Monday, October 27, 2008

Only the Brain Damaged Are Good at Risk Management


Risk & Insurance Online - Story October 4, 2007

It is an odd fact of life that you often find what you are looking for in the place you would not expect to find it. For example, the best risk management books--such as Moneyball, Bringing Down the House and Fooled by Randomness--were not explicitly about risk management. Similarly, I find some of the best ideas concerning risk management come from science magazines.
By Beaumont Vance

In the June issue of Scientific American, regular columnist Michael Shermer frames the issue of decision-making from an evolutionary perspective. Decision-making, he suggests, is the product of the evolutionary drive towards efficiency:

"Life, like the economy, is about the allocation of limited resources that have alternative uses. ... It all boils down to energy efficiency. To a predator, prey are batteries of energy."

He then quotes Read Montague: "This doctrine mandates that evolution discover efficient computational systems that know how to capture, process, store and reuse energy efficiently."

The real problem with the state of our decision-making is that the environment for which these programs were designed no longer exists. There might be some nasty people at work, but it is unlikely that they are going to attack your village in the night.

In that confusion we can be tricked. Coca-Cola, Shermer points out, was shown in experiments to trick the human brain: "The Coke brand has a 'flavor' in the ventromedial prefrontal cortex, a region essential for decision-making."

A soft drink ad can impact the decision-making region of our brains.

Shermer's article is yet another arrow pointing at how our brains are not as well equipped to make decisions as we think. ...[B]ehavioral scientists have so far identified about 200 biases that cause us to make illogical decisions. If we are not aware of these, they will trip us up (see the July 2007 issue of Risk Management Reports on correcting biases.) But if we understand them, we gain an advantage when dealing with decisions involving risk....

Aside from allowing one to escape from unwanted cocktail party conversations, this little part of the brain is also very important for decision-making. According to Time (the January 29, 2007, issue, P. 116), this has been demonstrated in an oft repeated experiment in which researchers pose hypothetical situations to the subjects and measure how long it takes them to make a decision.

Josh Green of Princeton has made a career out of strapping people into an fMRI machine (or functional Magnetic Resonance Imaging machine), which can graphically depict brain activity, and asking them to make decisions involving the death of others.

For example, he proposes a situation in which a runaway train is headed for a group of five people. The train can be diverted to a side track, but there is one person standing on the side track. Therefore, by diverting the train and saving the five people, the guy on the sidetrack will be killed. In other words, it is a choice of letting five people die, or killing one person.

The researchers then measure how long it takes the subjects to decide. Readers should remember that in such a situation, lengthy deliberation will mean that five people will die; no decision is indeed a decision.

Then, to spice up what would be an otherwise tedious experiment, the researchers change the scenario. This time, the train is still heading for the five people standing on the track. But instead of pulling a switch to divert the train, the subjects are told that they can stop the train only by pushing a bystander in front of the train, thus derailing it and saving the five people. It is the same problem: Let five people die or kill one person.

In the first scenario, where the switch can be pulled, 90 percent of the subjects have no trouble deciding to pull the switch. However, in the second scenario, when they have to push an innocent bystander in front of the train, 90 percent choose to let the five innocent people on the track get killed.

According to Josh Green, the subjects use completely different parts of their brain on each of these decisions. In the former, logic tends to win out. In the latter, an older part of the brain takes control. It is the same decision, but our brain takes over and throws logic out the window when the circumstances change a bit.

This same experiment was conducted with people who have damage to the ventromedial prefrontal cortex (VMPC). It turns out that people with damage to the VMPC are three times more likely to advocate throwing a person to certain death to save five lives. They made the decision much more quickly than people with undamaged brains.

Human decision-making is a complex process that we still fail to grasp completely. While it is hard to see any practical applications to this study--aside from screening all decision-makers, board members and political leaders for VMPC damage--it is a great demonstration of the folly of believing that decisions can be entirely rational. ...

There are, however, some practical implications to these studies. ...[T]hey also demonstrate how differently people make decisions if they have to get blood on their own hands.

Think of this in the context of layoffs. A senior executive might make a rational decision to fire 500 workers. All managers in the company could agree that it is necessary for the survival of the company. After all, if the cuts are not made, the entire company could fold and then everyone would lose their jobs.

However, when the individual, lower level managers are asked to actually fire some of their own people, they balk. In other words, senior management is simply pulling the railway switch, while the line managers are pushing innocents under the train. No wonder layoffs are so difficult to execute!

We should be careful about our assumptions of what constitutes "rational" thinking. If ads for Coke can affect the VMPC, one has to wonder what else can rewire our decision machinery. The better we understand how our brain works, the better we can become at correcting deficiencies. Who knows, perhaps advertising Coca-Cola in the office place could nullify the problems caused by that pesky VMPC. Have a Coke and a RIF!*

(* For those of you not familiar with American corporate slang, RIF is the acronym for Reduction In Force. It is a euphemism for a layoff.)

BEAUMONT VANCE manages risk for Sun Microsystems Inc. This column was a complimentary excerpt from one of his latest "Risk Management Reports" newsletters, which he edits and publishes. For more information on how to subscribe to the full version of the newsletter, please visit .

Click here to read Beaumont's latest Risk & Insurance® columns.

READ MORE: Features | Special Reports | Industry Risk Reports | Columnists | In-Depth Series