Page 54 - Developer

STUDENT gamE PROFILES
educated play!

RefRaction http://games.cs.washington.edu/RefRaction/RefRaction.html

The hisTory of “eduTainmenT” is liTTered wiTh failed games ThaT weren’T fun enough To deliver Their educaTional payloads. The posT-grad Team behind fracTion-
Teaching puzzler refracTion is deTermined To noT be anoTher hisTorical fooTnoTe—which is why They’ve also creaTed “playTracer,” a piece of analysis sofTware
ThaT Tracks players’ acTions and TranslaTes Them inTo visualizaTions ThaT The developers can use To balance refracTion’s difficulTy, educaTional poTenTial, and fun.

Alexandra Hall: What inspired and try to make as large an to iterate and improve the game,
you to make an educational impact as possible on this specific and study its effectiveness. We’re publisher/developer
game about fractions? topic before tackling other areas. actually working on an entire set center for Game Science
Yun-En Lu: i’ve always liked our goal is to create games and of fraction games, two of which Release date
September 2010
teaching. Helping out in an tools to supplement existing are nearly ready for initial release. development time: ~1 year
elementary math class is a classroom curricula and make each of these games can tailor development budget: $150,000
lot of fun. every student has a teachers more effective. the play experience to each player # of lines of code in the game:
slightly different way of looking individually through procedurally ~100,000
at things, and trying to come up AH: You’ve worked on RefRaction generated levels and progressions a fun fact: Players have commented
on the difficulty of getting the coin on
with examples that students can for a couple years now—unusual along with sophisticated analysis level 2-5 more than any other. in fact,
reason through to find conceptual for a student game. Have your tools to measure individual Michael John (senior creative director
truths is a really interesting goals shifted? student learning. at ea) thought it was impossible to
challenge. Games are a natural YL: as Ph.D. students, creating obtain when he first played through
the game.
way for this to happen, since RefRaction and its sister games AH: Is it tricky to find a
they’re all about exploration is more like a job than school— balance between fun and
of strategies and interactions we’re paid, we have funding, we educational value? created a tool that shows how
between complex rule sets, so it take very few classes, and so Erik Anderson: Difficulty hundreds or thousands of players
seems only natural that we could on. Since we can devote a lot balancing for educational games move through a game’s state
let students experiment with math more time to these games than is not really that different from space by creating graphs in which
other games. all games involve circles represent game states and
learning, even those without an arrows represent player actions.
explicit educational purpose. PlaytRaceR uses multidimensional
However, one key difference of scaling to ensure that similar
designing educational games is game states are shown close to
that our learning goals restrict each other, and vice versa. this
our choices of game mechanics. allows game designers to rapidly
although game designers find where players are getting
sometimes bend reality in stuck and what strategies they
order to improve usability or are using.
playability, we have no choice but
to implement target educational AH: Are you planning to offer
concepts exactly as they appear Playtracer as a tool for use by
in the real world. other developers?
EB: Yes, we plan to make our
AH: Did Playtracer spring from a Playtracer tool publicly available
desire for finer-grained difficulty in the future. in the meantime,
and experience-tuning? we have published the technical
in the same way using games. it’s undergraduates taking a game EA: We developed Playtracer details of Playtracer in a few
my hope that through our games course could, it also means we because we needed a tool that different research papers,
we can convince a generation of can go much deeper into the really could effectively visualize player allowing developers to integrate
kids that everyone can learn math interesting problems, such as behavior in games without a these ideas and methodologies
if they put their minds to it—and running experiments to see how virtual environment. Heat maps into their own tools. none of the
even find it fun. players react to different tutorials are highly effective for games ideas are particularly complicated
Eric Butler: Why fractions or optional rewards, trying out such as first-person shooters to implement; we know a few
specifically? the original plan for algorithms to predict what people because a game designer can developers who wrote their
our group was to generically make are likely to do so we can adjust mark the map each time a player own version of Playtracer to
educational games about science, the game experience for them, dies, for example, and determine analyze their games. We are also
but we quickly discovered that creating new data visualization the most dangerous sections of currently working with some
basic algebra and mathematics and analysis tools, and so on. that map. However, for puzzle external developers to plug
was a huge stumbling block for a EB: the project cannot really be games like RefRaction, there is their games into our in-house
lot of students. So we decided to called a success until we make no equivalent “map,” so we need infrastructure. (You can read
focus very narrowly on the early a measurable impact to early some other way to visualize more about Playtracer in this
education problem of fractions math education. So we continue player actions. therefore, we paper: bit.ly/playtrcr)

52 game developer | october 2012