Sunday, 17 March 2013

Puzzles and Causal Histories

In the last post I brought up a few reasons why puzzles should not be dismissed. In this I will bring up another one: making the player feel as an active force. I refer to this concept as having a causal history. My hope is that it provides a new way to view and evaluate puzzles.

First I must go over the term "causal history". Basically, it means the sum of actions that players feel they have caused; any previous actions remembered as "I did...". Having finished watching a scripted sequence, the causal history will just be "I triggered a scripted cutscene". But if participating in a gun fight, the memories may be something like, "I first shot that guy,  dodged a flying knife, threw a grenade ..". Causal history is a tightly related to agency. A rich causal history increase the feeling of agency. It creates a personal experience for players; an experience where they feel much more present.

Consider these three different gameplay segments, as retold by a player.

1) As I walked up to the door a cutscene started. I watched the protagonist search for a key. Rummaging a few nearby drawers and boxes, she managed to locate it. She unlocked the door and entered the next room.

2) As I walked up to the door, I found a guiding cursor pointing at a box. Searching the box I located a key. The cursor now indicated the door as my next objective. I interacted with it, unlocked it with the newly found key and entered the next room.

3) I walked up to the door and found it locked. I had recently overheard that the orderlies hid spare keys and proceeded to search nearby boxes and drawers. Turning an dusty crate upside down a key was revealed. I picked it up and unlocked the door. I was now able to enter the next room.

All of the three have the same thing happening, but involve the player in the different ways. The dividing factor is the amount of agency provided. The player has, to various degrees, been shown or participated in a happening,

In the first example, the game takes care of the whole situation, and the player does nothing. In the second and third examples, the player does the exact same things; the third, however, provide a much stronger sense of causality. In one retelling the player is explicitly told what to do, in the other the player is implicitly hinted what do. The third one manages the latter by being designed as a puzzle, resulting in a rich causal history.

I think most will agree example three is the kind of experience one wants to strive for. This sort of gameplay set players smack in the middle of the happenings. They are present and responsible; causal agents in the narrative.

By paying attention to this concept an aesthetic for designing puzzles emerge. One wants to have puzzles that  provides the strongest and most detailed causal histories. It fits neatly with the idea brought up in the previous post; that puzzles should give the player a feeling of coming up with solutions on their own.  By focusing on causal histories puzzles become a means to implicitly guide the player through a set of actions. It entails setting the player in a certain frame of mind, to give hints and provide relevant exposition.

While trying to think up complex chains of actions for the player, one cannot get too carried away though. The rules set out in the previous post must still be adhered to. If not, there is a big risk players get stuck, counteracting the intended experience. It is a balance act, and one of  taking risks,when it comes to determining the complexity of puzzle. It should provide strong causal histories, but at the same time it can not break the flow of the narrative.

Another aspect is that simply adding more actions is not good in itself. The actions that make up the causal history must fit the narrative being told. For instance, if a puzzle-lock is added to the box containing the key, it adds nothing to the experience. It just halts the flow and gets in the way. Padding will not improve the experience, but only detract the player from what is intended. Actions should be meaningful or not be included at all.

Having puzzles is not the only way of creating causal histories. Earlier I gave a gun fight as an example of a detailed sequence of actions. This is a form of gameplay that arise directly from the underlying mechanics. If possible, it is an excellent way to create agency.Many situations are however impossible to create in this manner or just too expensive.  Classical gameplay also come with a lot of problem that often break the sense of immersion (outlined here), making it at times undesirable. It all depends on the situation. For any given segment, it is crucial to make sure that a puzzle is the best approach.

Let's summarize. Following this aesthetic one designs a puzzle in a way that gives players rich causal histories. The retelling of a puzzle should be dense with the player performing actions, not passive spectating. One must also make sure that the puzzle does not block the flow, and the actions involved must support the intended experience, not hinder it.

Even though I have not used this approach much for actual work, it has already made me see a few puzzles in a new light. I think there is a lot of potential in this concept and look forward using it more. It might of course turn out to be the wrong way of thinking, but so far so good.


  1. Release the pigs already! :D

    1. Yes, can't wait much more!

  2. As always, a very impressive blog post!

  3. Can't wait for AAMFP!

  4. An excellent post on puzzles. It was rather enriching to see the same sequence in three different manners. Puzzling in storytelling has always intrigued me as does casual histories. I agree with the puzzlebox situation as well. When making a game that is meant to tell a story and have the player progress in a manner that makes them feel invested, that will clog the process. Now I could see the puzzlebox working if the story lends to it. An example would be Safecraker: The Ultimate Puzzle Adventure.

    Completely fantastic blog post though. Will be bookmarking this for another reading.

  5. Great post. That second example is exactly the problem I had with Silent Hill: Shattered Memories. Enjoyed a lot of that game, but the exploration segments were diminished by the over-use of arrows pointing out the locations of puzzles and items.

  6. After the swine rise, I look forward to seeing this approach in action.

  7. "I think most will agree example three is the kind of experience one wants to strive for."

    Unless you cannot get the player to care enough to want to engage in solving the puzzle. Then, the player may prefer a more passive option. So, example three is optimal, but requires a well-designed imperative. If the imperative is too long-range or too weak (as by poor storytelling), the reward of completing the puzzle may be too remote to motivate the player.

    1. I agree with that one. Also, sense they don't you cut scenes, the story telling they use in game will have to be extremely clear for this effect to work properly and for it to enhance the narrative.

      Thomas, In amnesia I had to play the game at least 3 times before I fully understood what was happening in the story/plot. I think its safe to say that some people did not understand that the servants in amnesia where not the shadow, and there were other elements as well that people did not get. In some game reviews people had it backwards. The story wasn't always clear, but sometimes it was confusing. I was wondering if you are aware of this and if you are going to do anything differently this time around.

      I remember while playing amnesia, I was one of those people who did not like reading the notes unless they were voice acted. I was always very immersed in the environment and listening to the flashbacks, but I always felt that having to read a note broke the flow. Was that just me, or did others seem to experience that as well? I know I was not alone, but I am wondering if it was a big thing that you noticed. If so, what do you think about it?

      -Jesse P

    2. Getting rid of massive note reading is an aim for our next game. That said, many have reported liking the notes, so it is not always a bad design choice.

    3. I think notes aren't a bad choice either. Its obvious it can be used to enhance certain experiences when done a certain way. I think a method that can help with that issue is when you make the note a part of the environment, like making it an actual object with some texture on it in game and not simple flat 2D image when you pick it up and read it. When you turn the page have it could go through a flip animation (Or since you have the physics engine control method, maybe you could literally turn the page which I think would be fantastically neat, cool, and very fitting! In fact, I think the page turning physics would be A VERY wise thing to add. IN FACT I'm willing to be blunt and a little impolite to say that it would be very very stupid if you didn't consider that as a possibility to add that *chuckles*). It helps with the immersion in certain cases and can used to enhance story telling when used properly.

      Also, not all the notes were immersion breaking, I can't quite remember which ones did break it though. I think what caused it was when I couldn't make the connection in seeing that notes were attempting to answer questions that I had at the time and things of that nature. Like when it just felt like there was no reason to be there (even though there was a reason for pretty much all of them, which I eventually figured out in later plays.) Like, that note in old archives, I think was called local folklore. I pushed that one aside because the note at the time seemed to be out of context with the information that the game was sending me at that time. That part of the game was all about Daniels past experience in the tomb. Now, I learned later on that the note was actually quite valuable to the story, but its in the last few pages of the note. I saw that most people who read the notes in some of the lets play videos would read the first couple pages and then drop it because it seems like it sort of going on a tangent and wasn't important, but little did they know the last parts had the good stuff which is mainly the part about Alexanders age . I know what you were trying to do with the gathers, (the writing there was really spooky!) but that connection you tried to make wasn't very clear the first time you play.

      Also, You can give a note a lot more depth if you give it some art design. The only thing I've seen it used for though is environmental story telling, and it makes sense. I can't really think of anything else it could be used for.

      I also have some more thoughts about interaction between doors and objects in game that I don't think you might of heard already and some ways you could maybe improve them and make your environment a bit more immersive. I know you are busy, but I know that feed back can be important so I have some more thoughts to share if you would like to hear them.

      Thanks a lot for your response, I hope all is going well with your game, you have all my support! :)

      -Jesse P

    4. Probably my favorite "note" ever was "There was a HOLE here. It's gone now" It was part of the world and the way it was incorporated fascinated me in a way that it is up to this day one of the biggest things I remember from that game. I know many other games use writings in walls but in a lot of cases they are rehashed and they lose the feeling of being special. Also, the first time in Portal I came into contact with "The cake is a lie" I got shivers down my spine because it was a sudden realization that the came is so much more than just a puzzle game.

  8. How about a mechanism that adapts the player involvement to his receptiveness? It would be great to go from situation 3 to situation 2 when the player is not in the mood for looking everywhere (or simply when he is stuck). I remember a demo (from "The last of us", maybe) where the items were starting glowing only after having searched for them a bit (or when the items were near and on the screen for few seconds). I liked the idea of having with a less active fallback gameplay.

  9. Very interesting, but please! Launch the release date for AAMFP!!
    I don't care if the game will launch in two weeks or in five months... I just want to KNOW how much time I must wait! I think an approximate date is not asking too much.

  10. I agree with the post, Thomas.
    The third example is the best way to make puzzles for an experience. But what do you mean a puzzle should not break the flow? I mean it is right, that a puzzle has to be placed comprehensible with the situation.

    But what do you think about the welded metal box in Overture where the generator battery is included? That's the same as you said where the key box is locked. But does it really break the flow? I think not, because it is always good to have an extra puzzle more to make the player thinking. Well, this works better when a puzzle is fitted in plausible.

    I like puzzles very much and I think it does the most impact to the experience. So, the worst thing you can do "as Frictional Games" is to make a game without or easy puzzles. I really hope all your puzzles would be like in example three and that they are much more complex and various without getting hints. To stuck in a game does not break the flow, in contrary, it will improve the experience (thinking of the player, explore the location etc.).

    I really enjoyed to stuck in Penumbra when I don't know what to do next. Amnesia was too easy compared to that. The only challenging place was the control room, but the puzzles there were not really complex, but hard to solve.

    What I want to say is, that as harder a puzzle is as better the experience will be - even when you stuck and don't know what to do. Because this is what challenge the brain and take the best impact to experience.

    Do you remember some puzzles of Silent Hill franchise when playing on "hard"? You need an extremely big brain to get through, but this is what I always enjoy.

  11. Oh, to the note reading.
    Of course, notes are very good for the experience, especially whan the content is interesting. But let me explain the difference between Penumbra and Amnesia here.

    In Penumbra you have to read notes as "Copenhagen Post Article" for example with interesting stories about the situation/location. And what I liked very much is that you also have to read notes like the "Explosives Book" or "Generator Manuals" or "Morse Codes" or someones diary etc. what you need to solve puzzles.

    For example, a note will bring an interesting sense when you need it to operate a machine or get information about a situation you couldn't solve without reading that note before.
    Also the game felt not like a "note grazing" how it was in Amnesia. In Penumbra the notes always were placed well and not too much at once.

    A bad example is Amnesia, where you have to collect notes every single minute in the archives, and then find all three parts of the diary. That was simply too much.

    Notes should not look like "collect items" you have to read one after another...

    Notes should be more like plausible stuff left in the environment. And not blinking up!!! Flashed items looks just like "fun collecting" where the game wants to show the player where a next item waits to get picked up.

    Please do give up such elements what just makes the game easier and kill the immersion. Do objects never looks like game elements placed for picking up. Do it more like plausible content of location. Same with the notes. Do never place notes just as elements to pick up and read. Make notes more important, but not impose on the player as a game's goal.

  12. I agree with what was said about the servant/shadow question. It took me going through the super_secret.rar to realize this. I understood the rest of the plot fairly well.
    MORE HALLUCINATIONS! That would be awesome. Like, when your sanity causes you to pass out you see a servant(or pig thing) walking toward you and when it gets to you everything goes black for a moment and after this whole ordeal, you stand up. The problem would be fitting this sequence into the short time that one is out.

  13. There's also a fourth puzzle example what I prefer to use:

    3) I walked up to the door and found it locked. I had recently overheard that the orderlies hid spare keys and proceeded to search nearby boxes and drawers. Turning an dusty crate upside down a key was revealed, but it was rusty and rickety after a long time lying in a dampness area. As I tried to unlock the door, the key got broken. Finally, I had to think how to elaborate that key by using some tools. I got it working, picked it up and unlocked the door. I was now able to enter the next room.

    The repairing process of the key also could be a further puzzle. Some tools (machines) needs power supply or mechanical components. To intertwine puzzles is a further step of experience. When there are more puzzles into each other, it is by far more complex than only to have one single puzzle.

    What I want to say is, that puzzles should challenge the player's brain and not only pick things up and use.

    1. If the intention is challenge. Sometimes it is not.

    2. What do you mean by that exactly?

    3. That sometimes the aim of a game experience is not to challenge the player, and then it is not a good idea to go beyond a certain complexity in a puzzle.

    4. But I think, Penumbra was much more complex compared to Amnesia, and that was the most important thing why I enjoyed Penumbra a bit more. When I remember the puzzles, Penumbra was by far a greater experience of logical thinking.

      Maybe you have right when saying "sometimes", but in general complexity of puzzles will take a big impact to the experience on how much you can find out in a game.

  14. I think you are right in many respects. But I think the traditional concept of a puzzle is to narrow here. What you strive for could be called "contextual challenges" - the main difference being that, unlike puzzles, these aren't there just so that the player would have something to do, but strive to be a natural part of the overall gameplay experience, and while they can be in the form of classical puzzles, this need not be the case.
    The kind of design involved here shifts away from traditional game/puzzle design to what I would refer to as "experience design". These are more or less uncharted waters, an art that is yet to emerge. Now, from a developer's perspective, striving for this kind of experience in your game sets a high-level goal, for which then the details can be worked out, that finally can be used as a basis for the features of the engine - the goal here is to invent and develop new game mechanics specifically to support various forms of these "contextual challenges" planned for the game - but in a way that enables them to be separated from the (generic) core engine and be modified, replaced and removed with relative ease (modular architecture). I think that it's very important to provide support for such things in the game code itself, because that opens up new possibilities and provides more maneuvering space, and more tools to express the gameplay. Because, so far, engines have mostly been about better graphics and AI and audio (and node/resource management), and made in such a way to support a wide variety of games - which is not a bad thing in itself; but what if more thought has been given to expandability and storytelling-specific features, narrative-building tools, and game-world-specific requirements from the start?

    P.S. About puzzle complexity - it all depends on what kind of a game it is and where in the game the puzzle appears. You're right that too complex, or poorly designed puzzles might break immersion or hinder progress. But puzzles of formidable complexity might have their place during exploration, in a setting where solving that particular puzzle is not crucial to beating the game, but if the player succeeds, there's a reward in the form of uncovering some new aspect of the game world (I don't like rewards in the form of some awesome powerup which unaturally shifts the game's balance - it's better when, say, the player finds out more about what is going on (wheres new knowledge might also bring new questions), obtains acces to new opportunities, learns something about the games fictional history, etc.)

  15. Hi Thomas,
    I enjoy this blog a lot, but I have a question completely unrelated to this... Have you considered adding Oculus Rift-support for Amnesia? or perhaps future frictional games? what do you think of this technology?

  16. I also think, that to pick up a key and unlock a door really can't be called as a "puzzle".


Note: only a member of this blog may post a comment.