The Behavioral Revolution
Reading David Brooks’ elegant summation of how behavioral psychology and economics can help explain the dynamics of the financial crisis — to explain, as he puts it, “why so many people could have been so gigantically wrong about the risks they were taking” — I couldn’t help but think of another area rife with questions of risk and decision-making, namely ‘the way we drive.’
Whether from personal on-road experience, or from reading studies, or from examining in-car footage of crashes and near-crashes, I am often struck by how often people seem to put themselves, and others, at great risk. Following closely at high speed on the interstate, or driving fast through a neighborhood street, they act in a way that suggests they believe that nothing could go wrong, or that they would be in control if it did. Over time, this behavior is typically rewarded, perhaps through sheer luck, until the ‘black swan’ event that they never expected actually happens. Then, as is often the case, begins a process of denial, an attempt to assuage the cognitive dissonance that has come between the image of themselves as a good and cautious driver and an event that was ‘beyond their control.’
Some quite literal connections can be drawn between the behavior of traders and the behavior of drivers. For one, both activities are prone to the ‘above-average effect’ — studies have shown how both large groups of traders and drivers define themselves to better than average. What’s also interesting is the gender question; research has also shown men seem to be more susceptible to the above-average effect. As Brad Barber and Terrence Odean showed in their paper Boys Will Be Boys, a study of a large brokerage house found that men made many more trades than women, per account, seemingly indicating a heightened sense of confidence, but that their portfolios on average earned less than women. Given the male dominance in the trading sector, it’s not hard to extrapolate these findings to the larger financial crisis. It also need hardly be pointed out that men are involved in more fatal crashes than women — overconfidence mixed with a greater propensity for risk-taking.
Another connection is the way we act on the information we perceive. “And looking at the financial crisis,” writes Brooks, “it is easy to see dozens of errors of perception. Traders misperceived the possibility of rare events. They got caught in social contagions and reinforced each other’s risk assessments. They failed to perceive how tightly linked global networks can transform small events into big disasters.”
This passage reminded me of a recent conversation I had had with a journalist in Abu Dhabi, who was telling me about the massive, fatal chain-reaction crashes that have occurred on fog-bound highways there. Fog is a classic perception problem: Differences in contrast affect how we perceive speed. Moving through fog, drivers actually feel as if they are moving more slowly than they are. So they continue to drive fast, much faster than they should. They may also drive close to the vehicle in front of them, thinking, falsely, that seeing the taillights of the driver ahead is safer than not seeing anything. All this is fine until the rare event happens, and that ‘tightly linked’ network, full of people reinforcing each other’s risk assessments and acting on what they think is sufficient information (but which may disguise hazards around the bend), find themselves in a calamitous crash.
Behavioral psychology isn’t part of the driver’s ed curriculum, of course, but there’s no reason it shouldn’t be, given that attitudes and behavior are as, if not more, important than driving skills per se. And the simple vision test that’s given is fine for testing the strength of one’s vision, but left unmentioned is the idea what we see of the world does not always represent the world as it is.
This entry was posted on Thursday, October 30th, 2008 at 9:36 am and is filed under Traffic Psychology, Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.