I think I've found my solution!
$$(-\frac{1}{4}\frac{\partial^2}{\partial \alpha^2}+\alpha^2 - g^2 \alpha^{2 \cdot 2} + \frac{1}{4} \frac{\partial^2}{\partial x^2}) \Psi(\alpha, x)=0$$
where $$x$$ here is our matter field in minisuperspace. The equation can be solved by a seperation of variables $$\Psi(\alpha, x)= \psi_{\alpha}(\alpha)\psi_x(x)$$ to give two coupled equations:
$$(-\frac{1}{4}\frac{\partial^2}{\partial \alpha^2}+\alpha^2 - g^2 \alpha^{2 \cdot 2})\psi_{\alpha}(\alpha)=E\psi_{\alpha}(\alpha)$$
$$(\frac{1}{4}\frac{\partial^2}{\partial x^2})\psi_x(x)=E \psi_x(x)$$
The definition of time now can be given as one of two solutions $$(\Psi_1,\Psi_2)$$ - to describe our time, we have also two choices, we can define our time as either the scale factor $$t= \alpha$$ or as the matter field $$\tau= x$$ which I have represented with a different description to identify the two $$\Psi(t, \tau)$$ - these two different descriptions could be given a unique transformation. We can view one trivially as an imaginary time dimension, by a wick rotation, and one as a real time, given by our Hamiltonian.
But the real question is which time reference do we make real and which one imaginary? Interestingly, we would run into all sorts of problems if we performed the wick rotation on the matter field, namely, imaginary mass descriptions. Performing the wick rotation on the scale factor will rid us of our matter field because real time calculations would have $$\tau=x$$ vanish due to the WDW equation. Though, we can simply say ''well, no use with that any more'' and resort to the final solution of an imaginary time reference on the theory.
http://arxiv.org/PS_cache/hep-th/pdf/9503/9503073v2.pdf