It seems to me that on the pragmatist theory of truth, which says basically that some belief is true if and only if it is useful given our intentions, one might be tempted to simply subscribe to a form of behaviourism which construes all beliefs as protocols. This is essentially a reductionism of the rational faculty’s apprehension to behaviour. Thus the belief that one and one is two is not true of the world itself, but rather is true with respect to my goals insofar as having the attitude that one and one is two will help me negotiate my way to my preferred ends as an organism.
However, consider how those who advocate for a pragmatist theory of truth, and who subscribe to behaviourism with respect to beliefs, have managed, albeit not to our satisfaction, to explain the appearance of ‘reason’ in man without the need for any rational intuition. As such, it seems that the theologian might take this model as a fitting epistemology of animals (sensitive souls). The animals have ‘beliefs’ in precisely this sense, and they have no rational intuition, and thus their beliefs do not have abstract content, as ours do.
This might have interesting implications if we subscribe to theistic evolution, and if we agree with Kant that 1) Teleological causes qualify as transcendental laws of reason/nature, and 2) Nature as a super-organism has a final end which is simply moral. In the first case, we want to know how the brain-structures belonging to physical species antecedent to our own relate to our own (might we have non-rational ‘beliefs’ of this kind as well). In the second case, we might want to know how the structure of our brains, allowing for ‘behaviourist’ type beliefs, might relate to the whole system of nature when considered as an organism (or super-organism, as Kant calls it in The Critique of The Power of Judgment) whose causes and effects are reciprocally related (the effects themselves cause their causes, etc.).