I intended to mean it was a similar mechanism in the broader sense that reducing unnecessary sensory stimulation is likely advantageous, using a specific example I was certain of as a token reference in the case of auditory development.

Retraining may not even be a good idea, as I can say with first hand experience being hyper-reactive to subtle emotional visual cues doe's not make functioning in society generally any easier. It does make you potentially better at reading people you are familiar with though.

I like your angle as a vestigial function, and after finding the study posted I think it's better supported than my speculation. What really interests me is the specificity, especially in the lack of reaction to happiness, but not entirely unexpected. Happiness, sadness, anger, disgust, contempt, fear, and surprise are all universal to our expressions. With that being said, I would like to at least have seen anger tested in addition, as we currently attribute the amygdala to processing fear and anger specifically.