Personal website
Home Research Mobile Science
Picture

Research team: Konrad Grabiszewski and Alex Horenstein
Available on: Android and iOS
Data Collected: August 2017 to February 2018


The Idea

Every day we make decisions that depend on what others might do, in business, politics, or social life. We try to think ahead, anticipate other people’s choices, and plan our moves accordingly. This kind of thinking is the essence of game theory, a field that studies how people make choices when their results depend on others.

Blues and Reds is a mobile experiment that gamifies a research inquiry into how people think strategically. The concept behind it is known as backward induction, a way of reasoning that starts from the desired outcome and works backward to find the best sequence of actions.

Blues and Reds was developed thanks to the financial support of the University of Miami.


How It Works

The app contains 58 puzzles where players compete with the computer. Each puzzle begins with a small robot called the RoboToken. The goal is simple: finish the game on a blue circle rather than a red one.

The player makes the first move, and then the computer responds. In easier puzzles the challenge is simple. In more advanced ones, both sides take several turns, and success depends on careful planning and understanding how the computer might react.


Gameplay Preview


Two Types of Challenges

Players have nicknamed the two kinds of puzzles Snowflakes and Rails.

Snowflakes: These puzzles show all possible moves as a branching tree. The player and the computer take turns moving the RoboToken along the branches. To win, the player must look ahead at the possible endings, reason backward, and choose the path that leads to blue. These puzzles show how well people make decisions when all the information is visible.

Rails: These puzzles use the same logic but in a more complex visual form. The underlying structure is not shown directly. This helps us see whether players still use strategic reasoning when the situation looks less clear.


Research Output

  • Understanding dynamic interactions (with Alex Horenstein), Games and Economic Behavior, vol. 149, pp. 96 to 111, 2025.
  • Measuring tree complexity with response times (with Alex Horenstein), Journal of Behavioral and Experimental Economics, vol. 98, 101876, 2022.
  • Profiling dynamic decision makers (with Alex Horenstein), PLoS ONE, vol. 17, no. 4, e0266366, 2022.
  • Effort is not a Monotonic Function of Skills: Results from a Global Mobile Experiment (with Alex Horenstein), Journal of Economic Behavior and Organization, vol. 176, pp. 634 to 652, 2020.