Science

TRACS is a Tool for Research on Adaptive Cognitive Strategies because the card games provide scientists with a laboratory testbed for investigating how (and how well) people make diagnoses and decisions under uncertainty. TRACS is currently being used for research in the "Mental Models" Lab at the MITRE Corporation and in the "CogWorks" Lab at Rensselaer Polytechnic Institute.

Institute of Electrical and Electronic Engineers, Symposium on Computational Intelligence
International Conference on Computational Intelligence and Natural Computing
International Conference on the Theory and Application of Diagrams
International Conference on Naturalistic Decision Making
American Association for Artificial Intelligence (AI)
Society for Judgment and Decision Making
Society for Mathematical Psychology
Bayesian Research Conference
Cognitive Science Society

TRACS games are novel because they are played with two-sided cards - where the backs provide clues to the fronts. This design captures the basic structure of many practical problems, such as identifying military targets (which are like the "truths" on the fronts of the cards) from radar images (which are like the "clues" on the back of the cards).

TRACS games are useful because the family of games is designed to present cognitive challenges that are similar to those of military, business and other domains, namely:

Risk Assessment - with uncertain information
Resource Management - on dynamic missions
Rational Engagement - in adversarial situations

TRACS is called "The Game of Confidence and Consequence" because the card games mirror the two sides of "Command and Control" in dealing with any domain, namely:

Diagnoses - to achieve some level of Confidence (belief) in uncertain situations, and
Decisions - to affect some desired Consequence (outcome) in evolving situations.

With these features, research on TRACS can measure and model the cognitive strategies that people will use to make diagnoses and decisions in real life. And that is the point: To investigate human thinking in the "low stakes" context of the lab (TRACS games) with an eye towards improving human thinking in the "high stakes" context of the world (real life).

Another advantage of TRACS is that the game states are more constrained than in games played with standard (single-sided) cards whose backs provide no information about the fronts. This makes TRACS more tractable to mathematical analyses and computer simulations of optimal strategies, which are needed to measure and model the limits and strengths of human thinking. Other card games are amenable to analysis only under gross simplifications that can destroy the essential character and practical relevance of the game (see Epstein R. A., “The Theory of Gambling and Statistical Logic”, Academic Press, 1967). For example, a typical game-theoretic analysis of Poker uses a 2-person game played with 1-card hands drawn from a 3-card deck (see Kuhn, H. W., A Simplified Two-Person Poker, Study 24, “Contributions to the Theory of Games”, Princeton University Press, 1950).

Another card game that has proven amenable to analysis (see Isaacs, R., “A Card Game With Bluffing”, Am. Math. Monthly, 42, Feb 1955) is played with a special deck of consecutively numbered (single-sided cards). TRACS offers additional advantages because it is played with two-sided cards and because the double-sided deck can be used to play a whole family of games.

A final advantage of TRACS is that the family of games allows scientists study the game (and to design new games) in a systematic fashion - ranging from simple solitaire games (like Straight TRACS) to complex competitive games (like Poker TRACS). This feature also makes TRACS useful for teaching people at different levels of ability about abstract things like probability and about concrete skills like decision making, in a manner suggested by Clark Abt (in his book “Serious Games: The Art and Science of Games that Simulate Life in Industry, Government and Education”, Viking Press, 1970):

“Intuitive problem-solving is an aspect of education neglected almost everywhere … Simulation games stimulate, reward and judge intuitions according to pragmatic standards rather than doctrinal ones. Enlightening intuitions are rewarded for their superior problem-solving speed over systematic analysis. False intuitions prove to be ineffective in game play. The ideal problem-solving strategy that emerges from most players combines intuition and analysis – analysis used to check intuition, and intuition used to extend analysis beyond familiar limits."