Preface
When I was first introduced to the concept of social impact games, it was a blue-sky moment. Someone could receive funding to make a social impact game without having to prove its impact before it existed. Funding was a catalyst to see what was possible.
Over time, funding shifting towards measurable outcomes, and I believe this change is reflected in the dominance of educational games within today's social impact game landscape. What attracted me to social impact games was the possibility that games could become a culturally significant medium for examining, challenging, and engaging with social institutions.
Early in my career, I designed and developed Dreamy alongside a health communication researcher. We wanted to create an intervention that would discourage players from using indoor tanning beds. We developed the game for $5,000, splitting the budget between design, development, and research. It took a year to make, a summer to collect data, and another semester to learn that it had produced an unintended boomerang effect.
A New Layer to the AnswerVerse
I really liked the idea of developing a game concurrently with a researcher, where the game's intended social impact could be measured. It's a slow process, though.
When I learned about Open Game Data, I immediately wanted to integrate it into Answer Campus. During the development of Dreamy, I had built a custom logging system for in-game events, and it wasn't something I wanted to rebuild for every game I developed. Open Game Data offered a way to make data collection reusable, allowing us to spend less time building research infrastructure and more time developing the game itself.
We were already approaching game production in a similar way. Rather than building a new game around each research intervention, NERDLab had developed a world, characters, and stories that could support many different experiences through the AnswerVerse Conversation Template.
Production
For Episode 2, we began by exploring what kind of intervention could fit within the existing Answer Campus storyline. I worked with Justin Jacobson, a graduate student in our design program with a background in counseling. We expanded the team to include faculty and doctoral students specializing in health communication and eventually settled on embedding motivational interviewing techniques into the game.
In the Text
Playtesting the original Answer Campus demo, we saw how single uncomfortable conversations could overwhelm a player. In the Text centered on power dynamics in classrooms. Leilani, a student in the class, confronts the professor when he reads a racial epithet aloud from an assigned reading. The player, as a bystander and fellow classmate, has to decide whether or not to support her and potentially risk the fallout that could come from challenging a professor's authority. In the demo, we never addressed the mental distress that could arise for Leilani post conflict.


We decided to take a cue from the work of Mary Flanagan and Geoff Kaufman, utilizing embedded design to think about how an individual episode might look.
We realized that we could embed one or two challenging conversations within a larger game world, using everyday interactions, mini-games, and campus life to give those moments context rather than making them the entire experience.
A New Character is Born
The AnswerVerse Conversation Template was being used by Matters at Play to develop Keys to a Dream during this time. For that game, a student created Dr. Purnell using the official AnswerVerse character style guide. Because the character already belonged to the shared universe, we could bring her into Answer Campus as a potential mentor and role model for Leilani.

Classroom as Writer's Room
Through a ULink grant, we had the opportunity to develop a course on story-driven game development. The idea was to turn the classroom into a writers' room for the second episode of Answer Campus. Because we already had the underlying game systems, much of the development work could focus on interactive writing. Students developed conversations that formed a five-act narrative, expanding the characters and storyline while creating a setting for evaluating motivational interviewing techniques.
Study Recruitment Experiment
As a highly interactive medium, I see a lot of potential in games as research vehicles. One challenge is recruiting enough participants to evaluate them, particularly when the games aren't designed for mainstream audiences. Looking at the impressions and downloads for the original Answer Campus demo, I wondered whether a game that developed an audience organically could use Steam's built-in playtesting features to make recruitment more efficient.
We opened a Steam Playtest for Episode 2 to begin exploring that possibility. The study itself hasn't started, so we're currently testing the recruitment approach rather than evaluating the intervention.
Modularization
We currently have a fully playable interactive narrative with a preliminary intervention. It's a streamlined version of the eventual episode, consisting of interactive conversations without mini-games or time-based decisions.
The plan is to use Open Game Data to study player choices and inform revisions to the intervention. Once we've evaluated that component, we can use what we've learned to build out the full episode, incorporating the additional gameplay and campus experiences established in Episode 1.