[jacques_chester]

Bodies aren't finite state machines and the concept of "modes" can be misleading. What you have is a (very large) number of independent processes that have differing amounts of flow under different conditions.

So, for example, you don't click from building muscle to consuming muscle. What happens is that catabolism has a higher flow than anabolism, so the net balance is negative.

As the body becomes leaner in starvation conditions, processes that recover energy from gluconeogenesis or from lipolysis cannot produce "enough" energy. So feedback loops that suppress the breakdown of organ tissue weaken and eventually that loop shifts.

[jacques_chester]

Clarifying some of what I wrote above --

* "independent processes" is better described as "interdependent, oftentimes antagonistic processes".

* In describing net balance, the key is that both processes remain active throughout, even if only the net balance of flow can be observed. Or more likely, the integration of the net flow over time (the stock).

* I say "that loop shifts", more accurately the loop dominance shifts. This is a term from system dynamics for describing how the behaviour of the overall system can suddenly change due to a change in flow of one or more loops that are working antagonistically.