High Wing - I too have my suspicions! Here's a paraphrase of what I found on the Coanda effect:
"The reason why air flows faster over the top of a wing, especially since the molecules don’t meet up with the ones they started with, is that it speeds up because the pressure is lower, and not the other way around - Bernoulli’s process may start things off, but something else must take over.
For example, the length of the path taken by air flowing over the top of the wing of a Cessna 172 is only about 1.5% greater than it is under the wing, so only about 2% of the needed lift at would be developed at 65 mph (indeed, 2% is the figure calculated by aircraft modellers as to the complete contribution to lift from Bernoulli). On those figures, it would appear that the minimum speed for this wing to develop enough lift to keep the 172 in the air is over 400 mph, or, looked at another way, the path length over the wing would have to increase by 50%. The thickness of the wing in that case would be almost the same as the chord length!
Assuming the Cessna weighs about 2300 lbs, and is moving at 140 mph with an angle of attack of 5 degrees, the vertical velocity of the air its wing deflects is about 11.5 mph. Taking half of that (from the lift formula), Newton's second law shows that the 172 in the cruise must shift about 5 tons of air (about five times its own weight), of air per second to keep flying. That means it must accelerate all the air within 18 feet of the top of the wing.
That’s a lot of air! What’s happening is that the air bending around the top of the wing is accelerating the air above it downwards, leaving a gap, or a lower pressure until the air pressure is equalised. This lower pressure will suck air from the front of the wing and shoot it down and back toward the trailing edge. It is therefore the top surface of the wing that is the critical part, and the magical force called “lift” is really the opposite of the downward trend of the air."
Phil