Maneuver Demand vs. Surface Position
Let me try another approach with slightly different terminology.
In the simplest terms, control systems can be classified into two groups:
1. Surface Position Control
2. Maneuver Demand Control
With the former there is a direct gearing from pilot controller input to surface position. This gearing may be non-linear. An example is roll control where usually small commands move just wing trailing edge surfaces while larger commands bring in spoilers as well.
With the later, the pilot controller input is treated as a request for a certain amount of airplane response (e.g., load factor, pitch rate, angle of attack, roll rate, yaw rate, sideslip angle). With this type of system the pilot's maneuver command is compared against measured airplane response and the surfaces are adjusted as needed to achieve that maneuver command. Usually such arrangements involve integral control such that the error between to maneuver commanded and the measured response is driven to zero long-term.
Both types of control systems can be implemented using FBW. It's a matter of how the control algorithm within the FBW equipment is defined.
When designing a maneuver command system, it is important to take into account what maneuvers the airplane is capable in each phase of flight. For instance, it makes no sense to have a pitch rate maneuver command system when the airplane is on the ground at low speed such that no amount of elevator or stabilizer will generate pitch rate.
It is for these reasons that airplanes employing maneuver command control modes when in-air must transition to surface command control modes on ground. That is what the chart provided above is all about.