Timothy O'Connor distinguishes two stages of cosmological arguments for theism:
1. The establishment stage, in which the existence of a being of a special kind, usually an independently and necessarily existing 'prime mover', is established.
2. The identification stage: the being whose existence has been established in the previous stage is identified with the God of traditional theism, or at least as having some of the attributes possessed by the God of traditional theism.
The establishment stage gets most of the attention in the literature on cosmological arguments. It has taken the form of various different arguments, the most well-known typically being deductive arguments in which one of the premises is a kind of Leibnizian 'principle of sufficient reason' (e.g Rowe 1968). In my view, the most persuasive route to the kind of being at hand is a kind of inference to the best explanation. Very briefly: take a dependent being to be one that depends on something outside of itself for its existence, and an independent being to be one that doesn't. If there are only dependent beings, we are left with an infinite regress of dependence, since each being relies on another being for its existence, and so on ad infinitum. But then, while each individual member of the regress has an explanation for its existence, there is no explanation for why dependent beings exist at all. If, on the other hand, we introduce an independent being to our ontology, we can have an explanation for why dependent beings exist at all, if it we posit that the independent being has the power to bring about dependent beings. Hence, the hypothesis of an independent being is more explanatory than the hypothesis of an infinite regress of dependent beings (Here I follow Ross Cameron 2008 and Ricki Bliss 2013 in the claim that the 'viciousness' of infinite regresses has to do with a lack of explanatory power).
Anyway, enough about the establishment stage. I came across a passage from Timothy O'Connor a few months ago, in which he suggests reasons to believe that the independent being arrived at in the establishment stage is a "necessary" or a "voluntary agent". To put it simply, voluntary agents actualize their potentialities through an act of the will, while necessary agents do so 'mechanically'. A panadol pill has a potentiality to relieve my headache. This potentiality exists by virtue of the pill's chemical structure, and is actualized by external triggers after I have swallowed the pill. I doesn't relieve my headache voluntarily. In contrast, I have a potentiality to take the pill, and this potentiality is actualized by my voluntarily taking the pill, i.e. through an act of the will. The pill is a necessary agent, I am a voluntary agent.
O'Connor writes that "the only available model we have of a necessary agent is that of an agent whose activity is in each case triggered by surrounding circumstances, and as such, is always part of a chain of events". Necessary agents, as in the case of the panadol pill, always seem to actualize their potentialities as a result of some external trigger. But this cannot be true of the creative activity of the independent being, since there would be no other beings to somehow trigger its potentiality to create, or to sustain things into existence.
Voluntary agents, on the other hand, need not act in the foregoing manner, not at least if they are free agents. O'Connor writes elsewhere that to be act freely is to "be an ultimate source of my activity" (2008, 121). Though a free action may be influenced by external stimuli (e.g. perceived states of affairs), it cannot be fully or sufficiently caused by such stimuli. O'Connor is well-known for his advancement of this agent causalist conception of free will. He states in the passage under consideration that the hypothesis that the independent being is a free voluntary agent, rather than a necessary one, is much more plausible and intelligible.
O'Connor doesn't pursue this line of thought much further, but I think it could be developed into an explanatory argument for the hypothesis that the 'prime mover' arrived at through the establishment stage has free will. If the independent being is a necessary agent, we have no explanation as to how it can actualize its potentiality to create or sustain dependent beings without this being sufficiently caused by some kind of external stimulus. If on the other hand the independent being is a free voluntary agent, we do have an explanation as to how it does so, since it does so out of a free act of the will, which by definition cannot be sufficiently caused by some external stimulus. Hence, the hypothesis of a free voluntary independent being is more explanatory than that of one who is a necessary agent.
This would go some way to meeting the requirements of the identification stage of cosmological arguments for theism. Free will is one of the attributes that the God of traditional theism is said to have. Moreover, possessing free will would seem to entail having conscious states, or at the very least a mind, since acts of the will necessarily are acts of the mind (and, I would add, conscious ones). An independently existing, free conscious Self does start to look quite a bit like the 'God' that we are all familiar with, at least in the West.
Another agent causalist, Roderick Chisholm, had already noticed the connection between agent causal free will and theism:
" If we are responsible, and if what I have been trying to say is true, then we have a prerogative which some would attribute only to God: each of us, when we act, is a prime mover unmoved. In doing what we do, we cause certain events to happen, and nothing — or no one — causes us to cause those events to happen."
If we have the relevant kind of free will, we are like little prime movers, causing events without being sufficiently caused to do so by things outside of ourselves, even if such things can influence our free actions, and hence perhaps be said to partially cause such actions.
It is important to note, though, that one doesn't need to believe that we actually have this sort of free for the above argument to work. Nor does one need to believe that having agent causal free will is the only meaningful sense in which an agent can be said to be free. All that it needed is for the idea of agent causal free will to be a coherent one. In that case, it will be open for someone who denies that we have agent causal free will (and perhaps affirms that we have a more diminished, say, compatibilist free will) to nevertheless agree that if the prime mover has agent causal free will, this would explain why it is capable of doing its job.
Chisholm, R. (1965). Human Freedom and the Self.
O'Connor, T. (1995). From First Efficient Cause to God: Scotus on the Identification Stage of the Cosmological Argument.