AA 757 off runway excursion at JACKSON, Wyo

[Evan]

Originally Posted by Gabriel

Evan,

You think that, because there is a procedure that was taught to somebody and that somebody was trained in doing it, this procedure MUST be done right, and if it doesn't, it's the the pilot's fault.

Yes. Of course. That is exactly what I think.

Would you blame someone for doing something wrong when it was beyond his powers to do it right?

No. Of course not. However...

Make no mistake: This pilot COULD and SHOULD have watched the speedbrakes lever before calling its condition.

Just as importantly, he should have manually extended the speedbrakes at the first sign of braking anomaly.

Is it within your powers, and can I honestly require you, NOT to make a mistake every few hundreds or thousand of decisions?

My answer for that is a definitive NO!

Mine is as well. Definitely not.

It is within your powers and your duty to act honestly, professionally and with commitment. The best you can do is simply the best you can do, and you simply cannot do better than that. So how can I blame you for for not doing something that you could not do?

Because (and please no more analogies) I was able to verify the speedbrake deployment and failed to deploy them manually. Therefore I am responsible for both the pilot error and the consequences. That doesn't mean I am a bad person, or even necessarily a bad pilot. It just means I am to blame. I got overwhelmed with the situation I was dealing with and forgot the procedures, because I'm human. It can be explained and maybe even excused, but I am still to blame for it.

Gabriel, I make stupid mistakes all the time. That's one of the reasons I'm not an airline pilot. Being a commercial pilot is a dreadful responsibility, and you have to be as clear-headed under pressure as possible. This does not come naturally. It comes from endless training, practice and self-discipline, which I believe is still being undervalued by some operators and CAA's. We can never eliminate the human error factor from the cockpit, but we can mitigate it by training pilots to act reflexively to sudden situations via black and white procedures designed to deal with the situation until they can get full situational awareness and work the grey area. Since no pilot can be expected to memorize and perform a long list of these procedures, they are limited to a short list of the most common and dangerous possibilities:

AFAIK, the 7 primary memory items for Airbus are:

1) Windshear/ windshear ahead: "Windshear, TOGA" 3.02.80 p19
2) TCAS: "Traffic, I have control" 3.02.34 p12 and p17
3) EGPWS: "Pull up, TOGA" 3.02.34 p15
4) Loss of braking "Loss of braking" 3.02.32 p11
5) Immediate actions of EMER DESCENT "Emergency Descent" 3.02.80 p7
6) Immediate actions of UNRELIABLE SPEED INDICATION/ ADR CHECK PROC "Unreliable speed" 3.02.34 p20
7) CREW INCAPACITATION 3.02.80 p9

We should expect every pilot to learn these procedures, rehearse these procedures and be prepared to execute these procedures in an emergency.

Not to mention standard everyday procedures...

Originally Posted by MCM

He would do so by looking at the lever. It is not just standard, but REQUIRED on every single landing for the PNF to not only verify the deployment of the speedbrakes, but to CALL that the speedbrake is up, or not up.

So, in this case, the Captain became focused on a single task which distracted his situational awareness. His discipline lapsed as a result and he failed to take the necessary actions. I understand that situations can quickly evolve into overwhelming scenarios beyond the pilots ability to fully cope with them, and errors will most likely result. That is what the memory procedures are designed for. To protect the pilots from themselves, basically.

My opinion is that we need to assign blame wherever blame is due, which might fall on the Captain for not performing procedures, or the company for not training him well enough. Then we have to understand what is behind these failures and remedy those issues.

How? In this case redesign the mechanical linkages if there is truly a problem in the design, modify maintenance procedures if the problem originated there, and place more resources on getting pilots to maintain a certain level of discipline (i.e always verify checklist items VISUALLY) and react correctly to the unexpected by becoming more instinctively familiar with procedures.

I believe this can be done, but not if we blur the issue by being ambiguous about who failed to do what and removing personal responsibility. This is the actual definition of 'blame', not the one you associate with pejorative.

But let me be clear on this point: in my reasonably well-informed opinion, this incident was NOT the result of a mechanical failure. Those mechanisms fail from time to time and are expected to fail, regardless of whether the failure is unfamiliar or unique. Pilots must always expect these failures and be mentally prepared to react to them by following established procedures where procedures exist. When they fail to do so, they must be held accountable. SEE: AF447...

[Gabriel]

Let me modify a bit your last paragraph to show an alternative version:

Originally Posted by Evan

But let me be clear on this point: in my reasonably well-informed opinion, this incident was NOT the result of a mechanical failure human error. Those mechanisms fail humans make mistakes from time to time and are expected to fail make mistakes, regardless of whether the failure mistake is unfamiliar or unique. Pilots Engineers must always expect these failures mistakes and be mentally have the hardware prepared to react to them. When they fail to do so, they must be held accountable.

It's not one or another. It's both... or none.
I am the of opinion that when a honest best effort is made, and a mistake still happens, that person should not be blamed, held accountable, or guilty (he could be and liable though)

The difference is that blame aims to a person, cause aims to a system. Even if you can ensure that his pilot never ever makes that mistake again (for example, you fire him, and I know you don't mean that, just making a point), that won't help prevent a future occurrence with another pilot, UNLESS you make the discipline action public to serve as example to other pilots, case in which I think you can much more constructively achieve the same thing with training and awareness campaign, for which you don't need to assign blame.

You've said it yourself:

My opinion is that we need to assign blame wherever blame is due, which might fall on the Captain for not performing procedures, or the company for not training him well enough. Then we have to understand what is behind these failures and remedy those issues.

We need to understand what's behind the FAILURES and remedy those issues. No need to assign blame to do that.

If you understand that this pilot failed but, with the current state of the system, any other pilot could have failed too, then it's worthless to name this pilot, and then you say that the cause was "the captain's failure to monitor and manually extend..." (I would have preferred "the monitoring pilot's...", because it was in his function of monitoring pilot and not captain that he made the failure). But saying that it was the blame of John Kent doesn't help if it was expectable that others could have committed the same mistake. It's a system failure, not a person's failure.

HONEST MISTAKES ARE SYSTEM FAILURES

So, unless you think that this pilot decided not to follow the correct procedures, or was acting with negligence, or taking his responsibilities lightly or showed lack of professionalism or commitment, or was not honestly making his best effort to do things officially right, I don't agree to assign blame.

Now, I respect your opposite opinion and can agree to disagree.

[Evan]

Originally Posted by Gabriel

Let me modify a bit your last paragraph to show an alternative version:

Wait... what!?

Gabriel, you can't make that circular argument here. The buck has to stop somewhere. I say it stops with the humans in the cockpit. You've said yourself that automation is a tool, not an excuse from flying the plane. The spoilers extend automatically as a workload-reducing measure, but as MCM has pointed out, they are no guarantee, and it is required that the pilot oversee that they deploy and manually command them if they don't. When flying in any degree of automation, the pilot (especially the PM) has the job of monitoring that automation. If it fails the pilot is there as the only redundancy. With the exception of a few key systems, like stall protection, automation is not intended to override pilot error. The pilot is ultimately responsible for the flight, no matter what mechanical/system failure occurs.

If I was the PM on this flight, I would blame myself, absolutely. I wouldn't beat myself up over it. I would recognize that the stars lined up against me that day, that I had my hands full, but I would also recognize that I boned it big time by not actually looking at the spoiler position or extending them manually as a matter of procedure when the braking was ineffective. Blaming yourself does not mean punishing yourself. It just means accepting responsibility (and I'm hoping he has). The next time something like this happens (god forbid) he has to do better or much worse things could happen.

I think the only problem we're having here is the definition of 'blame'. I'm using the dictionary. To me 'blame' is a key part of arriving at the facts. You're using your emotions, where 'blame' means calling someone an idiot or a criminal.

[TeeVee]

First off, thank you Gabriel for taking the time to write the "book" and help us understand. I for one appreciate it.

I'm not sure i completely disagree with evan here, but once again, i see him thinking in absolutes. one absolute is for sure: humans will always make mistakes. so do machines. fighter pilots are some of the most highly trained animals in the known universe and they screw up all the time. they train more than any commercial pilot, and have stricter rules to follow, yet, they make mistakes, crash and die. only difference is they are usually the only ones that pay the price.

F1 drivers are perhaps some of the most highly skilled drivers in the world. they too make mistakes and crash.

you can train humans till the cows come home, sleep and go out again, and they will still make mistakes.

to you, it may seem simple: be an automaton. follow your memory items. don't be distracted. don't improvise. don't guess. dont screw up. don't leave the cockpit to take a dump. don't talk--ever.

well guess, what, that ain't never gonna happen. as long as planes are built by humans, serviced by humans, and flown by humans there are going to be crashes and "events" and all the training and memory items will not prevent them 100%. NEVER NEVER NEVER.

[Fear_of_Flying]

Originally Posted by Evan

I think the only problem we're having here is the definition of 'blame'. I'm using the dictionary. To me 'blame' is a key part of arriving at the facts. You're using your emotions, where 'blame' means calling someone an idiot or a criminal.

Words have denotations and connotations. Surely you are aware of this? The word "blame" carries with it many connotations, many of which do have an emotive component. When the boss "blames" you for being late to work, he isn't just making a factual observation. He is censuring you.

Since the word "blame" seems so problematic (although I never thought of it as a very ambiguous word in the past, and even Gabriel seems to have an excellent sense of both its denotation and connotations), might I suggest making your points about this accident without using the word "blame" in the future? I think you will see that you are left somewhat wanting in the points you are trying to convey, because the word "blame" does in fact carry with it meanings - that is to say, connotations - that impact what you are saying, whether you choose to acknowledge them or not.

[Evan]

Originally Posted by Fear_of_Flying

Words have denotations and connotations. Surely you are aware of this? The word "blame" carries with it many connotations, many of which do have an emotive component. When the boss "blames" you for being late to work, he isn't just making a factual observation. He is censuring you.

Since the word "blame" seems so problematic (although I never thought of it as a very ambiguous word in the past, and even Gabriel seems to have an excellent sense of both its denotation and connotations), might I suggest making your points about this accident without using the word "blame" in the future? I think you will see that you are left somewhat wanting in the points you are trying to convey, because the word "blame" does in fact carry with it meanings - that is to say, connotations - that impact what you are saying, whether you choose to acknowledge them or not.

OK. Accident investigations are not generally colored by emotional censorship, but... fair enough.

In my opinion, this incident is the result of a monitoring pilot not being adequately disciplined in landing procedure (by calling out the speedbrake deployment without visually checking it) and unprepared to react to a system failure (by micro-tasking on the reverser deployment instead of following the primary loss-of-braking procedure), and i believe stronger and more frequent training on checklist procedure and memory procedure could prevent this scenario in the future.

I recognize the extenuating circumstances: a unique reverser linkage failure which might have produced the aural indication of speedbrake deployment. I am well aware of the human factors involved. Therefore I do not condemn the pilot for his errors. I do, however, assign him responsibility for the consequences. Call that what you will.

Now, if we take a standard investigation methodology, repeating the incident in the sim with fully qualified pilots (who are not expecting the event), my guess is that most of them will not repeat these errors because most of them, while performing the standard landing procedure, will visually notice that the speedbrake lever has not deployed and manually extend them. Therefore, the subsequent failure of the reversers will result only in the plane stopping further along the runway length, though not at the desired turnoff, and there will be no incident to speak of.

If I'm wrong about that, then we have a much, much bigger problem on our hands.

So, as I see it, pilot error on the part of the PM is the principal and deciding factor that resulted in the aircraft departing the runway.

[TeeVee]

has anyone determined that the failure to deploy the spoilers caused anything at all in this case? Gabriel's explanation of braking placed the lion's share of braking on dry surface on the wheel brakes while on slippery surface on the reversers. apparently what happened here is that spoilers failed to deploy on a somewhat messy runway (good for a challenger 300 means nothing for a 757).

i think that had the pm not failed to realize the spoiler problem they may still have overrun as the reversers would have been the primary braking power...and they failed too.

[Fear_of_Flying]

Originally Posted by Evan

OK. Accident investigations are not generally colored by emotional censorship, but... fair enough.

Actually, accident reports are self-censored to the extreme, I would say, and have their own vocabulary, very much like a legal document, in order to avoid just this sort of problem. They are an exercise in describing causes and effects, and I doubt very much that you will find the word "blame" ever showing up in a final report. "Failure of _________ to do _________" is not the same as blame, in my opinion. It is simply a description of cause and effect. The investigators themselves, being scientists, also have to guard against bringing emotion into their work, so yes, emotional censorship would not color, but un-color, what they do.

[Evan]

Originally Posted by TeeVee

has anyone determined that the failure to deploy the spoilers caused anything at all in this case? Gabriel's explanation of braking placed the lion's share of braking on dry surface on the wheel brakes while on slippery surface on the reversers. apparently what happened here is that spoilers failed to deploy on a somewhat messy runway (good for a challenger 300 means nothing for a 757).

i think that had the pm not failed to realize the spoiler problem they may still have overrun as the reversers would have been the primary braking power...and they failed too.

Reversers are designed to provide a credit that shortens the actual runway distance needed to stop. They are not designed as a stand-alone means of braking. Even if the runway was in poor condition and only providing a friction braking coefficient of .05, that is still far more then the reversers can provide.

If the runway was providing less adhesion than that, that would be an item for the pilot error bin. In this case either the runway is below safe minimum landing conditions or the spoilers are not deployed (as in this case), or both.

From the reports, it appears that the runway landing zone was in the wet/fair-to-good category.

Here is an illustration of the contribution of reversers compared to runway conditions (B737-800). As you can see, they become more important as runway condition deteriorates, but not effective enough by themselves in any case.

[Evan]

Originally Posted by Fear_of_Flying

"Failure of _________ to do _________" is not the same as blame, in my opinion.

And in my opinion it is the same, assuming _________ was able to do _________ but failed to do so. Blame is just the more vernacular and straightforward way of putting it. In my opinion it does not necessarily mean incompetence or malicious intent. But whatever... semantics... decorum... I'm over it.

[Gabriel]

Originally Posted by Evan

Now, if we take a standard investigation methodology, repeating the incident in the sim with fully qualified pilots (who are not expecting the event), my guess is that most of them will not repeat these errors because most of them, while performing the standard landing procedure, will visually notice that the speedbrake lever has not deployed and manually extend them. Therefore, the subsequent failure of the reversers will result only in the plane stopping further along the runway length, though not at the desired turnoff, and there will be no incident to speak of.

I fully agree with the above, but I'll add that that if you could make the same experiment with the incident pilot (what would require, before each trial, to erase the original incident and all the previous trials from the pilot's memory, and hence you can't), it is possible* that he would also do it right most of the times. That would show that the pilot did NOT perform sub-par but it was rather a normal "process variation" that is within the usual and expectable "process behaviour" (i.e. people do it wrong every once and then).

"Process variation" and "process behaviour" are terms used in statistical process control (FOQUA is an example of such). If you have a process that, even when running ok, makes one defective part out of 100, you don't see ONE bad part and say "something's wrong with the process". Now, if instead of 1% you find that the defective rate today was 20%, you do say it.

And even if today it was 1% as usual, that doesn't mean that you have to accept that 1% with resignation. It is perfectly legit to ask: how can we improve it, how can make defects to happen less often?

Then you have to study the process, find the causes for that 1% of defects, and try to correct them.

The same goes with the mistakes.

* I was tempted to say "probable" instead of "possible", but the fact is that there ARE bad pilots who don't take procedures and standard operative practices seriously and would commit the same mistake over and over (if we could erase each incident from his memory as to prevent learning). I don't know this pilot so I will not judge him.

[Evan]

Originally Posted by Gabriel

I fully agree with the above, but I'll add that that if you could make the same experiment with the incident pilot (what would require, before each trial, to erase the original incident and all the previous trials from the pilot's memory, and hence you can't), it is possible* that he would also do it right most of the times. That would show that the pilot did NOT perform sub-par but it was rather a normal "process variation" that is within the usual and expectable "process behaviour" (i.e. people do it wrong every once and then).

This is the important question isn't it... We have no way of knowing the answer here on this forum. Was this a random "process variation" error or had this pilot developed a habit of listening for the speedbrake rather than visually confirming it? It is very possible, and if this was habit, then yes, he would most likely repeat the incident in your sim scenario. How many other pilots might have developed this habit (it makes a distinct sound, so why bother to look, right?)? Is this possibility being addressed? Can checkrides reliably detect this? There have been other documented incidences of checklist items being called out without visual confirmation. There was also the pilot who cranked the rudder trim instead of the door lock switch, because he wasn't confirming his actions with his eyes. Was that a habit? That almost caused a disaster. How widespread of an issue is this?

Discipline has a way of giving in to bad habits if it is not regularly stressed.

ARE THESE QUESTIONS EVEN BEING ASKED?

[Gabriel]

Originally Posted by TeeVee

has anyone determined that the failure to deploy the spoilers caused anything at all in this case? Gabriel's explanation of braking placed the lion's share of braking on dry surface on the wheel brakes while on slippery surface on the reversers.

Not so. Maybe subtle wording difference but of high impact.

The wheel brakes are the main means of braking.
The max braking force that the wheel brakes can make is a product of two terms:
weight-on-wheels times coefficient-of-friction

On the other hand, the max braking force made by the reversers is constant.

Because of that, the RELATIVE effectiveness of the reversers increase as either the coefficient of friction or the weight-on-wheels diminish. But for reversers to contribute MORE than the wheel brakes the effectiveness of the later should be reduced to about nothing. I mentioned a THEORETICAL scenario of zero traction to show both extremes of the curve, but zero traction is unachievable, even wet ice and hydroplaning make some friction.

Now spoilers.

Spoilers are VERY important.
Spoilers by themselves make very little braking force (in the for of aerodynamic drag). But they do KILL LIFT and, since the weight-on-wheels is weight minus lift, they DRAMATICALLY INCREASE the wheel braking force, especially in the high-speed part of the landing roll (when the wings will still be making a significant amount of lift) which is the part that count most because most of the landing roll distance is consumed there.

Remember: loosing half of the wight-on-wheels because the spoilers are not deployed) have the same effect as loosing half of the traction (because of runway contamination.

As such, SPOILERS ARE AN INTEGRAL PART OF THE MAIN MEANS OF BRAKING, which is wheel braking.

[Gabriel]

Those are all very good questions, Evan, because in their answers there is the potential for safety improvement.

I don't know if they are being asked. I hope so and I guess that they are a big part of the human factors effort.

[MCM]

If the runway was providing less adhesion than that, that would be an item for the pilot error bin. In this case either the runway is below safe minimum landing conditions or the spoilers are not deployed (as in this case), or both.
From the reports, it appears that the runway landing zone was in the wet/fair-to-good category.

It is very very difficult to acurately know the braking conditions that will be experienced for a specific landing. It is something that the industry has worked quite hard on, but hasn't completely got its head around.

The most reliable reports are still from other pilots, who could be in different aircraft types, and experience it differently - the aircraft type in front might have particularly good braking characteristics in the wet. Yours might not.

You can't categorically say that there was good braking available based on an earlier report. Similarly you can't really question a pilot who choses to land on a runway based on such a report either. He has very little else to go on.

And fwiw, the "runway braking coefficient" as measured by a couple of different machines is a guide, but really only that. There are a number of reasons why the information they provide doesn't always correlate to the braking action experienced.

Was this a random "process variation" error or had this pilot developed a habit of listening for the speedbrake rather than visually confirming it? It is very possible, and if this was habit, then yes, he would most likely repeat the incident in your sim scenario. How many other pilots might have developed this habit (it makes a distinct sound, so why bother to look, right?)? Is this possibility being addressed? Can checkrides reliably detect this? There have been other documented incidences of checklist items being called out without visual confirmation. There was also the pilot who cranked the rudder trim instead of the door lock switch, because he wasn't confirming his actions with his eyes. Was that a habit? That almost caused a disaster. How widespread of an issue is this?

I'm not sure that we know the pilot called the speedbrake up based on noise. That was just a bit of a hypothesis of a complicating factor. I would suggest that no pilot would have it as a "habit"... its just one other little subtle thing. I don't think anyone would say "it makes a distinct sound, so why bother to look"... I would say that the sound, in a high pressure, high workload environment could be interpreted subconsciously by your brain as meaning the speedbrake is up.

Can the simulator check this? To some extent yes - or at least can highlight it. Oddly enough before this accident a simulator examiner I had gave us an RTO where the speedbrake lever moved slightly and then returned to stowed while it made noise. It never reached the fully deployed position. Its a real trap. It certainly made me pay more attention to it since though!

We're not talking about people who intentionally use other cues to decide whats going on, and we're not even talking about crew not paying sufficient attention. We're talking about the way the brain reacts during high/over workload scenarios, and its own form of loadshedding.

[TeeVee]

while i have ZERO experience flying any type of aircraft, i have jumpseated in my friend's challenger a number of times. takeoff, flight and approach are rather boring. final and landing however are a different story. i was particularly amazed one time when we landed during a nasty thunderstorm. i sat there dead quiet, possibly holding my breath, and was amazed at the concentration level of the two pilots.

of course, i have no idea what checklists they went through in their minds aside from the items they read out loud, i'm fairly certain that had something gone astray, with so many variables, i would have been fairly impressed if they had been able to 1) assess the problem, 2) mentally deduce the possible cause and 3) react within a split second.

yes, training helps, no doubt. but all the training in the world will not yield a perfect pilot. isn't this why humans are not permitted, under normal circumstances, to fly the aircraft at certain times? (i think this was discussed in the AF thread)

[Gabriel]

Originally Posted by TeeVee

i'm fairly certain that had something gone astray, with so many variables, i would have been fairly impressed if they had been able to 1) assess the problem, 2) mentally deduce the possible cause and 3) react within a split second.

Evan will say, and I'll support him, that there are certain problems for which you don't assess the problem, mentally deduce the possible cause and correct that cause, not initially at least. You do react in a split second with a memorized (and hopefully practiced until it becomes an automatic second-nature reaction) procedure that aims to contain the problem itself, regardless of the cause.

These are called "memory items", and are few. Evan posted those for the Airbus. Stall was not one of them, I don't know why because it clearly IS one of this kind.

For example, the plane doesn0t brake as expected. There are a number of reasons for that, but you don't waste time trying to figure it out. You jsut do everything that can be done to stop the plane.

In a Boeing, the loss of braking memory tiems could read something like (I am using just common sense):

Brakes: Apply full manual braking (the autobrakes could have failed or been forgotten to set) (don't even bother to check the status of the autobrake, any autobrake condition plus full maunal baking is still full manual braking)
Spoilres: Exted (the auto-spoilers could have failed or they could have been forgotten) (don't even bother to check the lever position. Just grab it and pull it all the way back, The worst that can happen is that it was already there)
Thrust: Set both thrust levers to idle. Set max reverse. If locked in-transit, bring the levers back to reverse off and retray (you could have forgotten or there could be a mechanical problem).

You don't care which of the causes caused the problem. You act to solve the problem and difer the troubleshooting for a less critical time.

I can mention a few incidents, some of them fatal, where just doing it instead of trying to figure out what's wrong would have corrected the situation. The TAM A320 in Conghonas comes to mind (a thrust lever was mistakenly left at the approach setting, whcih prevented the spoilers to extend afetr landing and the autothrottle to disconnect, so it kept adding power on that engine in an attempt to keep the selected speed)

[Evan]

Originally Posted by Gabriel

These are called "memory items", and are few. Evan posted those for the Airbus. Stall was not one of them, I don't know why because it clearly IS one of this kind.

Because Airbuses don't stall. of course...

Actually, that's a rather old list. Approach to Stall / Stall better be on that list by now.

[Fear_of_Flying]

Originally Posted by Evan

Because Airbuses don't stall. of course...

Actually, that's a rather old list. Approach to Stall / Stall better be on that list by now.

Except that if there's a grey area, memory items don't work. Events requiring memory items cannot be open to interpretation, and have to be such that the memory items will basically work 100% of the time. Either your plane is stopping or it's not. If it's not, you use the loss of braking memory items, and hopefully you will be successful. At worst, the problem is such that nothing would have helped, so you've lost nothing by applying the memory items. Same with unreliable airspeed. You either have it or you don't. If you don't have reliable airspeed indications, applying the memory items should get you by.

With a stall, as we have seen, there have been situations when the pilots were unaware that they even were stalled. So obviously memory items wouldn't do them much good. But more to the point, are there occasions when you might use approach to stall/stall memory items when they would not apply? The last thing you'd want is to train a pilot to react unthinkingly to a stimulus where the response did not always apply. What would be the unequivocal cue that would tell the pilot 'Ok, time to use the approach to stall memory items, no thought required.'? Would the stall warning be sufficient for this? Just asking.

[Gabriel]

Originally Posted by Fear_of_Flying

With a stall, as we have seen, there have been situations when the pilots were unaware that they even were stalled. So obviously memory items wouldn't do them much good. But more to the point, are there occasions when you might use approach to stall/stall memory items when they would not apply? The last thing you'd want is to train a pilot to react unthinkingly to a stimulus where the response did not always apply. What would be the unequivocal cue that would tell the pilot 'Ok, time to use the approach to stall memory items, no thought required.'? Would the stall warning be sufficient for this? Just asking.

Yes.