So I'm wondering if it's possible to do better, when unpacking the distinction between psychological and neurological forms of mental health problems, than:
a. Psychological problems are psychologically caused; neurological problems are neurologically caused.
The problem with this is surely that
b. Since all psychological factors are realised in the brain, and since there just isn't any substantial paramechanical psychological domain which causally impacts on the brain, and since talking therapy can affect brain function and neuroleptic therapy can affect psychological function, the very distinction - neurological / psychological - looks to be otiose.
It seems so very tempting to suppose that b. must be true - to suppose, say, that the very idea of distinguishing between neurological and psychological disorders reflects adherence to an unthinkingly dualistic metaphysics.
One salvaging move at this point might be to recast the distinction along structural / functional lines: neurological disturbances being disruptions in the 'hardware' of the brain, psychological disturbances being disturbances in the software being run on the neurologically intact system. That distinction might itself break down if we take too fine-grained a look at the hardware - i.e. at the level of the software implementation, but on the whole seems to work well enough. But still, construing the psychological / neurological distinciton in such terms might easily be seen more as a case of changing, rather than elucidating, the subject.
Despite the above problems, I want to make a case that the psychological / neurological distinction is well worth preserving. Why it should be preserved is, I think, an object lesson in the philosophy of psychiatry. The basic idea is that we do better to consider how these terms ('psychological', 'neurological') are actually used in clinical practice, rather than impose decontextualised meanings onto them from within philosophy's own metaphysical preoccupations. What I want to claim is that psychological disturbances are those mental disturbances which are a function of an individual's self-understanding; neurological disturbances are not.
This approach is metaphysically neutral about minds and brains, and about the ultimate causal origins of psychological disturbance (perhaps one particular psychological disturbance was originally caused by a one-off quirky mishap in the prefrontal cortex). What makes for its psychological character, however, is its having been caught up in the self-creating ongoing dynamics of a subject's self-understanding. Psychological disorders are, in this way, disturbances of inwardness: disturbances in aspects of those processes by which we become characters and steer ourselves around the interpersonal world.
A neurological disturbance is not first and foremost a disturbance in self-understanding. It may give rise to such disturbances, but the idea of the fundamental disturbance ought not to make mention of them.
Another way to try to make the distinction between psychological and neurological is on the basis of intelligibility: neurological disorders are intelligible only in the sense that the pattern of breakdown in a causal mechanism is intelligible given what we know about its mode of functioning. Psychological disorders, however, can be thought of as humanly understandable, in that kind of a way that someone's reasons for acting may be intelligible. Intelligible by being placed in the human order of rationality, that is, rather than merely in the natural order of causality.
In itself that seems just fine to me. What is perhaps less desirable is a commonplace way of spelling out what the basic ingredients of an intelligible response might be. The cognitive behaviour therapist, for example, tends to find distorted beliefs or thoughts underlying the psychological disorder. But whether the distorted forms of understanding are really captured, in anything other than a rather loose metaphorical way, through such notions as 'false or unhelpful beliefs', 'automatic thoughts', etc., is I believe doubtful; this is simply too intellectualist a form for the understanding to take. (The analyst's 'unconscious phantasies' seem closer to the mark, although this notion also risks conjuring up the idea of hidden occurrent imaginings, rather than of disturbed living dispositions.)
On the other hand it would not do to take too general a notion of disturbed understanding - to look, for example, for the kinds of 'understanding' that are contained in every unreflective perceptual act. For then, once again, we struggle to maintain our distinction between the neurological and the psychological. What gets lost here is what I believe is the essential reference to a disturbed self-understanding - a disturbed self-understanding which gets inscribed and reworked in the heart of the psychological disturbance.
I notice, to finish, that the notion of 'psychological disturbance' that I have arrived at is not very dissimilar to the old fashioned notion of a 'neurosis'. It occurs to me that this might not be such a bad thing, and I wonder whether that concept has simply gone 'underground' for a while, still shaping and unifying our conceptions of disorders from beneath whilst being temporarily erased from our reflective clinical consciousness.