I may have a concrete example of this regression to the mean. I did a lot of FEA on the pilot controls for the Boeing 787 back in the mid-aughts. The 787 is fully fly-by-wire, and I was a little astonished at how much expense and added weight Boeing was putting into making those controls seem like traditional hydraulically-actuated ones.
The yoke is essentially on a ~1 meter long lever, and there’s a big, somewhat complex linkage from that long lever to an actuator which provides feedback, simulating the “feel” of hydraulic controls. This is a good idea because it gives the pilot an idea of what’s happening at the control surfaces.
It’s analogous to power steering on a sports car: Porsche tries to leave as much “steering feel” as possible because it’s useful information for the driver, helping indicate that the car is about to oversteer or hydroplane.
The actuators and linkages add a whole lot of extra weight compared to Airbus’ sidestick, which is essentially a fancy joystick. When I did my little bit on the 787, the program was in big trouble for being overweight. It would be madness to redesign the pilot controls at that point, but it was clear that even though low mass was central to the 787 program, Boeing was willing to add significant extra weight to provide feedback through the pilot controls.
When you engage the autothrottle (cruise control, essentially) on a Boeing, the throttle levers move as the computer adds or removes thrust to match the autothrottle setting. The pilot can put a hand on the levers and feel what’s going on as the autothrottle adapts to small changes in wind conditions (or large changes such as wind shear).
My understanding from a pilot friend who flies A320s is that when you engage the autothrottle on an Airbus, the levers don’t move at all, so the pilot doesn’t get this haptic feedback about throttles and, by extension airspeed/wind trends. As far as I know, Airbus sidesticks don’t provide force feedback either.
So I can see why LSLGuy suggests that Airbus products make excellent pilots into mediocre ones. Excellent pilots make use of the yoke/throttle feedback to understand the state of the aircraft in a way that’s otherwise unavailable.
As an engineer, I dislike both approaches. Boeing builds a fly-by-wire plane while retaining much of the weight and complexity of a pure-hydraulic control format. Airbus “protects” the pilot from information it decides he/she doesn’t need to know about (and filters out pilot inputs it thinks the aircraft doesn’t need to know about).
The Boeing/Airbus divide is a false dichotomy; either could have the best of both worlds by implementing force-feedback sidesticks and moving autothrottles. That would be a lot lighter and (mechanically) simpler than Boeing’s approach while providing all of the haptic information Airbus throws away.
In reality, Boeing’s got a strong argument for using giant actuators and linkages to simulate its older controls: it makes it much easier for pilots to switch between different Boeing aircraft, reducing training costs and minimizing the opportunity for interface-based pilot error. Plus, since both yokes are mechanically linked together, both pilots know immediately if the other is trying to fly the plane.
I think Airbus is on shakier ground for not implementing force feedback/moving autothrottles. Consumer force-feedback joysticks have been a thing for decades; adding force feedback to sidesticks would be really easy (as aviation upgrades go). And although many Airbuses will sound an audio “DUAL INPUT” warning when both pilots think they’re flying the plane, stressed pilots often don’t process audio warnings. If Airbus made both sticks vibrate under conflicting inputs (or, better yet, used actuators to make both move synchronously) this wouldn’t be a problem.
Seriously, its very insightful and flows from your place on the inside of all this stuff where you can see what’s going on.