Asiana officially blames crew for flt 214...and the autothrottle.

[seagull]

What is the issue?

Confirmation bias. Same thing that got UPS in BHM, same thing that got AF 447, DL 191, USAir 1014, FedEx 80, Tiger 77 -- same thing that got a lot of other ones, too many to list here. Yes, confirmation bias is something that can be addressed in training, but nobody is doing it. Nobody. It is extremely powerful and difficult to overcome. Want proof? Look at any strong political or religious position. It is innate to our cognitive structure and once it has taken hold avoiding an accident is just luck.

 

[n57flyguy]

Confirmation bias. Same thing that got UPS in BHM, same thing that got AF 447, DL 191, USAir 1014, FedEx 80, Tiger 77 -- same thing that got a lot of other ones, too many to list here. Yes, confirmation bias is something that can be addressed in training, but nobody is doing it. Nobody. It is extremely powerful and difficult to overcome. Want proof? Look at any strong political or religious position. It is innate to our cognitive structure and once it has taken hold avoiding an accident is just luck.

What else do you know about human factors psych?

You are asking me to use a bias to confirm a confirmation bias. The theory really relates to political dispositions, self image, law, etc.

Not that I am in disagreement, but it appears to be a catch all excuse for ignoring visual cues and lack of SA...is a confirmation bias why pilot error occurred?

 

[seagull]

What else do you know about human factors psych?

You are asking me to use a bias to confirm a confirmation bias. The theory really relates to political dispositions, self image, law, etc.

Not that I am in disagreement, but it appears to be a catch all excuse for ignoring visual cues and lack of SA...is a confirmation bias why pilot error occurred?

In answer to your question, I probably know a tad bit more than the average person on the topic.

If a significant percentage of other people would also "ignore" cues given the same set of circumstances then we are seeing a human error that is within the bounds of normal human cognitive processes.

Let me turn your "catch all" statement around. It is exceedingly easy (and a lot less cognitive work) to just capture the hindsight bias (a type of confirmation bias, really) aspects and say "that guy screwed up". Perhaps you can simplify it to that, but at the very least, that does NOTHING to prevent a future occurrence. It just makes you all feel good as you walk away thinking "that won't happen to me because I (fill in the blank)". Not so easy to actually take a hard look at it. In a talk I gave on a related topic recently, I quoted Sidney Dekker on hindsight bias. I see these factors playing often in these discussions. Here is what I said:

Sidney Dekker (2007) elaborates on the problem of hindsight bias, stating it will:

· oversimplify causality ("this led to that") because we can start from the outcome and reason backwards to presumed or plausible "causes;"

· overestimate the likelihood of the outcome (and people's ability to foresee it), because we already have the outcome in our hands;

· overrate the role of rule or procedure "violations." While there is always a gap between written guidance and actual practice (and this almost never leads to trouble), that gap takes on causal significance once we have a bad outcome to look at and reason back from;

· misjudge the prominence or relevance of data presented to people at the time;

· match the outcome with the actions that went before it. If the outcome was bad, then the actions leading up to it must have been bad too—missed opportunities, bad assessment, wrong decisions and misperceptions (Dekker, 2007, pp. 66-67).

From: Dekker, S. (2007). Just Culture. Farnham, England: Ashgate.

 

[n57flyguy]

Excellent: So how to training departments attack confirmation bias related issues to reduce pilot error?

As you mentioned filling the gap (dissonance, really, which we are all subject to) between realizing "that guy screwed up" and doing something about the circumstances that lead to the guy screwing up? I might take a few decades to change a culture. I like the cognitive work required in figuring out these issues, that is why I ask and do not take "confirmation bias" or just "culture" as the answer.

 

[seagull]

I am not sure that dissonance is the right word for this, but as long as we are on the same page, it does not matter too much at this point. (Oh, never mind, read what you wrote too fast, right word for where you used it! I left my initial error in this edit for a reason as you can see as you read further).

I am concerned that it is not possible to separate out the confirmation bias in an airplane from the rest of our lives. Training in the sim is all fine and good but in a challenging situation we will revert to primacy, which would mean that to really change it we have to globally change our outlook on our entire belief system. To quote Lawrence Krauss, ..."the most exciting thing about being a scientist is not knowing and being wrong...". How do we teach people to hope to find out they are wrong, and aspire to that?

 

[derg]

I've had a captain, while it was MY leg, reach up and hit the Autopilot Engage button at 600ft AGL. Unannounced and unbriefed.

Last week I had a captain, who was really nice and cool otherwise, who would override me on the thrust levers and damn near firewall them every time the speed trend vector got anywhere close to touching the ref-bug. Me thinks that in the past an FO of his got a stick shaker or something like that. On every leg I flew, once we got into the terminal environment, he was constantly saying 'Don't get to slow... watch your speed watch your speed'... Dude, I got it. Your firewalling the throttles whilst I'm inside the marker isn't exactly the definition of stable approach!

Sounds like Mr. Captain was never a CFI.
I could be wrong.

 

[n57flyguy]

I am not sure that dissonance is the right word for this, but as long as we are on the same page, it does not matter too much at this point. (Oh, never mind, read what you wrote too fast, right word for where you used it! I left my initial error in this edit for a reason as you can see as you read further).

I am concerned that it is not possible to separate out the confirmation bias in an airplane from the rest of our lives. Training in the sim is all fine and good but in a challenging situation we will revert to primacy, which would mean that to really change it we have to globally change our outlook on our entire belief system. To quote Lawrence Krauss, ..."the most exciting thing about being a scientist is not knowing and being wrong...". How do we teach people to hope to find out they are wrong, and aspire to that?

Are you referring to primacy in culture taught by social orders and elders or by primacy in flight training and training events?

As we both know, primacy in teaching deals with teaching lessons right the first time, and learning right the first time, as it is much more difficult to un-teach a bad habit. It is possible that experience is not being taught, which could separate some of these issues into a manageable and productive means to rectify and enhance the "unshakable" first impressions.

Krauss is correct. Most people research a topic thinking they want to be right, turning out a poor result because they cannot leave the realm of their bias. Myself personally will research a topic telling myself "I want to be wrong." One leans more peripherally and can even change a bias on the subject. I did it the other day on a case study with a simple question that lead to hours of research only to realize I was asking the wrong question to begin with. From that what I learned multiplied to some irrelevant factor.

To generalize, humans want to self preserve their egos and do not like being wrong. But we learn from failures better (primacy could be argued) if we are confident in ourselves to keep going forward. To teach one to do this comes from the individual who wants to. Better asked, how do we motivate others to hope that they are wrong?

 

[Cptnchia]

It would be nice if Asiana would just come out and stop blaming the crash on everything under the sun, all except their training culture.

One word why they won't EVER do that. Liability.

 

[seagull]

Are you referring to primacy in culture taught by social orders and elders or by primacy in flight training and training events?

As we both know, primacy in teaching deals with teaching lessons right the first time, and learning right the first time, as it is much more difficult to un-teach a bad habit. It is possible that experience is not being taught, which could separate some of these issues into a manageable and productive means to rectify and enhance the "unshakable" first impressions.

Krauss is correct. Most people research a topic thinking they want to be right, turning out a poor result because they cannot leave the realm of their bias. Myself personally will research a topic telling myself "I want to be wrong." One leans more peripherally and can even change a bias on the subject. I did it the other day on a case study with a simple question that lead to hours of research only to realize I was asking the wrong question to begin with. From that what I learned multiplied to some irrelevant factor.

To generalize, humans want to self preserve their egos and do not like being wrong. But we learn from failures better (primacy could be argued) if we are confident in ourselves to keep going forward. To teach one to do this comes from the individual who wants to. Better asked, how do we motivate others to hope that they are wrong?

It is a great question, and that is what I am wondering. In terms of primacy, I am talking about both. However, I do not think that what we do in flight training can change someone to slow down and take a more difficult cognitive approach to decision making, and in that regard, the real primacy is in their baseline thinking. Flight training is great for training immediate action items, Gary Klein's "recognition primed decision making", but falls short for the other part, which is the corner point that leads to a large percentage of today's accidents.

 

[BobDDuck]

I've had a captain, while it was MY leg, reach up and hit the Autopilot Engage button at 600ft AGL. Unannounced and unbriefed.

Last week I had a captain, who was really nice and cool otherwise, who would override me on the thrust levers and damn near firewall them every time the speed trend vector got anywhere close to touching the ref-bug. Me thinks that in the past an FO of his got a stick shaker or something like that. On every leg I flew, once we got into the terminal environment, he was constantly saying 'Don't get to slow... watch your speed watch your speed'... Dude, I got it. Your firewalling the throttles whilst I'm inside the marker isn't exactly the definition of stable approach!

I'm flown with guys like that. Normally I'll give them a warning that it's my leg and if they want me to do something or fly a specific way, just ask. After that, if they keep it up, it's their plane for the rest of the day. Much safer when only one guy is trying to fly and I get paid the same either way.

I've actually got an 8 landing day with a guy like that starting in 2 hours...

 

[n57flyguy]

It is a great question, and that is what I am wondering. In terms of primacy, I am talking about both. However, I do not think that what we do in flight training can change someone to slow down and take a more difficult cognitive approach to decision making, and in that regard, the real primacy is in their baseline thinking. Flight training is great for training immediate action items, Gary Klein's "recognition primed decision making", but falls short for the other part, which is the corner point that leads to a large percentage of today's accidents.

Not that the complexity of the issue is a bad thing, complexity is often needed.

The other part needs to be defined. Is it the culture from the country? Is it the culture of the airline? Is it the culture of the pilots? Is it a sub-culture of pilots? Which parts?

It is not an end all answer, but if airlines and training departments could learn to teach experience better the "gap" we speak of could be closed. While certain cultures with a "captain is God" complex will undoubtedly remain, failure to cognitively process visual cues and a failure to maintain SA does imply a factor of of poor overall piloting skills. That being said, it might not be changing a culture, but working with an existing one to mitigate some of the problems going forward.

 

[z987k]

Not that the complexity of the issue is a bad thing, complexity is often needed.

The other part needs to be defined. Is it the culture from the country? Is it the culture of the airline? Is it the culture of the pilots? Is it a sub-culture of pilots? Which parts?

It is not an end all answer, but if airlines and training departments could learn to teach experience better the "gap" we speak of could be closed. While certain cultures with a "captain is God" complex will undoubtedly remain, failure to cognitively process visual cues and a failure to maintain SA does imply a factor of of poor overall piloting skills. That being said, it might not be changing a culture, but working with an existing one to mitigate some of the problems going forward.

There are some things that just don't work. The culture is "broken" and no amount of mitigation of related problems is going to fix it. Airlines that have a captain is god culture will continue to crash airplanes at significantly higher rates than those that don't. I don't think any amount of duct tape can patch a hole that big, certainly not long term.

 

[n57flyguy]

There are some things that just don't work. The culture is "broken" and no amount of mitigation of related problems is going to fix it. Airlines that have a captain is god culture will continue to crash airplanes at significantly higher rates than those that don't. I don't think any amount of duct tape can patch a hole that big, certainly not long term.

So this discussion ends with we can't fix the issue of pilot error because certain culture prohibits it from developing? What exactly doesn't work? Why can't it?

Cultures have been subverted for decades, in recent time, right now, and will continue to in the future. I don't see why that cannot happen regarding certain pilot cultures and sub cultures.

 

[seagull]

No, this is not talking about a particular culture, and no "one culture" is better than others in this regard. This is a very basic human trait that is part of our cognitive architecture utilized to simply the world around us and enable us to make faster decisions. Worked great on the Savannah! The problem is that those simple mental models don't work so well in a complex environment. Our brains just did not evolve for that, and unless those that are unable to shed that bias start having fewer children than those that do, (the opposite is actually occurring) we are pretty much stuck with it as a species.

That said, it can be taught, but it has to be truly global. That is what Krauss was talking about. The problem is that "global" means challenging your entire belief system, including religion, etc. That's big stuff.

 

[n57flyguy]

seagull, could you define more of these concepts to me?

Are you talking about MRT? Temporal Perception? Like I said, I am interested in the complex portion of this, but starting with simple questions is just that, a start.

I found an interesting parallel to "changing an entire belief system." I will have to do a write up.

 

[seagull]

seagull, could you define more of these concepts to me?

Are you talking about MRT? Temporal Perception? Like I said, I am interested in the complex portion of this, but starting with simple questions is just that, a start.

I found an interesting parallel to "changing an entire belief system." I will have to do a write up.

I assume when you write MRT you mean magnetic resonance topography? Temporal perception involves time, so I am not sure how that is related either? Read some of Daniel Kahneman's stuff for a place to start.

 

[n57flyguy]

I assume when you write MRT you mean magnetic resonance topography? Temporal perception involves time, so I am not sure how that is related either? Read some of Daniel Kahneman's stuff for a place to start.

This is a very basic human trait that is part of our cognitive architecture utilized to simply the world around us and enable us to make faster decisions.

Multiple Resource Theory, dealing with auditory, verba, visual, and spatial orientation our responses to those and how we encode them.

Well, TP is a basic trait of our brain, we are essentially clocks in a sense. That relates to the issue on pilot error on how pilot time such phases...how we know what to do when we do it.

That is what I thought you were alluding to.

 

[seagull]

Wow, ok, old stuff! I am looking at the research published in the last 5 years or so. No, I do not view it as being related to either of those concepts. Perhaps indirectly, but certainly not as a base for it.

 

[Jimflyfast]

"The airline also said test pilots had trouble landing under the same conditions in simulators."
My brain is hurting. Somebody please help me.

What the report actually said was that the test pilots had a trouble landing from the high energy state Asiana was in without exceeding stable approach parameters. None of them had trouble landing.

 

[SlumTodd_Millionaire]

One word why they won't EVER do that. Liability.

And that's okay. Their responsibility is to their shareholders, so admitting fault would be irresponsible. It's the job of regulators, investigators, and courts of law to get to the bottom of it and hopefully prevent it from happening again. We shouldn't ever have an expectation that a corporation is going to do anything different than protecting their business and their shareholders from liability.