It's the idea of some immediate psychological response to the environment like that of the stress level of a passerby to a facade that bothers me, because it reminds me too much of stuff which IMO hasn't been particularly useful in urban planning, such as the broken windows theory. There's no immediate inference that can be made from the immediate psychological response of a passerby to the behavioral response of a urban population to the same environment over long periods of time. There plenty of neighborhoods in this world that might look like shit but which aren't actually shit, and most often neither do the plainer houses in a given neighborhood house people any worse than their neighbors. But in so far as lots of money might be wasted trying to "solve" these issues, the whole idea to me sounds to me like a massive red herring.
Still, I see your point: are they getting to the right answer in the right way? I like the idea of breaking up blank facades, but I agree with your point that an area can look ugly or frightening to an outsider but fairly pleasant to a local -- and that if this way of doing things grows beyond the blank-facades question, it could become a serious problem.