thermostat. This is not an AF 447 thread. Most of what you state either contradicts the currently accepted facts – location, ice-crystals; or is unsubstantiated speculation.
In terms of man-machine interface there is reasonable understanding of the environment and of the machines reaction to it; what is unknown (unknowable) are the factors and thought processes contributing to the crew’s behavior. The latter point, as discussed previously, is the most difficult aspect of the interface to predict and control.
Linktrained’s point ‘… every mile saved was worthwhile in fuel and flight time …’, is very relevant. Humans are goal driven, and as much as technology improves, the human finds new, unexpected, ways of interpreting the interface (Murphy skills). Thus with modern WXR, crews choose to interpret radar colours literally – green is ‘go’ - 'let's cut the corner'. However, this depends on the situation (machine–situation interface), thus at high level and particularly near Cbs with icing potential, green may be best level of detection for ice and ice crystals, i.e. a weakness of technology.
Training – education and knowledge can restrict some of these weaknesses, but not always.