I am having a confusing issue with a simulation study and the application of gravity. It seems to me like when gravity is applied to a solidworks simulation that basically the software is applying a point load at the center of gravity of the assembly. But this doesn't really reflect what would happen in reality. 

Here is a very simplified example that demonstrates the problem I first observed on a more complex model.

The setup is simply two stacked blocks and the only external load is gravity:

What I have done to test my theory is vary the density of the bottom block (did not modify any other material properties). When I make the bottom block much denser, I notice that the stresses in the top block actually change. I do not think that this would happen in reality. 

I believe that the stress in the top block will not change as the density of the bottom block changes. Am I missing something?

In summary, I do not think that the mass of the bottom block should impact the stress induced in the top block, but I am seeing that happen in my study.