Skip to content

Skinner and Obsolescence

October 16, 2013

Dear Reader,

There are several shared discussions going on right now around the blogosphere.  One I’m particularly interested in is taking place between Balkoth of his eponymous Word and Rowan of I Have Touched the Sky.  While their discussion is far more in depth about game design, a subject I like to tinker in but know nothing practical about, I found at the core of their disagreement a topic of much interest: the relevance of gear.

Obviously, we can all agree that from a psychological perspective, getting gear is about rewards.  However, what does that mean when you don’t get a reward?  In psychology, there are four basic “motivators” discussed in operant conditioning, made famous by B.F. Skinner and his various boxes.  We hear all the time how MMOs are like “Skinner Boxes,” meaning that you hit a button and get a reward.  The proper term for that reward, though is a positive reinforcement.  The positive portion indicates that the subject is being given something.  The reinforcement indicates it’s meant to increase a behavior.

The problem with viewing loot in this very outdated psychological model is that the other three types of conditioning are handled so badly.

Negative reinforcement is taking away something unwanted to increase a behavior.  Cleansing debuffs is an example of this.  When a player has a crippling debuff, their encouraged to remove it.  The problem with this is that it’s often not on the player in the role of dps or tank to deal with this.  As a result, only the healers have to increase their behavior.  Standing behind a boss is another example of a negative reinforcement.  Doing so removes the block and parry chance, so moving behind the monster is reinforced.  Again, this only affects one of three roles.  Since the “negative stimulus” is often only a problem to one-third of the roles, the responsibility for dealing with them – the behavior to be increased – is not shared equally.  That ends up creating “dps,” “tank,” or “healer” fights, which, while there’s nothing the matter with asking one role to push extra hard, can still breed animosity in a raid team.

Positive punishment can easily breed the same animosity.  Positive punishment is giving someone something that they don’t want (a spanking or a restriction) to reduce a behavior.  Many of the game’s positive punishments come in the form of avoidable damage.  Again, this does not share responsibility equally.  Ideally, if you receive damage, you should want to avoid whatever caused it in the future.  However, as we’ve seen in LFR, often what dps or tanks will do is simply rely on the healers to deal with it.  The behavior isn’t really changing, then, as the punishment is actually affecting someone (a healer) who cannot control the behavior of the other players.

This often occurs because of the nature of negative punishments, where you take something away from the subject to reduce a behavior.  An in-game example of this is interruption of cast or channeled abilities from movement.  This discourages moving around a lot so that your opportunity to do more dps is more often available.  By discouraging movement, they’re creating the problem of people getting tunnel vision and standing in fire because they’ve been taught that moving is somehow worse than standing still.  Since the immediate outcome of standing in fire is that the healers have to do more work, the dps just reward themselves by standing still while punishing the healers.

The new loot system interacts poorly with the Skinner model of operant conditioning, as well.  If you’ve taught people that killing a boss (a behavior to increase) provides a reward (a positive reinforcement), then start making killing a boss only maybe provide a reward, that sends a very confusing signal.  Research tells us that this variable positive reinforcement is actually a more powerful force than always providing a reward, but to many players who’ve come to expect the other, it feels much more like a negative punishment; that something that we want – loot – has been taken away.  The result, of course, may be a decrease in behavior.

All of these “bad” Skinner designs breed animosity between “team” members or confuses players’ existing expectations, both of which hurt the overall experience of playing the game.  So I propose, then, that gear as a reward, while an obvious view, isn’t the intentional or, at least, isn’t the only view of gear.

Balkoth comes into this because he mentioned in the above linked comment that gear was “to nerf content.”  I liked that thought immediately, though I want to take it one step further and say that it’s actually to obsolete content.  While I see and understand from a design point of view why this is necessary, I don’t like it, even though I think it’s accurate.

From a design standpoint, it’s necessary to obsolete content for two reasons, both of which mostly relate to hardcore players.  First, if there was no obsolete content, the competitive element of cutting-edge raiding would be lost.  If everything’s fair game, then nothing’s really “cutting edge” and the competition will be too spread out among different activities (raid zones).  Secondly, it allows those same hardcore raiders the chance to not play every hour of every day.  If even a full expansion pack of content is “relevant,” it’s going to encourage hardcore players to do all of it all the time, which will undoubtedly burn them out more quickly.

So gear obsoletes content both as a marker – “this gear is beneath what you now need, so stop raiding here” – and as an decrease of difficulty – “This content is no longer challenging due to your gear, so it’s time to move on.”

What bothers me is that there’s an enormous amount of obsolete content that must come to years of man (and woman) power just sitting collecting dust.  Blizz has finally realized this, too, and has added pets and transmogrifying to encourage people to run old content, but that’s still a small subset of players doing only a few bosses and raids.

Previously, they dealt with the obsolescence issue by updating old content into new modes, but that rightly met with a lot of resistance, and I hope they stop it, frankly.

Instead, why not use old content as part of the gear ladder?  Right now, you can skip all the dungeons (or all the zones) towards the end of each content pack because the next area’s gear will be so much better with little more effort (or even less, in some cases).  Instead, a rebalancing of that gear distribution where the end-of-pack gear is significantly better and the new content pack is harder would encourage people to do all the dungeon content.  Old raids could be designed to act as leveling dungeons (much like they did with Upper Blackrock Spire).  This encourages people to enjoy what’s already there without any incentive to skip over it while also preserving its place in the story of the game (unlike what they did to ZG and ZA).

So, I hope Blizzard starts to really consider how to deal with the problem of obsolete content, and while they’re at it, straighten out their Skinner formulas, too.


Stubborn (and thoughtful, for once)

16 Comments leave one →
  1. October 16, 2013 4:23 pm

    The only problem I see with using end-expansion dungeons and raids to gain, essentially, required gear for the new expansion starting zone is that you are basically locking new characters out unless they have friends or guildmates who are willing to run them through the content. Speaking as a tank I’ve done a lot of that sort of thing in Outlands, Cataclysm, and Pandaria dungeons – but only because the people being run wanted to nail the instances while they still paid good quest XP. If they had to run each five or more times just to get the gear to do the next zone I think we’d just form a group and let the over-geared 90 destroy everything.

    I’d much rather they applied the Flex technology to the old content to scale the dungeon/raid and awarded Valor for each boss downed each day or week. That would keep people running them as leveled-up content without needing to invent new gear. Of course, make it a switch to run at character level or run at the original level for those of us who are just looking for the pets.

    Personally, I’d rather a system that improves character power but isn’t gear dependent. I think that would add a lot of interest in crafting (especially if all gear degrades with use) and would make things feel more organic. In most games it feels like most of my abilities are the things I carry and not the things I can do.

  2. Samus permalink
    October 16, 2013 7:30 pm

    “Obviously, we can all agree that from a psychological perspective, getting gear is about rewards.”

    Actually, I couldn’t get anyone to agree on that. But I’ll let you fight that battle this time.

    As to your post, I think you are missing the main reasons behind why Blizzard went to making content obsolete patch to patch rather than expansion to expansion. Primarily, they needed to have everyone doing roughly the same content so it would be easier to find a guild. The other thing is, what happens with new players who weren’t around when the expansion first started? All the raid guilds are long since done with the bottom tier.

    Seeing content was the other big thing. During the Burning Crusade days, fewer than 1% of raiders (at a time when raiding was far less popular) had killed Illidan when they released the Sunwell stuff, fewer than 1% finished the Sunwell by the time WOTLK came out. etc. Many players complained (not unreasonably), why are you focusing most of the patch resources on content almost no one ever sees?

    The reality is that as long as raids require at least 10 players, they need to push everyone into the same raid tier. The population just isn’t large enough to spread across multiple raid tiers. So one option is to offer 5-man versions of raids, allowing you to keep more raid tiers as “viable.” The reaction would be similar to the introduction of 10-man raiding for all tiers, some would scream and cry about how it was the death of raiding, while the change would be overwhelmingly popular amongst the vast majority. Unfortunately, I think there are a great number of bosses that are just too complex to easily scale down.

    Another option would be to have some kind of monthly bonus old school raid, in which you get some sort of special reward for completing. But I think it is more work that people realize to scale past raids up to current item levels, and I think Blizzard would argue that time is better spent just making new raids.

  3. October 16, 2013 8:02 pm

    Reblogged this on Blog about games and commented:
    on skinner box loot and rewards

  4. Cain permalink
    October 17, 2013 8:46 am

    The other thing that happened during BC when you had to run all the past content in order to be able to run the current one was a major pain in recruiting. If you were working on BT it sucked to have to run a new person through previous content to be able to bring them to BT. You spent enough time in those raids when they were current, that once you cleared them you wanted to move on and not look back. In Vanilla we still ran MC almost the entire expansion because of legendaries. We still ran BWL and AQ for a long time even when Naxx was out. All of that didn’t take that long because we outgeared it, but it still cut into the time we had to work on bosses in Naxx. Wotlk finally had less of needing to go back to old content, but it still had some due to some really good items in Ulduar. WotLK had the problem of needing to run the same raid multiple times per week though which just made burning out on the current one that much faster. Running ICC 2-4 times per week instead of 1 was just brutal.

    I think the selective scaling that they are working on is going to be a nice change so that you can choose to run old content as if it was current without having to.

  5. twilitsoul permalink
    October 17, 2013 10:50 am

    I think tying health to damage would be a very interesting experiment. A “wounded” mechanic might finally make dps care about safety.

    • October 18, 2013 1:34 pm

      I agree, but I’m afraid Blizz would never so radically alter their game play so much. Better interrupt systems, like on Dark Animus, rather than damage, might be a solution, or greater pushback (which used to be the case; you’d never get a spell off if you were taking damage) may be a solution.

      Thanks for the comment!

  6. Beshara permalink
    October 17, 2013 11:20 am

    I personally like the flex loot system. It avoids guild loot drama, and it does seem to give you gear about the same rate as normal would, except the gear drops you get will be tailored to your spec of choice. Granted, you may get duplicates, but you won’t be stuck with a piece that your class can’t use. The only thing I don’t like about it is that you can’t share duplicate drops with others in your group, like you can on Normal. I have luck with weapons but not trinkets, and my husband has had the opposite. But this comes from a person that has seen guild groups go weeks without upgrades in Firelands due to unlucky duplicate or non useful boss drops.

    The positive and negative punishments described definitely need some change. We saw healer mana pools become a flat amount due to raiders relying on healers to heal through avoidable mechanics. That doesn’t seem to have fixed the problem, because the people receiving avoidable damage are not the ones dealing with the consequences. A healer could let that person die, but the raid group suffers that loss of dps from death, and the blame game starts. Avoidable mechanics need to have consequences for the player hit, not the team’s healers. Not just one shot mechanics though. Either placing a debuff on the affected player, like the miss chance debuff from Putricide, or even granting a buff for moving the right way, like the last boss in MSV. Examples do exist, but not enough for that behavior to become the norm.

    I would love to see the old dungeons and raids become scalable 5 mans or even scenarios you can run for your weekly valor. You see the tech Blizzard has with challenge modes, proving grounds, and flex raiding. Certain fights would need tweaking, but I think they have the potential to do it. Like the council fight in Black Temple, you could have an item any ranged player picks up that gives them the ability to range tank the Mage add. It has potential, but I don’t know if Blizzard would devote the manpower to do it.

  7. October 17, 2013 10:10 pm

    You’re brave to pick up the Skinner knife 🙂 But I wholly agree. I think where you say they apply it in the case of gear is somewhat different when we take the other mechanics into account. Or rather how they apply them.

    Maybe it’s just the way the model interacts with others. For example, in my mind gear is about empowering players. This interacts with the things Samus mentioned such as group finding and tiered content in general. There’s a lot of systems intermingling here, all of which try to encourage players to spend more time inside the game. And that’s exactly the extent to which I think gear is employed in this way.

    I think we see Skinnerian models more strongly in the daily quests (or questing in general really), where it’s all about giving players a way to get their reward for the day, guaranteed. Their greatest net effect is to keep players logging in, which in turns keeps the dollars flowing. As a content feature, quests are there to hook you into continuous gameplay.

    On the other hand, gear is there to stratify players into groups based on their relative power. It’s the pursuit of this power that drives the player. The rewards are more like a cherry on top in my mind, though again I do agree with what you’re saying here. I guess I just think this intersection of systems of compulsion is interesting.

    • October 18, 2013 1:33 pm

      I think this heavily depends on the type of player, which I suspect you mean, as well. For instance, the Secret World quests I enjoyed long after I’d outdone the gear I was getting for them. My group and I did every quest we could find in every zone, even after we had our 2 “best builds” and top quality gear. I went back later, too, to play the new issues just to enjoy the stories. While I agree that the quests are about sustaining gameplay, the reward may not be the quest completion but the quest itself. That, I think, is something Blizzard has only begun to explore (the day Deathwing Came, for exampe), where newer generation games are really using this better.

      As long as “kill 10 rats” is a base quest – or heart or whatever else you want to call it – objective with no greater meaning or enjoyment, the reward has to be the completion, but when you have quests like the investigation quests in Secret World or The Day Deathwing Came, then you’ve made something that’s truly fun.

      Thank for the comment!

    • October 18, 2013 6:54 pm

      I disagree that it depends on the player. Daily quest design is skinnerian regardless of player type or reward. Remember, the kind of reward here is irrelevant as long as its desirable, whether gold or a new wand. And to be clear, while I agree gear does act as a carrot on the stick, I think daily quests are more skinnerian and the two models interact in interesting ways.

      To give a stronger example, raiding (gear hunting and conquest) wouldn’t even be in the game if it wasn’t the entire point of the game itself. It’s also the primary source of the best gear so from your perspective its a giant skinner lab. However, raiding “ends” the game and thats contrary to the skinner model making the gear hunt something ancillary, not central, to the “push button” mechanism. The goal for MMOs is continuous, ever-lasting gameplay. Daily quests facilitate this. Raids absolutely go against it. This is one reason I think daily quests are far more skinnerian than raids and much more problematic. The other reason is that most players dont play the gear hunt game (raiding) while everyone quests.

    • October 19, 2013 11:17 am

      Go back and watch the PA video I embedded on my post (Linked in Stubborn’s 1st paragraph). Dailies are no more Skinnerian than your typical 9-5. RNG loot drops, on the other hand, are exactly the sort of thing Skinner found particularly compelling (not in a good way, imho).

    • October 19, 2013 8:48 pm

      @Rowan: I’ve seen the video. It doesn’t negate anything I’m offering here. Skinner model is about operant conditioning and is not the same thing as the use of variable rewards (a type of game loop to compel players to continue playing). Skinner conditioning is about rewards for repetitive actions and specifically, rewards for doing the action as instructed. In fact, my exact point is that these two models (skinner and RNG) interact in interesting ways.

      The reason the skinner model is better seen in the daily quests than RNG gear hunting is because dailies lack the RNG; you push button and get a guaranteed reward. The gear hunt isn’t operant conditioning so much as good ole fashioned compulsion mechanics (the gear hunt is analagous to playing slot machines, not a skinner box).

    • October 21, 2013 9:47 am

      I have no desire to get in the middle of ya’ll’s discussion, and usually I don’t respond to people responding to one another, but I want to jump in here. Skinner and operant conditioning absolutely do relate to “variable rewards,” though he termed them “variable reinforcement schedules,” as both I in my post and Rowan refer to (though I don’t name it; I just refer to research he’d done).

      He found that a variable interval (after the first and then more randomly) reinforcement schedule was a far more powerful force than a fixed interval (every time) reinforcement schedule. He suspected this was because in a fixed schedule, once the behavior was learned, it would only be repeated when the subject wanted the reward, whereas in a variable schedule, the subject would have to continue doing the behavior because it could not be sure when the next reward would come.

      Now I’m not applying any of that to either of your arguments, but I did want to put that out there for accuracy’s sake. I think you two may be talking about different ideas with a similar name (I assume), and not actually arguing about the same thing. Of course, I could be wrong about that, but it seems R was talking about variable reinforcement schedules (variable rewards), and D was talking about variable rewards as a kind of game mechanic. I don’t know if the two are in any way related and will let you folks battle that out.

      Thanks for the comments!

    • October 21, 2013 10:45 am

      Nope, you made my point, Stubborn. That “variable reinforcement” is more effective for operant conditioning was shown in the video mentioned. While you are are right, Doone, that people will log in just to do dailies, it will only last as long as they have a need being fulfilled (i.e. XP, gold, etc.) This falls within Skinner’s observations. But in that sense, dailies are not really much different than regular progressional questing, which also provides for those needs—with the added potential bonus of novel content. The power of variable reinforcement is evident when people are pursuing a reward they have no guarantee of receiving (hunting the firefly in Zangarmarsh, for instance) and then talk about all the “work” they went through to “achieve” their goal.

      I fear that I come across as belittling end-game raiding. I don’t. I know it requires a lot of individual skill and team coordination. But what end-game raiding is not, is something I enjoy doing. I was involved in it for at least a year, total, between TBC and WotLK. I decided it was not how I wanted to spend my evenings.

      I was almost awestruck at the incredible world Blizzard created, when I first stepped out into Coldridge Valley. The novelty and sense of adventure I experienced through Vanilla and two expansions was wonderful. Raiding is not that for me. Repeating content over and over for weeks and months—regardless of how easy it became before we moved on to the next challenge—lost its appeal long before I came to the realization that my raid group didn’t really need me. I was totally replaceable.

      And that was OK, because I moved on to other things, other games. I may be a bit of a gaming nomad now, but I am not alone in my wanderings.

    • October 21, 2013 1:00 pm

      Ok. I don’t think we’re disagreeing. I think we’re pointing out different things. The key word “reinforcement” is the point I was dissecting and I see that wasn’t clear from my responses. I also still think daily quests are much more manipulative, but that’s certainly up for debate and probably splitting hairs (they all have the same net effect).

      The original WoW was definitely very different from today’s WoW as you describe, Rowan, and used these devices in pretty different ways to get players to continue to play …and that’s interesting on it’s own. Lich King was the end of an era and Cataclysm the beginning of a totally different game (literally). It’s almost as though the game today is designed for stronger retention instead of stronger fun. No one is surprised at that though, right?

    • October 21, 2013 1:13 pm

      Cata did change the game. My time playing the open world post-Cata can be measured in weeks. For social reasons, I continued to run lowbie dungeons for another several months until just before SWTOR came out.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: