I'm a cognitive psychologist, not a pilot. I've just read (well, skimmed parts of) the BEA report, and I think it's interesting that it emphasizes a few issues:
- Diagnosis : did the pilots diagnose "unreliable air speed"?
- Stall warning: did they comprehend it?
The report suggests
- Diagnosis : the pilots never diagnosed "unreliable air speed". Analysis of the dozen or so similar incidents where pitot tubes froze suggest that most of those crews did not diagnose it either.
- Stall warning: did they comprehend it? The report provides references suggesting that in a confusing environment, humans can be cognitively deaf to aural stimuli, and tend to prefer and respond to visual stimuli much better.
So, from a cognitive perspective, the accident makes sense. A big part of this was the human-machine interface, which did an extremely poor job of letting the pilots know what was actually going on.
Could they have done better? Of course. Are they entirely, or even primarily to blame? Far from it.