Externalists typically hold a strong interpretation of the truth connection. They hold that there must be a logical connection between justification and truth, that somehow or in some way, whatever factors make for justification must also make for objective likelihood of truth.
Internalists could subjectivize this requirement and hold that whatever factors make for justification must also make for subjective likelihood of truth, but they typically don’t go this route. Here the paradigms for internalism I take to be Chisholm and Lehrer, since they don’t endorse a view that they admit entails the strongest version of skepticism as does BonJour. These paradigm internalists don’t endorse the objective truth connection nor do they endorse the subjective version either. Chisholm wrote about the hope that our justified beliefs would be true (and Keith has endorsed a similar idea at least in conversation), and sometimes talked of iterational principles, such as when he claimed that justification doesn’t imply likelihood or probability of truth, but at most the likelihood or probability that our beliefs are likely to be or probably true.
There is now, however, a large body of literature on the connection between coherence and truth, expecting an increase in coherence to be correlated, at least under certain circumstances, with an increase of probability. To the extent that a coherentist takes a cue from Chisholm, as Lehrer clearly does, I don’t see why an internalist coherentist would find this assumption plausible. Instead, I think what one should have expected to see on the formal side is an investigation of the question of the conditions under which coherence is confirmation-inducing. In this respect, one would have expected formal investigations more like the nice paper by Luca Moretti with the fitting title “Ways in which Coherence is Confirmation Conducive”, rather than the truth conducive investigations that I have seen to this point.