Meet the UO professor who studies the serious side of games

Maxwell Foxman says his career is “a classic academic one” and “a little bit of a boring story.”

Neither of those statements appears to be accurate.

Foxman is an associate professor of media studies and game studies at the University of Oregon School of Journalism and Communication. His research focuses primarily on how games and play impact nongame industries and media professions. 

So it might be appropriate that the once-popular social media app Foursquare played a role in jump-starting his career. Here’s how that happened:

In the 2010s, Foxman was in a master’s program at New York University and was taking a class in video-game theory, “because it was interesting, and I had played video games all my life. I wouldn’t say that I was a quote-unquote gamer.”

In the class, he noticed that many of the theories he was reading about “applied to things other than gaming.”

And then there was Foursquare, “which allowed you to check into places using your smartphone, and I was obsessed with it. I loved it. I was checking in everywhere. I still do, actually. I’m one of the few people still using it. … I was like, ‘Wow, this is interesting, and it’s interesting that social media has all these different components that are playful.’ And so that’s what I ended up writing my master’s thesis on, and then I went to my Ph.D. program to study this, and I haven’t really looked back.”

At UO, Foxman continues to research “this relationship between games, play, and how they come up against other forms of media, because I noticed it in that sort of early, early moment, and I just keep seeing it in all sorts of other things that I do, and that’s probably what keeps it interesting for me.”

The work also has led into other areas — including game journalism, artificial intelligence, virtual and augmented reality and game production.

In this question-and-answer session with Lookout Eugene-Springfield, Foxman talks about his career and research — which lately has included creating a six-question test readers can use to assess the authenticity of images. 

Here are highlights from the conversation, which has been lightly edited for clarity and style. 

Lookout Eugene-Springfield: Are you tired of people saying, “Oh, so you study games for a living? How did you manage to get that gig?”

Maxwell Foxman: I like to say that I have every 13-year-old’s dream job, at least on paper. But I don’t get tired of it because it is always interesting. … I think when I finally stop getting that question, that might be the sign that games are truly mainstream.

How did you get to the University of Oregon in the first place? Was it the job?

Essentially the job call was for someone with experience in games, virtual reality and journalism, and that’s exactly what I studied, and so it was too perfect for me not to apply. …

One of the great advantages from the academic side of the SOJC is it is a departmentless school, so I collaborate across the entire school with many different colleagues on many different projects, which I’ve always really appreciated because my Ph.D. was also pretty interdisciplinary. 

And then the town I just really love, mostly because I am a cyclist, and this is perhaps one of the best places in the country to bike.

One of your many projects is a book in progress that in part probes the intersection between games and technology. 

(The book) basically argues that games are … often the canary in the coal mine for new and emerging technologies. Whether you think about AI or you think about immersive media or anything in between, (games) … allow people to play with them early, and so the kind of joy and exuberance and experimentality that is inherent to play — which is very different than when you’re tasked with something as a job — that is precisely where you can see emerging technologies being figured out.

One of the things that’s interesting about having game studies in the School of Journalism and Communication is that you increasingly see news outlets using games to explore deep issues and as part of investigative projects.

I mean, this is something that’s been happening for 20 years in journalism under the rubric of news games, and I’ve always thought it is a valuable tool in a newspaper’s toolset.

I actually teach a course here at UO called Game Journalism and News Games. The first half of the class is, as the name suggests, about writing about games, and then the second half of the class is taking a news topic and turning it into a game. And students are asked to use off-the-shelf tools and their own journalistic skills to build a game or interactive experience, and I love them. … The other thing that I always tell students (is that) a game is only good if people want to play it. 

Seeking ‘fidelity’ in images

You’ve been researching about how worries about images created or manipulated by computers or AI may undermine people’s faith in those images. But, in some ways, this isn’t a new issue.

I wrote recently about how journalists can rethink their relationship to computer-generated images in general … the same way that when photography first became a medium for journalists, it was less trusted than woodcut blocks or prints. I think we’re at a critical moment where images that are computer-born are going to become even more ubiquitous, and how journalists can understand and articulate the truth of those images, while understanding that they are manufactured in a way that is not a perfect … photographic capture of reality, is, I think, a really important question for the next generation of newsmakers.

People sometimes forget that photographic images can be manipulated — and have been for years.

Part of the scholarly work that I did was really focused on sort of thinking about how objectivity as a journalistic norm has shifted, and how it was an authoritative claim really in the 20th century, and now how we have to start maybe rethinking that relationship, particularly because some of the hallmarks of particularly visual journalism are changing. Now it’s so normal to manipulate images, and frankly, so easy to both generate and manipulate them. That doesn’t mean that they’re not truthful. It just means that … the hallmarks of objectivity might not work in the same way that they used to. 

That question — is this image fundamentally truthful or fundamentally false? — is important for news consumers as well as journalists. Some of your recent work focuses on the notion of “fidelity,” a concept from computer science. How do you think fidelity can be an important tool for news consumers? And you came up with a list of six questions that people can use to judge the fidelity of images. (See the bottom of this story for a list of Foxman’s six suggested questions.)

When we’re thinking about fidelity as opposed to objectivity, we’re asking a very simple question, an important one, which is: To what degree is the thing you are seeing approximating reality? And of course, that question, depending on how you decide to look at it, is complicated, but just asking that question, I think, is a start of a useful heuristic.

So, when you think of the most ridiculous AI slop, it clearly has low visual equivalents. It often doesn’t look realistic in terms of the people portrayed, but also if you compare it to a real place, it won’t look right. So it’s an easy enough way to start to take that apart, and just a few Google searches can easily satisfy that.

Another one is thinking about whether something is high fidelity or low fidelity. There, I think it’s really important, when you’re thinking about the graphic quality (of the image), to understand that both can convey truthful information. This is one of the reasons why I like comics journalism as one of my examples, because comics are a low-fidelity medium in many cases, but that doesn’t mean that they don’t convey the truth. They often convey very truthful things to people.

I occasionally speak to groups about the state of journalism, and I sometimes start by showing one of those New York Times quizzes about which images are real and which ones have been created by AI. The groups get it right about half of the time. But basically, the advice I’m trying to give is if there’s a voice in your head saying, “This is something I need to check out. I need to ask a couple of questions about this,” you need to listen to that.

I completely agree with that, and I think … this is giving people a set of six easy questions that (they) can start to ask.

The only other thing that I’ll tag from my top six is thinking about context matters so much, both in terms of how the image is presented and in terms of how it’s made. So some of that … (is) really a call upon media makers to be transparent about their process.

But from a consumer point of view, if you are consuming a piece of AI imagery in a context that you cannot trust, that matters, and I think we often tend to forget that because of the way that we consume news through social media, through interpersonal communication, etc. 

I think we instinctively do a lot of this, and it’s maybe giving a name to a process that many people do already. I was surprised, as someone who studies emerging media, how quickly people have been able to easily identify AI-generated images. … Yes, the image generation is getting better, and yes, that that relationship will continue to be complicated. (But) the ability for information consumers to start to identify things based on things like context, on quality, on an understanding of this question of fidelity, I think is, hearteningly, pretty good. 

With this work on fidelity, and your book projects and everything else, it looks like you’re able to keep busy.

Yes. I try to overcommit.

The six ‘fidelity’ questions

Foxman suggests asking six questions to determine whether an image is high- or low-fidelity:

1. Does the image accurately convey the reality it claims to show?
2. How was the image made, and what are the technical limitations?
3. Does the level of detail reflect the story or message told?
4. Is the image clearly tied to its context and source materials?
5. How much has the original information been changed or altered?
6. Can the process and documentation behind the image be traced and accessed over time?