In a crisis, your brain may be your enemy

We can learn a lot about the human brain from today’s release of the report on the crash of an Air France jetliner into the Atlantic in 2009.

In its final report, the French civil aviation authority’s Bureau of Surveys and Analysis said two less-senior pilots at the controls of AF 447 were “completely surprised” by the failure of cockpit instruments. All 228 passengers and crew died in the crash of the plane, which was on its way from Brazil to Paris.

Without any data to tell them what was happening to their plane, the pilots pulled up on the controls, which only made things worse.

Even Chesley Sullenberger couldn’t refrain from criticizing the pilots…

I’ve written about this crash before, and specifically about the problems with an overloaded brain.


There’s only so much data the human brain can handle. Keep in mind it’s night, the cockpit is dark, there are no references outside to tell you where you are, lights are flashing, alarms are going off and a simple airspeed indicator isn’t working (declining airspeed can tell you you’re going up, increasing airspeed means you’re probably going down), the noise from passing air is changing and telling you something, but what? Your body — which lies to you at times like this — is telling you one thing, some working instruments might be telling you another. Which one do you believe?

Oh, and you’ve got one minute to get all this sorted out.

A warning to the pilots that the plane was losing its lift (stalling) went off 75 times. Seventy-five times! And yet, the pilots didn’t pay attention to it, apparently, because there’s only one way to prevent a plane from stalling: push the nose down, the opposite of what they were doing.

How could this be?

Two paragraphs in today’s report tell us:


Numerous studies have been conducted on insensitivity to aural warnings and they showed that the aggressive nature, rarity and unreliability of these warnings may lead operators to ignore these signals [1, 2]. In particular, in the event of a heavy workload, insensitivity to aural warnings may be caused by a conflict between these warnings and the cognitive tasks in progress. The ability to turn one’s attention to this information is very wasteful as this requires the use of cognitive resources already engaged on the current task. The performance of one of these tasks (solving the problem or taking the warning into account) or of both would be affected [3].

In addition, studies on the visual-auditory conflict show a natural tendency to favour visual to auditory perception when information that is contradictory and conflicting, or seen as such, of both senses is presented [4, 5, and 6]. Piloting, calling heavily on visual activity, could lead pilots to a type of auditory insensitivity to the appearance of aural warnings that are rare and in contradiction with cockpit information. A recent study in electrophysiology on a piloting task seems to confirm that the appearance of such visual-auditory conflicts in a heavy workload situation translates into an attention selectivity mechanism that favours visual information and leads to disregarding critical aural warnings [7].

In other words, even when the answer to a problem is blaring in your ear, your brain’s concentration on solving a problem makes you not hear it.

The pilots didn’t have a lot of time, but the situation underscores the importance of an old axiom: Sometimes the best thing you can do, is step back and take a breath.

Here’s the full report.

  • jon

    Ever watch “Children of Men”?

    When they were driving the car, there was a HUD display alerting that there was something about to hit the car.

    Would this be a better choice for these types of warnings, to present them visually?

    TJSwift you are trolling.

  • Caleb

    Bob,

    I remember Kerri talking about a book called the checklist manifesto that deals with the medical practice adopting the checklist strategy employed by pilots. Are you familiar?

    I would have thought that under that type of crisis there would be a simple algorithm or checklist that pilots would be trained to follow?

  • Bob Collins

    Pilot checklists are mostly for routine items — although there is a generic emergency checklist.

    The problem here, it seems to me is sensory overload. But at it’s basic, the problem here is the pilots didn’t know which was was up (See John F. Kennedy Jr. crash), and obviously at least one of them was paying attention to what his body was telling him and his body’s senses were wrong, which is a very common problem.

    But in that situation, the instrument that tells you whether you’re going up or going down is the airspeed indicator, and they didn’t have one that worked in this situation.

    I suppose one option here is to build a smarter autopilot that won’t just say “I’m out of here” when things get rough.

  • Jim Shapiro

    jon – and sorry Bob, for veering further off topic, but I’m compelled to comment regarding “Children of Men”: it presents a disturbingly realistic depiction of urban combat.