The Limits of Expertise: Rethinking Pilot Error and the Causes of Airline Accidents

by R. Key Dismukes, Benjamin A. Berman, Loukia Loukopoulos

"The Limits of Expertise" reports a study of the 19 major U.S. airline accidents from 1991-2000 in which the National Transportation Safety Board (NTSB) found crew error to be a causal factor. Each accident is reported in a separate chapter that examines events and crew actions and explores the cognitive processes in play at each step. The majority of all aviation accidents are attributed to human error, but this is often misinterpreted as evidence of lack of skill, vigilance, or conscientiousness of the pilots. Why would highly skilled, well-trained pilots make errors performing tasks they had successfully executed many thousands of times in previous flights? The approach is guided by extensive evidence from cognitive psychology that human skill and error are opposite sides of the same coin. The book examines the ways in which competing task demands, ambiguity and organizational pressures interact with cognitive processes to make all experts vulnerable to characteristic forms of error. The final chapter identifies themes cutting across the accidents, discusses the role of chance, criticizes simplistic concepts of causality of accidents, and suggests ways to reduce vulnerability to these catastrophes. The authors' complementary experience allowed a unique approach to the study: accident investigation with the NTSB, cognitive psychology research both in the lab and in the field, enormous first-hand experience of piloting, and application of aviation psychology in both civil and military operations. This combination allowed the authors to examine and explain the domain-specific aspects of aviation operations and to extend advances in basic research in cognition to complex issues of human performance in the real world. Although "The Limits of Expertise" is directed to aviation operations, the implications are clear for understanding the decision processes, skilled performance and errors of professionals in many domains, including medicine.

Genres: Aviation, Science

368 pages, Paperback

First published January 1, 2007

Reviews

[Alex · Rating 5 out of 5]

Every pilot should read 'The Limits of Expertise.'

Written by three scientists from the NASA Ames Research Center, 'The Limits of Expertise' is exquisitely researched and carefully composed. Best of all, it's well written!

The book consists of an introduction, nineteen case studies, and two chapters of conclusions. The case studies are riveting. As a flight standards captain with a major U.S. airline with over 30 years of experience, I could put myself in the control seat with the case study pilots every time. I, as have most other pilots, have been in situations similar to the ones described. I zigged. They zagged. Maybe someone else zogged. All of us can say, "There, but for the grace of God, go I."

In the case studies, the authors break down the steps in the causal chain that led to each of the incidents. Their great insight, in my opinion, is their repudiation of determinism. They understand that pilots, as human beings, aren't robots or algorithms. We have variations in performance, due to a wide variety of internal and external factors.

The authors combine their understanding of pilot psychology with statistical analysis in the concluding chapters, highlighting the scenarios that most airline mishaps have in common. I find this to be immediately, practically, usable: I'll brief these threats on my next trip.

Really, this is an excellent book. Its exploration of failure in high-stress, high-expertise situations is applicable to experts in a variety of fields. I can imagine versions written for physicians, first responders, and other experts who work in areas with tiny margins of error.

If you're a pilot, read this next. If you're an expert in a stressful, life-or-death field, read it soon. If you just think psychology is interesting, read it. I'll be recommending this book to my fellow pilots for years to come.

[Eric Cordina · Rating 5 out of 5]

An excellent book which gives insight into why "pilot error" happens. Each of the incidents reviewed is analysed and the pilots' actions gauged not using the standards imposed by hindsight but by asking the simple question "Would most other pilots act similarly in the same situation". The answers might surprise a few, but the analysis shows how much more can be done at organisational level, during training, during aircraft & system design to help prevent accidents. Reading about accidents that happened now well over 20yrs ago, one can already appreciate some of the changes that have been implemented to the benefit of the aviation community and the travelling public in general.

[Nick · Rating 5 out of 5]

"[...] when airline crews are not able to manage situations adequately, it is most often because of limitations of the overall aviation system rather than inherent deficiencies of the pilots. [...] crews are implicitly expected to make up for the deficiencies of the systems in which they operate by using skills and judgement [...]"

This was a fascinating read and I recommend it to anyone who works in aviation, or anyone who works in a field where experts in an operational environment are expected to work reliably in high-stress situations.

[Wayne · Rating 3 out of 5]

Rather extensive look into the limits of pilots to control aircraft in serious situations. While advances in aircraft systems have eliminated many of the root causes of accidents, the ability of the human pilots do have a finite limit.

[Lane · Rating 5 out of 5]

Excellent book for anyone wanting to better understand some of the ways accidents happen. It differs from many books on aviation safety by not focusing so much on what the individual pilots did wrong, but looking at it in more a a probabilistic and systematic way- asking the question "out of a large group of pilots how likely is it that some would make the same decisions and/or errors that were made here?" I found this subtle shift in focus to be helpful, and it lines up well with current airline AQP systems to better focus on procedures/habits that we can improve.

While some of the 19 accidents examined have causal factors that current technology (EGPWS, RAAS, digital doppler radar, etc.) or procedures may have prevented, others highlighted areas that we are still working on. One area in particular that showed up repeatedly was the performance/communication of the pilot monitoring. While we have improved in this area over the years, it is still a focus area and continues to challenge us.

My only minor complaint is that I occasionally felt the authors were quick to dismiss the way the pilots in the case studies operated as too high risk without a thorough examination of the environment and reasons why pilots operate that way. This only happened in a couple of the case studies, but those discussions felt incomplete and the conclusions considered the risks of choices without the reasons why those choices were made. Again, this was a minor point and I did not feel it detracted from the overall usefulness of the book.

Highly recommended for anyone wanting to better understand the (somewhat) recent history of aviation safety.

[Chance Barber · Rating 4 out of 5]

We can learn many things from these accidents. Interesting insight into how technology & equipment have been improved due to these accidents.