Chemical imbalance is what is called a working hypothesis. In absence of any proper knowledge, a working hypothesis can enable further research by reducing the subject to well understood components - even if those components ultimately fail to provide any useful explanations at all. A working hypothesis is rather understood to be essentially garbage, although it can technically still be or considered to be true.
Imagine a Windows 95 ATM machine was sent back 200 years into the past and studied by researchers of the time. After a while, naturally its integrated circuits start to break down due to its age. Also the PC would slow down considerably with time due to malware and disk fragmentation. The program will freeze quite often but very unpredictably and randomly so. It might even blue-screen and it will sometimes consume a lot of power and produce a lot of heat. The scientists are locked out of the operating system, so they can't study the PC from there. All that they really do understand at the time is voltage, which they can measure between the components. In fact they even discovered that raising the 3.3V rails from the PSU to 3.6V does somewhat improve the slow programs, but it will also cause more system freezes and more air will be pumped through the case by the fans. Likewise they discovered that lowering the 3.3V rail to 3V improves system freezes and less air will be "consumed". But putting it even lower causes crashes and slowdowns at the same time. Also the broken down ICs will function best in a temperature range of 50-70°C, but the scientists don't know that. They only observe that the PC runs more stable if the case vents are plugged shut - at least for the duration of their studies that is. When the PC freezes or displays other changes in behavior, the voltages they can measure all around the motherboard also change in a multitude of ways.
Forced to draw a conclusion, the scientists propose that the PC suffers from a voltage imbalance and furthermore that excessive consumption of oxygen from the air by the components causes ill effects through redox reactions. They conclude that it is the best course of action to raise the voltage to 3.6V and to partially tape the vents of the case shut.
This is neuroscience and psychiatric medicine in a nutshell.