Not your grandma’s Duck Hunt
I’ve said it here before: I’m not much of a gamer. My 9-year-old nephew gets exasperated every time he sets me up in front of the Wii and ultimately just takes the controller away from me so he can deal with both characters at once. But that doesn’t mean I didn’t get excited when I heard about the next Pop Up Open Lab Experience and Reception: Play + Innovate. Northeastern students and faculty members will be gathering in the Digital Media Commons at Snell Library on Monday afternoon from 4 p.m. to 6 p.m. to present 11 different game-related research projects. The event will feature interactive demos from the interdisciplinary research teams.
This week, I caught up with a couple of the teams whose research will be on display, just to get a taste of what we have in store for us. Here’s one thing I learned: games are getting smarter. When I was a kid, Duck Hunt and the Oregon Trail were fairly predictable. You knew you’d have to ford a river at some point, and eventually you could learn the pattern by which the ducks entered the screen. Well…some of you could. I couldn’t. I would get deeply frustrated because I was so terrible at the whole endeavor.
Russell Pensyl, professor in the Department of Art and Design, professor Magy Seif El-Nasr, who has joint appointments in the College of Arts, Media and Design and the College of Computer Science, and PhD candidate Bardia Aghageigi are developing a system that the child-me would have appreciated quite a bit (and actually, the current version of me probably would too).
A camera on the computer or the mobile device takes regularly-timed pictures of the player’s face. The pictures are compared to a huge database of photos of facial expressions that are associated with particular emotions. If you’re smiling and giddy, the computer will know it. Then it’ll dive into the back end of the game and take a look at what’s happening there. If you’re winning by a landslide, the game will adapt itself to be a little more difficult, making the game more challenging, and thus (hopefully) more rewarding. If you’re frowning, as I almost always am when I have a controller in my hand, it might make itself a little easier, tone down the number of ducks flying across the screen, that sort of thing.
This is a form of adaptive content generation. Seif El-Nasr is interested in the technique to help promote engagement with games for health and learning. “So while you’re playing, things can change to make it more engaging or get it to be more effective at a learning or health component,” she explained.
There are other ways a game can change to handle other sorts of challenges. Automatic content generation means a game is never static, each time you start it up, you’re faced with a different kind of task or a new puzzle. Instead of responding to a player’s experience, this is completely random. You would never be able to figure out the pattern of target entry if Duck Hunt incorporated ACG.
Assistant professors Casper Harteveld and Gillian Smith are combining ACG with community gaming in an attempt to promote interest and learning in a game called Gram’s House, which aims to promote computer science interest among middle school girls.
Grandma loves her house and doesn’t want to leave for an assisted living facility, so players of Gram’s House try to equip her home with assistive technologies that can help her live there for as long as possible. This kind of story is thought to be more engaging for a young girl than, say, one that asks you to shoot all the bad guys. It uses puzzle games to teach computer science concepts like mapping or the binary number system. But it’s currently a totally static game. Players can’t interact with one another and the puzzles are always the same.
Like many researchers, Smith and Harteveld are curious whether adding a community aspect and ACG will improve the outcomes of the game. They developed GrACE, or Gram’s House Automatic Content gEnerator, named for the famous computer scientist Grace Hopper, to find out. Northeastern second-year student Gregory Loden developed a puzzle game that asks players to identify the shortest path between important points in the home. The idea is that a robot will travel along this course as it helps Gram with her daily tasks. But it’s really teaching players about the concept of “minimum spanning time.”
These are just two of the projects that’ll be strutting their stuff on Monday. Another looks at how virtual improv can promote social intelligence (this one will be on full display, actors and all!), while others explore how games can help us deal with issues like sustainability and security.
There will be other adventures that I don’t have time to get into here. You’ll just have to come to the Digital Media Commons on Monday to see for yourself.