In The Norm Chronicles: Stories and Numbers About Danger and Death, Michael Blastland and David Spiegelhalter attempt to combine the objective norms of statistics with the intuitive flavor of anecdotes. They write a series of entertaining narratives on how laypersons and scholars conceptualize danger and respond to it. The final product of this attempt is a truly engaging and attention-grabbing narrative that guides the reader through a multitude of helpful and/or curious statistics.

In every chapter, the authors focus on the two facets of risk: (1) risk as a rather abstract and cryptic concept based on probability (e.g., the probability of creating a steady income from investing one’s retirement savings in the stock market without insurance against a market decline); and (2) risk as a phenomenon in people’s life (e.g., although the stock market is highly unstable, a person may be willing to gamble his/her retirement savings after reading about the successes of a trader in a news magazine).  As we all know, risk as a probability is less likely to shape people’s thoughts, feelings, and actions than anecdotes whose vividness, concreteness, and empathic power are difficult to ignore.  Anecdotes tend to be memorable not only because of these properties, but also because of their being often the centerpiece of news articles, repeated over and over by different media.  Repetition propels the illusion that anecdotal events are not only familiar, but also frequent occurrences. Thus, it is not surprising that the fear of being the victim of an airplane crash is much more intense and widespread than the fear of being in a car accident, even though the latter is much more likely to occur than a plane crash. 

In principle, it is easy to ignore or merely discount probabilities in reasoning, problem solving, and decision making. They are so abstract and impersonal. Anecdotes contain real people with whom audiences can identify. Sadly, the two facets of risk, and the guidance and advice that they may offer do not need to be reconciled, because one, probability, is largely absent from people’s mind.  Consider that the likelihood of an event (either desirable as a lotto win or undesirable as a car accident) is by no means certainty, even when the estimate of its occurrence is high.  Human beings search for predictability in everyday environments, mostly to maintain an illusion of control over their lives and destinies, and to nurture hopes that the future will be better than the present.  The likelihood of an event is a concept that is difficult not only to relate to the concreteness of one’s daily existence, but also to find convenient in predicting the occurrence of the actual event in one’s everyday life. Yet, experts in many fields use probabilities to estimate the likelihood of events in an attempt to ensure success in whatever endeavor they have chosen to pursue, whereas laypersons ignore probabilities as tools that are unable to produce sought-after certainties.

The authors discuss the two facets of risk in a variety of areas or domains, thereby making their narratives varied and potentially of interest to a broad readership. To ensure readers’ attention, they personalize stories by speaking of characters, actions, and outcomes while attempting to make the challenging connection between the odds of an event and people’s subjective views of it. Nevertheless, the perspective of probabilities and the perspective of individual cases remain an odd couple that even the skills of two clever writers such as Blastland and Spiegelhalter cannot integrate into more coherent storylines.  Perhaps, one way to bridge the gap is to focus on the information used by laypersons, rather than experts or computer programs, to estimate the probability of occurrence of events in everyday life so as to make decisions about present and future circumstances. That is to say, can the viewpoint of cognitive scientists shed light into the odd couple and make sense of incompatibilities?  Undeniably, this viewpoint is lurching in the background of the narratives developed by the authors as a stranger whose presence is recognized, but who remains distant from the intimacy of the actions being observed.  For instance, consider the distinction between subjective probabilities, which are estimates that are affected by the estimator’s temporary propensities (e.g., his/her mood) and/or stable dispositions (e.g., his/her extraversion or introversion), and objective probabilities, which are not.  It is reasonable to assume that a person’s mood may distort his/her estimates of the probability of success of an investment or of the probability of occurrence of a weather phenomenon (e.g., rainfall).  Mood, however, cannot distort the estimates of the probability of rainfall obtained from computer simulations of weather patterns used by the National Oceanic and Atmospheric Administration or the estimates of gains made by computer simulations at an investment firm.  Nevertheless, mood can distort the assessment of such estimates, leading a person to discount or attend to information selectively.  Furthermore, consider that objective probabilities are not always available. Thus, although discrepancies between objective probabilities and subjective probabilities have the potential to produce cognitive dissonance in one’s mind, hopefully leading people to give objective probabilities more serious consideration, discrepancies are avoided if objective probabilities are an unknown quantity.   

To understand the influence that anecdotes have in shaping the predictions that laypeople make, it might be helpful to contemplate some of the questions that cognitive scientists are likely to ask: How do laypersons gather the information that they use to make decisions about both minor matters (e.g., what time do I leave home to avoid being late for work?) and relevant affairs (e.g., shall I pursue a graduate degree?).  How do laypersons assess the relevance of the items of information that they have gathered? Do they use shortcuts?  These questions recognize the importance of an in-depth analysis of not only the biases of attention and cognition that people exhibit while processing events in their worlds, but also the reasons behind people’s persistent reliance on less than accurate devices and methods (Johnson, Blumstein, Fowler, & Haselton, 2013). 

Consider that humans’ biases in reasoning, decision making and problem solving tend to continue unabated for the simple reason that these biases can produce quickly and effortlessly ‘convenient solutions’ to everyday riddles. Not surprisingly, the account that people are ‘cognitive misers’ is based on a variety of phenomena whose desirable properties (e.g., being largely effortless and fast) equate them to bad habits difficult to discard or even control.  Often cited are phenomena that illustrate people’s tendency to selectively process mostly information that supports preexisting belief systems (i.e., confirmation bias; Schweiger, Oeberst, & Cress, 2014; Wason, 1960), their propensity not to consider all possibilities when estimating danger or other more desirable outcomes (Erickson, 1978), and/or their preference for heuristics (i.e., shortcuts or rules of thumb used to estimate probability, numerosity or frequency; Tversky & Kahneman; 1990).  The powerful influence that anecdotes have on people’s estimates of a variety of outcomes arises from the ‘normal’ functioning of the human mind which, as these and other phenomena demonstrate, is largely reliant on biases.  Knowledge of the human mind’s less than perfect, but utilitarian, information processing devices and methods may actually tell us whether specialized training can offset the gap between the perspective of objective probabilities and the perspective of individual anecdotes (as it would be desirable). It is the opportunity for an in-depth discussion of the human mind and of existing cognitive models of information processing that the authors of The norm chronicles: Stories and numbers about danger and death appear to have missed. Yet, Blastland and Spiegelhalter offer so many entertaining statistics that readers will be certainly willing to forgive.   

Works Cited:

Erickson, J.R. (1978) Research on syllogistic reasoning. In R. Revlin and R.E. Mayer (eds.), Human Reasoning.Washington, D.C., Winston.

Johnson, D. P., Blumstein, D. T., Fowler, J. H., & Haselton, M. G. (2013). The evolution of error: error management, cognitive constraints, and adaptive decision-making biases. Trends in Ecology & Evolution, 28(8), 474-481. doi:10.1016/j.tree.2013.05.014

Schweiger, S., Oeberst, A., & Cress, U. (2014). Confirmation bias in Web-based search: A randomized online study on the effects of expert information and social tags on information search and evaluation. Journal of Medical Internet Research, 16(3), 369-382. doi:10.2196/jmir.3044

 Tversky, A., & Kahneman, D. (1990). Judgment under uncertainty: Heuristics and biases. In P. K. Moser (Ed.), Rationality in action: Contemporary approaches (pp. 171-188). New York, NY, US: Cambridge University Press.

 Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. The Quarterly  Journal of Experimental Psychology, 12129-140. doi:10.1080/17470216008416717

© 2014 Maura Pilotti

Maura Pilotti, Ph.D., Ashford University