Thrust vectoring is used to extend the manoeuvrability of aircraft outside a `normal' flight envelope, by providing (optionally differential) vectored engine thrust, as explored in many previous questions (this and this).
How is the direction of vectoring actually controlled by the pilot during flight? Does a flight control computer automatically interpret "traditional" control surface inputs under extreme conditions and move the nozzles appropriately to create the desired effect? Or is vector direction a separately controllable interface?
I've noticed in my flight sim of choice, XPlane, that the configuration panel allows the mapping of an analogue axis on my joystick/pedal setup to 1D vectored thrust. I've no idea if this is representative of any real fast jets, but if it is, it seems that this approach could create a lot of extra "pilot workload" -- given that it's another one (to three!) control axes to be manipulated -- and manipulated at the same time as pulling potentially very aerobatic manoeuvres with little room for mistakes. This seems like a Bad Idea.
On the other hand, it might be the case that independently controlled vectored thrust allows for a wider range of aircraft manoeuvres than that which could traditionally be 'expressed' by control surface movements alone. Giving fast jet pilots the extra workload would, therefore, be worth the risk.