"According to a 2004 study, the average life expectancy of a non-instrument-rated pilot who flies into clouds or instrument conditions is 178 seconds."
This statistic is the aviation equivalent of the "Bill Gates is going to give you $5" email forward.
The study was in 1954, not 2004, and the pilots that were tested had zero instrument time (at least three hours are required to get a license now). They were put into Beechcraft Bonanzas, aircraft that none of the test subjects had time in; Bonanzas aren't tough to fly but you definitely have to transition into them, especially if you're not used to planes with more horsepower than a Cessna 172...they get ahead of you way faster.
That said, if you're a VFR pilot and you knowingly proceed into Instrument Meteorological Conditions, you're probably in for some unpleasantness.
While you are right that it is from 50 years earlier than stated, it seems like a solid study. No, it's not actually "life expectancy", but I was impressed on how well designed it was. It makes for sobering reading, but the conclusion is actually positive: with only six hours of direct experience training, they could teach people how to make a safe 180 degree turn with a high percentage of success.
They explain the choice of the Bonanza, saying it was the most complex plane that a non-professional pilot could be expected to fly. The emphasis of the study was that simple training could save lives, and thus they wanted a high bar. "In sort, the assumption was made that if the subjects, none of whom had soloed a Bonanza, could master the technique in this airplane, they could master it in any single-engine airplane under 3,000 pounds gross weight."
And it's not quite true that the pilots had zero instrument time at the time of test. The first lesson included a very small amount of time to familiarize themselves with the instruments, and were given a second try if they lost control within only a few seconds. How much of a difference this is from the current 3 hours, and how far short of the recommended 6, seems worthy of discussion and likely a more modern study.
It is, provided that context is given when statistics from the paper are mentioned...which rarely happens.
To fly a Bonanza you have to have a private pilot's license, a high performance endorsement (required to fly anything with HP > 200), and a complex endorsement (required to fly anything with a constant speed prop, flaps, and retractable gear). This takes some time...and good luck getting insurance coverage (the real factor in what you do/don't fly) if you don't have an instrument ticket.
Furthermore there's a world of difference between zero or a few minutes of instrument experience and three hours. I don't have any study to back this up but I'm a pilot and can tell you from experience that the first time you go on full instruments it's sheer sensory overload. Three hours doesn't prepare you to do a decoupled IFR approach or anything, but it give you enough to do a 180.
Here's another important factor about the study that never gets mentioned: in the study the attitude indicator (artificial horizon), heading indicator, and vertical speed indicator were covered, simulating a partial failure of the vacuum system as well as a partial failure of the pitot/static system. No pilot is going to take off these instrument in failure, and should they fail in flight most pilots are getting down, now. I know I would be heading to the nearest field if I saw partial failure of both instrument systems.
So instead of saying "VFR pilots are toast in 178 seconds if they go into clouds" quotes should read more like, "if you happen to be Dr Sam Beckett and you find you've quantum lept into a Bonanza that has partial instrument system failures and you've flown into a cloud, you better hope Al can jump your ass back out within three minutes."
Three hours doesn't prepare you to do a decoupled IFR approach or anything, but it give you enough to do a 180.
'Most noninstrument pilots can be placed in one of the three following categories: (1) the noninstrument pilot who knows he could not fly instruments and takes every precaution to avoid instrument weather; (2) the noninstrument pilot who "knows" he could not fly instruments, takes every precaution to avoid instrument weather, but believes his knowledge and experience would enable him, if caught, to fly out of instrument weather; (3) the noninstrument pilot who believes, primarily through ignorance of the problems involved, he could fly through instrument weather.'
Are you instrument rated? Have you taken further training beyond the requirements? What's your confidence that you could successfully perform a 180 degree turn out of cloud in the plane that you fly the most? Would you keep your hands on the controls or not?
in the study the attitude indicator (artificial horizon), heading indicator, and vertical speed indicator were covered, simulating a partial failure of the vacuum system as well as a partial failure of the pitot/static system.
Again, I think this makes sense in the context of the study. The goal wasn't to simulate failure, rather because those were (at least at the time) not required to be installed.
The interesting part to me about the study (I'm not a pilot, but have played with simulators such as XPlane and probably plan to get a license some day when I can afford the time and expense) is that none of the test pilots were able to perform a life saving maneuver when starting, and after 6 hours of direct training all were able to. And yet, oddly, this isn't (to my knowledge) part of the current pilot training. Why not?
Training for a PPL currently requires 3.0 hours of instrument time, including maneuvers like climbs, descents, and turn to a heading. It also requires unusual attitude recovery, a drill where you cover your eyes and the instructor puts the plane at a cockeyed angle and you have to recover using only instruments. You must successfully perform these maneuvers in order to pass your FAA checkride.
Basically in order to get your PPL you have to not just be able to turn a 180 on instruments, but demonstrate some level of precision in other maneuvers...so I think that while the study demands six hours, the FAA has determined that the job can be done in three.
As to your questions for me: I'm not instrument rated. I've taken about two hours of instruction on instruments beyond the required 3.0, most of that time was on the flight were I got my high performance endorsement...in a Bonanza, coincidentally. I'm confident I could do an instrument-only 180 in the Piper Cherokee I fly most frequently because I've recently done instrument work in a more challenging aircraft. A month from now with no further instrument work I'd be significantly less confident.
Hands off the yoke? Probably not all the way, but I've been trained to keep a very light touch as anything more, at least in cruise flight, means you need to check your power/pitch settings (eg throttle/trim).
The completely "hands off" (the yoke) startled me. They concluded that this was the safest approach to teach: all turns with rudder, and a predetermined trim and power. Post instruction, they allowed pilots to try their own approach, and say that the two who tried to control oscillations with the yoke "realized almost at once [that this resulted in more extreme oscillations] and both subjects immediately released the yoke and continued through the remainder of the steps with 'Hands Off'."
Having just read Langewiesche's Stick and Rudder, with the emphasis that all turns should always be coordinated, I found this very surprising. The study suggested that the instructor use this explanation: "Bear in mind that this whole procedure is a 'gimmick' designed to save your life. If it is easier and safer for us to make a 'sloppy' turn in order to get the job done, then that is the best way under the circumstances."
You're welcome. I'm like most pilots, my favorite subject is flying. There's an old joke: how do you know if there's a pilot in the room? Because he'll tell you.
You know, the hands off 180 definitely something I'm going to try next time I'm up with a CFI/I. Also if you get the opportunity in this life to learn to fly, do it. It's expensive as hell but it's also a very rewarding and fun challenge and is a great way to travel.
Highly recommended piece by William Langewiesche on the same subject:
The Turn
At the very heart of winged flight lies the banked turn, a procedure
that by now seems so routine and familiar that airline passengers
appreciate neither its elegance and mystery nor its dangerously
delusive character. The author, a pilot, takes us up into the subject.
Given the analysis of what happened in the recent Air France crash, I can't imagine why the industry does not write better software to help combat this. It seems as if the programmers writing the software are more concerned about writing valid code to fit the designed system ("the pilot signaled us to climb at exactly 40 degrees, so do it") rather than writing code that models the real world situations accurately. I found this bit of an article about the crash the most troubling:
It was at this point, after autopilot turned off and they
worked to change their course, that a stall warning
sounded, meaning that the airplane wasn't generating enough
lift. The report notes the co-pilot grabbed the controls
and lifted the plane, which, according to aviation experts
is contrary to normal procedure during a stall, when the
nose should in fact be lowered.
If pulling back on the stick should never be done when a plane is warning of a stall, why does the software allow a pilot to do it? This is surely simplifying the situation, but I can't believe we are incapable of writing better software to prevent human mistakes.
I wonder what kind of hard limits do exist in the current avionics. For instance, is it possible to roll a Boeing/Airbus upside-down? If yes, why?
What the pilot will do to keep the passenger and plane safe can't be foretold by software written in static situations on terra firma.
Sensors can lie, equipment can break. Your software may have no better idea of the true situation than the pilot.
In the end it is a judgment call and if you limit the ability of those judgment calls to a bunch of software then you are saying that those in a different time and place have a better idea of what to do than those with their feet in the dirt.
It looks like in this case someone made the wrong judgment call. False positives in warning systems in commercial aircraft are happening with some regularity, in the end the pilot has final say.
I think a more robust model of N completely different systems (both mechanical and electrical) to compute the same variable could be found so that the probability that it is computing a value outside of x% of the actual real value is lower than the probability that the human could make a mistake. Even if a speed indicator model in the event of pitot tube loss had a 25% margin of error, autopilot could have done the right thing or it could have possibly influenced their decision to pull back.
EDIT: You can down-vote, but this is already the direction that cars have gone. For example, there is no way to stop any Mercedes after 2009 from automatically applying full brakes in an impending collision.
I'm not downvoting you, but I think this is a kind of problem that seems solvable on the surface, but is very, very difficult to get right. You will always have situations where all speed indicators fail in a similar way leading the software to believe that the speed is correct. With all the erring on the side of caution that aircraft makes have used, we've had a couple of cases where the software "decided" something and caused incidents.
Studies in N-version programming (such as used in Airbus planes) show that software written by different teams often fail in the same corner cases (with differing outputs).
Eventually you might have software with "situational awareness" that can be trusted to make fewer mistakes than humans, but we're not there yet.
I agree it will take a long time to develop automated systems that can be trusted. It took 26 years for tech to develop from the first ILS assisted passenger landing to the first fully automated ILS landing.
If you were to take this specific speed sensor issue, I'm more concerned there aren't different types of systems being used to try to cross-validate the numbers. The pitot-static system has its own backup on the plane, but if there is an inherent weakness of that particular system (e.g. ice), there is no other way to measure speed that is not susceptible to icing. Surely that particular problem is within our reach, no?
To measure airspeed I guess something has to have air flowing through it or past it, and that something would be susceptible to icing. That said, current pitot have very good ice protection, and AF447 was an extreme example with a known problematic type that Air France were in the process of exchanging.
Perhaps GPS ground speed data with previous wind estimates could at least help with establishing that "something must be wrong with the airspeeds".
Something scary from a Boeing pilot in the comments:
Loss of all airspeed indications is not something we train
for. It's always assumed in transport category aircraft
that redundancies built in will always give us at least
attitude, heading, and airspeed.
There are meterological radars that use doppler shift to measure wind speed, I wonder if you could use the same basic idea to construct a radar airspeed indicator?
"For example, there is no way to stop any Mercedes after 2009 from automatically applying full brakes in an impending collision."
Which is sad.
From your link:
[quote]Brake Assist PLUS uses two radar systems. Both of these radar systems are very good at not only registering objects up ahead but also calculating the distance from these objects and their relative speed. If the distance between the vehicle and the registered object is so small that there is a real danger of collision, Brake Assist PLUS issues a visual warning on the instrument cluster display and sounds an alarm over the car speakers. While this is happening, the electronic system calculates the braking deceleration necessary to avert the collision. If the driver then hits the brakes, the braking pressure needed to bring the vehicle to a stop before impact is immediately provided automatically. Ensuring that an appropriate braking force is not exceeded gives those driving behind more time to react.[/quote]
A human would be quick to notice that the reason for why the car in front is braking is because of the impeding concrete wall. The computer however doesn't know that and would only brake in order to not hit the car in front and thus ensuring that both cars hit the concrete wall at equal speeds, great!
(I truly hope that it respects my choice if I really wan't full brakes).
What if there are slippery conditions? The collision might easily be avoided by changing lanes but changing lanes in muddy snow while braking could be suicide. What if I just started changing lanes when the computer feels like it needs to brake?
What if the reason for why the air-speed indicator is malfunctioning (reporting wrong, a complete loss of velocity is too easy to check for) is because the plane just flew into a flock of birds, it might be obvious for a pilot why the stall-warning was issued and the pilot might opt to to pull back the stick - just to avoid an collision...
Unless the sensors truly get the whole picture and the complete state of the aircraft/car (and I doubt we can do that before being able to implement a true AI) the pilot/driver must always have the last word. Assistance can be great, especially in cases where a human wouldn't have the time to react but you should always be able to override it.
The assistance should be clear and specific enough so that the pilot/driver really thinks twice before overriding anything (this requires that the pilot/driver respects the system). And if the pilot/driver can't handle the vehicle despite that then he/she shouldn't be allowed to drive it.
Aircraft are a lot more deadly than roads. An air collision is pretty much fatal, whereas automobiles hit things all the time, often with no real damage to the occupants.
If your software is making the decisions, then the pilot can't make an educated response to unusual decisions. Big commercial aircraft are already largely run by software, but the pilot is there for unusual situations because there's often a lot of conflicting issues.
Keep in mind that unlike auto accidents, incidents like AF447 are incredibly rare. That particular problem could've been fixed with better pitot heaters, no need for HAL-like software gimmickery. By implementing large-scale changes in an attempt to eliminate corner cases like that, it is all too easy to introduce vulnerabilities that will lead to completely different accidents, no less unpredictable.
There's a lot I don't understand about AF447, like, why couldn't a simple GPS receiver have saved them? GPS does a relatively poor job at reporting altitude, and is obviously useless at reporting airspeed, but it's certainly capable of telling you that you've dropped 10,000 feet over the past minute or two. I can understand how pilots could become disoriented, such that they don't know "which end is up," but I can't fathom why they shouldn't even know whether they are climbing or descending, which apparently was part of the AF447 scenario.
For example, in the case of Airbus airplanes, flying in normal law's flare mode provides protection against high angle of attack and bank angle protection.
However, when something bad happens, the plane is usually switched in alternate or direct law, based on the philosophy that the software can fail and the pilot is the only one able to see the big picture and compensate for possible faulty sensor data. It's all a question of who to trust in highly-tensed situations: is the pilot suffering from vertigo or it's just a matter of faulty sensor readings due to external ice build-up?
And if the software were to override pilot decisions, just think about liabilities in case something goes wrong, which is a Pandora's box for aircraft manufacturers.
Pilots are very reluctant to hand over control to any kind of computer. It goes completely agains their self image.
This is more true in certain company than other.
A very similar phenomenon is how the USAF crashes more drones than the Army since they use pilots that insist on flying them in manual, vs the Army operators that do not consider themselves pilot. http://www.theregister.co.uk/2009/04/29/young_usaf_predator_...
Commercial flight will become much safer when the persons in the cockpit will considere themselves plane operators vs pilots.
Very interesting.. Pilots that want to be rated for Cat II/III ILS landings have to perform them every now and then to stay rated, so I bet a lot of people have been through computer controlled landings without even knowing. Definitely whenever the weather is unmanageable, they have to turn over control already. Here's a really good example of something that would be very difficult for a human to do:
I think it's easy to say that without taking into account all the experience that pilots may bring to the table when an incident occurs.
Pilots (and airlines) seem more than happy to turn flying over to the computers, in fact FAA and Airbus are concerned that modern pilots lack hand-flying skills due to over-automation. The AF447 crew did not have training in high altitude hand-flying, for instance.
Until you have a situationally aware computer that consistently beats humans in accident scenarios, we still have to trust humans to do the job, for better or worse.
And still with AI systems taking care of the plane security there would be failures that will knock downd computers , lightning strikes, generator spikes, some kind of fisical damage to the plane, fire at the computer rack. All this is usual enough to keep comercial flights from using drones.
Hehe believe me computers fail and is not 1 in a million failure more one in a 50 flights or so, you reset them and nothing fatal happens, but they fail! Fly by wire only saves weight removing or reducing cables and hydraulic systems.
In fact adding Fly by wire adds more computers and more complexity to the plane. I has taken more than 15 years to fix to airbus to fix lots of bugs.
The airfrance crash INMHO was a lack of training( or missunderstanding of it) because the pilot flying on the captain seat ( actually a copilot, the captain was resting) was performing the maneuver for fighting windshear( you won't find this on any conclusions is my guess from his input on the sidestick).
In a stall you want to recover your speed asap , you dive and apply full throttle till you are able to level off with normal speed.
Windshear usually happens when you encounter heavy wind changes on approach or take off, usually caused by heavy thunderstorms. Your speed will increase and then sudenly decrease to stall levels. It can very easily throw you to the ground.
As you are very low, the stablished procedure is to apply full throttle ( as in a stall) but PULL your controlls full up ( the software limits the nose up position to around 25 deg nose up in airbus models). You flight at the minimum speed ( airbus airplanes will not allow you to fly below that speed at least if you are at low altitude, a plane using both engines at go around power will climb).
The point is that as they entered the thunderstom ( that previously was avoided by preceding airliners) and had that pitot icing, they reacted with a windshear procedure when it was unnecessary ( there is a procedure to flight without reliable speed indications just trusting your artificial horizont and thrust position for a given speed, they should have applied it in a perfect world[it is very easy to say what is the solution afterwards of course]).
At those altitudes the engines have no power left to perfor a windshear, so the speed started falling till they entered a deep stall. That stall was recoverable( this is a supposition) for lets say 2 minutes, but they still were aplying full up controlls thinking :why is this procedure that I have practiced thousand times in the simulator not working?( not understanding they were in a common stall)
It seems that the other flight officer may have known what was happening but failed to take the controls.
All this is my supposition based on the preliminary documents pblished and my experience with A320 aircrafts wich have pretty similar flight controls logic ( I am an airline pilot with 12 years experience flying boeing, douglas and airbus). Something completely diferent could be the final cause.
Windshear is a primary maneuver to practice and master due to the extreme danger that it represents for any kind of plane ( maybe not for fighters). Stall is seen as a improbable danger due to the airplane software protections. Both are practised but the windshear is more enfaticed at least in europe, also the stall is seen as a basic flying skill, something not practical to practice in an expensive simulator with limited training time for all that complicated failures.
To reply your question, you can not induce a roll to an airbus airplane ( not past 45 deg with normal conditions) unless you have lost some flight computers. You could for example with MD 80 or 737 as there are no computer protections. I don't know about 777 or other fly by wire boeings, but I guess they got some kind of protection.
Sorry for all the typos and spelling! I am writting on a van on my way to the airport.
the comment "Stall is seen as a improbable danger due to the airplane software protections." puts the suggestion that software do more to make things "safe" in an interesting light. thanks.
Software in airplanes is a huge help! Also if is not properly thought or has bugs it can kill you too. One of the problems right now is over reliance in the capacity of the plane to detect and combat properly ANY failure or emergency.
Seems like the perfect scenario for sensory augmentation. There was an article on HN a while back about attaching electrodes to the tongue to replace vision in blind subjects.
One of the precursor experiments was attaching electrodes to the hand that were tied into a compass. The subjects were able to eventually feel North.
A similar device could be attached to something like an altimeter or other instrument.
> According to a 2004 study, the average life expectancy of a non-instrument-rated pilot who flies into clouds or instrument conditions is 178 seconds.
That is a chilling statistic.
As a VFR pilot, I often wonder what would happen inside a cloud to make me react and rip my plane apart in under 3 minutes.
I think it's that so much of our orientation is dependent on peripheral vision. I know Canadas worst pile up was caused by a heavy fog. 87 vehicles piled up, they believe in a very short time from the fog rolling in. The main factor was speed, basically people were over driving their view distance. Visibility was estimated to be something like 50 meters, but driving at 110kph gives you a stopping distance of around 150 meters. What worsened the incident was that light fog generally reduces drivers speed, but heavy fog is known to increase speed as the lack of peripheral markers means people can't gauge their speed and too few drivers check their speedometer.
I'm sure a pilot in dense cloud has the potential to make a whole host of mistakes. First I'm sure without instrumentation a pilot would quickly lose gauge on how level they're flying, or second guess how level they're flying and try to adjust. If you get caught in a cloud with a very low ceiling, there's probably not much time between exiting the cloud and that "Oh shit, that stuffs the ground". Or a slight climb and not enough thrust could easily compromise your speed and put you into a stall.
I'm quite sure 3-dimensions of travel is a real bitch when you're blind and can't stop.
With an altimeter, an artificial horizon and an airspeed indicator, you would expect a VFR pilot to be able to get through any light cloud safely.
Untrained, never having touched the cockpit of a plane, I can get the meaning from these instruments. Is it simply that pilots fly without these, or that they simply disregard them during flight?
Neither. In US, your plane has to have the 6 basic instruments to be certified as airworthy by FAA. The reasons pilots die in non-visual flight conditions are twofold.
As the original article pointed out, the first reason is that the sensations are so compelling that doing what the instruments tell you will feel VERY unnatural and you will keep thinking that your instruments are broken or stuck (hence you HAVE to trust your instruments). And even if you remember not to trust your instincts, you will still often unconsciously apply slight pressure to the controls.
The second reason is that for every change in the plane's orientation and speed, several instruments will start moving at once. Without training it is very hard to integrate what the 6 instruments are telling you into a complete mental picture of what's going on with your airplane.
Ah, it's from a simulator, where you don't get the benefit of your own body responses - no inner ear cues, no tactile feedback, no proprioception feedback.
People driving cars in simulators also do worse than in the real thing - a simulated car doesn't have all the ways a car tells you that you're driving close to the envelope.
Not exactly. "Simulated Instrument Flight Conditions" usually means that your instructor has you put on a view limiting device [1] so that you can only see your instruments and not outside the aircraft.
One of the reasons it's done this way is to teach you to ignore inter ear cues and trust your instruments. Your inter ear senses acceleration, not absolute motion. Couple that with the centrifugal force during a turn changing your perception of which way "down" is, and your body responses become one of the main reasons you get into trouble.
It's similar to banked turns on car racetracks. When you're in a banked turn, "Down" for you is no longer towards the center of the earth, it's at an angle, perpendicular to the speedway. You also loose the sensation of turning once you're in that banked turn long enough (specifically once the fluid in your ear settles down and stops moving). If you were driving with a blindfold, you wouldn't be able to feel how steep your turn was.
Thanks for the info - I was unaware of those devices.
I guess what I was getting at was not so much that you could survive with your inner ear, but that you get a lot of feedback from a vehicle that you don't in a simulator. That's not relevant to this study if they didn't use a simulator, though...
Usually what will happen is that you'll think you are still flying straight but a small roll and dive will be happening, when you come out of the clouds you'll be in a very deep dive and/or almost inverted. If you don't have enough altitude to recover, you are done.
I've had a similar experience while driving on a highway at night in foggy conditions. All of a sudden my car started slowing down and was rapidly losing power. I checked and it wasn't fuel, so I figured the engine was dying somehow. It wasn't until I tried gearing down that I realized I had been on a steep hill and fifth gear wasn't powerful enough. But even afterward, I couldn't sense that I was on an incline.
I have a simple trick for that: I know the consumption of my car on a level road in 5th gear at 90 km/h, so when I'm in doubt about whether or not I'm going up, down or level I look at the fuel consumption indicator.
If it is higher than 6.6L/100km then I'm going up, less is down.
If you don't have a fuel consumption indicator then that will not work but lots of cars have them these days and I think using them as a level is a useful unintended application.
Another trick might be using star alignment. A fixed position camera on the top of the plane could sense pitch by tracking which stars are in view. Or similarly, if you boxed a sensitive GPS receiver so that it could not detect signals from the sides, the satellite IDs which are in view could possibly indicate alignment of the plane as well. The best model would tie in as many observations as possible to maximize the probability that it is right.
EDIT: What about fuel sensors in the wing? If the computer knew the volume of the remaining fuel and the altitude was not dropping, it could determine the orientation of the plane if there were sensors inside the tanks to detect which were submerged.
All car computers know this even if they don't display it to you. The computer directly controls the flow in the injectors and the computer knows their flow rate (unless you swap them out, of couse).
They are ~everywhere~ for a long time around me. Quite a lot of driving instructors (specializing in 'reduced fuel consumption' trainings) use them as a general guideline to "see" how your behavior drives the consumption up.
Heck, mum and dad use that thing for years and drive even slower since they first saw that they could make their car use less gas.
I don't see this article presenting the full picture of how this problem will be addressed in the future.
As ADIRUs (navigation computers) improve to better integrate GPS and inertial signals in all conditions even when all air data is lost, weather-related instrumentation failures will decrease. Next, putting a worldwide terrain map in the navigation system will let it compute whether it will hit the ground if the course is maintained - again, even when ground radar is not working.
Next, heads-up displays or helmet-mounted displays with better warning systems will make sure that the pilot receives the warning. And most importantly, when the pilot feels that they don't know how to react to the warning, they will be able to push a button to let the autopilot return the plane to wings-level flight.
On Sept 6, 2011, ANA Airlines had one of their flights flipped almost 180 degrees, because one of the pilots pressed the wrong button. The weird thing is that although passengers felt the turbulence, they didn't know they had flipped almost upside down, I guess because of the same forces that confused these pilots. It's a scary thing, and if I remember correctly, its the same thing that killed JFK Jr.
I don't see anything particularly confusing about that. The Wright brothers flew in 1903, and if we don't limit ourselves to pilots of heavier than air craft, Zeppelins started flying in 1900. It would be perfectly reasonable for a 1906 researcher to look into things that might effect pilots.
That said, the article may be mistaken in the sense that the research, while applicable to pilots, appears to have actually been motivated by his work with patients suffering from dizzy spells. (BTW, he received the 1914 Nobel Prize in medicine for this research).
This article includes a statistic so scary that I'm going to have to fight the urge to paste it twice:
According to a 2004 study, the average life expectancy of a non-instrument-rated pilot who flies into clouds or instrument conditions is 178 seconds.
At that rate, I'd imagine it didn't take long after the first plane flight for pilots to notice that there was a big problem with, say, closing your eyes or otherwise getting distracted while flying. Or perhaps a pilot or two tried flying into a cloud -- it's the sort of thing you'd try to do if you could, right? -- and their remaining colleagues then started wondering why flying into clouds was such a sure-fire way to die. Death sure does have a way of crystallizing such questions.
The other point I'd make is that fixed-wing aircraft research probably had obvious potential military applications even in the prewar age of 1906. Perhaps this research was funded so early for much the same reason that astronaut-physiology research was well funded even before the first astronaut reached low-earth orbit.
Your parent is probably skeptical because in 1906 total human powered winged flight time could reasonably be measured in minutes. Europe still believed it was a hoax. Pilots didn't even have instruments let alone a cockpit. If you were a pilot, odds are your name was either Orville or Wilbur.
Remember, the Wright brothers first flew in 1903.
As another reply points out, it was likely blimp, balloon or some other form of aviation that spurred the research on vestibular effects.
You forgot about Santos-Dumont, and many others besides.
Europe definitely didn't believe it was a hoax, in fact there were lots of people working at attempting powered flight.
That does not diminish the accomplishments of the Wright brothers in any way, but to make it seem as though Europe did not believe powered flight was possible is not true.
The time was simply ripe, lots of the bits and pieces had been developed by those that had gone before, such as Lilienthal (who built some pretty amazing gliders) and many others.
But it would still require action from a pilot, who is known to be disorientated, and probably quite upset about their imminent collision with the ocean.
Perhaps an alternative would be to use a belt that vibrates differently depending on the angle of the aircraft. Humans are good at integrating data without conscious thought. This has been used with compass direction on the ground to dramatically improve spatial awareness without any conscious effort http://www.wired.com/wired/archive/15.04/esp.html.
Even little planes have those. What they need is a switch to tell the computer that they expect no descent whatsoever below xy000 feet, to stop it before it starts.
I was going to suggest this in my post. A "dead-man"'s limit. A plane would lock you out and right itself it detected you dropped below some pre-defined altitude (e.g. within an acceptable angle without landing gear deployed or a greater than acceptable angle with or without landing gear deployed). Whatever altitude is necessary for the plane to right itself from the steepest descent. Something like that would make it impossible for someone to storm a cockpit and nose-dive a plane. It would need appropriate remote electronic and mechanical overrides, though to prevent possibly losing a plane if it was triggered incorrectly. But at least a plane flying level on lockout autopilot would buy some time opposed to it definitely being destroyed by the nose-dive.
A helium balloon suffers from the exact same flaw as your vestibular sense: it cannot distinguish between orientation and acceleration. There is already an instrument in every cockpit that shows the true orientation of the craft: the artificial horizon.
I'm not sure it would be that easy. During turns and such, it may send the balloon fluttering in odd or unreadable directions. During such an emergency, its highly unlikely that the forces acting will be so consistant.
This statistic is the aviation equivalent of the "Bill Gates is going to give you $5" email forward.
The study was in 1954, not 2004, and the pilots that were tested had zero instrument time (at least three hours are required to get a license now). They were put into Beechcraft Bonanzas, aircraft that none of the test subjects had time in; Bonanzas aren't tough to fly but you definitely have to transition into them, especially if you're not used to planes with more horsepower than a Cessna 172...they get ahead of you way faster.
That said, if you're a VFR pilot and you knowingly proceed into Instrument Meteorological Conditions, you're probably in for some unpleasantness.