2014-06-20 Rewarding a Thing

Recently Ian Borchardt wondered on Google+, what people thought of basing experience gains off of class. He was thinking of fighters gaining experience fighting, magic users gaining experience learning spells, and so on. Ian was interested in applying this to D&D solo play. I think the topic bears a wider discussion, however.

I see two things to consider. How will this rule affect gameplay at the table? How will this rule affect what characters do in the game world?

I remember playing Burning Wheel and related games from Burning Wheel Headquarters. These games usually tie advancement to successes and failures in tests. Therefore, every test you make takes a tiny bit of bookkeeping. How difficult was the test? Did I succeed or fail? Jot it down. That’s too much bookkeeping for my taste.

In the games of The Shadow of Yesterday, Solar System or Lady Blackbird, characters have at least one Key. Each key describes a very individual way of gaining experience.

Example Key from Lady Blackbird:

Key of the Paragon As a noble, you’re a cut above the common man. Hit your key when you demonstrate your superiority or when your noble traits overcome a problem. Buyoff: Disown your noble heritage.

Each key also has a buyoff. If the buyoff condition occurs, you have the option of removing the Key and earning two advances, which you can use to buy another Key or two.

Whenever I ran or played these games I liked this mechanic because it gave players the choice to pick whichever Key they desired, implicitly telling the GM what they wanted the game to be about.

As players get to select the Keys and they get to change them as part of their advancement, Keys can be more fine-grained than just the character’s class, and yet they don’t require as much bookkeeping as Burning Wheel and all those games because not every roll of the die needs record keeping. Players actively try to trigger their Key, and when it happens, they mark it off. Easy.

It is pretty free form, however. As the referee of such a game, you should have a list of Keys prepared that serves as an implicit indication of where you see the game going. By agreeing on a set of Keys beforehand, referee and players can make sure that the Keys stay within the kind of game they want play. It doesn’t have to be an anything goes kind of game.

Mazes & Minotaurs has the kind of experience system that Ian Borchardt was suggesting. I’ve never tried it, it seemed reasonable on paper, but I had trouble imagining it at the table. After every encounter, the fighter player speaks up and says, “that was worth 3 XP for me, right?” The mage player loots the lab and says, “two new scrolls found, 2 XP?” If it happens a lot, it could be a lot like the bookkeeping after every test in the Burning Wheel Headquarters games. If it happens rarely, it could be like a permanent Key in Lady Blackbird.

Then again, something I like in role-playing games is changing gameplay over time. It seems to me that making this Key basically permanent prevents this to some extent.

In a D&D game with dungeons and wilderness adventures I’d say that rewards based on class lead to a kind of reinforcement that I don’t like. Fighters will want to fight monsters that don’t need fighting, thieves will want to steal from people we don’t want to steal from, etc.

To elaborate – and this goes for solo as well as party play – I think that doing a thing should not be rewarded. That’s going to go Pavlov quickly is what I’m guessing. What I want is players doing things in order to get a reward, in other words, they are doing something else in order to get a reward. The action and the reward should be orthogonal. Fighters fight in order to get treasure. Wizards cast spells in order to get treasure. This is how they get to choose their approach, quietly or forcefully, quickly or slowly, talking or fighting, and so on. It allows for more ingenuity in my book.

Anyway, this is what I expect to happen without actually having tried it, and only based on my D&D 3.5 experience. There, fighting monsters granted most of the XP. Avoiding a fight and going about the mission quietly was always an uphill mental battle. It was going against the affordance of the rules. The reward structure did not invite players to push harder, it invites players to optimize harder (since combat appears unavoidable, in a way, combat is the reward).



In D&D, I generally favor granting XP for accomplishing goals, almost like getting XP for completing a quest in a computer game. How you reach those goals is unimportant – what matters is that you get them done. It might be dragging home piles of treasure from an ancient tomb, saving a princess, or stopping a marauding dragon. My only criteria are that the goals are non-trivial and matter to the characters and/or players.

In fact, I like to go a step further – characters level when the players have progressed the story arc or campaign some reasonable amount. Our current D&D Next game evolved away from totaling XP for monsters slain (though if we found a good way to avoid a monster or challenge, we were usually given identical XP for that too), and now just relies on the DM’s feel of the game to figure out when going up a level is appropriate.

– Adrian 2014-07-03 22:07 UTC

It makes sense to level up on achieving goals. Basically you’re saying: A new chapter is beginning. The characters have changed. I like how that ties in with my preference for changing gameplay over time. Enough sleep and magic missile. Time to see some lightning bolt and fireball action. The only difficulty I see personally is how a referee is supposed to handle this in an open sandbox. As the players adopt new goals, the table decides as a collective that upon reaching this or that goal, they will all level up? Or do referees decide themselves? As Courtney argues in his blog post, On Advancement Mechanics, Experience, there’s always the danger of taking away agency from players. If the table agreed on an adventure path, on an adventure arc, then that is not an issue, I guess. I never tried it. I’m glad to hear that it’s working out.

AlexSchroeder 2014-07-03 22:35 UTC

Another good article on the topic of rewards: Don’t reward your players for role-playing. I laughed when I got to “I have an intrinsic dislike of extrinsic rewards.” I totally agree with this: “Giving out an extrinsic reward destroys the intrinsic fun. When you’re rewarded for performing an activity you enjoy, you lose interest in performing it for it’s own sake.” The post also includes a link the abstract of A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation by Deci, Koestner and Ryan. The article is also available via Research Gate. So what’s the point of XP in the first place? It’s there to “make sure that the game you play tomorrow is different from the game you played today.” This is something I keep saying about long-term campaigns: There needs to be a promise of ever-changing gameplay. XP is part of this. (In D&D, I think the other part is due to how spells are structured.) And if you’re handing out XP to “make them role-play more,” Jack Mack has you covered as well: “It’s a type of behavioral conditioning, a skinner box made to get your friends to behave the way you want. You shouldn’t need this. If you have a player who’s shy and doesn’t role-play much, why use a passive-aggressive rewards system to punish them for playing that way? If you dislike the way someone plays, why not just talk to them about it? Extrinsic rewards are just going to make them enjoy role-playing even less than they did in the first place.”

A blog post full of win.

I ended up having an interesting discussion with Tim Franzke in the comments to my Google+ post and on his Google+ reshare. One of his points was that in BWHQ games (Mouse Guard, Burning Wheel, Torchbearer, ...) he’s having great fun with getting rewarded for role-playing, an activity that he enjoys. I said that as for providing both rewards (intrinsic and extrinsic), I could only reiterate what the blog post said:

Intrinsic fun AND extrinsic fun, that must combine to make the game more fun than ever, right? Well, research has found that’s not quite true. In the words of this literature review: “...expected tangible rewards made contingent upon doing, completing, or excelling at an interesting activity undermine intrinsic motivation for that activity.” Giving out an extrinsic reward destroys the intrinsic fun. When you’re rewarded for performing an activity you enjoy, you lose interest in performing it for it’s own sake. – Jack Mack

If it isn’t true for you, I said, then I’ll argue that Burning Wheel is different because of the following:

  1. Some of the rewards comes at the end of the session, so they are pretty dissociated from the act. Perhaps that undermines it.
  2. Most of the rewards don’t reward “role-playing” but succeeding and failing at tests, being useful, moving the story forward, moving the story in unexpected directions, making people laugh.

As for me, I feel like getting paid—I feel dirty!—when we need to talk about most valuable player, embodiment, mold breaker and workhorse. It’s not too bad because I don’t think about it during the session. I’d argue that the negative effects of the extrinsic reward don’t affect me that much because of #1 above, the rewards at the end of the session are far removed from actual play. It’s still an awkward situation for me, and one of the many reasons I don’t play Burning Wheel.

On Tim’s own thread, he summarized his experience as follows:

I like getting rewarded “for roleplaying” because it makes me do suboptimal things. It creates neat decision points. – Tim Franzke

In a recent old school D&D session run by Harald, my cleric lost an arm. So what to do? No more shield use? Use a spiked shield and switch between protecting myself and bashing people? I decided to try and run my cleric as a pacifist. No more attacking. Clearly a suboptimal decision, and I didn’t do it because I expected a reward, I did it because I think it will be fun. I still want him to go on adventure, fight monsters and take their stuff, because that’s how I’ll get XP and level up, but the absence of a reward doesn’t mean that suboptimal decisions will not be taken. In fact, I’d argue that being humble without expecting an extrinsic reward would be more humble. The player would feel humble. The other players would be astonished at the humility. It would be a more valuable experience for the humans sitting at the table. Sitting at the same table, I might feel the urge to tease you, saying: “But you just did it for Artha…” Perhaps we’d enjoy ourselves because of the irony as we are players sitting at the table making our characters do things we know the players don’t feel at the table. There’d be a lot of winking and eye rolling, of “my character is so humble, he cannot accept this bow!” and “of course he is, hahaha!” The meta level turns into the source of entertainment. Speaking for myself, I find this reduces my enjoyment, unfortunately.

On the same post, I got pulled into a discussion with Robert about “extrinsic rewards for X but not Y” vs “extrinsic rewards for X and Y”. Clearly, I’m in the camp of “extrinsic rewards for X but not Y”—the interesting question should be why I prefer to reward looting but not role-playing, right? The answer in the context of Jack Mack’s blog post is that role-playing is an activity I enjoy at the table, it has intrinsic value, where as looting is something that has no intrinsic value at all. I just do it because that’s what I need to in order change the game over time, or in Jack Mack’s words, “to make sure that the game you play tomorrow is different from the game you played today.” It’s one of the things that makes role-playing games a game. It’s also left to the table to decide how far we want to go in the pursuit of this goal. And the thing you’re rewarding doesn’t have to be looting in every game—but it should be something that isn’t intrinsically enjoyable or you’ll spoil it. Perhaps it doesn’t spoil it for you and others, and that’s fine, too. The research says that there’s a tendency for extrinsic reward to spoil your intrinsic enjoyment, however.

Robert said that that some people don’t find intrinsic pleasure in RP and that “extrinsic rewards give them an incentive to do something they wouldn’t otherwise do.” The original blog post has a whole paragraph dedicated to this:

In comparison, I always see role-playing reward mechanics recommended as a way to change how people play. You do it to make them role-play more. It’s a type of behavioral conditioning, a skinner box made to get your friends to behave the way you want. You shouldn’t need this. If you have a player who’s shy and doesn’t role-play much, why use a passive-aggressive rewards system to punish them for playing that way? If you dislike the way someone plays, why not just talk to them about it? Extrinsic rewards are just going to make them enjoy role-playing even less than they did in the first place. – Jack Mack

I think that’s exactly the problem I have when Robert explains that he teaches “within a system that has exam structures imposed on my class by the government, parents, the school management, etc. Within that framework, it is sometimes necessary to use extrinsic rewards to encourage students to develop skills and knowledge that are not valued by the system.” If I were sitting at his table, and he were offering me extrinsic rewards for something I wasn’t doing because of an intrinsic enjoyment I felt, then it starts to feel like work and school.

You might point out that my characters are looting for XP, so what’s the difference from role-playing for Artha? I think the concept of a game is important, here. We’ve come together to play a game and the game is about particular activities. These activities are driven by goals. Loot is a simple goal to work towards. In Burning Wheel, beliefs act a bit like individual goals. Sadly, when I encountered them, they were often not clear cut goals, but they certainly can be, and when they are, it works quite well. Using instincts to my disadvantage, using traits to drive the story in unforeseen directions, being funny, being the most valuable player, the work horse, and so on—these aren’t goals. At least, these aren’t goals in how I understand them. These aren’t goals that provide direction to my activities.

Tim said that these mechanical rewards for things that I don’t recognize as goals do motivate him to make disadvantageous decisions. I can’t argue against that. All I can say is that for me, the referee making failure interesting is the only thing I require. I don’t require a reward.

As to why I consider the use a trait to drive the story in unforeseen directions not being a goal that provides direction to your activities, the way I see it this: we come together, sit at the table, the last session we told the game master that we wanted to explore the Gnoll ruins. What will be looking for? Option 1 is “loot”. Option 2 seems to be “looking for a way to drive the story in unforeseen directions”? That doesn’t sound right. It looks like the difference between strategy and tactics. “Loot” or the belief “I’m going to kill the lord of Xitaqua” will tell the game master what we want to do, it’s about the big picture and it will give us a sense of accomplishment when we have done it. It will take a session. Me having the trait “Playful” and throwing some bones at the lions, possibly changing the course of the expedition is something that happens spontaneously, it changes the scene, it changes the story in unforeseen ways, it doesn’t give us a sense of accomplishment. (At least it wouldn’t do that for me?) “Loot” or a specific belief says what we want to accomplish. A trait might suggest a way of accomplishing said goal, or change how we’re accomplishing said goal, or complicate out attempt of accomplishing said goal. It will not tell us what said goal is, however.

Well, if anything, I don’t think I ever wrote that much on a Google+ thread. 🙂

AlexSchroeder 2014-07-07

I am going to respond to only one thing, your question about a sandbox. The way I see it (and this is just one point of view), in a sandbox game, the players are essentially “exploring” an unknown map. In some cases this is the literal act of mapping out unrevealed hexes on a map. In other cases it is finding out what is hidden in a particular wilderness hex. In other cases, it is the interaction with particular notable features on the map – like what do you do with that increasingly belligerent warren of kobolds, or the mysterious lone wizard in the mist-shrouded tower. I could see attaching XP rewards to all of these activities, if you really wanted a clearly laid out reward system – or you could play it by ear (not sure if I would want to give out XP for just exploring hexes, but I could see games where it would work). Dealing with the kobolds (slaying them, forcing them to leave, negotiating a treaty, defeating the chieftain and becoming the new kobold chief) might be worth a small XP reward, while exploring the lair of the legendary red dragon Ashfang (and slaying her / taking her treasure / appeasing her with sacrifices / becoming her agents) might be worth more. Bigger risk or more difficult task, higher reward, and so forth.

Now that I think about it, I think I am mostly reiterating what you said about exploring the gnoll ruins – the difference between what the goal is versus how we accomplish it in-game. I prefer rewards for the former, not the latter.

– Adrian 2014-07-07 18:31 UTC

Please make sure you contribute only your own work, or work licensed under the GNU Free Documentation License. Note: in order to facilitate peer review and fight vandalism, we will store your IP number for a number of days. See Privacy Policy for more information. See Info for text formatting rules. You can edit the comment page if you need to fix typos. You can subscribe to new comments by email without leaving a comment.

To save this page you must answer this question:

Please say HELLO.