Ludonarrative dissonance and harmony in visual novels

[Kailoto]

So today I have an interesting topic that's a little fickle when it comes to visual novel design. It is, as you can judge by the title, about ludonarrative.

For those that don't know, ludonarrative dissonance is a term coined by game developer Clint Hocking to describe a disconnect between a game's mechanics and narrative. For example, if a game is trying to express a theme of "valuing human life" through its narrative, yet the gameplay involves lots of combat and gives you experience and items for killing opponents, the narrative and mechanics conflict. One is telling you directly that taking human life is bad, while the other indirectly condones it by rewarding you for each kill.

The term itself comes from "ludology" or the study of games, and narrative. So "ludonarrative dissonance" means literally, "dissonance of gameplay and narrative."

Lots of games have problems with this - at least half of the critically-acclaimed games I've played in recent years fall prey to this. However, not all games are affected. Ludonarrative dissonance, by definition, requires the presence of both mechanics and narrative - take either one away and no conflict can ensue. On one end of the spectrum, games like Mario or Pac-Man have negligible narrative, so there's nothing for their mechanics to conflict with.

Near the other side of the spectrum, of course, is your typical VN. Kinetic novels exist at the farthest extreme, discarding the ludic aspect entirely, but most VNs do in fact have a mechanic, and it is the mechanic of choice. This is still a very, very minor mechanic, but it is there; and because it is there we must address it.

Just a little bit of thinking will reveal that visual novels are rarely at a risk for ludonarrative dissonance. There are very few cases where a conflict would arise - mostly in cases where your story has themes of determinism or lack of choice - but by and large, it's not something VN writers have to worry about. If anything, the existence of save scumming is probably more dangerous than the choices themselves: if you're trying to tell a story about facing the consequences of your actions, the player's ability to go back and choose a different option without much trouble poses a (still minor) threat to the cohesiveness of your narrative.

But every coin has two sides, and just as mechanics and narrative can conflict with one another, so too can they complement each other. It's something that I call ludonarrative harmony: situations where the mechanics and narrative not only stay out of each other's way, but actively enforce the themes expressed. To go back to the previous example, ludonarrative harmony would be the same theme of preserving human life, but paired with mechanics that reward you for not killing people and showing mercy.

So can visual novels provide ludonarrative harmony? Again, only in specific instances, but they certainly can. Imagine that dissonant VN about determinism, except this time, every choice you make leads to the same result. Suddenly, dissonance becomes harmony. Or, say you make that consequence-heavy game, except you add in persistent variables so that they player can't cheat their fate. What was once a crippling weakness transforms into a powerful strength.

I thought I'd bring this topic up because hey, it's been a while since I've posted anything lengthy, I find it a terribly interesting facet of interactive fiction, and I'm curious as to what ideas you guys might come up with. I already have a few story concepts that take advantage of this, but the sky's the limit with this one. If you come up with a unique spin on ludonarrative harmony, or simply have an interesting observation, I'd love to hear it.

[Darkmoonfire]

This made me think of the most recent episode in My Candy Love (episode 27). So My Candy Love is a fluffy otome game, and in this episode a new girl starts attending the same school as the main character, and she's supposed to be confident and mature and get along well with all the guys so that the player character starts to feel insecure and threatened by her. One of the ways they tried to make the player feel threatened by her so the player's feeling would match the MC is that in this episode all the pictures you can collect are of the new girl with the guys, instead of the guys with the MC like normal. She was literally usurping the MC as far as the image collecting element of the game was concerned. This would be an example of ludonarrative harmony.

Except that the pictures in this game are usually used as a reward for the player when they figure out what the best choices to make are. The game even gives you a littler alert saying "Oh, no, you failed to get the image" whenever you fail to get one. So in this episode it punished you for making the competent choices by not giving images. For example, at one point one of the guys looses his notebook, and you can find it and give it back to him. Doing this makes you miss the picture. But because the picture is of this other girl with the guy, getting the picture is a punishment of sorts and by being competent I was rewarded by not getting that. Except that getting a picture is always a reward in My Candy Love as far as the gameplay is concerned, and I wanted those pictures. This caused a dissonance that really bothered me when I played the episode.

It's kind of interesting because the CG Gallery isn't normally what I think of when I think of gameplay in a VN.

[E-night]

Oh, this is interesting. (And a subject I am interested in too.).

I think that killing is the most common culprit for dissonace. Only very few games dares to say, killing is okay, yet the protagonist (potentionally) kill thousands of people. (This can be a problem in movies too, btw.). Only very few games adress the dissonance, and those who do usually have it as a if not the major plot point.

That said, I do think that VN's have a easier time here than normal games and movies, because (just like with books) they can show the viewpoints characters inner thoughts easily, which makes all the killing easier to have an effect on the protagonist. (Of course that also mean that it sticks even more out when it doesn't).

I do think that VN's can easily fall prey to ludonarrative dissonace, though.
You can have choices that runs against the games themes. Let's say that you have a narrative about standing up for yourself, but all you can do is ask person X, Y and Z for help).

You can have a narrative that constantly stress that every choice matter and actions have consequence (something that is always dangerous to say) yet all choice simply leads to minor variations untill the big ending choice decide the ending.

And you can have a choice function mechanically in a way that makes the narrative just stop making sense. Forexample many VN's function by delaying the consequence (typically by racking up points). This can sometimes lead to that you get the bad ending where the world blows up just because your mc didn't eat their cereal for breakfast and thus is one world-not-blowing up - point short. There is very few narratives where there would be a logical chain between not eating cereal and the world blowing up.

[Parataxis]

I have been thinking recently about the mechanics of Visual Novels, along the lines of what I guess is a Game Idea striving to create interesting Ludonarrative harmony, though in a deconstruction-y kind of way.

Basically it occurred to me that VN mechanics are often more than just "the audience interacts with the story by making choices" there's also a lot of subtler mechanics which are more convention than direct programming. Consider for instance, there is often heavy emphasis on identifying with the protagonist in VNs--unless it is really stated otherwise it's pretty common for the player to act as though they *are* the character. This can have a lot of problems though, consider a pretty big flaw with a lot of beginners' visual novels: the choices are written in such a way where the player thinks they mean one thing, when the writer actually intended them to mean something else. (Player choses "Complement her speech" over "Complement her appearance" about a woman at a party and instead of actually complementing the content of her speech your character goes on to talk about how impressive it was that she made a speech...Completely undermining a player who was hoping to complement the content of the woman's character by instead being sorta patronizing.) Most people simply correct such errors as they mature as creators, but I think that these sorts of kinks in the process, when used with intention, are an untapped resource for creating some very interesting Ludonarrative effects.

Consider, for instance, a game where you are told not to trust anyone, because one of you is a murderer. Yes, making the player the murderer could be a cheap trick to shock and amaze people with you're cleverness, but it's also an opportunity to make a super powerful play experience. With careful writing and choices, imagine making a VN with your reader beginning with a standard whodunnit--no one to trust but your wits--where everyone has a motive and a secrets only have them to slowly detach to realize that maybe the person you shouldn't be trusting is yourself. I am personally not talented enough to pull of such a game, but I think that there is a lot of synergy to be found in looking in the subtler aspects of the narrative.

Or for a lighter example, consider making a stat raising Otome game. This is a really easy genre to make dissonant, because it is so easily game-able. If you have any variety of the "be yourself" theme it becomes almost inherently hypocritical. It would be really easy to make "raise my strength stat to woo the sporty guy", but some careful framing of the game could give it a ton of synergy. Consider forcusing on the theme "if you work to become the best version of yourself you will be happier and people will respond to that" instead. Immediately the stat raising has a place in the narrative.

It's a complicated problem though. After all, can you really make a game about the values of choice while limiting the player to a certain finite number of paths? I dunno. Books seem to do ok at exploring such themes, why not a KN?

[E-night]

@Parataxis. That is my most hated plot twist . Mostly because it kills the story 9 times out of 10. (Basically if I can't read/play the story again, I dislike the twist)

I like the idea for making the player vs mc destinction more concrete, though, and the idea of the MC directly disobying the player or not understanding the players "orders". It is an interesting concept and plays with the mechanics of an vn well.

[Parataxis]

@Parataxis. That is my most hated plot twist . Mostly because it kills the story 9 times out of 10. (Basically if I can't read/play the story again, I dislike the twist)

I like the idea for making the player vs mc destinction more concrete, though, and the idea of the MC directly disobying the player or not understanding the players "orders". It is an interesting concept and plays with the mechanics of an vn well.

That was an extreme example to get the point across.

The actual idea that started this pondering was more like "if you cut the player off from the main character's internal monologue could you make the player begin to suspect themselves by manipulating how the choices appear?" Consider, like, some one asks you a question about something that happened before the game, and after you answer, MC's friend turns to you and asks why you lied. This would immediately split the player from the character, and could be really frustrating and immersion breaking if done poorly. But! if you coupled it with a plot that should give you that alienation, even from "yourself" it could be a really creepy moment of synergy.

[E-l337]

I have to admit, this was one of the last places I expected to come across this topic... but boy am I glad there are folks thinking about this kind of thing. It's something I've been wanting to explore with my project - how to make the most out of the game mechanics to provide not just interesting play, but also meaningful play.

One thing I think that VNs can do that many other games cannot is the simulation aspect. This is a pretty important thing, because as has been pointed out before, a VN tends to have the player identify strongly with the main character. This can let you do some pretty awesome stuff that I'm not sure has really been appropriately explored in VNs yet.

I feel a lot of games are stuck on the topic of 'the illusion of choice'. This bothers me. I know it's the obvious route to take, along with the 'choices have consequences' aspects. But as I've been playing MGSV (a lot, admittedly... a lot a lot), it's struck me that there are some really interesting systems one can build in. Why do people play games like Skyrim or Fallout 3 for upwards of 300 hours on the same character? There's a good reason for that, but I'm starting to think that there may be another reason for this. It's not merely the wish fulfillment, or the adventuring aspect... but the sense of progression. The idea of your choices mattering - not because of how they affect the ending, but rather because of how they let you achieve that end.

It's kind of taking a page out of life. If all roads lead to Rome, and the joy in life is not the destination, but the journey, then shouldn't that be something which ought to be worth exploring?

Of course, using a purely choice-based system isn't necessarily the optimal choice either. On its own, choice can only do so much. But combine that with some dating-sim aspects (skills, relationship trees, so on), it should be entirely possible to provide an experience that would be very difficult to come by in other types of games. Because unlike the massive Bioware or Bethesda epics, your mechanics aren't based around combat - but the choices. The choices in your avatar for exploring this story, the choices in how you progress that avatar, and the choices you make in dealing with the people and situations you experience during the journey.

I might be rambling a bit at this point, but I find this stuff fascinating to consider.

[myheartisbeating]

This is a really interesting topic that has been brought up.

It is my belief that all the components of a game should contribute meaningfully to the final product. E-l337 brought up the fact they are bothered when games only have the illusion of choice, and they are rightfully bothered. In this instance the choice mechanic contributes nothing to the overall meaningfulness of the project as a whole. The storyline would have been more effectively portrayed if the creator put all their efforts into constructing a kinetic novel.

I feel a lot of games are stuck on the topic of 'the illusion of choice'. This bothers me. I know it's the obvious route to take, along with the 'choices have consequences' aspects. But as I've been playing MGSV (a lot, admittedly... a lot a lot), it's struck me that there are some really interesting systems one can build in. Why do people play games like Skyrim or Fallout 3 for upwards of 300 hours on the same character? There's a good reason for that, but I'm starting to think that there may be another reason for this. It's not merely the wish fulfillment, or the adventuring aspect... but the sense of progression. The idea of your choices mattering - not because of how they affect the ending, but rather because of how they let you achieve that end.

It's kind of taking a page out of life. If all roads lead to Rome, and the joy in life is not the destination, but the journey, then shouldn't that be something which ought to be worth exploring?

E-l337 brings up another excellent point here, though I would like to elaborate. The reason for the popularity of games such as Skyrim or Fallout 3 where you have the ability to carve out a separate alias unto yourself, is because you are essentially able to carve out a fictional characterisation of yourself. You are willing to spend hours upon hours refining an avatar because you view them as an embodiment of your own identity. Their achievements are your own, as well as their downfalls. The sense of progression enables this phenomenon to occur and makes the experience entertaining. A good example of a game like this is "The Seven Kingdoms: The Princess Problem". I've only played the demo but it is incredibly addicting because it does exactly what I've outlined above.

I would argue however that different endings are still essential to games which employ a choice mechanic. Having one singular ending in a game employing choice makes the player feel cheated, like their hours of time have been wasted. I can see the logic behind this - in reality we make different choices to create different outcomes. To have just one ending or even a fixed storyline can irritate or aggravate the player. Fixed storylines are effective when done well, but they must have differing endings. An example of a game with a definite storyline but many endings and choices is "SLAMMED!", an interactive fiction by Choice of Games. The heavy plot emphasis doesn't tug the reader in an unwanted direction but instead guides them through an amazing adventure, which then ends in a gratifying (or horrifically disappointing) ending.

Or for a lighter example, consider making a stat raising Otome game. This is a really easy genre to make dissonant, because it is so easily game-able. If you have any variety of the "be yourself" theme it becomes almost inherently hypocritical. It would be really easy to make "raise my strength stat to woo the sporty guy", but some careful framing of the game could give it a ton of synergy. Consider forcusing on the theme "if you work to become the best version of yourself you will be happier and people will respond to that" instead. Immediately the stat raising has a place in the narrative.

When I read the above passage I immediately thought of the Tokimeki Memorial Girls Side series. The series games are inherently "ludonarratively dissonant" (is this right? haha). You manipulate your stats and "suck up" to the "gettable" guy main characters, for the sake of initiating true love, while true love is not even nearly as contrived or planned. The game is still incredibly fun, despite this flaw. But why is this so? Shouldn't we feel disgusted or creeped out? How is this game even a thing? Behind the dating sim's fluffy exterior it is a cold, calculating strategy game. You manage your time, stress and funds a little like in games such as Sim City or Age of Empires. As such you are rewarded by fictional romance. How can we accept this process, though? How can I accept the fact by raising my charm statistic to 50 that Saeki will finally notice me (or something on those lines)? There are a mixture of reasons why. Firstly there is the "player avatar" aspect of the game. I can choose my name, horoscope, close friends, bedroom design and the clothes I wear - all according to my personal taste. I can choose what club I will join and what guy character I will devote my time to. As a result I become invested (and addicted). It's not the player character who improves their intelligence statistic, but myself! Secondly there is the sense of progression. In the beginning the guy characters only acknowledge me as acquaintances (or even ignore me, boohoo), but by my third year they are madly in love with and blushing with every step. This creates the illusion of progress and causes you to feel invested in the avatar and storyline. I still believe it is best to create games while aiming for ludonarrative harmony, but I recognise that games can still be entertaining despite their themes and mechanics being inconsistent.

In this forum post I haven't proposed any original ideas but have only attempted to further arguments that have already been brought up. What do you think?

[E-l337]

I would argue however that different endings are still essential to games which employ a choice mechanic. Having one singular ending in a game employing choice makes the player feel cheated, like their hours of time have been wasted. I can see the logic behind this - in reality we make different choices to create different outcomes. To have just one ending or even a fixed storyline can irritate or aggravate the player.

I should clarify on this point. Using Fallout: New Vegas as an illustration of my point - yes, there are different 'endings', but no matter what path you take, no matter what choices you make, there will always be the final showdown at the Hoover Dam. (Or alternately, the Lonesome Road, but let's go with the Dam example since it is much stronger.) You will always wind up at the dam, but your choices throughout the course of the game - the allies you made, the enemies you chose, all of them come to a head. You can pick one side, or no sides at all - and it doesn't change the fact that the final scene occurs at the dam. The only difference is how it all plays out. Did you play a murderous psychopath who still manages to convince everyone to leave 'your' town once you've finished murdering your way through two entire armies? Or did you cultivate as many alliances as you could, using their skills and abilities to your advantage to manipulate the ending to be what you wanted?

It's a combination of the skills you chose, the time you spent, and the people you interacted with all coming together in the end. The 'ending' is really just a way for the game to wrap things up. Sure, you might get some different cutscenes, but the important components are still there - two armies clashed at the Hoover Dam, and a lone wanderer walked in and changed the course of the battle.

In this case, the choices you made - or didn't make - during the course of gameplay directly shape the ending. But really, it all turns out more or less the same. The only thing that really changes is the flavor, but I don't usually see folks getting all up in arms over that.

tl;dr: It's important to have the choices have an impact in the end - either by making certain outcomes easier, making the ending easier to achieve, or even opening entirely new ways of solving a particular 'problem'. Done right, the player won't care about the ending more or less being the same, so long as the ending was still shaped in some way by their actions.

(I'd go off on a tangent about why the FO3 ending cheesed folks off so much... before it got patched, but even then it was pretty lackluster, and I think that I've said enough about that in this particular post.)

[trooper6]

E-l337 brings up another excellent point here, though I would like to elaborate. The reason for the popularity of games such as Skyrim or Fallout 3 where you have the ability to carve out a separate alias unto yourself, is because you are essentially able to carve out a fictional characterisation of yourself. You are willing to spend hours upon hours refining an avatar because you view them as an embodiment of your own identity. Their achievements are your own, as well as their downfalls. The sense of progression enables this phenomenon to occur and makes the experience entertaining. A good example of a game like this is "The Seven Kingdoms: The Princess Problem". I've only played the demo but it is incredibly addicting because it does exactly what I've outlined above.

Not everyone plays their sandbox games as self-inserts. Morrowind is my favorite of the Elder Scroll series because the world was so well drawn, as well as the conflicts within that world and your character's initial relations to that conflict.

My character was not the embodiment of my own identity. I role-played someone else. The person I role-played was a Dunmer from Cyrodil who had complicated feelings upon coming to Morrowind for the first time. I figured my character (who was a convict after all) never felt accepted in Cyrodil. He was treated poorly for being Dunmer, he was empoverished, he turned to petty crime. In prison, all of a sudden he gets shipped off to the land his parents came from...that he had all these fantasies about...a place where he'd be accepted...finally...and when he gets there? All the Dunmer there treat him like an Imperial outsider. Ouch! That character wasn't me. It was a character I played and explored. Similarly, my Greywarden in Dragon Age I was also not me. That character (Dwarven Noble) made choices that I wouldn't make if it were me personally, but I made those choices because they were true to that character.

So...not everyone plays RPG for the same reason or in the same way.

I would argue however that different endings are still essential to games which employ a choice mechanic. Having one singular ending in a game employing choice makes the player feel cheated, like their hours of time have been wasted. I can see the logic behind this - in reality we make different choices to create different outcomes. To have just one ending or even a fixed storyline can irritate or aggravate the player. Fixed storylines are effective when done well, but they must have differing endings. An example of a game with a definite storyline but many endings and choices is "SLAMMED!", an interactive fiction by Choice of Games. The heavy plot emphasis doesn't tug the reader in an unwanted direction but instead guides them through an amazing adventure, which then ends in a gratifying (or horrifically disappointing) ending.

I actually think it is an error to think that all choices are or should be plot changing choices. Choices can also be about defining your character and who they are...so rather than plot choices they can be characterization choices. You can have only one end point...but the journey to that end point can be radically reinterpreted based on the characterization choices made along the way. I could easily see a really interesting game where you play a game on the Titanic. The ship is going to sink at the end (no different ending)...but your experience going through the game ends up being very different based on what sort of starting character you chose and what sort of choices you make along the way. That is not a kinetic novel, because choice is still important...and players choices matter...but they matter in a different way.

I will say, my experience of the the Orzimar storyline was completely different that a friend of mine's experience of the Orzimar storyline in Dragon Age even though we made the exact same plot choices. Why? Because his character was a Circle mage and didn't have the sort of history with Orzimar my Dwarven Noble did. So same plot, totally different interpretive and emotional experience based on, not plot choices, but characterization choices. There is a reason why my Dwarf got really seriously involved with Morrigan and stayed true to her even though I personally would never have dated her...and it had a lot to do with the fact that he was a dwarf and dwarves don't follow the religion that the humans and elves do...a religion that Morrigan also doesn't follow.

I'm just saying that "choice" is not just one thing. And narrative is not just one thing. So there are many ways in which a story with a linear plot, can employ characterization choices and still have ludonarrative harmony.

[Carassaurat]

One of the difficulties of this is that the 'message' that the mechanics pose is never unambiguous, and a lot depends on how willing you are to go meta and be pedantic. And a great problem is that computers can't read your intentions and thus can't properly say if you're doing the right thing.

For example, Christine love wrote on dating sims: "Really, dating sims are inherently about manipulating other people — pick the right dialogue choices based off what you think they expect, learn about their interests so you can give them perfectly tuned gifts, make decisions based off whether you'll impress them or not — but like to pass it off as being about romance." [1]

Read that way, the message that every dating sim gives is that you should only be honest whenever that happens to align with someone's preferences. Even if a dating sim goes out of its way to punish you for slavish saying you like x when the love interest says they like x, it's still trying to get you to pick the answers the developer wants to hear.

Another example: a game might convey through its mechanics that "crime does not pay" by punishing the player for picking thug options. That's all well and clear. What you can't convey through ludic means is the message that "crime sometimes does pay, but you still shouldn't commit it", because the only way game mechanics can endorse something is by 'paying' you, so to speak. Any game mechanic -- that is, thinking about the act, not the text -- already implies the messages that 1) it's important to figure out what someone else wants you to do and 2) rewards are something you should strive for. Both of which are fundamentally at odds with quite a lot of stories we want to tell.

[gekiganwing]

Even if a dating sim goes out of its way to punish you for slavish saying you like x when the love interest says they like x, it's still trying to get you to pick the answers the developer wants to hear.

Yesterday, I resumed playing the third Tokimeki Memorial Girl's Side game with a fan translation. It's humorous and rather relaxing. At the same time, it got me thinking about a few things:

* The player routinely chooses which activities the main character does each week. If she has success on the volleyball team early on, then should she make it her #1 goal? Would it be better if she did multiple things for weekly activities, and her stats were balanced? It's not clear.
* NPCs have said things such as "You should get better grades" and "You should develop your sense of fashion." Should the player disregard one or both of these messages? I don't know how important they are.
* I remember playing Tokimemo Girl's Side 1, and not knowing whether the main character would get a good ending. The game let the player see affection levels. However, there were other variables such as stats, rivalries, and choosing college versus employment. I don't know if TMGS3 will improve on this.

The way I figure it, a way to respond to a video game is to create one's own game. I'm not sure if I am prepared to create a complex simulation game. However, it would be interesting to create a Tokimemo-esque game that would make goals more clear, and that would not pressure the player quite as much regarding activities or stats.

[Kuiper]

On the subject of ludonarrative harmony in visual novels, I've been having a grand time playing Black Closet since its recent release on Steam.

Black Closet is a noir mystery game with themes of politics and intrigue, and at your disposal, you have the ability to target persons of interest and either "question" (politely ask for information) or "harass" (interrogate or aggressively pursue information) them. Interrogation is the only way to get some guilty parties to confess, but if you interrogate someone who is innocent, you risk losing karma.

The "right" way to handle things, from a moral standpoint, would be to question everyone, and then interrogate individuals who act suspiciously when questioned in a friendly manner. However, if you start off by interrogating everyone, you can save time, since you don't have to question and then interrogate the guilty party, which requires an additional day of in-game time. Interrogating everyone certainly isn't going to make you popular, but time is a precious resource, and when you're on a deadline, sometimes you need to sacrifice a bit of goodwill in order to get the job done ASAP.

There's another thing about Black Closet: there's not always a guilty party. There are some cases where I started with the assumption that out of four suspects, one must be responsible for stealing some supplies that have gone missing. I began by questioning all of the suspects, but none acted suspicious. Then I began stalking and interrogating all of the suspects, but no one cracked during interrogation and nothing was discovered when tailing them. The case seemed unsolvable...until several days later, when I discovered that the "stolen" supplies were actually just misplaced due to a clerical error. I was seeing monsters where there were none.

I feel like both of these mechanics tie in really well with the theme of the game. Being in a position of power sometimes forces you to abdicate your moral responsibility to your followers in the interest of the "greater good," and sometimes that greater good is cracking a case on time to protect the reputation of the institution that you serve. But sometimes, playing that cold and calculating role where you try to crack every case as quickly as possible causes you to become paranoid and see everyone around you as a potential enemy.

There isn't really any point where the game's narrative comes right out and says "people in positions of political power gradually become soulless and paranoid and obsessed with furthering their own personal agenda at the expense of others," but it never had to. The game's mechanics told that story better than any bit of dialog or internal monologue could have, because I wasn't seeing a character becoming paranoid and making moral compromises, I was the character in that position who was making those choices.

[Tempus]

There isn't really any point where the game's narrative comes right out and says "people in positions of political power gradually become soulless and paranoid and obsessed with furthering their own personal agenda at the expense of others," but it never had to. The game's mechanics told that story better than any bit of dialog or internal monologue could have, because I wasn't seeing a character becoming paranoid and making moral compromises, I was the character in that position who was making those choices.

This is what Ian Bogost calls "procedural rhetoric"—discourse through systems playing out. Jonathan Blow has a similar term, "dynamical meaning", which refers to meaning conveyed through gameplay rather than narrative. Like you point out, the game doesn't come out and tell you something. Procedural rhetoric, rather than being a direct message from the developer, is instead questions and observations that arise naturally out of interactions with systems.

---

I'd suggest the idea of ludo-narrative dissonance falls more broadly into the idea of mechanics as metaphor. All facets of media can speak in metaphor, whether it be art, music, dance, writing, or film. Its the same with systems. Also, I'm working from memory here, so if anyone is interested in this I'd suggest checking out Jon Blow or Ian Bogost's lectures on YouTube. Blow is probably easier to process if you're new to this idea.

Incidentally, I made this post a while ago where I more or less describes LND without using the term using a different example. I can't remember why, but I recall actively avoiding the term for some reason. Also, this video discusses how win states contextualise gameplay.