Book of the month: Why we make mistakes by Joseph Hallinan

"Why we make mistakes" is another "light" read for this month following on from last month's "Set phasers on stun". Joseph T. Hallinan is a former writer for the Wall Street Journal and winner of the Pulitzer Prize. The Pulitzer Prize was for investigative reporting on medical malpractice in Indiana, USA, and so it seems apt that he has now written a book on mistakes.

Who's it for?

This book is written for a general audience and, as summer holidays draw to a close, could be squeezed into the last few days off, now that the kids have gone back to school. Hallinan covers a number of human factors terms and concepts such as hindsight bias (p. 5, p.65), different types of mistakes (p. 8), framing (p.92), anchoring (p. 103) and illusion of control (p.162). He also mentions some of the big names in human factors research such as Simons and Levin (p. 14), Kahnemann (p.93, p.206), Ericsson (p. 172) and Gaba (p. 192).

I haven't got time to read a whole book...

Read the introduction, chapter 1, chapter 5 and the conclusion. (For a description of these chapters see below)

What's good about this book?

Hallinan starts off well, stating:

"When something goes wrong the cause is overwhelmingly attributed to human error.... And once a human is blamed, the inquiry usually stops there. But it shouldn't - at least not if we want to eliminate the error."(p.2)

In the subsequent 13 chapters, he goes on to look at different causes of mistakes and offers advice on avoiding them.

The tabletop on the left is obviously narrower and longer...

In the first chapter, "We Look but Don't Always See," Hallinan refers to vision, perception and change blindness. The "door" study by Daniel Simons and Daniel Levin may be a useful example to use when discussing this aspect of human factors in a lecture or workshop. This chapter also uses the two tabletops image to show that even when we know that something is true we still cannot force ourselves to see the truth.

In this first chapter, Hallinan then goes on to provide us for another reason that radiologists miss dancing gorillas (and tumours). The less frequently something occurs, the more likely we are to miss it. Because tumours are infrequent radiologists tend to miss them (this is another argument for not requesting tests such as chest x-rays on the off chance that something may be picked up.) Baggage screeners fare little better for the same reason, as Hallinan states, in 2006 at "Los Angeles International Airport screeners missed 75 percent of bomb materials." A reassuring thought for the next time you're standing in line at the airport with your belt and boots in one hand and the other hand holding up your trousers.

The next chapter which has a decent shot at dealing with a human factors problem is chapter 5: "We Can Walk and Chew Gum - but Not Much Else". Hallinan explores the myth of multi-tasking and looks at task saturation, trying to do too many things at the same time. An interesting concept here is that we can walk and chew gum at the same time because neither of these tasks requires conscious thought. We can drive a car and have a conversation because (once we are proficient) driving the car no longer requires our undivided attention. This idea may be useful to present in a human factors workshop or talk; i.e. repeated practice of managing crises in a simulator will allow one to use less mental workload on the basics during a real crisis.

Would you work for this man?

A final concept which may prove of interest is that constant change at the top of an organisation may not be helpful to the organisation itself. Hallinan discusses Warren Buffett's company, Berkshire Hathaway, where none of the CEOs have voluntarily left for other jobs in its 42 year history (p. 160). Hallinan proposes that the CEOs stay in post long enough to receive feedback and learn from their mistakes. Might this have lessons for the NHS?

Hallinan's advice for making fewer mistakes is given in the conclusion and that is to "Think small". Unfortunately he doesn't then quite manage to explain what this means, but instead goes on to provide other, more useful, advice:

1. Calibrate yourself. Be honest when you consider decisions that you have made and what their effect has been. (We have a tendency to think we performed better than we actually did.)
2. Think negatively. By considering what can go wrong you can prepare for that eventuality.
3. Ask others for advice. A spouse, friend or colleague may provide insight into a problem.
4. Sleep. Fatigued people perform less well.
5. Be happy. Happy people make decisions more quickly and are more creative with their problem-solving.

What's not so good about this book?

Areas which could be improved include Hallinan's introduction to System 1 and System 2 thinking which is somewhat clumsy, quoting Paul Slovic: "Our perception of risk lives largely in our feelings, so most of the time we're operating on system No. 1" (p. 95) He also digresses somewhat to spend a significant number of pages discussing the direction sense of men and women (p. 143-148).

Additionally Hallinan shows a slight lack of experience when he discusses how time affects decisions, stating:

"Many factors can affect the way we frame our decisions. One of the least obvious is time."

Anybody who has worked in a simulator or in an environment where time is critical, will be aware of how great an effect time has on on decision-making. Hallinan also mentions looking for "the" root cause which Sidney Dekker would have something to say about, namely:

"Asking what is the cause (of an accident), is just as bizarre as asking what is the cause of not having an accident. Accidents have their basis in the real complexity of the system, not their apparent simplicity. (p. xiii)" (The Field Guide to Understanding Human Error"

A final criticism of Hallinan's book (and perhaps much of human factors research) is that so many of the conclusions are based on research carried out on American university (Princeton, Ohio Wesleyan, Duke, Carnegie Mellon, Yale...) students; are these truly representative of the rest of the population?

To conclude

This is another book which should be borrowed, not bought. It has some material that may be useful for those who prepare human factors workshops. For the human factors novice it provides an easy-to-read introduction to some of the concepts and big thinkers in the field.