Feedback

November 19th, 2022

This morning, Abdenour Bennani, one of the sharpest young practitioners I have met, asked some penetrating questions about the nature of feedback in storyworlds. 

"A question keeps bothering me while working on something inspired by LMD: Is there a way to let the player know that their choices are making an impact and provide feedback during play?

Do you think that this is even a real issue? Is it something players are complaining about a lot?

It seems to me that LMD is a paradigm shift in interactive storytelling. We turned from thinking in terms of the player changing in-game variables like relationships to him changing how well he understood the meaning of the storyworld (you may say changing the structure of the story either to one that fits the meaning or ones that don't). So players don't actually change anything inside of the game, be it relationships or things. As a result, I'm not sure how to make it clear that their choices are making an impact. Even if I changed the course of the story based on the global variable, how would the player know that he caused the change?

I thought of some ways to let players know that they're on the right track or not. For example they might have sleeping dreams that hint to that, or psychological issues like feeling guilty or depressed, or God or any all-seeing power talks to them, stuff like that. These things would happen at any time during play, triggered by the state of the global variable. But what's the point of letting the player know if he's on the right track if the ending serves the same purpose, and failing the entire thing is part of understanding the storyworld?

I don't know, just some thoughts. What do you think?"

I found the questions so challenging that I gave the matter some thought, and posted this response:

You raise one of the central issues in the design of storyworlds. We have always thought of them as rather like "games with drama". This in turn led to a deep architecture just like games: the player faces a series of challenges requiring them to overcome one challenge before proceeding to the next challenge. Each challenge, then, had to have some kind of dramatic import.

One of the fundamental paradigm shifts in the encounter system is the change from a sequence of challenges to a sequence of short stories. Most stories present some kind of dramatic challenge, but some serve only for contextual exposition. This shift changes the nature of the challenge. Game challenges are mostly boolean, life-or-death challenges. If you fail to steal the vorpal blade from the merchant, then in a later challenge you will fail to kill the purple dragon. Because each challenge is boolean in nature, victory is achieved only by success in each and every challenge. Victory = S1 & S2 & S3 & S4...

This provides the player with clear feedback. Fail on S2 and you have lost. Succeed on S2 and you have made a positive step forward.

But stories aren't a series of black-and-white challenges. When Obi-Wan teaches Luke about the force as they travel to Alderaan, Luke makes a tiny step into a larger world. If that scene had never happened, it wouldn't have ruined the story. It's one in a series of steps that ultimately lead to Luke heeding the advice "Use the Force, Luke!" THAT is the black-and-white decision that Luke makes -- and it is the ONLY truly black-and-white, outcome-determining decision that Luke makes. Luke's trust in Obi-Wan and the Force is slowly built up from "These aren't the droids you're looking for" through the fight in the bar and many other small steps. 

Thus, many stories follow a course in which a long series of small changes leads up to a profound change in a character. Luke transforms from a boy into a man and makes a hugely risky but truly manly decision.

But he didn't get much in the way of feedback as he traveled this path. Obi-Wan did congratulate Luke at one point that he had made a step into a larger world. Otherwise, Luke got no feedback on his decisions. He just blundered through them as best he could.

Because the changes are small, direct feedback is inappropriate. Congratulating the player for saying something nice is insulting. Many years ago I taught a course on game design, and when I graded the students' papers, I would stick colored stars onto the papers, just as our first-grade teachers might do. The students didn't get the joke -- they were insulted. Feedback that's too blatant is insulting. And a story isn't an exam.

Another aspect of this shift is the emphasis on intuition over calculation. Game players have developed an acute sense of "game calculation": a nose for sniffing out the requirements of the game and figuring out how to optimize their performance. That's fine for games, but stories aren't games. A story is not experienced as a series of levels; it is absorbed in its entirety. The audience does not calculate the significance of Luke's training with the light saber; instead, it builds its internal representation of Luke's personality as it goes. The storyteller's goal is to make the big transformation in the crisis of the story plausible and even inevitable.

I admit that some feedback would be of value, and I admit that LMD falls short here. That feedback, however, should never be specific to any of the player's decisions; rather, it should be strategic in style.

WRONG: "You alienated Joe when you yelled at him."

RIGHT: "You've been pretty harsh with people for the last few days."

In other words, engage the player at a higher level of abstraction. That's the difference between interactive storytelling and games.