Reminds me of Soda Play / Soda Constructor from way back

https://en.m.wikipedia.org/wiki/Soda_Constructor

> Potential energy minimization is one of the most basic forms of goal-directed behavior.

I'm confused by this statement. Potential energy minimization happens when my errant soccer ball rolls down a hill, but this seems not at all goal-directed. In this sentence the author has a link to a timestamp in a >2h youtube video, in which this opinion is expressed but I think not clearly justified. To me, this seems to obscure the difference between inert physical systems like my rolling soccer ball and systems capable of having goals (e.g. an animal walking presumably towards something). Further, I think it neglects that potential energy may be part of a goal-seeking strategy, like a bird of prey hanging out really high so it can dive quickly onto an unsuspecting rodent.

> Actuation mechanisms are modeled with an action-dependent energy function

And because the framework being described is intentionally flexible, I think maybe we need to be careful about distinguishing between a simulation of physical energy, and energy as a metaphorical way that control or influence is exerted on a system.

So why does the little guy gallop forward when you turn the brain on?

This is neat, it runs quite well. If you grab one of the middle dots and hold it just right, it's possible to get it stuck in a locked state