Extra Credits: The Feeling of Agency

This week, we kick off a series of episodes on Choice in games.

Episode video is on YouTube

Show Notes:

Nick Trujillo is the coolest. Check out his comic!

Be sure to check out the Extra Credits Store.

Would you like James to come speak at your school or organization?

For info, contact us at: kate@extra-credits.net

Apologies for any inconvenience, a better approach to new episodes will be live as soon as possible.

Recent Comments:


  • Actually - I do have one issue about your (and you admitted it yourself) controversial decision to make calculations not count as decisions: Doesn't this mean you've defined games with perfect information as ones that don't have any decisions because, given sufficient computational power, you could calculate the best possible move in any given situation? Even games containing good amounts of non-comparables, take the calculations far enough and you can work out which one, given perfect play by the opponent and yourself, will cause you to win and which will cause you to lose.

    Actually what it means is that games where one has perfect information and can assume perfect play don't impart a real sense of agency in play, because any decision is effectively predetermined and the only real choice is whether or not to play. When discussing agency, as the term is used in the social sciences which is the use here, there is a conventional understanding of the level of choice one has in any decision being limited by the structural factors beyond one's control.

    In a game those structural factors include such things as rules and the impact of directly given information and derivable information under those rules. With perfect information the decision becomes whether or no to play optimally, which isn't really a choice because you've already made the choice to play and can safely be assumed to be playing to win.

  • A choice is not a calculation... if there's a definitive right answer that can be ascertained through mathematics or [a] strict logical reasoning that's performable by the player in the time allotted to make the choice, the choice is no longer meaningful...

    Finally made my account just to disagree with this. Not the final conclusion, just the sentiment. I think this definition is biased against the role of analysis in games. I consider myself first and foremost a game designer, but my training is as a game programmer and I've published dozens of games over the past few years. Analysis is just as crucial as intuition and imagination in the design and play of games. To deny that choice is a calculation (among other things) is to be ignorant of this fact.

    For example, one working definition of intelligence (supposedly the thing that sets humanity apart from the animals and which gives us our unique agency) is: the ability to make optimal decisions to achieve complex goals in complex environments, using limited resources. While it is true that having a definitive, demonstrably optimal choice is meaningless, this is due to insufficient constraints on the system; any overly-simplified choice ceases to be a calculation by almost any definition. But this does not imply that all complex choices are not also calculations. Ask any AI designer and they'll tell you that the modelling of choices involves calculations, intuition, and imagination. From decision trees, to finitie state machines, to HTN planners, you name it...

    Here's a concrete example. In poker games, there's typically 4 types of AI (which are meant to model human-like decision making). There's the novice, who plays more or less (but usually less) randomly. There's the stone, who doesn't bluff and plays conservatively. There's the maniac, who bluffs constantly and plays aggressively. And then there's the shark, who knows when to bluff, and when not to. These aren't usually revealed to the player, but they're easy enough to identify over time from their playstyles. I've seen game designers create an easy mode with no sharks... this is a disaster waiting to happen. Especially if you allow players to win premium currency or such from this mode.

    To understand why, you have to appreciate the player's decision-making process as a calculation and see the golden path. I'll leave that as an exercise for the student, but suffice it to say, games aren't fun when there is only one meaningful choice. Conversely though, game designers who fail to analyze their game's choices as calculations are setting themselves up for failure. You've really got to recognize that if you want to avoid the pitfalls of golden paths or optimal builds (as the EC guys mentioned in WoW)...

  • I think any game that has a Good vs Bad morality system runs the risk of it's choices becoming arbitrary Black & White. Look at Knights of the Old Republic, most people when they played that decided early on if they were going to be a good guy or a bad guy for a particular play through, so naturally if you wanted to be a good guy you pick all the Lightside choices, didn't help that in almost all cases it was painfully obvious which choice was good and which was bad, meaning ost people pick the choice that suited their chosen alignment.

    Mass Effect improved on things as gaining one type of alignment points didn't deplete the other, so a player aiming to be a Paragon could go Renegade once or twice without any real problems.
    Dragon Age's idea of simply having companion affection is a good one (actually KOTOR 2 had that first, Obsidian beat Bioware to that idea) but DA2 actually did it better as with DA:O if you lost affection with a companion you just threw some gifts at them to fix it, I finished the game with max affection on all companions just by throwing gifts at them.
    Then there was Jade Empire which had an interesting philosophical twist on good/bad choices with it's "Way of the Open Palm" or "Way of the Closed Fist", the idea was that doing good or bad things didn't necessarily give you alignment points, rather it was the reasoning behind the decision that mattered, not what you did but why you did it.Sadly while on paper this was a great idea in practice it boiled down to "do bad things get dark alignment points" so it was not really any different from KOTOR
    Actually the old D&D games with their multiple alignments probably are the most intricate alignment system in games that I've seen, what with Good-Neutral-Evil and Lawful-Natural-Chaotic all being used in pretty much any combination.
    btw interestingly most of the games being mentioned here are Bioware games, does any other company even do choice systems in their games?

  • I consider myself first and foremost a game designer, but my training is as a game programmer and I've published dozens of games over the past few years. Analysis is just as crucial as intuition and imagination in the design and play of games. To deny that choice is a calculation (among other things) is to be ignorant of this fact.

    Or, in other words:
    XKCD

    There's not much utility in insisting complete, wholely-vocabularisticnessifntionism definitions. I think we're really all on the same page.

  • Guest artist, yay! I actually like guest artists because of what kind of new styles they bring to Extra Credits. This guy's style is pretty cool!

    You know, when I'm in a video game, I always make the kind of choices I'd make in real life... if I were a more heroic, brave, and athletically fit kind of person. Specifically, I typically made the good choice, the choice that would make the most people happy

    I feel like I should save this example for another time, but in The Legend of Zelda: Skyward Sword, this one side-quest would net me some Gratitude Crystals... now, I was given the choice to either deliver this letter, written by guy who's kind of a jerk, to a girl he likes, or to give it to a random hand that comes out of a toilet... Yeah. I didn't really want to upset the guy, but then again, I kind of wanted to see what would happen if I gave the paper to the hand.

    So I cheated. I tried both scenarios, resetting the game after seeing how each one turned out, and as it turns out, both situations ended witht he guy being miserable, as the option to deliver the letter only ended up with someone else confessing their love to the girl and leaving the jerk broken-hearted. I figured, since the two would end up together anyway, I'd give the letter to the hand, so that the hand would also be happy.

    Yeah, it sounds confusing, mostly because it is, but that's how my mind works. I doubt anyone else would think this hard about a situation like this... Or maybe they do? What choices did anyone else make playing this game?

Join The Discussion: