All characters in fiction are "along for the ride" - I don't see what makes you think Little Red Riding Hood is in a special category.
It's mostly a question of the logic used to drive action. You compare Red's actions with, say, the actions of Christopher McCandless. Red's a character without much agency in her fairy tale, so maybe Jack is a more useful point of analogy.
As Jack, with fairy tale logic, I can't create a giant beanstalk farm or feed my family on giant beans. I can't plant one magic bean and sell the rest to the highest bidder. I can't sell tickets to see the World's Biggest Beanstalk. This is fairy tale logic, so I am expected to act in accordance with the story's goals. The meta-textual awareness of what is "supposed" to happen is driving my actions. We're telling a story about an impetuous young lad and his travel through the magical world, not about how Jack's family never went hungry again because they ate giant beans the rest of their lives. If I defy the intended narrative, I'm ruining the mood (and the DM probably says the beanstalk never sprouts because it's a damn magic beanstalk and stop trying to wreck my adventure). Wonder and enchantment and, well, you can't
use that information. That'd wreck the wonder and enchantment of it.
So as Jack, playing with fairy-tale logic, I should act first and foremost in accordance with my knowledge about what this story is
supposed to be about, and all other concerns are secondary, including whether or not giant beanstalks and flying castles in the clouds "make sense."
As Jack, with naturalistic logic, I'm freer to determine what the story is
about myself, in play, because the world reacts in plausible ways to my interaction with it. I'm an impetuous young lad in a magical world, sure (the giant beanstalk is a supernatural thing!), but I don't need to worry about the meaning of my actions in a narrative. There's no metatextual goal here, only the goals that are internal to my character and how I, as a player, go about realizing them given the props at my disposal. This is naturalistic logic, so how my character reacts to the giant beanstalk isn't expected. It's an interactive object. I react to it however I want. Maybe I climb it. Maybe I try and grow a whole field of 'em. Maybe I go looking for the wandering old man to interrogate him with my fists about why he gave me these magic beans. These are all valid options for moving forward and all of them create very different stories. There's nothing I can do that ruins the mood because the mood isn't a goal set out at the beginning of play, it's something that emerges over the course of play. There's no adventure to wreck, just a bunch of levers to pull. Because the natural world isn't inherently strongly narrative, naturalistic logic cares first and foremost about nouns and verbs and only later about how those nouns and verbs combine to make a feeling. Which is the weakness of naturalistic logic - it's not likely to result in wonder and awe and a "magical" world, by itself. Any magic sufficiently normalized so as to be indistinct from the world's logic is just an object.
So as Jack, acting with naturalistic logic, I should ask myself "how do I pursue my goals with the objects around me?" Making sense is important for that - I need to be able to reasonably guess that these giant beanstalks might produce some sort of bean and that maybe there is a market for such beans before I try and become Jack the Giant Bean Merchant. So if the beanstalk develops pods that give birth to human babies instead of giant beans, well...that's information I can put to use in some way! (Maybe that mysterious old man was a Beanspawn...can I find a druid? Don't they talk to plants?)
When Little Red Riding Hood sees that grandma has big teeth, that is information she can use. When the woodsman tracks the wolf to Grandma's house, he can (as he does!) come in and kill it.
There are consequences for actions in Little Red Riding Hood - by disobeying her mother, Little Red Riding Hood takes a risk! By cutting open the wolf, the woodsman is able to rescue the swallowed victims. (This is a pretty standard D&D trope.)
In the game that [MENTION=6696971]Manbearcat[/MENTION] ran, my Knight Commander of the Iron Tower was able to heal the injured and rouse the frightened by speaking words of prayer and encouragement.
Just because consequences don't have a naturalistic causal logic doesn't mean that players can't engage with them, by activatig the game's mechanics.
It does mean it's not information you can use in other contexts. You can heal the wounded with prayer and encouragement, but can you do that all the time, or just in this particular instance? Is that information you can use to then bring the Army of the Iron Tower up against some other army in the world and defeat them because you have this magical divine healing ability and most other people don't? Or do all commanders in this world magically heal their troops with words of prayer and encouragement? Can you restore sight to someone whose eyes have been plucked out? Can you regrow limbs?
Or does none of that matter because those options are not even really on the table for your character, because your logic isn't really naturalistic?
When Jack's thinking about the world naturalistically, he's thinking that this giant beanstalk probably works like other beanstalks, more or less, and that he can use it to do what other beanstalks might do (like produce beans!).
The second bit - plus the fact that adjudication, even when it doesn't depend upon secret backstory, will still depend upon a GM's interpretation and application of naturalistic causal logic - tends (in my view) to somewhat negate the first bit.
The DM makes the interpretation, but the naturalistic and causal logic is shared. The DM and the player both agree that it's reasonable to expect the giant beanstalk to more or less behave like other beanstalks and so it's reasonable for Jack to try to farm its beans and when he farms human infants instead, both agree that this is a surprising result in the naturalistic logic. They don't just say, "Welp, *magic*," since that's not naturalistic logic.
Which is, again, why there's a benefit for both. Sometimes, in D&D, "Welp, *magic" keeps things interesting.
Too much, though, and you can't rely on a reasonable confidence of what the outcome of your actions will be.
Elaborating on my reply to Aenghus, I don't think these are very similar. D&D, and many other RPGs that emulate its approach to action resolution pretty closely, give the players a very clear range of options for responding to "a goblin approaches". There are rules for reactions (with CHA mods and/or Diplomacy-type skills); rules for closing the distance or running away; rules for shooting it with arrows; etc.
But (outside of 4e's skill challenges) there aren't rules for scoping out or setting fire to a field.
Don't need rules. Naturalistic logic. When thinking with naturalistic logic, burning fields is a thing you can do, just like swingin' a sword is a thing you can do.
Well obviously if the GM just says yes to every player action declaration then there won't be any issues - but that will be the case in any situation. Naturalism doesn't make it more likely.
It does mean that the job of the DM here isn't to decide what is possible, though. Using naturalistic logic,
players decide how to use the props at their disposal. The DM's job is interpretive, but under that logic, everyone at the table agrees that these are possible and permissible and viable actions. You realize you, uniquely in the world, can heal the wounded with your prayers so you give up your position as the knight-comamnder and go around healing leprosy because your heroic character is Good. Jack makes a deal with the giant to feed him humans and beanstalking becomes the favored method of execution for criminals in Jack's homeland. Yes, okay, this is all fine.
But as soon as there is adjudication, I think it's a different story. Saying there won't be much in the way of fiat doesn't seem plausible to me. For instance - have you ever started a grassfire? Or tried to put one out? I can answer yes to both questions (the events took place in the same order in which I frame the questions), but I wouldn't be confident in generalising my experience to a wheatfield in what is probalby a less arid environment than where I had my personal experience.
I remember in a tournament game years ago when our PCs were trapped in a space-base with a fire, and we started planning actions on the assumption that we had at least minutes to go before our oxygen supplies were threatened, and the GM had us asphyxiating within seconds. The GM thought he was reasoning naturalistically, not by fiat - but the chemical engineer in our group didn't agree!
But, things will burn. The plants won't cry out in mournful tones. You won't open a portal to the plane of elemental fire or piss off a Dryad. You've got a reasonable expectation about what might happen. This doesn't mean you have precise and exact knowledge, but you don't need it - naturalistic logic informs your plausible possiblity-space.
A D&D character is going to have a different "naturalistic" possibility space (that may very well include dryads and mournful plants), but they still can use the information that the elves have fields to determine their actions.
Compare the handling of the wilderness in
Firewatch with the handling of the wilderness in
The Long Dark.
In
Firewatch, the function of the wilderness is as a narrative device itself, and it invites you to ask what the wilderness means to the story. It's a fairy-tale logic (despite being grounded in a very realistic setting): your interactions with the wilderness are driven primarily by the broader metatextual meaning of those actions and that wilderness.
In
The Long Dark, the function of the wilderness is as a stage filled with props. The wilderness has a meaning in the (supernatural) story, but the logic in how one interacts with it is very naturalistic (because that's the play loop).
In the former case, there is a story to be told and your agency is as the secondary protagonist
of a story. The latter case, your agency is
as a being in that world.
