Veritas
An intervention to combat misinformation
What Misinformation spreads through social networks because people evaluate credibility through social trust, not source quality. A friend's re-share signals legitimacy more than the original publisher does.
Why Direct fact-checking interventions trigger psychological reactance. People resist being corrected. They need to build evaluation habits, not be told they're wrong.
How A persuasive mobile game that trains critical evaluation indirectly. Players care for a creature whose health depends on their information choices. Fictionalization and obfuscation sidestep the reactance that makes direct interventions fail.
Background
Misinformation spreads through social networks because people seek out and believe content that aligns with their existing attitudes — confirmation bias and motivated reasoning, amplified by the elaboration likelihood model: low-effort processing of content from credible-seeming social proxies like friends. Any intervention needed to build critical evaluation habits without triggering psychological reactance, the instinctive pushback people feel when they sense they're being corrected.
Research Tenets
A literature review of persuasive design, social media misinformation patterns, and virtual pet research surfaced three guiding principles for the design.
- Tenet 1People are susceptible to misinformation due to confirmation bias, motivated reasoning, and the sleeper effect.
- Tenet 2Employ fictionalization and intermixing to prevent psychological reactance.
- Tenet 3Personalization techniques increase the efficacy of persuasive messages and the user's connection with the intervention.
Game Design
We created a fictional world in which users care for a creature called a Worble, receiving information from friends in their pet community — correct information helps the Worble grow, false information makes it sick. The fictionalization serves a purpose: by embedding the misinformation challenge inside a creature-care narrative, users practice critical evaluation without the defensiveness that comes from being told their beliefs are wrong.
Paper Prototyping
Brainstorming sessions converged on the Tamagotchi-like concept. We created multiple fidelities for initial designs and conducted think-aloud sessions on paper prototypes with five participants. The feedback validated the core concept while surfacing what to refine.
- Participant 1I think it's fun and learning how to care for your worble seems like it'll be entertaining...
- Participant 2I loved decorating the creature...
Implementation
We chose React Native because the team's HTML5 experience transferred directly to JSX, and a native app's look and feel was essential to the game's immersion. I structured the app as three standalone pages with overlays managing transitions, using an emitter-subscriber pattern with RxJS since the prototype had no backend — an architecture that could later be wired to server events without restructuring the UI layer. Motion was applied throughout to direct attention and maintain fluidity, and I used native iOS accelerometer access to detect phone shakes as a game interaction, a small detail that made the prototype feel deliberately built for mobile rather than ported to it.
Prototype
After multiple design and development cycles, we tested the app with CMU studio participants under Prof. Geoff Kaufman. The reviews were favorable with constructive critiques from the studio audience.