World of Uncertainty

点击翻译按钮获取中文摘要

We want to build and test an educational computer game in which decision-makers can learn how to handle uncertainty.By experiencing a world of uncertainty, and by explicitly estimating uncertainty in that world, they can develop the skills and attitudes needed to make decisions under uncertain conditions, instead of trying to eliminate uncertainty from the problem. Too often, decision-makers, journalists and members of the public discuss scientific issues in terms of competing stories (e.g. global warming is/is not a threat). In simplifying the science, all uncertainty has been removed from these narratives. Decisions can be distorted by our natural instinct to believe in stories. That is why PR and spin can be so successful. It isn't a case of lying with statistics, but the narrative lie is better understood than the statistics. But in this world, unlike the Discworld, million-to-one chances do not happen nine times out of ten. Truth is stranger than fiction.So how can we get more people involved in decision-making to understand uncertainty? Instead of trying to eliminate uncertainty, lets teach them how to explore uncertainty, handling it with the aplomb of a bookmaker or professional gambler. If we can get decision-makers and the public to better understand uncertainty, and scientists reporting to them to better understand stories, then we stand a chance of bridging this gap in understanding.There is one space where linear narrative does not dominate: the computer game. Unlike films, books, or even TV science documentaries, a computer game does not follow a single narrative thread. Instead the player creates his or her own narratives as a result of his or her decisions and actions while playing.We will build a prototype computer game in which a trainee decision-maker can explore uncertain data, make estimates of how uncertain she is, and get feedback on how well she handles the uncertainty. Imagine a game in which veterinary students can explore the uncertainties in making decisions about an epidemic, be it BSE, foot and mouth, or bird 'flu. Inside the game will be an uncertainty engine, which handles the mathematics of scoring players not on getting answers right, but correctly estimating how uncertain they are.Then we measure the effects this game has on groups of learners, to see if they can improve their skills in handling uncertainty, and their understanding of scientific uncertainty. If it works, this could be the model for a new kind of educational game.