When making a game, replayability is often one of the most important things to consider - enabling players to still have a fun experience playing through a game they’ve already completed. But while some games have replayability built into their most basic gameplay, others can find it hard to achieve it at all. The kinds of interactions that underly these types of games - learning, making connections, and more - are the basis of many amazing games and gameplay mechanics, but it’s not as easy to see how they could be used to bring replayability and life to a game. I think that trying to understand what might make these “knowledge-based” mechanics replayable could give greater insight into how those mechanics work and what makes them so interesting in the games we love.
In the introduction I mentioned the concept of “knowledge-based” game mechanics, but didn’t really elaborate on what that means. There are many different ways to divide game mechanics and other aspects of games, but when talking about knowledge-based mechanics I think the most important distinctions come from how they develop through player interactions.
In my mind, when we divide game mechanics in this way we end up with 6 groups:
Each of these deserve their own long post (or posts) but I don’t want this one to be too long so I’ll only touch on them, but I think understanding what these other types of mechanics are can help us talk about knowledge-based mechanics too.
In my opinion, the simplest kind of mechanics are flat mechanics. These are ones where nothing changes over the course of the game - not the player’s interactions with them, and not the result they have in the game world. Think about a door as an example - a door with a flat interaction will always be unlocked or always be locked, and the player can always interact with it in the same ways.
I think the other side of the coin from flat mechanics are progression-based mechanics. These are mechanics such as xp/levelling systems, where the results of the mechanic change over time but the way the player interacts with it stays the same. To use the door example again, a door with a progression-based interaction might require the player to reach a high enough level or gather fragments of a key from across the map to unlock.
Skill-based and legacy-based mechanics are both ones that change dynamically based on how much time and experience the player has in game. I think of skill-based mechanics as mechanics that rely on some player skill other than analysis (which is big enough to get its own category!), like twitch reflexes or perception. Rhythm games use skill-based mechanics all the time, and developing a player’s skill is often part of many games - a skill-based door could need the player to pass a quick reaction test like rhythmic movement or accurate shooting to pass.
I think of legacy-based mechanics as in some ways the opposite of skill-based ones. While skill-based mechanics rely on the player’s abilities outside of the game and can even use skills taken from other games, legacy-based mechanics use the player’s “legacy”: their previous actions and interactions in the game world. These actions change the scope of what the player’s next options could be, expanding it in some ways and shrinking it in others, but the crucial thing is that they all change based on the player’s past actions in the world. A door based on legacy-based interactions might need the player to do good deeds and build goodwill in a nearby village, then the villagers will come down and open the door for them.
Analysis-based mechanics are one of the most common in games and can be thought of as any mechanic where the player needs to analyze a situation and come to a decision based on information they have. I think this describes puzzle games really well, but some other genres such as grand strategy and RTS games also rely on the player analyzing a situation and making a decision. A door that the player would use analysis-based mechanics to pass could require the player to solve a puzzle to unlock.
Just like how flat/progression-based and skill/legacy-based mechanics are related, I think analysis-based mechanics are closely related to knowledge-based ones. But the difference between them is that analysis-based mechanics rely on knowledge the player already has, while knowledge-based mechanics need the player to find new knowledge.
In my opinion this is the core of knowledge-based mechanics: the player needs to learn new knowledge and make connections with what they already know. This usually leads to the player developing an understanding of how the game world works and how different elements relate to each other, and often they’ll base this understanding off of the real world. Because they use the player’s knowledge and understanding of the game world, knowledge-based mechanics can be used to give incredibly impactful experiences by building off of the player’s understanding of the world in interesting ways or by revealing new and revolutionary information.
As I say this, I think the reason why achieving replayability with knowledge-based mechanics is hard becomes clearer: once the player learns some information they can’t unlearn it. Since many knowledge-based mechanics rely on having the player find new information (such as solving a mystery) to give a fun experience, those experiences are ones that you might only be able to have once. However I think this isn’t the only way we can use these mechanics to do something interesting, and this might open the door to creating knowledge-based games that can be truly replayable.
In Videocult’s amazing 2017 game Rain World, the player explores an environment which they (at first) know almost nothing about. One of the first and most evil dangers the player might encounter is the pole plant. The game’s movement involves a lot of pole-based climbing, and pole plants mimic these poles to catch the player if they get too close.
Pole plants’ sneaky nature can result in unsuspecting players being easily lured to their deaths, especially when trying to escape a larger and more mobile enemy. But although pole plants look similar to regular poles, they aren’t exactly the same. A new player will still learn this quickly, since pole plants periodically flare up their red gills and clearly reveal that they aren’t regular poles. Throwing a rock or spear at them will also make them recoil which normal poles of course don’t do. But an experienced Rain World player will learn that even when hiding pole plants actually still have some giveaways - they’re smoother than regular poles and have a slightly different colour, not to mention they’re usually at suspiciously convenient positions for navigating a room.
Once the player notices this, it can actually transform the ways they interact with the world. Take the room above as an example - if a player doesn’t know these differences, they might be limited in how they can discover the room’s pole plants. They could try throwing spears, waiting for the plant to move, or just hoping it’s a regular pole and jumping; but this probably isn’t great. At best you might lose a spear or some valuable time, but at worst the player could be killed by a pole plant they just didn’t know how to identify.
On the other hand, for a player who knows how to spot pole plants this is an entirely different situation. By slowing down to take a closer look at the poles they see, they could quickly identify which are actually climbable and which are deadly pole plants. What I really want to emphasize here is that this isn’t just a better solution (in that it saves time and resources), but it’s a completely different kind of interaction than the one the unexperienced player has. The unexperienced player has a choice between wasting their time/resources, a kind of transactional interaction that results in losing something valuable; or just guessing, a flat interaction which (for the player) is basically just random chance. The experienced player is doing a kind of analysis interaction where they use their skills of analysis to determine what a possible pole plant is before they move through the room.
What’s even more interesting is that even the limitations of these interactions are changed. For a player running from a dangerous enemy, a convenient pole can be a lifesaver - but if the player is unexperienced they would have to either waste one of their weapons checking for pole plants or just take a chance and hope that the pole is safe. In the case of taking a chance this is actually a totally flat interaction, so (in the player’s mind) their chances are the same regardless of if they’re frantically running or slowly puzzle-solving! With an experienced player in the same situation, they’re limited by not noticing that they’re walking into a pole plant - even though they always have the ability to tell just by looking at a plant, they might not be considering that possibility or not paying attention to where they’re going.
But I think that this is only part of what makes knowledge-based mechanics so interesting. Knowledge-based mechanics also give the possibility for the player to develop an understanding of the world as they learn more about it, enabling them to discover new knowledge by using what they already know.
To go back to the pole plant example, imagine a player were to suddenly discover a new enemy. The environment of Rain World is full of useful spears, which look like short poles scattered at random angles across the ground. Maybe that gives us the possibility that, just like with pole plants, we could have “spear plants” (to be clear there are no spear plants in Rain World, I’m just using them as an example here). A spear plant would behave similarly to a pole plant, but disguise itself as a spear rather than a pole. When the player encounters a spear plant for the first time, they might not be expecting it - but if they know how to spot pole plants, they might be able to quickly guess how to find spear plants even after just meeting them. The player could apply their knowledge to start looking for spears that appear a bit “off”, such as looking too smooth or discoloured compared to regular spears - the same ways that pole plants are different from regular poles.
This might seem like a natural connection to make, but what I think is really amazing about this is that the player can use what they’ve learned from previous experiences to predict how the world works without needing to learn it from a separate experience. One possibility of this is that two different players could gain the same knowledge in different ways. A player might encounter pole plants first and learn how to recognize them, then apply that knowledge to spear plants when they first encounter them (like we’ve been exploring so far), but you could also have a player that didn’t realize how to spot pole plants until they noticed it when encountering spear plants, or even a very unlucky player who encounters spear plants first and learns how they differ from regular spears before they even see a pole plant for the first time.
When the player begins developing an understanding of how the world works, they draw from knowledge both inside and outside of the game. To use Rain World as an example again, the game is made of things that players will recognize from the real world: plants, animals, poles, walls, tunnels, water, spears, and more. Even if you’ve never played the game before, you can still have a good understanding of what these things might do; tunnels are narrow passages you can travel through, water lets you swim but requires you to hold your breath, large and dangerous-looking animals will try to eat you, etc.
I know that it might sound as though I’m just pointing out obvious properties of knowledge-based mechanics, but I think this is really important to understanding how these mechanics work. Grounding aspects of a game in the real world can greatly reduce how much the player needs to be taught, - but what’s more, they can even use knowledge from other games. A great example of this is in standard control systems for different game genres - for example, most shooters will share the same control inputs for all of the fundamental actions of their games such as moving, running, shooting, looking down sights, changing weapons, and more. There’s a good chance you can guess exactly what keys I’m thinking of, which is amazing when you think about it! When making a game that fits into any preexisting genre, I think this external knowledge is absolutely necessary to consider; developers can rely on players already having a huge amount of knowledge and experiences with the real world and with other games, but also having developed an equally huge amount of expectations and ways of understanding aspects of your game.
I’ve recently been playing Outer Wilds and it’s quickly becoming one of my favourite games (don’t worry, there won’t be any spoilers). But as I progressed further into the game I realized that it’s the kind of game you could never replay, at least not with the same experience. Once you learn the important secrets of the world you just can’t unlearn them, and eventually this means that you’ll run out of new things to discover and lose the experience of learning that makes Outer Wilds such an electric game in the first place. I won’t say that you can’t replay it after completing it once - there’s enough depth in Outer Wilds that you could definitely still have an incredible experience one a second playthrough, but I’m not sure you could play it a third time. Eventually you’d just have nothing left to learn, and learning is what the game is about.
Realizing this is what caused me to start writing this post and to start thinking about how knowledge-based games could keep giving more to the player after multiple playthroughs.
One possible workaround could be to create a knowledge-based game with lore so deep and plot intricacies so subtle that you could play it over and over again and always find new things. And you could actually do that - stories like Lord of the Rings have enough happening under the surface that you can read and reread them and always find new things. But I just can’t say that this is the best way forward, since writing incredibly nuanced stories and worlds like that is just really difficult. For every Lord of the Rings that succeeds in making a story and a world that you can keep returning to, there are countless stories that are just too vague to understand, too derivative to be interesting, or complex in ways that are just boring to most audiences.
Another important thing to remember is that games aren’t just lorebooks, and every part of a game should work together to give an experience beyond what they could be apart. This might not work if you’re replaying a game to find new details: the music soaring as you discover a new area or character for the first time might make that discovery even more impactful, but not so much when you’re just trying to speedrun through the area (or dialogue tree) to reach a specific part you think you’ll find secrets in - the music could lose its emotional impact, or even becoming annoying and repetitive as you continue to revisit it. I think that when relying on this kind of intricacy to give depth you would want the other parts of the game to bend around the player and heighten their experiences, but for some games that might be an impossible task relying on impossible knowledge about the player.
That’s not to say that intricate stories and worlds can’t be a part of knowledge-based games - I mentioned Rain World and Outer Wilds before, and I think the depth of their worlds is a core part of what keeps them alive for me - but I feel that it’s not enough by itself.
If writing subtle plots and complex lore is the way to make a game deeper, then randomness might be the way to make a game wider - to expand the amount of content in the game beyond what the player could experience in one playthrough. And just like adding depth, I think that adding more content to a game is a crucial part of making it replayable, but I don’t think it’s something that can make a game more replayable by itself. I think there’s two main kinds of randomness we could look at.
One way is what I think of as “branching paths”, which actually needs the developers to create more content than the player can see in a single playthrough. The player only sees a fraction of the content each time, with what they experience being determined either by their own choices (consciously or unconsciously) or at random by the computer. A good example of this is in Mass Effect, where you answer multiple distress calls to build an alliance - however, you don’t have enough time to answer all of them. This is a really interesting decision, since the developers could have gone the other way and have you answer every distress call - by only having a limited number be answerable, the player is both forced to make complex choices and granted the ability to replay the game many times to see the different possible combinations. In this case it’s not random, but we can imagine a game that randomly determines what content the player experiences - such as a game where you play through the story with a randomly-chosen group of characters.
The other possibility might be thought of as “endless” randomness, where the player can just keep playing the same game with new, randomly-created content. Since this kind of randomness is (theoretically) infinite, the developer can’t create all of this content beforehand and they’re limited to just determining how the random generation works. I think this is a really notable limitation and it might be why we don’t see this kind of structure in games very often - I actually can’t think of even one example. This kind of randomness could have different levels of impact: the most limited version might just change the names of some characters, while the most impactful version would be entirely changing the fundamental structure of the world and story. There have been some experiments with randomly-generated stories in some other media, but it’s rare and can face difficulties in communicating the author’s intended themes and messages. I think the limitations of this kind of randomness are a bit clearer to see - too shallow and the game would be essentially the same experience, but too deep and the game would lose the artistic understanding needed to create interesting characters and stories (not to mention how hard it would be to randomly generate characters or storylines that make sense at all).
You might be thinking about roguelikes as the shining example of randomly-generated game experiences and I agree! Some of my favourite games are roguelikes! But roguelikes usually aren’t meant to be story games, and they usually randomize level layouts rather than characters or stories. As a result, roguelikes usually don’t rest on knowledge-based mechanics as much as other games, and those parts of the game aren’t randomized when they do.
All this said, I don’t want to imply that random generation is a bad thing which should be kept out of knowledge-based games. I think that it can be absolutely essential to an interesting experience, but can’t create that interesting experience by itself.
In my opinion, to find what makes some knowledge-based games so replayable and dynamic we have to step outside of the games themselves. We’ve been talking about knowledge-based mechanics for a good 3000 words so far, and I hope that it all shows what I feel the true heart of what they are and what makes them so interesting in games: knowledge-based mechanics enable the player to make connections and they transform other interactions. I think it’s only right that the way to make them replayable is also rooted in these two pillars, and that’s why I think that the path to replayability has to start with the player.
I think it’s important to always keep in mind that the player is a real person, and they bring their own perspectives and interpretations into every game they play. Every player’s experience of a game will be different, but not only different from other players; a player’s experience will also change over time as they change as a person. This is why you can reread a book or rewatch a movie you’ve already seen and have a completely different experience - what you’re seeing hasn’t changed, but you have. When it comes to knowledge-based games the rules are the same: adding more content increases the breadth of the game, making the content more intricate increases the depth, but giving a more complex meaning and a more interconnected game gives more dimensions - different ways to understand what you’re seeing.
I feel this way because this is actually what I’ve experienced personally. Rain World is one of my favourite games but I’ve actually only played it once, completing the standard “survivor” campaign. And if that was all that I’d ever experienced from it, I would probably just think “that was cool” and quickly forget about it - but this isn’t what ended up happening. Rain World is a game that you can approach from many different perspectives and interpret in many different ways, not just in terms of story but in gameplay as well. Some areas of the game feature little blue plants and I still don’t know what they do, even though my friend constantly teases that they have some secret use. Different players can develop different understandings of the world and how it’s connected and there isn’t a single “right way” to think - all of those different understandings can bring you to the end of the campaign. What’s more, the game’s support of many different interpretations extends to its story and characters as well. It’s hard to talk about this without spoiling the story, but the world and the characters you interact with are written with a kind of depth, emotion, and interconnectedness that lets players interpret them differently - the game doesn’t declare that any single interpretation is the “right” one. I feel like I can look back on parts of the game’s story that I personally haven’t had such deep interactions with, but they can be so much more alive and vibrant to me because I can see them with other players’ experiences and interpretations.
I think that this is what can give knowledge-based games the kind of replayability and life that other games can achieve - the ability to be interpreted and understood in many different ways, both in a gameplay sense and in a narrative sense. It can create a community where people share their own experiences with a game and the understandings that they’ve developed, and enable individual players to learn from and incorporate those other understandings into their own. I think this is also what makes some other media (like books and movies) so rereadable/rewatchable even though the content doesn’t change: the audience’s interpretation of it does.
For some types of knowledge-based games this can be a nearly impossible change. Games based on solving a mystery (like Outer Wilds or a detective game) tend to be inherently structured on the player finding the one right interpretation of everything they’ve learned, and this means it’s hard to apply replayability in this way. That’s not to say it’s impossible though - one idea is that a mystery game could have multiple suspects or multiple right answers, and challenge the player to think of all the different possible interpretations to find every suspect - but it’s not something that every game can do, and that’s fine! Some games just can’t support replayability, but that doesn’t mean they can’t give interesting experiences.
When replayability is an achievable goal though, I think that the way to achieve this kind of multiple interpretations changes from game to game - it’s something that every developer has to find for their own games. But in my opinion, there are some factors that replayable games and other repeatable media that I like have in common: they draw from real emotions and experiences to create characters and worlds with the kind of depth and interconnectedness that reality can bring. I'm not sure if this is what brings a sense of dimensionality and replayability for everyone, but my experiences with these games show that it does for me. If you want to discover what can make knowledge-based games exciting and replayable, I can only encourage you to think about the games that you love and the experiences you've had.
Did you like this post? Tell us
Leave a comment
Log in with your itch.io account to leave a comment.
I'm not sure if they count but I consider games with deep systems that offer a lot of variety such as an RPG knowledge based games too. And your first playthrough in a way is you trying to understand the mechanics of the game, the strengths and weaknesses of your build, gear, and whatnot, And at some point you reach a mastery of your current build. And in your second playthrough when you have a much better understanding of the mechanics it's a completely different experience since you don't have to learn the mechanics again and you might even be inclined to play or build your character in a way that was unlike your previous attempt like say going from a full melee build to a fully ranged one. the playthroughs after are always a lot more fun and usually less time consuming and more efficient. And hell you might even discover things you didn't in your last playthrough about the mechanics.
P.S. I forgot to reply to the post back when you posted it oop.