Behaviour and Motivation — Page 176, Lesson 278

I want to walk you through a critical area of human performance that directly affects decision-making on the flight deck: behaviour and motivation, and specifically how we evaluate evidence. This is not just theory—the book makes a very strong statement: the ability of the pilot to evaluate evidence and come to conclusions will, in future, be the only reason for keeping him on the flight deck. That tells you how seriously we need to understand these biases. Let's start with the list of factors that degrade situational awareness. First, lack of situational awareness itself—that's the state where you no longer have an accurate mental picture of what's happening around you. Then we have confirmation bias, which we'll come back to in detail. Next is frequency bias—this is the tendency to call to mind frequently encountered experiences or scenarios and then apply them to an inappropriate situation. In plain language, if you've seen a certain problem many times before, your brain will reach for that explanation even when the current situation doesn't fit. Then there's inference in accord with wishes, hopes or desires—that means deducing that a problem is one that is known and whose solution is within your capabilities, simply because that's what you hope it is. You're essentially convincing yourself it's a problem you can handle, rather than facing the possibility it might be something worse. Now, let's look at how pilots evaluate data. A decision requires using mental models—these are your internal frameworks that organise all the information available from your environment. But evaluating evidence is not straightforward. The weight you place on individual pieces of information will be biased by several factors. The first factor is probability of occurrence. For example, a bang heard on take-off could be a tyre burst, a bird strike, or an engine failure. A burst tyre is the most probable cause of a loud noise at this stage of flight. So pilots may carry out the initial drills for that event first. That's logical, but it can also trap you if the less probable event is actually happening. The second factor is previous experience. If you've had a certain component failure recently, you'll look immediately to that equipment as a source of warning. Your recent memory biases your search. The third factor is ambiguous data. A red light in the sky could be the setting sun reflecting off cirrus cloud, but it could also be the beginning of a volcanic eruption with the consequent danger of flying through volcanic ash. The same visual stimulus can mean two completely different things, and you have to evaluate which one it is. Fourth, people are keen to structure information and make inferences from it. Once you've formed a theory about what's happening, you become reluctant to move away from that interpretation and start again with a new theory. This is a natural human tendency, but it's dangerous in the cockpit. Fifth, and this is the core of confirmation bias: even if you are presented with contrary evidence, you will tend to ignore that evidence, but you will seize with alacrity—that means eagerly—any small detail that reinforces your original idea. If you try to test your hypothesis, you will tend to try only those instances that may reinforce your original thoughts, and you will not try negative instances that could disprove it. So you're not really testing your theory; you're just looking for evidence that supports it. Finally, there's a mechanism related to prepared responses. If you expect a stimulus and prepare a response, you will respond more quickly if the expected stimulus occurs. But if an unexpected stimulus occurs, you are more likely under pressure to carry out the prepared response anyway. For example, a pilot notices engine instrument variations showing parameters approaching out of limits. He or she mentally prepares the engine shut-down drills if the limits are exceeded. Then any stimulus—perhaps as simple as the noise of a tray falling—may be sufficient for the pilot to shut down the engine. That error is known as an error of commission, and it's also sometimes referred to as a response error. You performed an action you had prepared, but you performed it to the wrong trigger. That figure shows the biases we've just discussed in a visual format. So to summarise: evaluating evidence is not a clean, logical process. Your brain is influenced by probability, past experience, ambiguity, the desire to stick with a theory, confirmation bias, and the tendency to execute prepared responses to the wrong stimulus. Recognising these biases is the first step to managing them on the flight deck.