That isn't to say there isn't any logic to the design of great games but also something much more intuitive to their design decisions that doesn't follow known principals or science.
A framework like this does, however, help you make better artistic choices. It helps to identify why something isn’t having the impact you thought it should and gives you some insight on how to fix it. It also helps you to deconstruct other works and understand why they do or do not work.
which leads to
https://www.raphkoster.com/2014/01/15/a-vision-exercise/
and its critique counterpart:
https://www.raphkoster.com/2014/01/06/how-i-analyze-a-game/
and crucially, the observation in here that you can start with either end -- the experience you want or the systems -- but you gotta make them meet in the middle.
These days, game design is generally taught as "decide on your experience, and fit systems into that." But I favor being open to starting from either end, and also in general think that focusing so strongly on the experience has a LOT of dev pitfalls:"
https://www.raphkoster.com/games/presentations/rules-of-the-...
Lastly -- starting at this end is just as artistic as starting from a chord progression, a cool synth sound, a color palette, or a piece of wood with interesting grain. Just as with any other craft-centric view on things, it's fine to start at a formal or an experiential end -- both are artistic.
FWIW, I have an MFA. :)
This and the other scenarios you mention are deliberately created to make the player have fun. They are all engineered to manipulate the player’s emotions, the intention is to trigger dopamine and other neurological reactions. As I said, that doesn’t have to be a bad thing!
You don’t have to think about it in terms of chemical reactions, but artificially creating fun is the goal, if you boil it down.
You do get that dopamine hit when you achieve a goal in Shenzen.io, or even a self-directed goal in Shenmue, whether the designers thought that way or not.
As Raph Koster says, fun is linked to progression and learning.
Progression applies to self directed goals too (you’re setting yourself a series of minor goals when driving the forklift in Shenmue).
Ironically, motivation theory tells us that the intrinsic fun of doing undirected chores in Shenmue or mastering facts about the systems in Shenzen.io is stronger than the onslaught of mostly extrinsic rewards generated by Cookie Clicker. You had less fun playing that game, that’s one of many reasons why.