How accurate are our evaluations?

If I were inclined to gamble, I'd have risked a month's income that the two gentlemen who boarded the bus Monday were more than slightly acquainted with prison life, and probably were still gang-affiliated.

Demeanor, clothing, something about the expression and the way they looked at people--I can't pin down the details, but put it all together and it added up to "people who found the trouble they went looking for."

The newspaper publishes mug shots of those arrested or convicted, and most of the time the perp looks the part. You'd avoid him, or not trust her checks, just because of the way they looked. "The eyes are the mirror of the soul"--and maybe "a lifetime of bad character" puts a stamp on the features.

Not always. The bank robber trio consisted of a man who looked suitably violent and two younger men who looked, for a change, quite harmless. False negatives.

And my driver's license photo shows me as a malevolent zombie, which I trust is a false positive. I'm not shambling, I just have a bum knee!

But a lot of the time it works--at least within a single culture. Skillful manipulators can fool pretty much anybody, and some people (and animals) have unfortunate features that give a terrible first impression. (I'm thinking of a prognathous and jut-fanged dog that looked aggressive, but tended skittish.)

I don't know if this was replicated, but the claim that people are good at spotting the naturally empathetic was interesting. I don't think I'd put much faith in studies involving people looking at pictures--face to face evaluations include far more information. Can we put numbers to this with any reliability?

Somebody must have done some studies of this for use in jury selection. Guilty or innocent? Hmm. He's in a suit, and smiling almost normally, just as he's been coached, but we're not face to face and not talking together.