What Can You Learn from Accident Reports?
by Gerry Binnema, Civil Aviation Safety Inspector, System Safety, Pacific Region, Civil Aviation, Transport Canada
A lot of people who work in aviation like to read accident reports. The reports serve as good reminders that aviation can be dangerous and that we always need to be vigilant. But if we were all being completely honest, sometimes we read them because it makes us feel smugly superior to the people who messed up. So, how much do we really learn from reading accident reports? Surprisingly, there has been very little research to see if accident reports actually have any positive effect on the people reading them. It seems very obvious that accident reports would be helpful, but there are a number of things that interfere with our ability to learn lessons from them.
Our brains process information and organize it before it is presented to our conscious attention. This processing follows certain relatively predictable patterns, which serve to help us understand the world around us. However, this processing can also distort our view of things, as information gets processed in such a way as to protect our self-esteem and our confidence. The patterns of processing that are very relevant to our understanding of accident reports are hindsight bias, attribution error, and invulnerability.
Readers will recall Heather Parker’s series of articles on the "new view" in the past three issues of the Aviation Safety Letter (ASL). In these articles, she described hindsight bias and attribution error. These concepts also apply when we are reading accident reports. By way of a brief review, hindsight bias refers to our tendency to look back at events and believe the events should have been predictable beforehand. A classic example of hindsight bias is the Monday-morning critique of the weekend’s sporting events by armchair athletes. The coach should have anticipated the other team’s strategy. They should have known that the goalie would get re-injured if they put him in so soon. In reality, as we try to anticipate what will happen next, there are many different potential outcomes and we make the best decision we can with the information that we have available. As we read an accident report, we already know how the flight ends, and so we tend to judge all the decisions that led up to the accident with hindsight bias, believing that the pilot should have known better.
Attribution error refers to our tendency to overestimate the contribution of personal factors when we observe other people’s errors. This means that when we see other people making a mistake, we tend to believe that their errors are a result of their own inadequacies (ignorance, incompetence, laziness), rather than a result of situational factors. Even when a situation arises over which the pilot had no control, we still tend to believe that they were at fault for allowing themselves to get into that situation.
Invulnerability refers to our tendency to believe that bad things will not happen to us. Of course, there are hazards all around us, so in order to enjoy life we suppress our fear and deny the possibility that anything will happen. But an unrealistic sense of invulnerability actually places us in danger. Young people, especially males, have higher levels of invulnerability, and this can be observed in the number of accidental injuries and deaths among young males. A strong sense of invulnerability will prevent us from taking the lessons of an accident report to heart.
In combination, these three factors make it easy to read an accident report and learn very little. It would be almost natural to believe that the pilot should have known better, that their errors were caused by their own ignorance or incompetence, and that this kind of thing could never happen to you.
I recently had an opportunity to conduct some research to see if accident reports were having an impact on readers. Eighty-nine college aviation students participated in the study by completing a questionnaire, and then six weeks later reading an accident report and completing another questionnaire. The questionnaire was intended to measure invulnerability and attribution error.
The participants’ responses to the questions on invulnerability showed very clearly that they did not believe they could be in an accident. The participants also clearly demonstrated a willingness to place the entire responsibility for an accident on the pilot, even when a number of situational factors contributed to the accident. However, the most interesting finding was that there was a remarkably consistent, but small, decrease in the measures of invulnerability immediately after reading an accident report. This means that reading an accident report does have an impact on the reader and does help to make a pilot think about their vulnerability to an accident.
The participants read one of two accident reports. One was a typical accident report format, while the other was written in a narrative format, describing the unfolding events from the pilot’s perspective. Both report formats achieved the same level of change in invulnerability. However, the latter format was able to build sympathy for the pilot so that participants who read this style of report were more likely to believe that they could commit the same errors and be in a similar type of accident.
This is good news for those of us who read a lot of accident reports. It really does give us a more realistic sense of the fact that we could be in an accident if the wrong set of circumstances hit us. In addition, earlier research (see
http://psy.otago.ac.nz/cogerg/Rememb...ses%20Past.pdf) conducted in New Zealand, and repeated here in Canada, demonstrates that we do recall lessons from accident reports while in flight. However, in order to make the most of these lessons, we need to keep some things in mind. Here are some practical suggestions for reading accident reports:
• Be aware of the fact that hindsight bias and attribution error do alter your perspective on an accident. As you read a report, think about how the unfolding events might have appeared to the pilot. Think about the decisions the pilot made, and try to ignore the fact that they resulted in an accident. Could you have made the same decisions? What circumstances might have led you to those decisions?
• Be aware of the fact that the majority of people have an unrealistically optimistic belief about the probability that they will be in an accident. Ask yourself if you are really being as cautious as you should be.
• Finally, as you read an accident report, remember that the pilot’s actions made sense to them at the time. If you cannot make sense of the actions, you do not understand the situation as the pilot understood it. Try to step into the pilot’s shoes and see if you can build sympathy for their predicament. Could you fall into the same trap? Could some external pressures or stresses cause you to behave in this way?
• If we all use this kind of strategy as we read accident reports, we are more likely to learn valuable lessons from them, and this may prove to be the critical piece of information in some future decision you need to make. In the next issue, I will look at how to apply these same ideas to the way organizations think