You are here:

Data

Data is Cheap.

A well instrumented game can provide a steady stream of meaningful data. If the game is widely deployed, the streams become a torrent. Most commonly our data stream is tracking in-game player experience, learning outcomes, usage and demographics, in order to optimize learning.

Insight is Precious.

Great analysis turns raw data into actionable insight. Simple stats don’t always distinguish correlation from causality. To get it right, we’ve learned to suppress our hunches, and go instead where the math leads us. We build hypotheses to fit the analytics, not the other way around. Surprise results are always more valuable than confirmation, even when they are uncomfortable.

Visualization

Insight is squandered if it is not communicated. We develop innovative data viz and useful dashboards that are tuned to the needs of users, not statisticians.

Case Study

Most often the goal of analytics is to optimize learning and tune playability. Or to increase market penetration or player adherence.

Sometimes the goal is different. We built a game-based testbed for the Human Performance Directorate of the Air Force Research Lab. The purpose of the testbed was to generate experimental data showing how humans interact with autonomous drones in a realistic convoy protection mission. As these UAVs become self-directing, the responsible humans first become more productive – until thresholds are crossed where the human is disengaged and demotivated. The testbed allowed experimenters to precisely calibrate autonomy in four domains of drone behavior, and aggregate performance data in order to help design future systems.

Players direct a team of robots to find the optimum level of autonomy.
Players direct a team of robots to find the optimum level of autonomy.

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.