Having an inoperable brain condition would not affect how we used a person's actions to predict whether they were likely to do similar things in the future

I've always viewed there as being a third theory of morality: People who do bad things, are more likely to do other bad things. If my friend lies to me, they're more likely to lie to me in the future. But they're also more likely to steal from me, assault me, etc..

A brain defect (such as compulsive lying) therefor needs to be accounted for - the person is likely to commit domain-specific actions like lying, but this doesn't generalize out to other domains. So, I might not believe my friend who is a compulsive liar when says anything outrageous, but I won't be worried that he's going to rob my house or blackmail me.

It's questionable how valid this metric is, but I'll confess it feels "intuitive right" and emotionally satisfying to me, as long as one is able to identify and adjust for domain-specific flaws.