No contradiction, but you obviously haven't read my previous posts, as each of your comments have already been addressed.
Forget outside references, or ground track. Fly 90 degrees to the wind, and turn hard. Some of the previous commentary on inertia and other factors does play here, as you can certainly see the "mythical" effects of the downwind turn.
As I've previously stated, on my shelf, I have the stick grip a pilot was holding when he impacted the ground in a pecan grove, after executing a downwind turn and being unable to recover.
Yes, the illousion makes a difference, but one must presuppose that some of us are able to see past that. Perhaps another 20 years more...but perhaps not. Perhaps we see things rather clearly now...hard not to, when you see it all the time.
Again, ignoring ground track, keeping the eyes strictly inside the cockpit and ignorning ground references, one can indeed see the difference in airspeed when turning downwind. The airplane, while in a fluid medium, simply does not move with the wind faster than it can turn. This is especially the case with a hard steep turn...a turn (for the purists) that would NOT result in a stall or appreciable loss in still air.
Someone had previously asked if this were the case in the jet stream; if one complete's a 360 degree turn, why is there not a point at which an airspeed loss, and an airspeed gain is noted. In such a case, the aircraft does not experience any change in indicated airspeed. Groundspeed, of course, changes, but this is not relevant. Make that a hard turn, especially when loaded heavily, when the airplane may be made to turn faster than it accelerates over the ground (ignoring all outside visual cues), and the airplane may experience an airspeed loss or gain.
Now I appreciate the illousion factor, and the gradient factors that have already been discussed...I discussed them myself in previous posts. But these are germain only to the concept of perceived loss...not actual airspeed loss (which does occur). Believe me, I had several occasions this summer in 30-50 knot breezes, turning downwind and into rising terrain while descending toward it for a drop, that the illousion was all but overpowering. And downright intimidating. But it's neither here nor there.
Of a greater interest to me at the time was the attendant airspeed loss. I could mitigate that by making a slower turn, but that resulted in unacceptable drift and difficulty obtaining the proper run-in line around a smoke column. The only choice was a harder turn, and to account for the loss by descending to make up for it.
All the same, it's a real phenomenon.