Streetlights and Shadows: Searching for the Keys to Adaptive Decision Making

In making decisions, when should we go with our gut and when should wetry to analyze every option? When should we use our intuition and when should werely on logic and statistics? Most of us would probably agree that for importantdecisions, we should follow certain guidelines--gather as much information aspossible, compare the options, pin down the goals before getting sta

In making decisions, when should we go with our gut and when should wetry to analyze every option? When should we use our intuition and when should werely on logic and statistics? Most of us would probably agree that for importantdecisions, we should follow certain guidelines--gather as much information aspossible, compare the options, pin down the goals before getting started. But inpractice we make some of our best decisions by adapting to circumstances rather thanblindly following procedures. In Streetlights and Shadows, Gary Klein debunks theconventional wisdom about how to make decisions. He takes ten commonly acceptedclaims about decision making and shows that they are better suited for thelaboratory than for life. The standard advice works well when everything is clear, but the tough decisions involve shadowy conditions of complexity and ambiguity.Gathering masses of information, for example, works if the information is accurateand complete--but that doesn't often happen in the real world. (Think about thecareful risk calculations that led to the downfall of the Wall Street investmenthouses.) Klein offers more realistic ideas about how to make decisions in real-lifesettings. He provides many examples--ranging from airline pilots and weatherforecasters to sports announcers and Captain Jack Aubrey in Patrick O'Brian's Masterand Commander novels--to make his point. All these decision makers saw things thatothers didn't. They used their expertise to pick up cues and to discern patterns andtrends. We can make better decisions, Klein tells us, if we are prepared forcomplexity and ambiguity and if we will stop expecting the data to tell useverything.