32

There has been debate in healthcare simulation communities around whether or not to allow a simulated patient to die during training because it might negatively impact the intended training outcomes.

Do flight simulation instructors similarly stop the simulator before the simulated plane crash to avoid the student being negatively impacted (traumatized)? Is this common practice?

Update

There were many great responses to this question, to summarize, indeed simulators are stopped before the crash, but this is not related to a fear of traumatizing the student. The simulator is stopped in consideration, typically, of expensive training time (resetting motion and rebooting sim takes time).

BtureP
  • 755
  • 1
  • 4
  • 22
  • 20
    As an aside, historically flight simulators used a camera that "flew" over a physical model. Which meant that a crash was very annoying because things could physically break...! – user2705196 May 24 '21 at 14:04
  • 1
    Mildly related, but speaking through such a case before it happens real life and giving help how to process it is a valuable exercise, I can imagine. Not sure if mixing it with routine drill is good, but then again recognizing the situations where a decision could be the last one seems like a important teaching? – eckes May 24 '21 at 14:34
  • 36
    That's interesting there is thought about ending a medical simulation as to avoid the trainee from being traumatized. But a patients death will eventually happen in a persons career, so why is it better to shield someone during training, knowing they will have to face this in real life? If anything, it's better that a trainee gets an idea if they can handle that kind of.... result, before they go through a lot more training/expense. – Issel May 24 '21 at 16:29
  • 13
    @Issel, the debate over whether to allow a simulated patient to die in a healthcare simulation arises because health care workers will encounter death on the job. The question is not whether to train or not train them on encountering death (though it is generally reported that they do not receive enough training in this area), the question is how and when they receive training regarding death. The debate is whether letting the simulator die, without structurally accounting for it as part of the training exercise, leads to an overall positive learning outcome. – BtureP May 24 '21 at 16:45
  • 42
    Anybody traumatized by a simulator, (medical or flight...) probably ought to consider a different line of work. – Michael Hall May 24 '21 at 19:43
  • @DarrelHoffman, yes, within the consumer realm, the representation of a crash is significantly banal, for both the reason you suggest (glorifying the spectacle), but also it not being the primary purpose of the sim, as well as, similar to the full commercial flight simulators, the complex physics of managing a wreck/explosion. I have heard Microsoft Flight Simulator (2020) users complain of this. Here is an example of a crash in MFS: https://youtu.be/MunbxlO0ctk – BtureP May 25 '21 at 16:33
  • @user2705196 - Correct! My Dad showed me such a machine at Farnborough when I was about 10 – Rich May 25 '21 at 20:57
  • 6
    @MichaelHall Not so sure. The simulation may be really good, and you can intentionally immerse so deep into the simulation that the simulation causes trauma, it will be less than the real situation of it, because you can leave at the moment that "smells like PTSD"... – Volker Siegel May 26 '21 at 13:38
  • 2
    @Volker Siegel, realistic as they might be, there is still the knowledge that it isn't "real". The reactions might be entirely real, stress, elevated heart rate, sweaty palms, etc. but that isn't the same thing. The reaction of a reasonable person following a simulator crash ought to be one of two: 1.) Disappointment in their performance with a learning point having been reinforced, or... 2.) "Well, now I know what an unrecoverable situation looks like". – Michael Hall May 26 '21 at 14:11
  • 1
    @user253751: On the flip side, you don't want someone in a real plane that seems destined for an unrecoverable situation to get used to "oh well, looks like we'll need to restart the simulator again". – supercat Jul 29 '21 at 16:47
  • @supercat that's the same side, not the flip side – user253751 Jul 29 '21 at 16:59
  • 1
    I'm not sure whether it was on here, but I did hear a little story about a student pilot asking the instructor to restart the simulator after the student screwed up, and the instructor said there were no restarts in the real world. – user253751 Jul 29 '21 at 17:00
  • @user253751, yes, really... I stand by my earlier comment. This is of course my personal opinion, but it is based on years of flying and instructing in hi-fidelity full motion simulators. I flew a really crappy single engine missed approach at minimums just the other night. It was self induced, the instructor intended for me to land. I was very bummed out and hard on myself. The accelerated heart rate and sense of initial shock/disappointment was very real, but still I knew it wasn't. My point is that rational people are able to discern the difference. – Michael Hall Jul 31 '21 at 17:14
  • "The debriefing should probably point out it's a simulator and nobody got hurt and that this is why they were training in a simulator in the first place." Expanding on my earlier comment, if this much is truly necessary to explain reality and ease trauma for an aspiring professional pilot then you may want to reconsider the IQ, emotional maturity, and overall suitability of the trainee... – Michael Hall Jul 31 '21 at 17:16

4 Answers4

46

I would say that crashes happen fairly frequently on initial type courses where new pilots are moving up to jets, and mainly on what are called "V1 cuts", where an engine is failed right at rotation or just after.

Many are not ready for the hard roll you get with a swept wing airplane when it yaws hard and you are slow to counteract it with rudder. Before you know it, the wing tip is on the ground. The instructor has to let the pilot try to recover and usually has to let it go all the way.

Otherwise, instructors will normally try to freeze the sim just before a crash because of time wasted waiting for the machine to reboot and having to reconfigure all the settings the instructor has set up for the session. This wastes very valuable and expensive sim time (I was also told it was hard on the equipment, but I'm not sure why).

It's not done to avoid traumatizing the students though. One basic tenet of training is to make things memorable, so overall, a crash-in-the-box is a plus.

You can get a bit lost in the moment and forget you are in a simulator if the pressure is dialed up high enough (after 4 hours of abnormals and emergencies, you get wound up pretty tight; I used to tell people it was like spending hours in a cement mixer filled with bowling balls), but that effect is pretty superficial and when a crash happens, you know you are on an amusement park ride in the end, and you're not going to wet your pants in your last simulated moment of life.

In fact, aside from the visuals, which on the latest ones are pretty good but still not like a real view, after you get used to it, you start to become more and more aware of the physical differences from a real flight, like the motion sensations that don't quite fit what you are seeing (they really only simulate accelerations and decelerations and short term jerky up/down and rolling/yawing movements with any realism, which is good enough most of the time), and the sound of the hydraulic gear working below you, and that tends to kill the "lost in it" effect even more, and it becomes more and more like an amusement park ride as you go back for recurrent training.

John K
  • 130,987
  • 11
  • 286
  • 467
  • 30
    The "hard on the equipment" is likely because of the motion system. The mathematical model is normally not meant to simulate a crash, or even an impact properly. If the software doesn't abort automatically, its behaviour (in terms of simulation) may be basically unpredictable. This may result in large abrupt demands for motion, which can be stressful for the equipment as well as for the occupants. – Zeus May 24 '21 at 03:44
  • 11
    @Zeus I would assume that the software is set up to limit loads to those that won't significantly damage either the simulator or the occupants thereof. A hard landing sure can get the attention of everyone standing around in the room where the sim is, though. :) You can't see what's going on from the outside, so you go from just seeing very light, gentle motions to the thing suddenly slamming against the stops. - haha - It feels like a pretty big bump from the inside, too, but not enough to injure anyone. Not that I've ever had a hard landing in a sim or anything... – reirab May 24 '21 at 16:52
  • 2
    @reirab, Not "significantly damage", of course, but "hard on equipment" it can be. Like 101% N1 will not damage the engine, but will shorten its life... Yet artificially limiting actuators to very safe motion would prevent simulating certain effects in the infrequent cases when they are actually helpful. Often, things are more complicated than just a poweful stroke: for example, motion system can induce vibrations that are not healthy for the cabin, the equipment or people. – Zeus May 25 '21 at 00:29
  • 11
    instructors will normally try to freeze the sim just before a crash because of time wasted waiting for the machine to reboot and having to reconfigure all the settings - If it is such a big problem, why is there no automatical system to do so? I would think it should be trivial to restore a saved previous state of simulation, regardress of the current state. Why would whether the plane crashed or not affect this? – Neinstein May 25 '21 at 06:34
  • 1
    How old are these simulators that take a while to reboot? Presumably airliner models that are nearing retirement still have operational simulators; they might be 25 years old or so. – Anonymous Physicist May 25 '21 at 08:17
  • @Neinstein In a pure-software simulator, that would indeed be easy. The moment you start combining it with hardware, both controls and actuators, it gets a lot trickier. – Mast May 25 '21 at 09:37
  • 1
    @Mast It shouldn't be if the controls/actuators have an absolute location/state monitoring system, which I would assume to be the case in any moderately advanced skinetic simulator system... – Neinstein May 25 '21 at 10:01
  • 2
    "You can get a bit lost in the moment and forget you are in a simulator if the pressure is dialed up high enough" --- Does the opposite ever happen where a pilot thinks that he is in a simulator when he is in a real plane? – DrMcCleod May 25 '21 at 10:25
  • 4
    @DrMcCleod I doubt it. When you are in the real plane it's obviously a whole new world. – John K May 25 '21 at 13:25
  • @Neinstein, not necessarily. Many don't have absolute position sensors (I know for a fact). But this is not even a big problem. Modern simulators are a multi-computer (and multi-unit in general) system. If you restart it, it may need restart, reinitialisation and self-test of all sub-systems, etc. Sometimes (intentionally or not) the avionics will lose the settings/flight plan/etc, and it will need to be set up again. In many cases restart could be optimised to be faster, but as usual in aviation, it is not practiced if it has even 1% chance of incomplete or inconsistent recovery. – Zeus May 26 '21 at 02:35
22

In the rare cases in the Shuttle Mission Simulator when a training session entered a loss-of-control or other irremediable situation, it was standard practice to freeze the sim rather than run the case to the point that it auto-froze due to ground contact or excessive acceleration.

Once it was clear that the situation was irrecoverable, it was a waste of expensive simulator and personnel time to continue.

Organic Marble
  • 2,146
  • 11
  • 24
  • That reminds me of the Space Shuttle Simulator training scene in the 1986 movie 'Space Camp'. "...ahhhhhh, no survivors!..." – user57467 May 25 '21 at 00:27
17

In my experience, often the pilot taking the training is in the process of attempting to manage the aircraft during a maneuver (e.g., engine-out go around, rejected takeoff, etc.) and the loss of control/crash takes place so quickly that the sim instructor does not have time to freeze the simulator beforehand. If it is obvious that a crash is imminent and no recovery is going to be possible the instructor will freeze the simulator before the crash.

However, in my experience (with professional pilots), it is not common for a crash (usually resulting in a full-flight simulator to come off motion and return to its pre-session status [occasionally requiring a reboot] ) to actually occur in simulator training.

If the sim does crash the time it takes to reboot (if necessary) and re-energize (return to full motion/reset) takes valuable time away from the training/checking session. Full flight sim time is normally scheduled very tightly and in order to complete the planned training/ checking events any sim down time can have a significant negative impact.

  • In those cases where a crash is imminent and no recovery is possible, and the instructor freezes the simulator, why do they freeze it? For expediency in training? Or, as is suggested, to avoid a negative training experience? – BtureP May 23 '21 at 20:58
  • 1
    @BtureP - in my experience, for expediency in training. (Professional pilots). –  May 23 '21 at 21:04
  • 10
    @BtureP: I would expect that in many cases, simulators would be designed to stop if the scenario goes outside the range that the simulator is equipped to model. A video game should generally keep going even if its behavior wouldn't be realistic, but a training simulator that keeps running in circumstances where its behavior wouldn't match reality may be worse than useless if e.g. it makes people think a situation would be recoverable when it actually wouldn't be. – supercat May 24 '21 at 04:54
  • @supercat obvious question: why doesn't the simulator auto-freeze itself instead of rebooting or whatever? – user253751 Jul 29 '21 at 16:41
10

In my part 91 (VFR, general aviation) training my CFI was more than happy to let me crash in the sim. Others have spoken about sims that took time to reboot/reset, but in the case of the single-engine simulators I used it's all managed via tablet more or less instantly. I definitely "tested the terrain system" in the sim a few times without it being treated gravely.

To directly answer your question - no, in my experience there was no special treatment about the sim. It was understood that it wasn't real, that this was training. If i'd done that in a real plane I (and any passengers) would have died, but that's the point of the sim. The flight school did the same for part 141 and part 91 pilots, so presumably those on their way to the airlines would go through the same experience.

However, given the subtext of your question (training medical workers), it's not quite the same real-life situation between aviation and healthcare.

In an aviation incident, pilots and passengers are both similarly likely to be injured or killed from accidents. Crashes are not expected or common, and if it happens you'll have more than psychological trauma to worry about. Whereas in a medical situation you're expected to see death routinely. A nurse or doctor might see such a thing on as much as a daily basis, through absolutely no fault of their own or anyone else's.

Knetic
  • 209
  • 1
  • 3
  • 1
    As you say, if you crash a plane in real life, you probably die yourself. But in a sim, you know, on a deep internal level, that didn't actually happen to you because of the fact that you aren't dead. Losing other (simulated) people you were trying to save may be different, because you don't have that gut-level "that didn't really happen" feeling. (I haven't done either medical or aviation training so this is just guesswork on my part). – Peter Cordes May 25 '21 at 17:15
  • 1
    Training to remote-pilot a hypothetical crewless passenger plane could be the middle ground here: you're still thinking of aviation factors, not directly about the human lives you're trying to save as it slips away from you, but what you experience would be exactly like if you'd lost a real plane full of pax. If real planes losing real pax this way was a real thing that you were worried about, I could imagine that being a bit more potentially traumatic. – Peter Cordes May 25 '21 at 17:18