The Elephant Test

As those of you who have read my previous columns know, I am great admirer of the work of the Nobel Prize winning psychologist Daniel Kahneman. Dr. Kahneman won the Nobel Prize in Economics for his work on Prospect Theory -- a theory that challenged decades of assumptions that people made economic decisions in a rational way. He followed this up with a best-selling book, "Thinking Fast and Slow," which sold more than 7 million copies, written when he was 77 years old and he recently released the sequel, "Noise," written when he was 87 years old. I firmly believe that Dr. Kahneman is one of the great intellects of the 21st century.

On Aug. 25, I had the incredible good fortune to join a small group of people on Zoom and speak with Dr. Khaneman for about an hour to learn more about his background, the work that led to his important discoveries and his thoughts about how these theories affect our lives. Dr. Khaneman is an alumnus of the Hebrew University of Jerusalem (which boasts eight other Nobel Laureates) -- the event was arranged by the American Friends of the Hebrew University, one of many groups worldwide that raises money for the university. As for the discussion, let's just say that it's not every day that you have the chance to chat with a Nobel Laureate.

Dr. Kahneman's new book came about as a result of an assignment given to him by a large insurance company. The company was interested in knowing whether there was consistency in its underwriting process. An analysis by Dr. Kahneman led to the startling conclusion that while an insurance company would usually assume and accept a roughly 10% variance in underwriting opinions, the actual variance in the analysis was greater than 50%. The "noise" of the title is Dr. Kahneman's term for this type of irrational and inconsistent variance in human decision making.

One of the examples in "Noise" is a study of sentencing by federal district court judges. In this study, more than 200 district judges were asked to determine the appropriate sentence in a hypothetical case. It turns out that the process is far more noisy than you might imagine. Some of the judges imposed sentences more than four years longer than other judges. Further research determined that the outcome of the sentencing process was affected by many non-rational factors, producing wildly differing outcomes. Dr. Kahneman also analyzed the effectiveness of interviewing job candidates using the traditional one on one process. He found that the process was essentially useless. The process was, in his words, "noisy," meaning irrational and inconsistent.

What does this have to do with the process of mediation? Quite a lot. In "Thinking Fast and Slow," Dr. Kahneman outlined his theory that human beings make decisions in two basic ways: "fast" and "slow." To illustrate, most adults if asked "what is two plus two" would answer "four" without really thinking about it. Their response would be intuitive. But what if the question was: "what is 17 times 17"? This would take some thinking, probably aided by pencil and paper or a calculator. This thinking would require an analysis based on basic principles of arithmetic and other learned means of resolving problems. This is "slow" thinking.

As human beings, most of our thinking is fast and we devote time and energy to slow thinking for only a small portion of everyday tasks. One reason is that fast thinking is convenient and slow thinking takes real effort. If you want to experience this for yourself, take this simple test: First, think of an elephant. It probably took you only an instant to picture an elephant, or think something about an elephant. Now, time yourself and try, for one full minute, to not think of an elephant. You will find that it's extraordinarily difficult.

What Dr. Kahneman learned is that while fast thinking is essential to daily life, it is also unreliable, not simply because it doesn't take much time, but because it's influenced by many other factors, mostly various kinds of biases. Thus, for example, confirmation bias drives us to reach unreliable results because we make our analysis based only on facts and arguments that support our position and we rebuff contrary facts and arguments. Attribution bias creates unreliable and irrational results because it prevents us from accurately understanding the behavior of other people -- instead of carefully understanding and analyzing their behavior, we attribute their behavior to "bad" things and thus we reject their positions. For example, if the defendant in a mediation were an hour late to the proceedings the plaintiff might attribute this tardiness to the defendant's disrespect for the proceedings and thus denigrate the defendant's arguments throughout the rest of the mediation -- but in fact the defendant was late not out of disrespect but because his car had a flat tire and his arguments might be rational deserving of respect and careful consideration. The plaintiff's summary, attribution bias-driven rejection of those arguments might lead to a disastrous result when those arguments later resonate with a judge, jury or arbitrator.

Back to mediation. The decisions of many people to either settle a case or instead proceed to trial or arbitration are greatly affected by these methods of analysis. People make irrational decisions because it is difficult to see that there are two sides to an argument -- they "like" their side, so they form the belief that "their side" is the only side (this is called "What I See Is All There Is" thinking). People accept the risk of a jury trial because they believe that the jury is "rational." Thus, they believe that the "rational" jury will see the case the same way that they see the case, when in fact the jury may be much more willing to listen to both sides and the jury may also be affected by other, more intangible, issues. In short, making an analysis of an important economic decision using fast thinking, or making assumptions that the system to decide your dispute will be at least mostly free of noise may be a dangerous way to proceed.

So, how do you and your clients bring a more rational approach to the process? One way is to use mediation as what Dr. Kahneman would call a "noise audit." In practice, in the context of a large organization like an insurance company, a noise audit would focus on identifying and then reducing or eliminating the inconsistencies in results, such as the 50% difference in the opinions of the underwriters.

How would this work in the context of mediation? First, it's vitally important to understand the concept of fast thinking and the perils to which it exposes you and your clients. Second, it's equally important to bring slow thinking discipline to the process of analyzing risk in the mediation process. Third, it's extremely important to be consciously aware of how bias may be driving your analytic process off a cliff. Fourth, it's helpful to listen carefully and openly to the thoughts of the third party neutral. In this regard, the idea of the third-party neutral is often used in large organizations as a form of noise audit when a company is considering a major decision, such as the launch of a new product. Dr. Kahneman's research disclosed that in such situations companies routinely underestimate the time, effort and cost and overestimate the probability of success and he also learned that a review by an independent person or persons can have a profound impact on the reliability of the analysis. Fifth, don't just listen. Instead, take the information and try to calculate how the analysis of risk really affects the case. For example, perhaps the mediator suggests that you have a statute of limitations issue that wasn't apparent before the mediation. Statutes are a zero-sum outcome -- they don't reduce the damages that you might get in court, they eliminate them entirely -- thus they have a profound impact on your case. Or, perhaps you have spent years with a particular client and your relationship with this client has blinded you to how this client is perceived by others -- you think of your client as a personable honest person, but the mediator suggests that others may view your client as arrogant and shady. Think about how this might affect your case if the trier of fact has the same perception as the mediator.

How do you help your client participate effectively in the "noise audit"? One very effective way is to say these simple sentences to your client: "Let's listen to what the mediator has to say." "That's one of the reasons why we are here and why chose this mediator -- we respect his opinion." Think about it for a moment -- you have spent a great deal of time and money for the mediation. You have prepared a brief. You have met with your client. You have set aside the day for the mediation. Is it more effective to say "Let's listen to what the mediator has to say," or is it more effective to say to your client once the Mediator leaves the room: "Don't listen to that -- all mediator's say that to try to scare you into settling."

I close with a more personal note about Dr. Kahneman. For all his formidable intellect, his Nobel Prize, his financial and academic success, his worldwide reputation, and his ground-breaking research that fundamentally changes our perceptions, he could not have been more personable, warm, engaging, or down to earth and patient. He is a remarkable and delightful person.