Meanwhile the subject T7 has remained grounded since the incident. There may be more than regular maintenance to perform in spite of initial statements that there was nothing wrong with it ?

A face-saving method for the crew and the airline.

After the aircraft arrived at the terminal It wouldn't have taken two minutes to download the FOQA and maintenance computer which will have shown there is nothing wrong with the airplane.

Air France, having lost billions of dollars due to covid, has taken the plane out of service, and is sacrificing hundreds of thousands (or millions) of dollars in revenue, to save face?

Depends on how busy the rest of AF’s fleet is. Aviation in Europe, whilst it is on the up it is still well down on pre-covid levels. If they’ve got AoG anyway, might as well make it one that is the subject of an investigation. Any other time, I’d agree with you though.

It would seem fatigue played a role there, from internal AF rumors on another site. not sure the BEA will go too deeply into that based on their past reports but who knows , maybe this time.

Yes, the startle effect is real - amongst the average human being. However, pilots should not suffer it, or should be trained out of it; That is part of what it is to be a pilot - you are not startled by events, but revert to a pre-programmed and pre-practised response to things that don't go according to plan. Isn't that what we practise and are selected and paid for ?

The 'startle effect' is the industry term for amygdala or limbic hijack. In very simple terms we react to things in two broad ways:

1) Stimuli - senses - sensory cortex - amygdala - response. This is the cognitively engaged way; we are thinking about our reaction and it can be as low as 250 milliseconds for racing car drivers or professional athletes etc. But normally it's 500+ ms.

2) Stimuli - senses - - - - - - amygdala - response. This is amygdala hijack; it's reflexive and happens as quickly as 14 milliseconds. 14 ms!!! Too fast to stop or comprehend.

All mammals have this instinct and pilots are mammals. Some folk may be more resilient to amygdala hijack than others but this it not assessed directly at any point in pilot training/recruitment.

The problem is that this all occurs basically instantly HOWEVER the effect on cognitive performance can last for up to 60 seconds. Especially at a time when pilots are required to perform complex tasks with multiple stimuli - go-around with ATC instructions and conflicting traffic.

Also the research indicates that those who are highly proficient at a skill actually react better during amygdala hijack and those who are low in proficiency react even worse. So all the below average pilots who are probably less resilient to startle and don't mentally prime themselves well are affected even more than the average by 'startle'.

Also also; current training regimes can do almost nothing to really replicate real world startle effects. So in the sim you might see a pilot react poorly for a few seconds but fix it. And in real life you see what happens in this case....

Just because a pilot doesn't exhibit the symptoms of being startled doesn't mean they aren't, they've learnt to manage it. I think that's what you mean. By and large pilots are average human beings. Every generation wants to put their own spin (aka terminology) on what has been a fundamental concept in aviation since the Wright brothers - sh1t happens. It isn't a new concept.

And that is one of the worst case scenarios for fault finding an incident, the aircraft system not registering a fault, but a fault witnessed by multiple sources outside of the system, like the human component. That's when a plane will sit for long periods to work out the why and how.

Absolutely , if I remember correctly that was the case here : Martinair B767 Boston 1996
Where it took quite some time to understand what happened.

Uplinker,
I understand your sentiment. What I mean by “startle effect” isn’t a hysterical reaction to a situation, Instead, it’s when an event occurs outside of its “normal” or expected flight regime. For example, a go-around occurring at 1500 AGL as opposed to at decision Altitude where most pilots are expecting and thinking about executing a go-around. I’ve had two “aborted or discontinued approaches” which resulted in a go-around in the last few months. In both instances the F/O was flying. In both instances, the request by ATC took us by surprise and the go-around was not your typical scenario and some confusion ensued. In Seattle, flying into Boeing Field KBFI, you actually have to DESCEND while executing a “missed approach” or aborted approach due to crossing traffic ABOVE you approaching Sea-Tac Intl. Send 100 crews into a sim and give them that scenario and you’ll see many examples of “startled” crews.

The aircraft in question is currently about an hour into an airtest CDG-CDG.

It may have nothing to do with the incident on its last flight, but on the other hand ...

Um JFK-CDG isn't exactly a long trip. Have the crew had a reduced roster due to AF coming out of Covid? Maybe its sleepiness (the need for sleep) rather than fatigue?

Well true the JFK - CDG sector won't have been a long one (by Long Haul standards) but what was the entirety of the trip flown by the crew?

I certainly recall occasionally feeling pretty sub-optimal on the final stages of many an eastern seaboard -LHR sector due to lack of sleep/length of time since any previous sleep....down to "sleepiness" or "fatigue"?

I know the vocab used matters to the regulator/operator ("but our rostering system means you can't be fatigued") but to those on the line does it really matter?

Ok, thought it strange yesterday to see an AF 777 doing two subsequent go-arounds on 9R as we taxied out and held at the rwy. But this being a test flight crossed my mind.

Were the BEA on board this "test flight?"

Well, I don't want to make a big deal of this but all I can say is we must be using different dictionaries. In mine "startle" is "to be surprised or frightened, so as to start, [i.e. jump] involuntarily", (like someone suddenly shouting Boo! in your face).

While flying an airliner and having to do a go-around, or being told to do a go-around, that you have briefed and programmed for, does not startle me. Instead, I think, (in a bored tone of voice); 'oh bollocks, here we go'......"go-around flap" etc. I don't think 'oh **** !..........what.......what's happening?.............what the hell do I do now?', which would be more my definition of startle.

All approaches could be a go-around until you pull reversers on the runway. If an engine exploded, or there was an explosive decompression during the cruise; yes, we would all probably be a little startled initially, but then we would know what to do next.

If crews are being startled by a go-around, then frankly they are in the wrong job. Sorry.

Missed approach:

From an aircraft perspective - Single push TOGA switches = leisurely 2000 fpm clb, (thr toga toga) if LNAV available it will auto engage at 50’ RA or above (thr lnav toga). VNAV is available at 400’ baro but whether or not you can use it is up to your company.

From a human perspective - What Uplinker said..

BTW, If you select toga at less than 2’ ra (when rollout is engaged) then nothing will happen so it’s manual thrust only …


// edited to add ra and baro.. . //

All reactions mentioning here determine startle "reaction" as controlled, however the main idea is that the physical reaction in a startle is "uncontrolled".

Whether "fight or flight" principles make you go-around or not, the action remains uncontrolled by the brain. Your body might be doing things, in a startle the brain is not moving along.

Those people thinking your go-around will be controlled are missing the point of the startle. You personally will only know if you go-around or not if you're experiencing it. And if you're experiencing it, the fact you go-around or not is not that important. The most important thing here is that apparently some people think they can control it. You can't. Live with it. Don't think you are different.

Uplinker # 97 :ok:

A safety issue in modern communication is in the use and intended meaning of words. Also, with some forms of training - define something, then that is what it is; no need to consider alternatives.

Surprise involves the unexpected; part of normal human activity which enables learning - ‘I didn't expect that’; thence evaluation, action, refection, experience. Surprise is within the process of awareness, and involves situational risk assessment and decision making; our reactions, activities align with the perceived risk.
As posted elsewhere, surprise can be ‘fundamental’, an exceptional situation which cannot be comprehended, unimaginable; a rare but not impossible situation.

Startle is generally accepted as an outward reaction, often physical, sometimes automatic depending on the extent of surprise.
Startle is often misused as something to avoid, like ‘error’, inappropriate action, judged after the fact.

In the 777 event or similar, crews might be surprised at the aircraft’s response to an activity.
No response to a manual input is surprising. Such situations require concentrated effort for evaluation and understanding, more often with time pressure.
A reaction (or not) to a button press or mode change might not be so surprising, even if not seen before or available from memory based on training or someone else's experience. A typical automatic reaction is to reverse the last action - backspace, undo, Ctl Alt Del, but even these should be done consciously.

Conscious activity is critical, this requires mental effort to change the mode of thinking; this is often lacking because of the surprise and demand for mental resource.

Training:
- “sit on your hands, count to four” this requires a change of thinking mode - slow down, a forced change to a conscious thinking mode.
- improving situational experience in simulation to develop agility and flexibility in situations without SOPs; unfortunately when surprised in our SOP dominated world, the activity is more often to search for an SOP (or something similar) before fully understanding the situation.

Some deeper / associated reading below; - also note the indications as to why CRM and monitoring might be less effective than we believe or rely on.
‘The Impact of Styles of Thinking and Cognitive Bias on How People Assess Risk and Make Real-World Decisions …’
https://www.spe.org/media/filer_publ..._179197-pa.pdf