The primary source of energy for the motion of the lithospheric plates
is dissipation of energy from the earth in the form of heat transfer.
This heat transfer causes convection within the mantle and this
convecting mantle material is hypothesised to exert a drag force on the
base of the lithosphere. This is known as Basal Drag and was
thought to be an important driver of tectonic plate motion until
recently when advances in geophysical imaging techniques (specifically
3D seismic tomography) allowed geophysicists to form a more detailed
model of the structure of the mantle but were unable to locate the
necessary large scale convections structures. Further to this,
geophysicists no longer believe that the asthenosphere has the necessary
stiffness or rigidity to cause sufficient friction on the base of the
lithosphere to be a significant driver of plate motion.
Gravity is also believed to play a role in the driving of plate motions, both at mid-ocean-ridges (MOR) and at subduction zones. Both these mechanisms are also linked to the dissipation of heat and convection in that they are reliant on variations in the buoyancy of the lithosphere as temperature changes occur as described below:
When hot (and so low density, buoyant) mantle material rises at an MOR, it is intruded into the pre-existing oceanic lithosphere as well as extruded at the upper surface. This newly intruded material is at a high temperature and cools relatively slowly. As such the newly formed oceanic lithosphere is buoyant and rises relative to the older cooler oceanic lithosphere further from the MOR. Even though this younger lithospheric crust is buoyant it is still significantly denser than the underlying upwelling hot mantle material and so "slides" down away from the crest in a process known as gravitational sliding. This is considered a secondary driving force as it only acts in close proximity to the ridge and does not transmit load into the surrounding lithosphere (it does not "push" the lithosphere away from the ridge, it is instead dragged - the tectonic regime is tensile, not compressive).
Another more significant gravitational driving force of plate motion is that caused by the old, cool and so low buoyancy oceanic crust that sinks back into the mantle at subduction zones and acts to drag the attached oceanic lithosphere towards the plate boundary. This is known as slab pull and is thought to be a significant driver of plate motions as it has been observed that tectonic plates with subduction boundaries tend to move with a higher velocity than those without. Subducting oceanic slabs also promote another driving force of plate motion known as trench or slab suction. This is a process where the movement of the downgoing slab promotes flow in the nearby mantle. This flowing mantle material is thought to exert a traction both on the downgoing slab and the overlying non subducting slab, pulling them both towards the subduction zone.
It should be noted that not all plates have significant subduction boundaries and yet still undergo plate movements so it is possible that a combination of all these mechanisms influences plate movement (as well as the possibility that there is some as yet undiscovered driving force that has a significant role).
The above is a simple summary of a complex topic and as has been stated the causal mechanisms that drive the motions of tectonic plates are not fully understood and as such the reader should remember that this is a current and active area of research in geophysics and so the hypotheses summarised above may be accepted, rejected or modified over time!
Content will be erased after question is completed.