QnAs with Daniel Kahneman.

The science of human decision-making has long been a stronghold of psychologists. Among the voices that abound in the literature on how people make choices, one scholarly voice has remained strident through several decades. By studying human behavior through the lens of economics, Princeton University psychologist Daniel Kahneman has shown how inherent biases might influence our choices, overtly and subliminally. For his influential work on prospect theory, which holds that people make decisions based on their perceptions of losses and gains rather than on final outcomes, Kahneman shared the 2002 Nobel Memorial Prize in economics. Working together with late psychologist Amos Tversky, a longtime research partner, Kahneman demonstrated that decisions believed to be the result of deliberation often stem from educated guesses, rules-of-thumb, and pattern-recognition. In his book Thinking Fast and Slow, a critical success published in 2011 and targeting a broad audience, Kahneman reports on a wealth of experimental research that point to two modes of decision-making in the humanmind:Whereas system 1 is automatic, effortless, and rapid, rushing to judgment based on heuristics, system 2 is deliberate, effortful, and slow, plodding through reason before reaching conclusions. Understanding how we toggle between the two systems, Kahneman says, can help us make sense of a broad swath of human behavior. On the occasion of the 2012 Sackler colloquium The Science of Science Communication, where he was the keynote speaker, Kahneman discussed his work with PNAS. PNAS: What motivated you to write a popular book for a general audience? Kahneman: I’ve had a long career and worked in many fields. This book, which is not an autobiographical account, is a review ofmany of thosefields. I wanted towrite a book that would speak not only to the public but also to future scholars. Articles may become dated or less easily accessible, but a scholarly book that appeals to a wide audience might endure. PNAS: In the book you describe two fictitious systems in our mind—systems 1 and 2—that underlie our automatic and considered decisions, respectively. Would an acute awareness of the existence of these two systems allow people to suppress system 1 while making decisions? Kahneman: Most of the time we follow system 1; our impulses are generated automatically, but they are also continuously monitored. This, for example, is how people almost alwaysmanage to refrain from swearing in polite company. However, the selfmonitoring is fairly lax, andmost of the time the suggestions that arise automatically from system 1 are simply endorsed by system 2 and expressed in behavior. As a result, impressions turn into beliefs, and impulses into choices. However, if you wanted to monitor yourself closely all of the time, it would quickly become impossible because system 2 is much slower and less efficient than system 1. So the idea of replacing system 1 with system 2 is infeasible. PNAS: You write that system 2, the part of the mind that helps us make considered decisions, tends to be less active in monitoring system 1 when a person is happy. What are the implications of that finding? Kahneman:Thefinding thatmood has subtle but pervasive effects on people’s thoughts has emerged in the last 15 years or so. Generally, people become less vigilant when they are in a good mood, when system 1 impulses are more likely to express themselves as beliefs or choices. It turns out that there is a close association between vigilance and the degree to which system 2 supervises system 1. What’s more, the relationship is reciprocal: when the fluency of associative processing is low, it tends to result in bad mood, which, in turn, affects associative processing. So people are more likely to make more superficial mistakes when they are in a good mood; it’s not an enormous effect but a fairly consistent one. PNAS: You also distinguish between two kinds of well-being: experienced and remembered. You note that memories of an experience are influenced by its intensity and ending, not by its duration. And to support the assertion, you cite experiments with people’s recollection of unpleasantness during colonoscopies administered under different conditions. Can you elaborate on your findings? Kahneman: Associative memory, or system 1, works by producing chunks of narrative. Certain aspects of the narrative, such as causal relations, tend to be emphasized. However, the duration of an experience is generally not very important in assessing its overall value. This is similar to narratives where events are critical and the uneventful passage of time is ignored. We are aware of how long experiences last, and we certainly know that it’s better for good experiences to last longer and bad experiences to be brief. However, our spontaneous evaluation of past experiences is generally quite insensitive to duration. Neglect of duration would make sense from an evolutionary perspective: after a threatening episode, it is important for an organism to remember how bad the threat was and how the episode ended; how long the episode lasted is essentially irrelevant. PNAS: The book also touches upon the importance of familiarity to the acceptance of a message or conclusion. To wit, you describe the mere-exposure effect. Can you explain the effect and its implications for communicating scientific messages to a lay audience? Kahneman: The mere-exposure effect holds that when we are repeatedly exposed to something, we tend to trust or like it more. The evolutionary justification for the effect, as proposed by the late Robert Zajonc, is that if you’ve been in prolonged contact with something and it hasn’t killed you, then it must be relatively safe. The increasing sense of safety with the familiar is thus an important aspect of our everyday lives. However, scientific messages don’t always lend themselves to easy repetition, so I’m not convinced of the value of repeating slogans to get scientific ideas across to a wide audience. On the other hand, science communicators should realize that if themessage is intended to lead to action, they are effectively addressing people’s system 1, which thrives on stories, individual anecdotes rather than statistics or evidence. And most people’s beliefs are shaped not by arguments but by the beliefs of others they trust. Counterintuitive as it may seem to scientists, most people believe in conclusions before they accept arguments. So stories and source credibility are at least as important as the quality of arguments when it comes to the public acceptance of scientific ideas. Daniel Kahneman.