Posts about: "Muscle Memory" [Posts: 65 Pages: 4]

Again, noting that all commercial aviation accidents involve one in a million, one in a billion events, I don't find it implausible on its face that a crew member flipped the fuel cutoffs while intending to do something else entirely, even given the deliberate and well studied ergonomic differences designed into cockpit controls.

It could have been an intentional, malicious act, but I think, for now, "brain fart" or "extremely inappropriate reaction to perceived anomaly" makes more sense for now.

Muscle memory and performing an action at the wrong time is quite possible. Back when I was flying piston twins and had been on the same type for over a year, I was sitting in the cockpit starting up and my hands had got ahead of my brain. They were doing an action in the start sequence ahead of what I was thinking about at the time. Mental note to self to make sure that hands and brain were in sequence from then on.

Having now been on the same type for nearly 20 years, I could do the PM after landing flows with zero thought because I've done them thousands of times but deliberately slow myself down and read the labels on the controls and switches before actioning them. It's a bit like tying your shoelaces.

THE LANDING GEAR

There are many missing jigsaw pieces here that are highly relevant to building an understanding of what these guys were going through.
One being that the gear was not raised. The report ( very oddly in my opinion ) is NOT declaring any voice data on the prompts/commands to raise it once they achieved a positive ROC after lift-off.
At lift off, it's pretty safe to assume ( yes, I know ) that everything was going according to plan....so why, oh why wasn't the gear raised.....?

This is quite extraordinary because:-

1. This sequence is muscle memory for every commercial pilot. It’s drilled into simulator sessions, route checks, and emergency training. Even during non-normal scenarios like an engine failure, the gear is typically raised before handling other procedures.
   
2. The delay is not explained in the preliminary report. The gear lever was found in the DOWN position in the wreckage. No mention is made of any mechanical failure or crew discussion about the gear. That silence is notable.
   
3. There was no declared emergency prior to 180 kt. No CVR audio so far suggests anything unusual before the sudden power loss—so there was no obvious reason to skip this SOP.
   
   
   
   

Almost as though the hand which reached out to raise the gear while the brain was thinking about something else found the fuel cutoff switches and managed to manipulate them to off without the brain ever becoming involved?

If the guards to the stab trim switches were left open by maintenance, I could see a situation in which muscle memory might lead to closing the adjacent fuel control switches. This would also explain the "I didn't do it response", as he believed he was merely closing the guards instead.

It might be 3.01 seconds, due to the sampling rate. Still not quick, but a lot quicker than 4 seconds in the context. If it was being done by PF while trying to fly the aeroplane, then it wouldn't be as slick as the shutdown routine (and it would be against muscle memory of that routine as the switches are being moved in the opposite direction).

I'm pleased you said this as I've realised that my image is wrong, as the timestamp is not +/- 1 second, but only -0-1 seconds.

Corrected (FWIW):

It may equally well be 4.99 seconds.
MrShed will have to redo his drawing.

I would suggest everyone that wants to analyze the likelihood of particular actions on the part of the crew, whether referring to "muscle memory" or "brain farts" or whatever, simply add the phrase " during the initial climb " to their thoughts and see how that sounds. Without callouts for raising the gear and other than sleepwalking, what explanation is there for flipping the fuel cutoffs down in sequence? And further, doing so precisely in the time window where it was most likely to make the situation unrecoverable?

The absence of evidence is not evidence. And, in any case, adding "during the initial climb" to a brain fart/muscle memory/seizure of some kind/psychotic break or whatever doesn't make make one of those possibilities any less likely.

That status message is quite common.
When trouble shooting, it normally resets through a BITE test. I don't remember ever having to touch the stab cut off switches as part of line trouble shooting. The aircraft is on a turnaround, if the test doesn't clear the message further trouble shooting may be lengthy, most airlines would want the aircraft dispatched in accordance with the MEL.
Deactivating the relevant transducer involves entering the stab bay, disconnecting a plug and writing up the deferral.
If it's true, as someone above suggested, that the aircraft was signed off an hour before departure, that is not last minute by the way, then I would guess that the BITE test cleared the message.

Or did the F/O call for the gear up, the Captain still looking outside, did an action with his right hand, both pilots felt and heard the engines wind down, The F/O looking down realised what had happened, the Captain looked down and couldn't reconcile his intended action (bringing the gear up) hence the "I didn't do it " comment, then selected the FCO switches back to run but it was to late for the situation to be recovered. The only action they could take that they felt they had some control of was to broadcast a MAYDAY.

For me the prelim report just reveals an unintended consequence of relying on muscle memory to carry out an action that has been performed multiple times without confirmation. It happens a lot but rarely with such a tragic consequence. I have turned the ignition switch to Normal during an engine start when asked to set the park brake during a pushback. There have been multiple occasions where an A320 park brake was set when a flap setting was commanded. On more than one occasion the flaps have been raised when "gear up" was commanded. This may not have been the first time the FCO switches have been selected but definitely the first time it wasn't picked up early enough to reverse the action.

As to the CVR recordings, I will repeat what I have often stated previously. There is no inherent right of the public to receive a full transcript of the CVR in order for them to form their own opinion of what happened. It is up to the Indian AAIB to conduct an investigation under the requirements of Annex 13 and possibly a fuller transcript of the CVR will be published in the Final Report to help the reader gain an understanding of what happened.

My belief is that CVideoRs, with robust protections and legislation around their use, will help accident investigations immensely by answering some of the what questions that the FDR and CVR don't seem able to. It doesn't have to be set up like the many Go-Pro images that are on social media. All that is needed is an image of the center console and the engine display and EICAS/ECAM screens .There would be no need to have images of the pilots faces.

There has been discussion about sim instructor muscle memory translating to the live cockpit. We of course do not know if that has any relevant to this yet, but I believe that is a hazard of training in an environment (eg FFS) that has psychological impacts on the candidate by design.

When running a sim, on repo', setup or end of session I never let any candidate push any buttons or do any actions that would not be considered not part of a normal flow that would normally be conducted at that point. On a reposition to a threshold, I might get the candidate to cycle the FDs or LSK in the MCDU once we are lined up and visuals back but that's about it. Everything else, start lever reengagement, flap resets etc are all off limits to the candidate and I brief as such.

Furthermore when I do the odd simulator in seat instruction exercise and do OEM prescribed "non normal" training (like a "gear up" -> Flap react incorrect selection) I will clench my non operating hand or squeeze my toes until they hurt and consciously tell myself I am about to do something incorrect... then do the (incorrect) action. That is my way of - hopefully - preventing muscle memory habits.

The report doesn't give us the words one pilot used to ask why he did the cutoff, nor the words used by the other pilot to explain that he didn't, nor any indication of the tone of either pilot or even who said what. Some people *believe* that it was co-pilot Clive Kunder who asked why the switch was cutoff, but only because he was PF and would/should have had both hands on the control column during take-off, leaving the captain/PM Sumeet Sabharwal to be the one who cut the switches, but this is very far from conclusive. The investigators will know exactly who said what and when, so it is very strange that it is left out of the report, but is not evidence of anything in itself.

The only distinction the report makes is that the question was 'why did you cut off' rather than 'did you cut off', which suggests one of the pilots saw the other cut the switches, which in turn suggests either (a) the pilot who did the cutoff had done it accidentally and wasn't aware of it, or (b) the pilot who did the cutoff was well aware of what they did but then lied about it despite having clearly been seeing doing it.

Given the question was 'why', it seems equally plausible that it was the PF Kunder who transitioned the switches accidentally as part of some muscle memory blunder, and it was the PM Sabharwal who asked Kunder why he cutoff, who then replied that he didn't, because the fact he'd made the mistake hadn't registered.

OTOH, it might just be a sloppy transcription.


I totally agree with this, it makes no sense at all, especially after his 'just one or two more flights' statement. The guy had already moved cities to help look after his father, and was considering retiring early to have even more time to dedicate. Again, for me, this points to a blunder by Kunder.

So far so good.

Three seconds after lift off, setting initial climb attitude, you suggest that the PF is/was moving his hands ...where?
Walk me through how and why that happened/happens. The training and CRM implications of what you propose are pretty serious...for that airline.
Yes, the vague (in points) language used has elicited a lot of comment.
Not sure why you appear to want to throw the FO under the bus.

What basis do you have for asserting that the switches were missing a locking spring? You are apparently making this up.
Why?
(We had a similar switch on an aircraft I flew some decades ago, and if the spring wasn't working You Replaced The Switch!)

Your previously expressed hypothesis - long sleeve cuff catching on them, and then one at a time this sleeve cuff pulling on each, and then (despite the physical sensation of catching on them and the CM checking to see what the hang up is) changing the position of the switches accidentally - could be easily tested in the simulator.
The investigation team has had between 20 and 30 days to look into that possible explicit accidental switch triggering path.
You seem wedded to it.
Why?
Do you believe that they have, or have not, considered it?
(Part of the reason that I ask this is that the next expected hand movement would probably be forward, not aft, toward the gear handle to raise the gear, in anticipation of a positive rate of climb being called out since they just took off and the PF was setting the climb out attitude...)

I'm not surprised though it gets frustrating when folk start defaulting to the most remote possibilities instead of the more likely ones. The flight data recorder logged the fact that the switches were moved to cutoff, 1 second apart, very soon after airborne, then some seconds later moved back to run. None of that fits with any realistically conceivable mechanical failure nor automated system fault. Both of those are physical actions taken by a pilot. The first action was to shut them down, and the second action was an attempt to start them up again (tragically close to being successful). Nor does it fit with a muscle memory mistake or confusion with another switch.

Which leads to a very unpleasant possibility that many people would rather not consider, and some even refuse to consider, despite there being at least one historical precedent that I can think of (a pilot for a particular airline many years ago who had an undiagnosed mental health condition and admitted to struggling with impulsive thoughts of shutting down all 4 engines on a B747).

Fully agree just seems crazy to me that a non-insignificant number on this forum are chasing a 0.00000000001% chance over the 99.99999999% very obvious answer looking at you square in the face. We might never know if it was intentional or not, only the full CVR may give insight but even then who knows. While its fun to try and dive into the system details, but main problem is no currently employed Boeing engineer who knows the system well enough or could cross check could even comment here without losing their job. All of us who no longer work there can comment but without access to the internal details their is no way any of us can exactly remember the systems 100%.

There are ways to have gates that can be closed in either position (and are perhaps spring-closed), but either they are going to be left open, are a PITA to operate a switch single-handedly, or operating the guard will become part of the muscle memory for operating the switch.

Both gear and flap retraction requires upward movements of the lever while the fuel control switch requires a downwards movement to select Cutoff. And in the case of say gear or flap retraction, only one lever movement is required. In this case both switches were commanding cutoff within 1 second of each other.

With this in mind, the muscle memory / fatigue theory is difficult to understand. I would find it easier to accept if only one FCS was accidentally selected to Cutoff.

Isnt the first action on the FCO switch to lift it? Haven\x92t flown the 787 but every other lever-lock switch I\x92ve used worked that way.