about the flight simulator. Why do you have a flight simulator? Because when we teach people we encourage them to make mistakes. That's how you learn. You make a mistake and you get feedback, and that's why we have a flight simulator. Interviewing someone should be the same way.

So when I interview someone, maybe I ask the wrong question. I asked a different question from person A than person B based on how that person looked or maybe I looked in the wrong way or I leaned in in a way that was not the proper way. And what we've built is an interview simulator which is extremely realistic. These are very photorealistic avatars. And you go through and you interview people and you ask questions, and we record your movements, we record your utterances and we give you feedback. And then you repeat it.

And the National Football League is using this in a number of places. The example I'll talk about is there's something called 'the combine' in the NFL. And the combine is where players go to try out and they don't only try out with their athletic skills, they do interviews to see if they'd be a good fit interpersonally on teams, and there's interviews that go on between general managers and scouts with players. And for a couple of teams, we trained their scouts and GMs—the NFL as an organization encouraged us to do this—so that they could be better trained on how to do an interview.


Related