Matthew Fendt

What is the role of gaming in higher education? Matthew Fendt, Baylor lecturer of computer science, is an expert on artificial intelligence, the gaming industry, and the role of gaming and interactive digital environments in education. In addition to teaching classes like Baylor’s gaming capstone, in which students create their own video games, Fendt has partnered with professors across campus to create tools for data collection. He recently partnered with Baylor Public Health and the Family Health Center in Waco to create a decision support tool for primary-care physicians. He shares how these various threads tie together in this Baylor Connections.
Transcript
DEREK SMITH:
Hello and welcome to Baylor Connections, a conversation series with the people shaping our future. Each week we go in depth with Baylor leaders, professors and more discussing important topics in higher education, research and student life. I'm Derek Smith and our guest today is Matthew Fendt. Dr. Fendt serves as a lecturer of computer science in the school of engineering and computer science, an expert in artificial intelligence. His research focuses on using data science to create learning games and interactive digital environments for education. Dr. Fendt teaches classes in the gaming concentration of computer science major and was recently awarded a grant to partner with the Family Health Center in Waco to create a dynamic decision support tool for psychiatric care for primary care physicians. And he's with us today here on the program. Dr. Fendt, thanks so much for joining us. Great to have you here.
MATTHEW FENDT:
I'm happy to be here.
DEREK SMITH:
Well when we just described a little of what you do, but it's an area that I think a lot of people don't have a lot of knowledge about or aren't aware of Baylor's involvement about that. So I was curious, when you tell people what you do, what are some of the responses or reactions you get? And I'm also curious if they vary by age.
MATTHEW FENDT:
Well, I think there certainly is a lot of interest since everyone is at least passingly familiar with games. There's an immediate buy in from younger people since they're more likely to play games and we can immediately make that connection and share our favorite games. For older people, I emphasize the fact that games are really engaging environments that we can use not just for entertainment but for education and training as well.
DEREK SMITH:
And we will discuss that a little further on on the program, what that looks like. I'm curious as a kid, were games something you enjoyed from a young age? Is this a path that you always knew you might take?
MATTHEW FENDT:
Yeah, sure. So my first gaming system was the Atari that my mom gave me and my dad brought home a Dos computer that we played games on as well.
DEREK SMITH:
Okay. Yeah, we might be around the same age then with the Atari and early Nintendo and some of those top game consoles. But what should people outside of the discipline know about the academic study of video games and how it's growing here at Baylor and at other institutions around the country?
MATTHEW FENDT:
Sure. Well they're great testbeds for research, especially artificial intelligence. You have these really interesting domains and problems in games, which are challenging for an AI. Of course the end goal is to move beyond just the game domain and apply the AI to challenging problems in the real world. Like chess for example, had been the gold standard for AI for some time before a machine was made to beat human professionals. Now researchers hadn't recently moved on to a board game called Go, which is vastly more complicated. But in 2016 an AI called AlphaGo beat a high-ranked Korean professional. So that's now it's a solved problem. And I've heard the term that AI is just basically problems that are unsolved and then they just become standard computer science. So the current AI game challenge is in video games with less discreet moves, like a real time strategy game called StarCraft and a multiplayer game called Dota.
DEREK SMITH:
For people who hear the term AI but are aware of artificial intelligence but it's maybe a little nebulous. What would you like people to know just to sort of wrap their brains around that more broadly?
MATTHEW FENDT:
Well it's really just a process for computers to solve problems. And I think the challenge is you can program a robot to assemble a car in an assembly line, right? But when you move out into the real world, you don't have these constraints that make things standardized. So it's applying AI to novel situations and task solving in the real world.
DEREK SMITH:
Visiting with Dr. Matthew Fendt and let's talk about the work you do with Baylor students in the classroom. One of the more popular things that you do and I've seen a lot of media attention certainly is your work with students in classes to create their own video games. What goes into that?
MATTHEW FENDT:
Well, our students usually come into the program with a drive to game development and they played a lot of games or some of them have even designed games themselves. So as part of the computer science major, of course we're interested in the programming aspect. But it's great because it's a good exercise in the discipline because games require such a breadth of knowledge in different areas of computer science, like artificial intelligence, obviously. Databases or storage and networking for multiplayer and software engineering for project development. The list kind of just goes on. But of course in our program we also teach good game design fundamentals that are transferable to the development of games of all kinds. So what makes games fun, how to properly iterate and play test and the industrial and social side of game development as well.
DEREK SMITH:
Obviously you need to have good computer skills and you talk about those design skills to help it be designed well. But how challenging is it for students? I got to think when you're putting something together, putting a game together, you've got to really have a pretty strong breadth of knowledge about whatever the topic of your game is. Because do you have to cover just about everything in one way or another? I mean obviously if it's a sports game you've got to account for just about every conceivable scenario in say a football game or a basketball game. How do you help students think through that process?
MATTHEW FENDT:
Sure, yeah, great question. So I like to think of games not as like simulations of the real world, but these experience engines. So I always tell the students to have the experience come first and the game implementation mechanics kind of follow that. Games are just another form of media like film or books or something like that. So I would say don't replicate football in its actuality in the real world. But what are the fun aspects of football and use that to drive your development.
DEREK SMITH:
Talking with Dr. Matthew Fendt. Can you tell us just more broadly about the gaming concentration in computer science and how that all fits together? That's certainly a growing area here at Baylor.
MATTHEW FENDT:
Sure. Well, so our concentration is a basically a curated set of electives. So we do this over having a major in game development so the students can get the broadest range of training for both gaming and traditional computer science jobs. So the students take four classes with us in gaming and three over at the film and digital media. So they come into us freshman year and intro to game design and they make games in the game maker software. We also talk about the design and industrial side of gaming. And one cool thing that I do in this class is bringing the students to the Chillennium Game Jam at A&M. So it's this yearly, 48 hour a game programming competition where the students make games from scratch over the course of the weekend. And we've been doing this the last four years and they had over 400 people this year. So that was great. The students get to see artists and musicians working alongside the programmers. And actually for the last couple of years we've been bringing Baylor music students to join our students teams. So that's freshman year. And then they go off and do their core computer science classes. Along the way they take the film and digital media classes and they learn about the production side of games. They come back to us junior year for a three class sequence, which is graphics, frameworks and the gaming capstone.And in graphics they learn about the math underlying game software and in frameworks they make a game engine from scratch. But the most exciting class in my opinion, is the gaming capstone. So this is the senior project that they do that's a culmination of their research. So we're partnered with Aaron Tebow who is an executive producer at Gearbox Software and that's a video game design studio that makes the Borderlands series. He comes in and advises the students and brings all kinds of great speakers. And last year the students got to tour the Gearbox campus, which they really enjoyed. So in this class, the students work on a group project, which is one game they make over the course of the semester. And at the end of the semester they publish this game on Steam, which is the biggest online video game distribution platform. And last year the students got to collaborate with art and music, film and digital media, entrepreneurship. And it was great getting to work with people from other majors. And the gaming industry, it's so interdisciplinary that I really value the collaborations the students get to do. So we had this big end of the year demo, about 150 people showed up, including 75 high school students, Baylor students, industry professionals, and the public. We had giveaways with signed art prints and a hand painted at 3D statue of the game Protagonist. So it's a great way to show off the gaming program and computer science to the local high school students. And for the seniors themselves, they can come out with a published game that they can show off to recruiters. So I mean if anybody's interested in checking out our games, they're available at baylor.edu/gaming.
DEREK SMITH:
Baylor.edu/gaming, we will have to to check that out. And I'm thinking, you think about a capstone project. What are some of the challenges that you have to help students work through? I picture some of those students when they're there late at night in the wee hours like you do sometimes when you're getting a big project done. What are some of the moments that are most pivotal in creating that, those moments where maybe sometimes they have to come to you to help kind of think through something. What does that look like?
MATTHEW FENDT:
Sure. Well, so again, we're driven by creating the experience in games and the mechanics come secondary. So I think a lot of it is when the students have to shift maybe an idea that's not working, something that they've grown very attached to. Because it was part of their initial vision, but they need to kind of maybe move away from or pivot. So it's that big break through that, it's the process and not just one idea that will make it to the end. So I think a lot of play testing and iteration and just a lot of hard work from the students.
DEREK SMITH:
You mentioned that idea of maybe they had an idea that they have to cut for the final project. And I think anyone who's written something or certainly in FDM produced something. There's that moment, which maybe leads to a question. You talked about film and digital media, you talked about students being around, working with music students and others. We talk about computer science, but there is an art to that a little bit as well. What does that look like? Those interactions in FDM or with musicians or other artists. To add that, to see that artistic side of it fully lived at as much as they can in their projects.
MATTHEW FENDT:
Right. Yeah, it's been great getting to work with other majors. For example, last year in the capstone we got to interface with artists who created assets for our game and title sequences. So and they think very differently. So I thought it was very valuable for the artists to learn how the programmers work and vice versa. So just because in an industry you're going to be working with people with all kinds of different skills. So we also are now beginning to have regular collaborations with music and entrepreneurship as well. So it's been great having the students understand what goes into other students' work and contribution to the project.
DEREK SMITH:
What are some of the jobs that either former students had have gone into or that your current students aspire to? Both right out of college and maybe looking down the line?
MATTHEW FENDT:
Well our employment rate for our graduates is almost a hundred percent, so they almost all come out with jobs ready to go. Or if they're going into higher education, they have graduate student positions. We last year had one of our gaming capstone students graduate and go off to Gearbox, which is great. So placed directly into the industry. But kind of as I mentioned since we give them the broad training, a lot of students go into general programming jobs and then maybe end up in the game industry down the line.
DEREK SMITH:
We are visiting with Dr. Matthew Fendt on Baylor Connections. Fendt serves as a lecturer of computer science in the school of engineering and computer science at Baylor. And let's try to transition now more to your research and we talk about that interdisciplinary nature and certainly that's a big part of what you do. But could you give us just a brief overview first of those research interests and how games can be used as you mentioned to educate.
MATTHEW FENDT:
Right. Yeah. So games go beyond just entertainment and there's such engaging environments that we're exploring them as a supplement to traditional means of education, such as lecturing or reading. And it's one thing to be reading about some great historical landmark and maybe see a picture, but it's a lot more impactful if you could move around in a digital space and explore it.
DEREK SMITH:
So when we talk about learning games and those digital environments, what does that look like for maybe those of us who grew up very much in that transparency and VCR wheeled into the classroom age.
MATTHEW FENDT:
Right. So VR is a great way of experiencing things. I remember back when I was in North Carolina, we went to Duke and they had this great digital environment where they spent tens of thousands of dollars to create this 3D space where all six surfaces were screens basically. And you can interact with it. And obviously that is limiting in how you can offer an experience to the user. But now with the increase in improvement in technology, you can basically get a lot of that capability with a couple of hundred dollar VR headset. So it's becoming more and more accessible. And since it's getting out to the public, we want to make sure that we're keeping up with the learning content we can provide.
DEREK SMITH:
What are some of the ways that you conduct that research that you're able to glean your findings?
MATTHEW FENDT:
Right. So we've had some great collaborations with the Family Health Center as I mentioned and I've got a interdisciplinary collaboration with Baylor and Bangalore Baptist Hospital in India. And what we're trying to do is create apps that provide these interesting engagements through gaming experiences, but also provide educational content as well.
DEREK SMITH:
With Family Health Center, I know you know Dr. Kelly Ylitalo, Word says she's conducting research there on health and aging and physical activity. How are they able to utilize those and what was that development process like of kind of learning their needs and incorporating them?
MATTHEW FENDT:
Absolutely. So I've done a number of collaborations with them in the past, but we have a current partnership that maybe I can talk about. So this is a partnership with the Family Health Center, Harvard, Massachusetts General Hospital, their Department of Psychiatry and myself at Baylor. And what we're trying to do is expand their psychopharmacology decision support tool. And this tool is designed to give guidance to primary care providers for the management of the majority of psychiatric illness. So some guidance does exist in various forms in psychiatric literature, but a lot of it's not transferable to primary care practices and that's who now treats the vast majority of illness. So we're providing a tool to help in diagnosis and treatment. So my contribution to the project would be to create an app that uses data science and machine learning to provide a dynamic version of this tool to the primary care physician. So we're going to use something called a recommender system and that allows the tool to be dynamically customized and adapted to the patient based on their past treatment and their similarity to others. So imagine there are various medicines a doctor could prescribe for a treatment. What we're going to do is make suggestions based on what's worked in the past for them. And many patients have connected illnesses as well, so that will influence treatment. And for this thing called a recommender system, it works a little bit like what Amazon or Netflix do when they make a product recommendation. So if you and another person like item A and that person also likes item B, then maybe you like it as well. So as you mentioned, we've got some great collaborators at the Family Health Center, so doctors Ryan Laschober, Kelly Ylitalo, Lance Kelly and Matthew Foskey are collaborating with me on the project and I'm supervising several undergraduates as well. So we hope that this decision support tool will allow for better and more adaptive care for patients.
DEREK SMITH:
What does it mean to you to be able to, you think about it, what it's doing is making life easier for busy people. It's making it easier for patients, easier for doctors or researchers here at Baylor to access in and accumulate knowledge. What does that aspect of it mean to you? I think the gaming is the exciting part and you're helping students do some things longterm, but it also goes beyond just your immediate discipline.
MATTHEW FENDT:
Right. So with the current state of the art for the psychiatric care that the doctors basically examine the literature, but we're hoping to streamline that process. Right? So that could be the level one, just the streaming line with the decision support tool. But now when we get into the patient interaction, right? Maybe they need to be prescribed treatment and it's something that you want to provide regular interactions. So how do we entice them for these well engagements through something like gaming.
DEREK SMITH:
We are visiting with dr Matthew Fendt on Baylor Connections. We had into the final few minutes of the the program. As you study the effectiveness of digital tools, whether it's more traditional lecture or book learning in schools, whether it's more traditional methods of data collections. What is your research showing you and does that leads you down other paths?
MATTHEW FENDT:
Absolutely. So we find that there's a lot of increased excitement about gaming and the topic as well. We developed one project where we went into middle schools to teach about Texas Civil War history and we found that the students were a lot more engaged and interested about learning the subject and wanted to share the tool with others. We also made a tool to talk about the Armstrong Browning Library, which is on Baylor campus and we compared this to more traditional means of learning such as reading a website or a self guided tour packet. And we found that those that played the game learned a lot more and were a lot more excited about wanting to come to visit.
DEREK SMITH:
So far on program, we've talked about areas of your work. We started off talking mostly about gaming, but we've delved into the arts and FDM to health matters to music to American history.
MATTHEW FENDT:
Right.
DEREK SMITH:
Armstrong Browning Library. So it kind of paints a wide picture of where your work leads you.
MATTHEW FENDT:
Absolutely. And I think it really goes to show that gaming isn't just sitting in your room playing an entertainment game by yourself, but if you expand what gaming is capable of, you find that it's applicable to a wide variety of domains and it's something that there's a lot of interest in and a lot of people are interested in collaborating.
DEREK SMITH:
Well finally, Dr. Fendt as you look into the future and think about Baylor's role in data science research, in the video game industry, what challenges or opportunities with that focus are most exciting to you?
MATTHEW FENDT:
Well, I think that data science has a natural overlap with gaming. So data science is roughly the intersection of computer science, statistics and business. So you will get large amounts of data and use it to draw conclusions. And since gaming is this huge industry, we have access to large amounts of player data. We can use that to learn what the players are doing and how to make their experience better. For me in particular, I want to use this data to increase the effectiveness of the learning games. I'm working towards games that are customized to the player and give them the best learning experience possible.
DEREK SMITH:
Awesome. Well, we'll look forward to seeing more about that. You mentioned baylor.edu/gaming. People can see a little more about the program and those capstone projects, and there's also information online if people want to Google you, Matthew Fendt or Baylor gaming. There's plenty there to learn more about that. Thank you so much for your time. Thanks for joining. It's great to have you on the program today.
MATTHEW FENDT:
Thank you for having me.
DEREK SMITH:
Dr. Matthew Fendt, lecturer of computer science in the school of engineering and computer science. Our guest today on Baylor Connections. I'm Derek Smith. Reminder, you can hear this and other programs online, baylor.edu/connections. Thanks for joining us here on Baylor Connections.