Prototype evaluation and Skinnerian loops

Comments

  • edited
    Ok so I haven't read the Pete Collier in as much depth as I would like, but I felt I needed to reply before the thread disappears into the abyss known as the second page.

    The underlying principle is actually something I strongly believe in. Although Brenda Romero outlined my views on the matter rather succinctly:

    Focus on second-to-second play first. Nail it. Move on to minute-to-minute, then session-to-session, then day-to-day, then month-to-month (and so on). If your second-to-second play doesn’t work, nothing else matters. Along these lines, if your day-to-day fails, no one will care about month-to-month, either.

    So I believe that in order to progress to a meaningful meta-game you need to have an initial core aesthetic.

    I'm going to talk about the meta progression, but not in the context of Broforce. 'Cause I'm going to bash on the notion of the Skinnerian loop, and the need for the meta progress.

    So I recently got GW2 and they did this really odd thing that your skills, initially, are bound to your weapons, and you have to kill a certain number of things in order to unlock the next skill. I realized that SCB did the weapon unlocks in a really similar manner. So I ask myself why. In GW2 I found the weapon so why do I still have to unlock the skills. The answer is gated progression and mastery. By forcing me to use earn the skills in a progressive difficulty it is forcing my to learn, and optimise the use of the skills I had already unlock, which in turn created a mastery over my use of the skills. The designers did not make the unlock there as part of a psychological enticement, but rather as part of making sure I could handle and use all the skills at my disposal before I earned new ones. SCB does the same things with the weapons. Even the fact that you start out with the really dinky pistol serves as a gated progression, although it is far more of an incentive to pick up a crate and get a better gun.

    Mastery is a really important factor in game design to me because it uses a player's own progression as a source of extended play generation. I have to game which to me have very similar play aesthetics.

    So Jetpack Joyride and Canabalt are really similar in terms of what you do you dodge deaths and run the greatest distance. But I feel Canabalt is a much better game. Firstly Canabalt has the sweet tunes of DannyBStyle. But that aside there is no meta progression to Canabalt you have simply yourself, and everyone else on the highscoreboard. There is no achievement just the intrinsic fun of jumping, and the non-intrinsic fun of trying you beat your own high score.

    Jetpack joyride is full of meta progression. At one point I forgot there was actually a game beside the meta progression. The game had lost it's intrinsic fun because there was just extrinsic motivation to do things.

    Oh, and just to clear up intrinsic and extrinsic. Intrinsic is I do something because I enjoy doing it. Extrinsic is I do something because there is a reward waiting for me at the end.

    I'm going to draw a lot from Chris Hecker's 2010 GDC talk. WHICH EVERYONE SHOULD READ http://chrishecker.com/Achievements_Considered_Harmful?

    But basically if you take something that was intrinsically rewarding, and apply and extrinsic reward it loses it's intrinsic value. This is something that behaviourist and non behaviourist psychologists. Listen to the Chris's talk to see how much these guys hate each other.

    The best example that Hecker gave of how the occurs was: take a girl who enjoys drawing. Give her a reward for the drawing. She will do more drawing of worse quality because her focus is no longer on the fun of drawing but now the obtaining of the reward.

    This is why Zynga has such huge churns in their games. Eventually they run out of extrinsic reward and players find the game boring and move onto the next social clone. And the weird thing is if you strip out the shit in a social game you land up with management games like SimCity (the old and cool ones).

    ---

    With the above in mind I'm going to talk about Broforce again. Ok so the unlocking of Bro's isn't something I've put some thought into, and I wouldn't actually say that it is extrinsic. Since Bros are when rescued are resource for the player. There is just an additional exploration aesthetic added in, which combines into enhancing challenge for the player by providing more items for them to master. It is similar to SCB in you would still feel fine just picking up boxes, but it's a nice bonus that I felt I got a new weapon, and sometimes not 'cause fuck that disk launcher to hell.

    The key difference here is the the freeing/unlocking of a bro is not tied to the progression of Broforce where the collecting of crates is the progression of SCB. In fact in earlier versions of Broforce the freeing of bros was almost discouraged due to the fact that certain bros just kicked way more ass.

    So yes I would say that SCB motivates the same amount of play without the weapon unlocks. Although I would not say without unlocks in general. Since I was initially presented with a selection of maps before a selection of weapons I wanted to get to the next map. So that was pick up 20 creates (I can't remember the exact number) without dying. Seemed easy. So that was my initial goal. The fact that I unlocked weapons was just a nice side addition. And eventually when I unlocked the next map I realized I didn't actually care because I was having a but ton of fun just trying to beat my own score. So as long as I have a score (which I cannot state how big of a deal that I only see my own score) to beat I still feel motivated to play. It's the same in Canabalt I just want to beat my own score while listening to really cool music.

    Broforce. I've beaten the levels. Jamestown, one of these days. Ones of these days I will beat you on judgement you hear me!

    Oh ye Jamestown forgot to talk about how it does amazing progression in mastery since at certain points you have to increase the difficulty to continue. And you will die A LOT. But you learn there is a certain fun in dying, and you accept that.

    Actually if there is one thing I would want in Broforce is that when your player avatar dies nothing spectacular happens. Limbo did dying great since there was squishy death sounds and all that stuff Broforce you just kinda fall down. Canabalt has some humours quips when you die in interesting ways. Also make changing bros cool: So your new bro high fives the old bro with such awesome that the old bro is blown away.

    tl;dr IMO the best driver for a player to continue playing a game is to give them a challenge where they can better themselves. And make dying cool, and more high fives.


    [edit] Looks like someone else posted, and I should check the forums more often. Exams suck.
  • edited
    @Rigormortis My understanding regarding "intrinsic" and "extrinsic" motivation is that it revolves around the question of self motivation.

    So fighting a monster in Diablo 3 starts off being intrinsically motivated (because killing monsters is cool) and ends off being extrinsically motivated (because killing monsters doesn't get more interesting but loot drops acquire a lot of value for players).

    The action of picking up loot is intrinsically motivated, but killing a monster to get loot is extrinsically motivated. http://en.wikipedia.org/wiki/Motivation#Intrinsic_and_extrinsic_motivation

    (It's blurrier than that of course, and my feeling also is that when it still part of the game it feels more meaningful and more intrinsic, whereas points and achievements almost always feel extrinsic. Jesse schell has some of the best thoughts on it I feel. http://www.gamasutra.com/view/feature/134594/peering_at_the_future_jesse_.php?page=1 )

    Though I think "External Reward" and "Extrinsic Motivators" need to be distinguished. An external reward (like an achievement) is an extrinsic motivator, but a loot drop in Diablo is not an external reward (exactly). I think?

    Anyway. I never disputed that the depth of the gameplay in Broforce keeps the game enjoyable well after the first play through (I've played it a bajillion times), and neither did I dispute that that depth doesn't compel players to play more on its own. Just that I believe there is an additional Skinnerian effect as well as a lot of intrinsic motivation. In the OP I didn't really think the intrinsic motivation was worth mentioning only because that's pretty easy to anticipate (the game is derived from Spelunky after all, so what is true in Spelunky should be true in Broforce).

    Super Crate Box is far easier to analyze in terms of a possible lingering Skinnerian effect (most of its content is unlockable). And while I'm certain Super Crate Box still has interesting gameplay at the 1000th play, I don't think either Super Crate Box or Broforce could achieve that without the unlockables. I'm fairly confident that what is true here in SCB is also true of Broforce (though I'm still trying to work out how exactly it functions).

    Anyway. I think I've covered this well enough. Seems the power of Skinnerian loops is doubted by some. Or at least, no one seems to be able to discuss them without ending up talking about mastery, exploration, relatedness etc. (disappointed face)

    I mean, I'm not trying to undermine the value of good gameplay. No-one here is. So why all this standing up for it?

    [Edit] Ack! Karuji wrote a huge post before me!

    I mean, I understand, this community doesn't favour meta-games, but I feel a little bit like people here are undervaluing them by taking a gameplay purist stance, and are doing so at their peril.

    Why when I say: "You can add compulsion into your game in a way that staggers difficulty and a bunch of other stuff that benefits the core gameplay" does this devolve into a conversation about "IMO the best driver for a player to continue playing a game is to give them a challenge where they can better themselves. " (And I'm sorry @Karuji if this seems like an attack, but that goes without saying).

    Are we unable to talk about compulsion?

    If we do want to have a discussion about the many many awesome benefits of intrinsic motivation then maybe this thread should be split?

  • edited
    @Karuji: I feel like you're missing a key element of SCB, which is actually a Skinners Box type reward cycle: Picking up crates changes your gun. Sometimes that's good, sometimes that's very, very, very bad. The chaos that ensues is an intrinsic reward for an extrinsically mandated action. You could just sit in a corner with the minigun for hours, but then the game tells you that you're not performing well, pushing you to get that next crate, which then changes your gameplay. The random reward/punishment of the contents of that crate are a huge driver in SCB's success, way more-so than the unlocking of new guns/levels over time.

    The new guns and levels add wrinkles to short term gameplay and make the whole thing juicier. Like salt makes a good meal better, but doesn't make shit taste great. That's what MSG is for...

    -Edit- post @BlackShipsFilltheSky post: I want more people to experiment with metagame stuff. We'll get a comp that pushes that eventually. Maybe you'd run it? ;)
  • edited
    @BSFtS I owe you a hug! You hit on what I was trying to get as with the bros and weapons. So while the bros may be an extrinsic reward they are an internal game component. Achievements are an external game component. So if the extrinsic motivator is a further internal working of the game, and if that additional component can be used to enhance gameplay and mastery it is a good thing for the meta game.

    I don't believe that anyone is simply anti-meta game, but more of a with what we know we know to be careful of these type of things. I mean the the skinnerian loop in Diablo 3 is a lot like that in Diablo 2, but the meta component does not seem to be as satisfying.

    Also I'm trying to add addition things which make the game compelling besides a psychological conditioning. Challenge mastery forms a similar kind of meta-game and game prolonging effect.

    Edit: ack @dislekcia post:

    I don't think I missed that since I stated that there was an explicit driver of needing to pick up boxes to progress in SCB. Since you don't progress by killing enemies but by picking up boxes. We could all post a really detailed deconstruction of the gameplay in SCB.

    I was focusing on the effect of self challenge being a driving force in SCB that you naturally want to beat your own best score. But I fully agree on the effect of the box pick ups. I was actually trying to show that without mentioning the word Skinnerian or something related.

    Also I totally love the salt and MSG metaphor and I hope you don't mind if I use it when I describe achievements in the future.

    Edit edit: on a meta thread level has anyone noticed how often it seems to happen that people will post something only to go argh x person posted while I was typing?
  • You guys are building this city on rock & roll.

    If rock were wall & roll were text :)

    Interesting discussion, I'd just like to point out that using anecdotal evidence (especially "I did this and I would do that") is really the least evidential of evidence. If we could put out two identical games at different audiences who didn't know about the existence of the other game, and one had Skinnerian loops (rewards) and the other without, then maybe we can conclusively say something.

    But even without that experiment, I think the experience of game devs who has put out games WITH loops can also tell us that it's valuable and that it works.
  • @Tuism I would ask you to clarify the context of skinnerian loops before I would posit and answer. 'Cause the term Skinnerian Loop has been used in about 3 different contexts in this thread, and yes I and guilty of that :P (On a side note when I write articles they are insurmountable walls of text so all this is light typing and reading ;) )
  • edited
    @Dislekcia Hey. I'd love to try run a Skinnerian comp.

    Of course I do feel like I'm learning this stuff still. I feel like there's a lot of techniques/implementations to creating compulsion that ALSO improve the game (as apposed to overshadowing intrinsic motivation) that I'm not aware of (and there probably are some radical implementations that no one has tried yet). So, super, amped.

    @Dislekcia I really love that compulsion loop in Super Crate Box. I guess that's a part of why, despite a lot of complaining from players, bro selection in Broforce is still entirely random. Though there are a host other reasons, and a lot of considerations of alternatives. In the face of players and colleagues pushing for more player control over bro selection I generally have just argued "But we want to encourage playfulness", but now that I'm fully aware there is a Skinnerian loop there I have a stronger argument (possibly the one I was clutching for but didn't have the words). So much thanks for bringing that to my attention!

    [Edit]
    Although, I'm suddenly wondering about it's Skinnerianess... or at least how the brain processes it... Changing bro in broforce is not the reward. The reward is resupplying your special attack and either a free life or rescuing your buddy. The reward then isn't random. But getting the reward changes your abilities in game in a random way (just like Super Crate Box). I wonder if the player conflates the reward with the randomness (and as such feels more anticipation/compulsion because of it) or if the randomness rather functions to make the game more interesting (and motivate the player through mastery or exploration etc) (or both, especially because some bros are so wonderfully imba that they can only be considered rewards, like Blade).

    It's kind of like a Skinner Box where the pigeon pecks the button and a piece of food drops out the hatch AND the pigeon becomes an ostrich. So the question is: Does the possibility to become some other kind of domestic bird motivate the subject to press the button even faster?

    And: Does this work on humans?

    (I suspect so because look at SCB).
  • @BlackShipsFilltheSky, you found the article I was talking about earlier. It was this talk that I listened to. That was what I found in my limited research about skinnerian loops. I looked at the definitions as well and will try to use them in that context. :)

    Also, I'm not against compulsion loops(in case it seemed like it). I think they are a great way to reward the players for playing the game. I think that rewarding your player for playing with more things to play with is great.

    I agree that there is some type of lingering compulsion because of the unlockable content in BroForce. But I have to ask, is this something that worries you, or would you like to extend it's effects? Not trying to judge or anything, just trying to figure out exactly what we need to get out of this discussion.

    My understanding has grown while typing this post as I reread everything discussed. So most of the stuff I set out to type has basically become useless :P I think this is a good discussion to continue. Hopefully we can find some more articles and stuff to post here so we get better at this.
  • @Rigormortis My understanding is improving too!

    I'd definitely like to expand whatever compulsion effects there are in Broforce. And if possible make them so freakin compelling players won't be able to stop (ever).

    But I want to tread a line where the rewards (that cause the compulsion) add to the game and don't make the moment to moment gameplay more grindy in the slightest.

    So I'm trying to figure out how to do that.

    Dislekcia mentioned possibilities for rewarding players for exploration through non-linear (or secret) level progression. It's not exactly Skinnerian, although if the player stumbles upon a special exit to a level it might be (especially if that exit wasn't always there or something). That kind of system only adds to the game and makes players more curious (and more willing to play again). (Although lots of non-linear content is a lot of work obviously).
  • @Karuji errrrr I think I'm talking about the loop of rewarding people like in Super Crate Box, seeing as that's the example being used around this topic? My point was that the game designer tries to capture what the most people would want from the game (unless they specifically state otherwise, games are usually made for others to play and enjoy after all) and so anecdotal evidence from the few that goes against the majority are less useful.

    That is all :)
  • The unlocking of weapons in SCB is a fixed reward schedule. The changing of weapons on create pickup is a skinnerian loop.

    The fixed reward schedule serves a dual purpose of enticing players to discover new things, and gating the tasks, and knowledge, required of the player while they are starting to play the game. I think about it as a form of tutorial as game. Since you are actively playing and discovering you don't really notice you are learning. There also may be a sense of psychological conditioning which causes players to continue playing past a certain point. Although from personal experience I am really hesitant to give this much weight game which use exploration as a key aesthetic grow dull once the exploration has ended unless they have other key aesthetic.

    A Skinnerian loop, by the looks of things, seem to be most effective in short term rewards or moment to moment game play. This is the reason for the one more turn in firaxis games. There is a constant uncertainty, and a looking to see the return of previous actions. Game Dev Story is, to me, the ultimate in this Skinnerian loop being integrated in to short, medium, and long term game play. Also due to the human nature of optimization there is a constant flow of loops which allow.

    The skinnerian loops feels like it is something more on a mechanical level where a fixed reward schedule feels more like it like it is on the aesthetic level. Although I would love to see someone research the Skinnerian Loop, Fixed Reward Schedule, the time frame of them, and present the answers in the using terms from the MDA framework. I would also like people to stop using the word genre when the mean style. And neither of those seem likely :P

    ---

    @BSFtS

    If people never stopped playing Broforce then how would they tell other people to play Broforce :P

    Also looking at Broforce I have to ask is unlocking new bros really that compelling? As I said before it's not even essential to the progression of the game.

    Personally I think something like the difficulty progression in Jamestown would be of greater benefit to Broforce. I mean a great deal of reason why I go back to Jamestown, SCB, Canabalt is because it is stupidly fucking hard. Jamestown because I have yet to actually finish it, and the other two because I want to beat my own score.

    I'm not saying don't or skimp on the meta/exploration component. But you rock at making action games so why not jam as much action and difficulty in ;) it would bug the crap out of me if there was a level in Broforce I couldn't finish.
  • edited
    Unlocking bros isn't essential to the progression of the game... Catching pokemons you haven't caught isn't even essential to the progression of the game either :) Different strokes for different folks, the extreme difficulty of Jamestown makes me meh about it while catching all the Pokemon has kept me up many nights.

    But I'm not everyone, and neither are you! (but I *am* saying that Pokemon was wildly successful) So I think all forms of reward have their place. It depends on which bunch of people the game are catering for... Heck it's usually the creator themselves most of the time :) And there's nothing wrong with that as long as the creators take care to venn diagram themselves with an acceptably sized audience (according to their research and only if they're trying for commercial success)

    It just bugs me that anecdotal evidence from a person is being flouted to advise how a game should be made, a game that would be played by many many people :)
  • @Tuism I am very sure that I have stated that these are just my opinions and experiences. And given that @BSFtS have had a number of design discussions before I believe he would understand what I am saying since my comments were directed at him.

    Pokemon is a far more diverse game than broforce. You can play it for story, competitive play, or just collecting everything. The notion under which I say the collection of bros is not necessary for the progress in Broforce was likening it to the collection of crates in SCB, which was the progress metric for the game. I did this to show that there is a potential flaw in the unlocking of bros being as deep a driving mechanism for the game. Essentially I am being the devil's advocate here.

    The reason I mentioned Firaxis, Game Dev Story, SCB, et al is that they all have a form of Skinnerian loop as a deep core mechanic of their game. The core mechanic in Broforce is shooting people. The saving bros is secondary to that. And the unlocking a meta component to that. Now what that should not infer is that I think that the bro-change (we need a brophrase for saving bros) is bad. When I put Broforce in the MDA framework I stated that exploration is one of the core aestetics. What I simply posited in the last post was that @BSFtS focus on combat as a driving force of meta exploration. The reason I did that is for me the confrontation and challenge were a major part of the enjoyment of Jamestown, and given that both can be played in co-op and have a similar development team. (Pokemon is an order of magnitude larger than Broforce.)

    Given that this thread is about the about the 'Skinnerian Loop' it would seem odd for @BSFtS to post about the combat progression. The reason I went on about the difficulty is because I know @BSFtS is really gifted with action games so suggest difficulty plays to his strength where he has stated that this meta-game is new territory so why not posit some advice. It is completely up to him to take it, leave it or make a funny picture out of it. And he is a good enough designer for that, I've told him before I'll be harder on him than just about anyone else because his games can be worse than crack (I assume crack is something like when you play wow for 48 hours straight.) And I want to see Broforce as a bastion of what can emerge from the community.
  • edited
    @Karuji Is SCB weapon unlocking a "fixed reward schedule" as you put it? I presume by this you mean "not a Skinnerian Loop"?

    But the weapon unlocking goes like this: Collect A Specific Number of Crates -> Receive Random New Weapon -> Repeat. (Apparently a "Fixed Reward Schedule")

    How is that that different from: Pull Lever -> Receive Random Amount of Money -> Repeat. (Obviously a Skinnerian Loop)

    Both involve the repetition of a task to receive a type of prize (and both involve a random quality of that prize).

    I'm not sure I like this term "Fixed Reward Schedule", it seems to have made things more confusing.

    Remember that Skinner didn't just come up with the Skinner Box. He came up with what he called "Radical Behaviorism" of which the concept of "Reinforcement" was central. The idea is through positive or negative reinforcement (from a source outside the subject) you teach the subject to behave in a certain way. http://en.wikipedia.org/wiki/B._F._Skinner#Theory

    Loot drops in Diablo reinforce killing monsters, saving bros in Broforce is reinforced through a random chance to become Blade. And I've been arguing that playing longer, and picking up crates, in SCB is reinforced by the large number of crates you need for each unlock.

  • @BlackShipsFilltheSky: I'm pretty sure that the weapon unlocks are a fixed reward schedule because they tell you how many crates you need to get to, plus it's not a random weapon, the progression through the game's weapons is exactly the same each time you start over. It's a designed sequence.

    The unlocks actually feel like a very strong counterpoint to the randomness of each crate pickup (which to me is the actual skinnerian loop element in the game) because they progress no matter what, you don't lose those when you die (well, you do for the "score X crates" unlocks, but those are more for levels and modes than weapons) which means that the fixed weapon reward schedule helps keep the game meaningful when you're dying a lot, plus it directly adds to the random crate reward system by making one of the possibilities per pickup slightly cooler because it's new.

    Personally, the way that you have to go and pick up a crate to get the new weapon you've unlocked, instead of simply starting with it, is genius. It ties both loops together brilliantly.

    P.S. Note that I'm not saying that there are specific codified ways to do this sort of stuff or implement skinnerian loops in a game, I'm just analysing what worked in SCB so that I can understand it better myself. Sometimes things work in games because they're the opposite of what worked before, games carry far more context than almost any other medium, so I feel like we're never going to have a "simple" set of design rules around pretty much anything to do with them. It's always going to be fuzzy and exploratory, which is part of the fun ;)
  • [quote = dislekcia]It's always going to be fuzzy and exploratory, which is part of the fun[/quote]
    This is how I feel as well. I think something that hammers this home is when you look at definitions for games. In books/articles I have read most people usually say that the definition they use is relevant to the current content and might vary from person to person. Which I always found to be both strange and totally sensible at the same time.

    I really need to play super crate box, dont I? :/
  • edited
    Dislekcia said:
    I'm pretty sure that the weapon unlocks are a fixed reward schedule because they tell you how many crates you need to get to, plus it's not a random weapon, the progression through the game's weapons is exactly the same each time you start over. It's a designed sequence.
    It doesn't tell the player what the next weapon is making it effectively random. The internal workings of the mechanic are not what the player sees. The fact that "the progression through the game's weapons is exactly the same each time you start over" is entirely irrelevant surely? Players don't start the game over very often. Players don't have prior knowledge of the order.

    Yes this means SCB is less compelling the second time through... but... so what?

    I'm not even debating whether its a "fixed reward schedule" or not. Yes it is. But again, so what? A "Fixed Reward Schedule" can be compelling. Surely?
    BlackShipsFilltheSky said:
    Collect A Specific Number of Crates -> Receive Random New Weapon -> Repeat.
    Which is a very effective compulsion loop. To deny that is just plain ridiculous.

    Is the problem here that I'm using "Skinnerian Loop" instead of "Compulsion Loop" or something?

    Surely if someone says something is not Skinnerian that also means it is not compelling? Surely "Skinnerian" means that its root is not in "Behaviorism", the science Skinner pioneered. (That's what's confusing me here, where did the word "Skinnerian" acquire the "Exactly like a simplistic Skinner Box in its mechanics" definition? I don't see the word being used like that outside of MakeGamesSA).

    http://www.gamasutra.com/view/news/39341/Opinion_Compulsion_Loops_In_The_Short_Medium_and_Longterm.php#.UJ6NRobxaUk
  • @BlackShipsFilltheSky, I think the difference between the "fixed reward schedule" and a "skinnerian loop" that @dislkecia and @Karuja are trying to get across is that the skinnerian loop has random rewards which made it more compelling than a fixed reward schedule. I might be wrong though. If I am please correct me.
  • Yeah, when I talk about the skinnerian-ness of something, I'm particularly talking about the involuntary dopamine dumps that go with performing the action that has a chance to trigger a reward, not the reward itself. That's the key difference for me. A scheduled reward, even if you don't know what it is, is far less compelling at the raw brain chemistry level.

    When you pick up a crate in SCB you get multiple good feelings, one is from knowing that you're getting closer to the next scheduled unlock, another is from your current score increasing, yet another is the involuntary skinnerian response to the random change to your weapon. It's a multi-layered thing.

    For me "skinnerian" means a repeated action with a random chance at a reward, a "compulsion loop" could be a different kind of compulsion. A skinnerian loop is a kind of compulsion loop, so layering together many types of compulsion, forming a compound compulsion loop, is something that is eminently possible.

    P.S. I'm now seriously considering writing about designing around randomness as a way to respect your player. If anything can happen while playing, you have to design much better and more robust systems to allow players to survive, yet as those get more complex, you're acknowledging that your players are skilled enough to plan through and with that randomness. If you've got a 100% designed, predictable interaction, your players can be dumb as bricks because there's only ever 1 thing they need to do.
  • edited
    @Rigormortis I think you're right. But it seems a specious definition.

    I mean, outside of this conversation, would any person understand "No that's not a Skinnerian Loop, that's a Fixed Reward Schedule."

    And it doesn't even seem a useful distinction:
    A "Fixed Reward Schedule" describes the rate at which a player receives rewards (which is useful in planning how players receive rewards).
    A "Skinnerian Loop" describes the effect of a game loop on a player (i.e. A Skinnerian effect).

    It's like saying "No he cannot be tall because he is orange".
    Dislekcia said:
    For me "skinnerian" means a repeated action with a random chance at a reward
    I agree.
    BlackShipsFilltheSky said:
    Collect A Specific Number of Crates -> Receive Random New Weapon -> Repeat.
    Though I'd contend that "Skinnerian" doesn't imply random. Just that random is better at compulsion.

    [Edit, Dislekcia beat me to it] Ack.
    I'm not being constructive here.
    I think we only disagree in the terminology here (guilty face). I'll just read your posts baring that in mind (hopeful face).

    I absolutely love randomness. I know I've linked this a million times, and its just a basic introduction, but: http://www.wizards.com/magic/magazine/article.aspx?x=mtg/daily/mm/37

    I'm sort of still getting my head around: random rewards > fixed rewards. I know it's true, and I know as a player how that feels better, but there's also this magic line where if things are too random it's just disorientating. Why wouldn't things just get more and more rewarding?

    (Obviously it's about players desiring to figure out useful patterns, and elements of those patterns shifting giving players partial information to use (which is an interesting problem if done right), and the randomness is facilitating mastery etc... But it still feels more intuitive for me than something I can express as a set of rules... so I'd be very interested in a thorough randomness article/talk).

  • BlackShipsFilltheSky said:
    I mean, outside of this conversation, would any person understand "No that's not a Skinnerian Loop, that's a Fixed Reward Schedule."
    Probably not. But we are in this conversation right? :P

    I agree with you when you say that Skinnerian doesn't imply random. But I think that is why someone suggested that we call it a fixed reward vs skinnerian loop. Maybe it would have been better if we said that there are two types of skinnerian loops: The fixed reward loop where you are always guaranteed of getting a reward after a specific action. And the non-fixed reward loop where you are either not guaranteed of a reward, or not guaranteed of what exactly the reward is. The relevant distinction between the two that the non-fixed reward loop is better at compelling the player.
  • Collect A Specific Number of Crates -> Receive Random New Weapon -> Repeat.
    But that's not a random reward chance per crate pickup, which is the key element in skinnerian loops/systems to me. The whole point of skinners boxes and all that is that you don't KNOW when a reward is going to come, so your brain actively incentivises the activity that has the potential for delivering a reward. You actually need non-rewarding instances of that activity for the skinnerian stuff to trigger, that's fascinating shit!

    A fixed reward schedule, even if you don't know what the reward is, will always come to you after X actions. Especially when it tells you that explicitly (which I'd call a ratcheted reward, because you can't really "fall back" in your progress towards it) doesn't trigger the same brain activity - the action is largely irrelevant because we're looking forward to the reward. So the first thing you need to do is establish some sort of scale of relevance for the rewards so that the player wants them. That's quite different in implementation to initially disappointing the player with a commonly performed action. So I think the distinction is quite relevant from a designer perspective, at least.
  • maybe this link might help clarify some things?
  • edited
    Dislekcia said:
    The whole point of skinners boxes and all that is that you don't KNOW when a reward is going to come, so your brain actively incentivises the activity that has the potential for delivering a reward.
    Skinner boxes also work without the randomness. http://en.wikipedia.org/wiki/Operant_conditioning_chamber

    @Rigormortis Yeah, a Skinner Box with "Continuous reinforcement" would still be a Skinner Box, just not the best kind.
    Dislekcia said:
    You actually need non-rewarding instances of that activity for the skinnerian stuff to trigger
    Are you sure about that? http://en.wikipedia.org/wiki/Operant_conditioning#Factors_that_alter_the_effectiveness_of_consequences
  • @dislekcia So I've just started following this very deep conversation, but what I got from all of it is pretty much what you have summed up here. This made me wonder about something though. If I have a game where some action of the player is rewarded randomly, say fishing for instance, but the player can do it as many times as he wants, is this still considered a Skinnerian? In the example, the player can fish for a while and not get anything, yet he knows if he keeps going at it he will get something. I was going to say I don't think it is, but now that I've put it like that, it sounds exactly like a Skinnerian
  • Like I said way earlier, I'm probably misusing some of the nomenclature, but I don't have a distinct name for the dopamine-dumping effects of a Variable Interval reward schedule, other than Skinnerian Loop.

    Skinner's stuff isn't actually what I'm referring to either, so I should probably just stop using the name altogether.
  • edited
    I think we only disagree in the terminology here (guilty face). I'll just read your posts baring that in mind (hopeful face).

    I think also my perspective is coloured by myself being less susceptible to Variable Interval reward schedules. I stopped playing Diablo 2 and Diablo 3 right after I'd finished the story. I stopped playing Super Crate Box right after I'd unlocked all the weapons. New content excites me, the chance to get something mildly useful (like loot or a little money from a slot machine) doesn't as much. Though I'm not suggesting that Variable Interval reward schedules are not astoundingly powerful.

    Though I absolutely love randomness. I know I've linked this a million times, and its just a basic introduction, but: http://www.wizards.com/magic/magazine/article.aspx?x=mtg/daily/mm/37

    I'm sort of still getting my head around: random rewards > fixed rewards. I know it's true, and I know as a player how that feels better, but there's also this magic line where if things are too random it's just disorientating. Why wouldn't things just get more and more rewarding?

    (Obviously it's about players desiring to figure out useful patterns, and elements of those patterns shifting giving players partial information to use (which is an interesting problem if done right), and the randomness is facilitating mastery etc... But it still feels more intuitive for me than something I can express as a set of rules... so I'd be very interested in a thorough randomness article/talk).

    @Denzil I think you answered your own question. I know those fishing games are very addictive, so that loop must be very compelling.
  • dislekcia said:
    Skinner's stuff isn't actually what I'm referring to either, so I should probably just stop using the name altogether.
    Yeah, bad @dislekcia. :P

    Seriously though, when I think about something like Super Mario, I feel that there are so many "mini skinner boxes" in the game. Mini in the sense that it addresses on thing in the game. Like when you jump over gaps, you don't die. When you jump on goombas, they die. When you hit a block, you might get a coin. I feel that all these things are skinner boxes even though people would probably describe them as good game play.

    I think games are inseperable from skinner boxes, because it's the main way we teach our players to play the game. How you would use that to your advantage, I'm not sure yet.
  • @BlackShipsFilltheSky, I actually have an answer for the too random thing, apart from the recognizing patterns. :) I think when it comes to the point that the randomness feels hopeless, that is when the player loses engagement. Most games that are competitive are based on choices you make. If you hoose to pick up a random weapon you at least control when you pick up the weapon although you don't know what you will get. But if you get a random weapons at random times. You are not left with any choice. And the whole game can basically be reduced to each player rolling a dice and the highest value wins. I say competitve games because I read a lot of Maro's stuff too and he usually writes from a MTG perspective.
Sign In or Register to comment.