This is a hypothesis that recently came to my mind but I was not able to quickly prove or reject it with a Google search.
=> Body-brain mass is highly correlated on a logarithmic scale
=> Bigger bodies have a higher inertia than smaller ones
=> Bigger brains have potentially more neurons and on average longer conduction delays
My question is: Is one reason for the correlation between body- and brain mass that more neurons and longer conduction delays are needed to computationally incorporate the inertia of mass during motor sequences (which results in temporally extended sequences?) More neurons, for example, could be needed to implement a finer acceleration and deceleration of body parts. Longer conduction delays could contribute to a better timing in learning motor sequences in bigger bodies.
If this is true, this relationship would point to a very general principle of learning and encoding motor sequences that would require an evolutionary perspective to understand certain brain functions.
And yes, of course I know that there are other factors as well determining the body-brain mass ratio :)