In discussing the moral virtue of prudence -- i.e., right reason about a thing to be done -- St. Thomas identifies three acts of reason:
Judging of what has been discovered.
By taking counsel, he has in mind an inquiry or analysis conducted by the reason to determine the means to achieve some end.
The evidence presented in Blink suggests two important, and perhaps underappreciated, things about how our faculty of unconscious judgment plays in all this.
First, unconscious judgment is capable of doing some mighty fine analysis, even on matters we ordinarily think are more suited to discursive or logical analysis. Unconscious Judgment isn't just the ditzy blonde sidekick saying whatever pops into her head while our hero, Rational Intellect, solves the puzzle by careful deduction. They both operate on the same sense data, memory, and experience.*
It's because we usually don't, and often can't, know how or why our snap judgments are made that we might ordinarily think of them as random guesses (if we devalue them) or quasi-mystical intuition (if we value them). It's also because we usually don't, and often can't, know how or why our snap judgments are made that it's very difficult to know how much to value them -- how to judge our judgments in that second of three acts of reason I began the post with -- in any particular case.
Which brings me to the second underappreciated aspect of unconscious judgment: It can be trained. Our memory and experience affect how we unconsciously respond to new sense data, and we can consciously add to our memory and experience in ways that improve our unconscious response. This is basically what happens when we master a subject, when we reach the point where we don't need to analyze a problem step-by-step but can just leap to the solution.
Now, as St. Thomas famously admits in the very first article of his Summa Theologiae, even the best of human reason can involve "the admixture of many errors." A trained unconscious is no more infallible than a trained conscious. But memory and experience can improve both, and by that I mean memory and experience of correct analysis and judgment.
Which is to say, the person of virtue, who habitually chooses good and avoids evil, who has mastered the subject of right moral action, will have a relatively reliable unconscious judgment. He will be prepared, consciously and unconsciously, to do the right thing.
On the other hand, the person who lacks virtue will be unprepared unconsciously, even in cases where he consciously desires to do the right thing. His discursive counsel will have to work against who knows what sorts of unconscious impediments to produce a sound analysis. In short, he's up the creek.
The take-away, then, would be something like this: We should do all we can to train our unconscious judgment to operate correctly, both over time, by showing it what's correct (in churchy terms, by being virtuous); and from moment to moment, by priming it to incline toward the good (in churchy terms, by praying always).
* If I can be forgiven a computer analogy, unconscious judgment might be compared to a neural network, which produces an output by combining the input data according to rules that don't have any real meaning outside the network. A weight vector in a neural network doesn't correspond to anything in the real world; it's just an artifact used by the network to come up with an answer.
Discursive judgment, on the other hand, is more like -- oh, say, artificial intelligence or formal methods. You perform a sequence of analytical steps, at each of which you can explain what you're doing and what it means.