When it comes to calculating risk, we are hardwired to respond to the risks of a pre-modern world. As STATS Senior Fellow Maia Szalavitz points out in the January/February issue of Psychology Today:
The human brain is exquisitely adapted to respond to risk—uncertainty about the outcome of actions. Faced with a precipice or a predator, the brain is biased to make certain decisions. Our biases reflect the choices that kept our ancestors alive. But we have yet to evolve similarly effective responses to statistics, media coverage, and fear-mongering politicians. For most of human existence, 24-hour news channels didn’t exist, so we don’t have cognitive shortcuts to deal with novel uncertainties.
Still, uncertainty unbalances us, pitching us into anxiety and producing an array of cognitive distortions. Even minor dilemmas like deciding whether to get a cell phone (brain cancer vs. dying on the road because you can’t call for help?) can be intolerable for some people. And though emotions are themselves critical to making rational decisions, they were designed for a world in which dangers took the form of predators, not pollutants. Our emotions push us to make snap judgments that once were sensible—but may not be anymore.
Take 9/11: In the wake of terrorists hijacking and crashing airliners into the World Trade Center, there was a dramatic drop in air travel. Some 1.4 million people altered their travel plans, and many of them chose to drive instead. But terrorist attacks notwithstanding, traveling by car is riskier than air travel. Driving, on average, kills around 38,000 people each year in the U.S., or approximately 1.42 fatalities for every 100 million miles traveled.
The annual rate of airline fatalities fluctuates wildly depending on whether there’s a crash or not. But over a ten year period from 1992 to 2001, and including 9/11, researchers at the University of Michigan’s Transportation Research Institute at Ann Arbor calculated that the risk of death for a passenger on a non-stop domestic flight by a U.S. carrier was less than 1 in 10 million, or about one fatality per 9.4 billion miles traveled.
(There are difficulties though in directly comparing the risk of each mode of transport: for instance, driving on an interstate highway is safer than driving on city streets, but while the risk of an accident is evenly distributed in most car journeys, which is to say, that all things being equal, there is no part of the route that is typically safer than another part, the risk of an accident in flying is greatest at take off and landing.)
Researchers at Cornell University found that because more people drove instead of taking a plane in the three months after 9/11, there were an additional 725 fatalities from car crashes than during the same period in the previous year. A study by Michigan’s Transportation Research Institute took a different tack and came up with a higher figure -1,018 more traffic fatalities than would have been expected based on earlier trends.
In explaining these results, the Cornell researchers cite Cass Sunstein’s comment in “Terrorism and Probability Neglect,” that “when strong emotions are involved, people tend to focus on the badness of outcome, rather than on the probability that the outcome will occur.” Moreover, as Szalavitz notes,
Because fear strengthens memory, catastrophes such as earthquakes, plane crashes, and terrorist incidents completely capture our attention. As a result, we overestimate the odds of dreadful but infrequent events and underestimate how risky ordinary events are. The drama and excitement of improbable events make them appear to be more common.