I've been using the idea of coherence for some years most recently in my recent post on the complex domain of Cynefin. The way I normally explain it is we can say that evolutionary theory is coherent to the evidence, while creationism is incoherent. This does not involve us in saying that evolutionary theory is right, because we know its incomplete and aspects of it are probably wrong. But it is coherence enough to explore further. On the other hand creationism requires mental gymnastics that place any person educated this century on the border line between self-deception and psychological aberration.
Now while I cite Thagard on this, I don't want my position to be confused with the coherence theory of truth within philosophy. I find a lot of value there, in particular when it draws on the Pragmatists but in some ways it goes too far. I am not arguing that something is coherence because people believe it to be true, at least unchallenged. My definition of coherence here relates to the idea of a sufficiency of evidence to allow us to progress and remember that progression in the complex domain is based on parallel safe to fail experiments. Coherence here does not mean that we know this particular path is correct. It means that this path is (to use a colloquialism) worth a punt.
Coherence from this perspective is linked to a method for gaining evidence, as much as providing an evidence based approach to validate the experiment. Tomorrow I will talk more about the tests for coherence that can be used (and it will not be exhaustive).
Cartoon from this wonderful source.