Probably the closest analogy possible is computer repair - due to the close interaction of hardware and software, troubleshooting for problems is often difficult. And software being a somewhat emergent property of the machine, tracing the hardware interactions that cause software to err are difficult to work through. Not to mention that the brain's properties are to a large extent emergent from its structure, which we hardly understand well enough. I'd say that the correct analogy would be to convince people that their computer's GPU has gone bust due to some untraceable power problems, and to live with that till the computer gets junked. Still a shitty situation, but the lack of options and clarity on the repairman's part is more clear.
Nah, while I would love to adopt this analogy in part due to my desire to be a cybernetic life form, I disagree that it's a reflection of lack of knowledge but rather an ignorance of knowledge (and consensus) that varies from individual to individual due to heuristic issues in the field. That's why I used another medical analogy, because there is not a real political incentive influencing the model adopted for - and the science done for - most basic computer problems. When we talk about low blood pressure and volume in the context of massive haemorrhage, it's not the same thing as talking about it in the context of other ailments, even though the physiological mechanisms can be similar. 'Chemical imbalance' almost becomes a layer of abstraction because it ignores what we do understand about developmental impact on children and about the brain's responses to external stimuli, as well as serious social problems that contribute to alterations in psychology, and focuses on the effect in a way that essentially obfuscates the issues.
If you overheat your computer, it has protective mechanisms on multiple levels designed to shut itself down before it's irreversibly damaged. Now if your laptop isn't properly ventilated and it shuts down, to many end users it's just crashing, and a nuisance. But it's different to recurrent crashing due to corrupting memory, of course. Depression is extremely well-studied and we can say heuristically as not end users that we understand there are many issues with attributing it to even one physiological cause, let alone one cause at all. What's semi mystical is some of the details and the sheer variety of mechanisms, which then resist treatment.
Additionally, much of the evidence for the different biochemical models of depression comes from studying how treatment of it works, which has very little to do with the cause of depression and only to do with the pathophysiology of it. There are other physiological issues where treating, say, serotonin dysregulation may not make sense, and many people show no response or only get worse. And thinking more broadly, someone who is in an exploitative social situation but has no other issues may develop depression, and these problems may get worse from depressed behaviour: self-isolation, ideational kindling, addictions, anhedonia. All these factors have been well-studied, too. And then of course it is absurd to act like that person is not on some level developing predictably and to attribute it simply to the imbalance that arises is to obfuscate the links to the treatable maladaptations and the causes. And there is real incentive to ignore what these typical causes are and the populations that are at risk or predisposed (which again makes it sound like something that arises from the patient as an individual rather than a manifestation of the experiences of a social condition - dangerous when dealing with psychological illnesses and destigmatisation).
We understand the developmental or lifestyle role on some level for most diseases; the brain is a very complex organ and consciousness itself is not greatly-understood but this is actually a form of mind-body duality in itself, to focus on it separately from the systems above it. And while most people will agree that there is a social or environmental component to depression - in fact they will often batter people with the idea that mental illness isn't a real thing that happens in the brain because you can't see it - they downplay it or misunderstand that it's not necessarily an individualistic component. And people end up thinking it's their brains that are majorly predisposed to being 'imbalanced' as opposed to it being imbalanced by the complex trauma and stress of their situation, especially since we see results (functioning, clinical outcomes, etc.) as so important.
And that is why therapy is usually recommended along with drugs, only many of the dominant frameworks in therapy take similarly obfuscating approaches. For many situations they can only help you cope in order to try to solve your own problems or deal with the symptoms of your mental illness precisely because therapy is typically individual-centred. And the failings of this model become really evident to me because I can see where it has succeeded for me and also where it will just never be able to do anything because I am, for example, dealing with economic, social, and colonial subjugation. Is it correct to teach people self-blame in order to cope with being alienated from their labour (not a rhetorical question)? etc.*
And that is why my analogy is sound imo. If we could just replace brains or otherwise have a more comprehensive way of treating depression it might be different, although it might have different ethical issues, like the fact that being able to treat an issue almost certainly doesn't justify inflicting it.
* in writing this I have to mention that I spent years attempting to get DBT, and while my therapist inexplicably stopped giving me DBT in exchange for random talk therapy so I need to sort that out, it's got a better basis
Sorry for the long post lol