Too much prosopagnosia

Dave Munger reports that "face blindness," prosopagnosia, might be found in ~2% of the population. Extraordinary claims demand extraordinary evidence, so I'm not convinced, but, if it does pan out this seems to have two primary implications off the top of my head 1) It is a signal for mutational load, and the genetic pathways that lead to face blindeness are sensitive to developmental stress (due to bad environment or genes) or the loci that control them are subject to abnormally high mutational rates 2) There is something else going on and face blindness is a pleiotropic side effect In other words, 2% is just way too high for something that seems so clearly maladaptive. This is only a few steps short of finding out that 2% of the population was incapable of language, recognizing faces is a necessary precondition for much of human sociality. Of course, this might be a normally distributed quantitative trait on some level (this seems plausible, to an extent), and this "2%" are popping up on a very sensitive measure, and only a far smaller % are extreme enough to qualify as pathological. An analogy might be between Asperger's Syndrome and Austism.