How to use players' emotions

I'm targeting a few game developers with this, but welcome input from anywhere.

Let's say that there's a technology -- a module that you can plug into your projects that allows you to access the internal emotional state of the user of your game. How would your game use that? Any game. Any kind of game. What kinds of games would use it in what ways? Are there ethical considerations? As a designer/developer/whatever, would you like to have access to that extra stream of data? And why?

In email, Patrick Dugan wrote:

I'm currently playing with relationship-focused multiplayer interaction and how imperfect information can balance that out, kinda like what Between did but from a different angle. So naturally I'd like to experiment with making that data known to the other player, as well as leaving it unknown and doing some fancy back-end stuff with it. The easiest thing to do, of course, is leave "traps" for players to express emotional extremes, like a sudden gap-up in a market, and reveal things to their counter-player and to themselves that they may not have been aware of. That's the trick, the cognitive construction of choice matrices is powerful enough to mold behaviors behavior and take metrics on it, you don't need direct B-CIs. Of course, the B-CIs would be useful, but the control grid will come in more subtle forms.

That's good stuff for me. I hadn't been thinking at all about multi-user reveals, only how to manipulate the player with adaptive algorithms.

In email, Craig Perko wrote:

The problem with a question like this is that it's a very unsatisfying
thought experiment. The answers are all too basic and uniform because
the question is too wide-base.

Anyone with an ounce of brain can see that it would allow you to make
better games, and they can imagine up the kinds of ways it can happen
for all the various kinds of games. Anyone with an ounce of brain can
also clearly see the ethical considerations.

In order to be an interesting thought experiment, it has to be about
something more specific, something that requires thought. Something
that ten different designers would come up with ten different answers
for.

But if you really want an answer, okay. I would make games scarier,
more touching, more adrenaline-filled, obviously with an eye to player
health and privacy. You could easily develop a feedback system for
stressing the things that affect them most: if Anna makes them more
happy than Bert, you can show them more Anna, or you can strive to
make Bert more likeable. And, of course, that's a simple intro
example, it can get very complex as you apply it to level design, NPC
scripting, and variations in player capacities for emotion...

The problem is that it's a straight run: everyone will give you more
or less the same answers, perhaps flavored with different levels of
paranoia. I imagine these are the same answers you came up with.

Uh, yeah. What he said is what I'm finding.

There are little exceptions. Patrick's multiplayer thoughts are great. James Brown suggested a Rock Band game where audience appreciation matters to your score -- which is a different, easier kind of multiplayer use. He also suggested using the tech to gauge the effectiveness of product placements. That's neat.

Party game design

1:30 - Using Brenda Brathwaite's article, Easiest Game Design Exercise Ever I decided to make a silly little game quickly. Also, I need text-heavy stuff to do in order to practice working with my new keyboard.

Right off the bat I'm aiming for a somewhat risque party-game since I don't have any experience inventing such a thing.

1:39 - Step 1 - So let's say there are twenty spaces on our track.

1:40 - Step 2 - Must...choose...narrative... OK, so I really prefer an abstract, but I want to follow this exercise. I want to have some social play elements ( so I want it to be social like running for class president or something but kind of more grown up. Um...like a successful party host(es) or something. Yeah, that'll do -- Party Time.

2:10 - Step 3 - My first impulse was to violate the rules here, but I've gone back and decided to *really* follow this as closely as possible and see what I get. I'm going to just "go with dice" but I want there to be three or four sources of movement in step 4 and I want them to be relatively more powerful than the random movement of the dice. I want the dice to result in a distribution with a low kurtosis. What I'd really like is 3d2 or something but I think that's impractical -- flipping coins sucks compared to rolling dice. After goofing around for a bit I think I'll go with Roll three dice (d6) and take the lowest roll as the mechanic. It gives a slight chance of extraordinary movement (six spaces) but you're about 70% likely to move only one or two spaces. This also allows future developments to alter the die-mechanic.

2:49 - Step 4 - Conflict. I'm not sure about calling this conflict but I want a few position-modifying mini-games. Maybe once they're fleshed out the board will just have one of them on each space or maybe they're the result of a card-draw. The trick now is to figure out what they are. I'd like to have at least four mini-games and ideally, more. I'd like them to extensible -- fitting expansions and player-generated content in seamlessly and allowing certain Challenges (I think that's what I'm calling them) to be removed to fit the tastes of the play-group. I'd also like for the Challenges to result in a net movement of <2 spaces -- so some player(s) might advance or fall back up to two spots or maybe some of each would happen but only up to one space in each direction -- and if this were occasionally violated, no biggie.

Challenge 1: Who here has (never) done...? The way this works is that the Focus player (she who triggers the Challenge) selects a question about past deeds for which her truthful answer is "me" (indicated with a raised hand) and presents it to the crowd. (E.g. "Who here has driven over 100 MPH?" or "Who here has drawn a lover's blood with your teeth?" or "Who here has never skied?") Every other player either raises their hand, indicating that the statement is true for them too, or does nothing indicating that it is not. If no hands are raised, the focus player advances two spaces. If fewer than half the hands are raised then the focus player advances one space.

Challenge 2: Color-Match When using this Challenge, a pile of colored cards or tokens or something must be available. (For testing, I'd like to get a pile of color chips from the paint department at Home Depot or something.) The focus player select a color randomly, drawing it and revealing it to all the players. If that player is wearing a garment that is pretty close in color to drawn color (as determined by a consensus of fellow players) then the focus player has two options. If not, skip the first option below and execute the second.

2a: If the focus player is wearing a garment of close color she may opt to apply the color test rules to herself rather than seeking out the closest match as per the paragraph below. If she happens to wear the closest match in the game, then the option is moot.

2b: The player wearing a garment that is closest in color to the selected color must go back two spaces. If that player would rather, he may remove the garment and instead go forward one space.

Challenge 3: Kiss Your Neighbor When playing with this Challenge module, there is a play aid -- a card with five strata each listing one of, in order: hand, arm, lips, neck and tongue, upon which a pawn rests and moves. The pawn begins the game on the lowest (hand) space. When the focus player lands on a KYN space, he must consult the play aid and kiss one of the players adjacent to him -- the one he has kissed the fewest times if applicable, otherwise his choice. If he will not, he loses two spaces. If he kisses the neighbor in a place and manner appropriate to what is indicated in the space with the pawn, he advances one space. The focus player also has the option of kissing his neighbor as per the space below or above the pawn's current location, subtracting or adding one to the number of spaces gained. Further the pawn is moved up a space on the play aid each time a kiss is conducted earning one or two spaces gained on the victory track. The point of the kiss-location track is to escalate the passion and intimacy of the kiss. To play this appropriately, the meaning of each space should be apprehended with this in mind.

Challenge 4: A Round of Greetings The focus player is to address each other player starting with the player to her left and continuing in clockwise order as if she were the host of a party and were greeting guests as they arrive. Each greeting should include physical contact where practicable (maybe only with adjacent players if your game space doesn't permit freedom of movement), direct eye-contact and a compliment. The focus player receives two space-advances minus one for every fellow-player who thought the act implausible.

Challenge 5: Art Contest The focus player names a subject for the other players to draw. Each other player is given a small piece of paper and a pen or pencil and attempts to draw the named subject. This should take just a few minutes. The focus player then collects the drawings and judges them -- selecting a winner. The drawer of the winning picture advances one space. Then, the focus player's judgment is judged. If every player agrees with the judgment, the focus player advances two spaces. If not, but every player agrees that it was a reasonable or plausible judgment, the focus player advances one space. If not, and every player suspect the focus player of gaming the judgment then the focus player loses a space.

etc.

I'm now thinking that the way to do the board is to just assemble a small deck of cards/spaces based on the modules you're playing with and the relative frequencies desired and then shuffle them up and lay them out making a dynamic board.

5:07 - Step 5 - Oh, I'm done. There you go!

MMO class design: reaching balance

I just read the article at Gamasutra, MMO Class Design: Up With Hybrids! An Economic Argument in which the author analogizes the ecology of classes and player-competition within that to several economic factors. It was a good read. And it struck me that it would be really, really easy to provide one kind of balance to the class ecology by simply applying a reward multiplier (maybe just on XP if that's how your game rolls) based on nothing but the popularity of the class. If everyone detects that the fire/fire tank in CoH is uber and picks that, they might choose something else when they realize that the invul/mace tank earns xp at a rate 1.5 times as fast as the fire/fire. Right? And you could have that dynamically tuned as a weighted average based on the previous week on in-game play time or something. Then, barring stupendous gaffes in design which you've hopefully caught in testing, you can just completely avoid pissing people off when you nerf their class.

Is any game doing this? Why not?

Metaplace dev blog fundaments #2 response

So, I'm trying to grok game design. I've been reading RPG theory off and on for five years at the Forge and diaspora sites. I read stuff written by board game designers off and on and I know several of them in person. And recently, I've been reading computer game design stuff. Several blogs -- you can see some of them in my sidebar at the right. I'm in line to buy Raph Koster's book pretty soon. And I'm watching his new project, Metaplace eagerly.

This morning I read part two in a series of game design fundamentals on the Metaplace developer's blog written by Raph and I thought about it and did some background reading and then posted some questions to the Metaplace forum.

I'm writing this mostly to myself -- under the assumption that no one is reading this. :) If that's not the case, feel free to respond to me either here or at the Metaplace forum. Here's what I wrote:

Do you have to prepare for the challenge?
…where prep includes prior moves? …and you can prep in multiple ways?

Does the topology of the space matter?
…does the topology change?

Is there a core verb for the challenge?
…can it be modified by content?

Can you use different abilities on it?
…will you have to in order to succeed?

Is there skill to using the ability?
…or is this a basic UI action?

Are there multiple success states?
…with no bottomfeeding? …and a cost to failure?

You have to answer yes to all of these for your game atom to be fun. And yes, we mean every atom in the game has to meet these criteria.

There's a lot of this that I think I get, but could be wrong. There's some of it that I just don't. First, I'm just going to ask some questions and then I'm going to try to analyze something using this framework.

Are 'atom' and 'challenge' synonomous to the extent that they bear a one to one relationship?

I sort of understand the basics of mathematical topology -- is that what we're talking about? Or is it being used more generically to suggest some kind of state change? Does the opening move of a Chess pawn change the topology of the board? And is it because of the seperation of the front of pieces or because it changes the way in which the next player can engage the situation? What is the topological significance of the fact that that pawn's move is not reversable? Any? Or does topology in this sense only care about loci of in-game actions and how they're connected?

I'm pawing through Raph's blog archive looking for game grammar stuff to understand core verbs. I found this presentation there (which helps me with topology a bit, too). So I see that e.g. Chess pieces are verbs. Cool. I get that. But uh, "each has associated topology?" Does that merely mean that each verb potentially interacts with a different subset of the game's topology and acts differently (rooks travel from node to node on different paths than do bishops) or is it something more essoteric? Also, what's the significance of a core verb? Am I right in expecting that many atoms will use several verbs?

Are abilites verbs? Is the following true: skills belong to the player and abilites belong to the game (or the player's presence in the game).

What's bottomfeeding?

Then I think that Raph's saying every atom should answer yes to each of those questions. But I think that's impossible and even inconsistent with stuff like "fundamental atoms will be ones demanding no skill, providing no risk of failure" from that same link above. So what is really meant regarding the scope of the application of these questions? Also, how fun is fun enough? How do you account for niche games where people obviously think it's fun but lots of others do not? Are you saying that at whatever the appropriate scope is, failure to meet the above criteria will result in a game that isn't fun for (almost) anyone? (This is what I'm guessing.)

I think I understand atoms. They occupy many different scopes/scales. Double-clicking the armor in a vendor's shop to tell the game that you're buying it is an atom, right? Maybe a "fundamental" atom? Also, the game of knowing which armors work well against what opponents is an atom. Right (I'm less sure about that one)?

So in the case of just using the mouse to buy the armor: Are you getting yes or no to those questions? Are you preparing for the double-click by looking at the screen and figuring out what coords to settle on? And I'm not sure on sophisticated topologies in examples like this -- are there nodes, per se involved? I guess there must be but it's awfully big to grasp. What's the core verb and how can it possibly change? What's the challenge even? What is an ability at this scale? Are multiple in use during this purchase? And I think there's only one success state unless I'm more confused than I'd have guessed. Is the cost of failure the quarter second to redo the double-click?

But it all seems easier in the bigger atom (where analysis should be much harder): There are lots of preparations -- gathering knowledge through experimentation or social means. I guess topology is turning out to be the hard thing for me. Topology must matter in as much as it effects e.g. information gathering, access to armors and enemies using various damage types, etc. But where the information exists in the player's head (or notes) I don't see how it can change the topology (unless we claim that topology can extend out of the game -- which might be useful, but I'm unprepared for. :-). Is the core verb combat or social networking or the avatar or something else entirely? You can, and possibly must, use different abilities to get a complete understanding (the way I'm envisioning things). There are definately skills involved in this kind of analysis. I'd say that there are a very large number of success states if there are even several armor and damage types in this game as incomplete understanding can result in nearly infinite states of grasp. And certainly there are costs of failure even if they are transparent to the players prior to their realization that this atom/game exists.

So, I think the first atom can't be fun but can be necessary to a fun game. The second atom, if sufficiently large, is quite likely to be fun (if not crazy-fun).

Are these analyses along the lines of what the article is suggesting?

thanks a ton,

Chris