Quantum Ogres, Trolley Problems, and How to Make Choices Matter
There is only one Quantum Ogre
Picture the following: you are travelling in a dark forest when you come to a fork in the road. Left, or right? You take the left-hand path, and before too long a ferocious ogre leaps out at you. So far, so straightforward. Now imagine that you somehow go back in time, take the right-hand path, and Lo! the same ogre appears before you with malintent. It would appear that the ogre exists in a superposition of states that only becomes determinate after you have picked one of the two paths – hence, ‘Quantum Ogre’.
The Quantum Ogre is a classic example of when a player’s choices obviously should lead to different outcomes, but don’t actually do so. It’s a way of fostering the illusion of player agency without having to work for it: design one destination, but present a choice of two paths. Of course, in Table-Top RPGs – where the Quantum Ogre originated – the Game/Dungeon Master might just about get away with it, since the player usually can’t go back in time and take the road untaken. In a video game, however, the player can take both paths, on top of which someone will eventually upload a list of all the choices in your game that do and don’t matter.
The case of the Quantum Ogre illustrates a perennial problem for all games that fall somewhere between sandboxes and linear narratives. On the one hand, you want to give your players interesting choices that feel like they matter (the choices, that is, although I suppose the players by extension). On the other hand, the moment you start introducing branching story pathways you face the problem of bloat, since those pathways grow exponentially. Even if you find a way of reconnecting those paths somewhere later down the track, you’re still investing a huge amount of resources in creating parts of the game that the player may never see.
With 11 branching choices you get 2,048 different outcomes and a migraine
Obviously, the Quantum Ogre (and his good friend, ‘Seven Different Dialogue Options for Saying “Yes, I’ll do the quest”’) provides a way around this problem, and there are plenty of games that have made use of it. However, since people tend to hate hypocrisy more than sin itself, the shattering of the illusion of agency can come at a high price. Fortunately, there are ways of making choices ‘matter’ without needing to change the game narrative that the player actually plays through. They’re not mutually exclusive, and choice-heavy games tend to use a combination (with the occasional Quantum Ogre thrown in for good measure).
First off, we have choices that affect how the game is played. These tend to give the player different items or abilities, but by and large keep the prime story pathway unaffected. Basically, they’re a narratively dressed-up way of letting the player change their ‘loadout’. In inFAMOUS (2009), for example, the player’s conduit powers change depending on whether they make predominantly ‘good’ or ‘evil’ choices. In Pathfinder: Wrath of the Righteous (2021), the player is frequently presented with choices such as: 1) transport the bones of a priestess to a shrine to boost morale; 2) nick the priestess’ powerful bow from her tomb; 3) resurrect the priestess as an undead companion for their party. In Mass Effect (2007), the player can end up with a different set of squadmates to bring on missions (and very different relationships with those squadmates), but the missions themselves are largely the same. The list could go on, but the point is that these kinds of choices affect the gameplay while keeping the prime narrative confined to a single pathway. Minor offshoots (like NPCs being enamoured or disgusted with the player’s actions) and a variety of endings embroider the tapestry without requiring a whole new tapestry.
The age-old choice between good, evil, and optimal lightning customisation
Then we have the choices that ‘matter’ because they’re designed to weigh heavily on the player’s conscience. These often revolve around the fates of NPCs or parts of the larger gameworld, but rarely affect the playable parts of the narrative. There might be a brief cutscene to explain the far-reaching consequences of the player’s choices, or they’ll get a mention in the final wrap-up before the credits, or (for double the fun) they’ll be intertwined with ‘loadout’ choices. Yet what makes these choices impactful is the fact that they present the player with philosophical or ethical problems. In the firstWitcher game (2007), for instance, the player must decide whether to protect a witch from an angry mob by (eventually) slaughtering the ringleaders of said angry mob. It’s a variant on the classic Trolley Problem, with additional complications coming from the fact that both the witch and the ringleaders appear to be guilty of some pretty heinous crimes (although there’s also a deliberate ambiguity around all of this that exacerbates the problem). The choice the player makes here doesn’t affect the main plot at all – they still get the pass to move on to the next area – but the situation, the stakes, and the uncertainty of guilt all make for a compelling and memorable moment.
Plenty of other narrative-heavy games throw these kinds of choices at players (especially CRPGs like Divinity: Original Sin 2 (2017) and Pillars of Eternity (2015)). If we take The Witcher’s example, it seems that the effectiveness of these choices boils down to a few essential elements. First, immersion is key: if we don’t care about the fate of the gameworld or the people in it, or if we start viewing these things as constructs rather than people, then the choice and its consequences aren’t going to matter. Closely related to this is the emotional investment the player has in the situation: a choice that will affect ‘people’ that the player has come to know is much weightier than a choice affecting vague and unnamed NPCs.
The lack of pitchforks suggests their hearts aren’t really in it
Another important part of what keeps us immersed is the extent to which the situation we’re placed in seems natural or contrived. Despite its fantasy elements, the situation in The Witcher is not only believable (what with its actual historical counterparts), but also reveals itself gradually over the course of the level so that when the final choice is presented it seems all but inevitable. Contrast this with the standard ‘The tentacular god of the deeps demands its sacrifice of puppies or it will flood Manhattan’ (OK, not a choice I’ve actually encountered in a game, but there have been some that weren’t far off). The point is that presenting a grounded, situated, and organically emerging problem is a large part of what separates a game (or any form of narrative media) from an ethics class.
Finally, there’s the matter of the choice’s ambiguity. Obviously, ambiguity isn’t always desirable, especially not repeated ad nauseam. However, when it’s properly deployed it can make a choice feel like it has a significance beyond the bounds of the game. In The Witcher, the player can gather as few or as many clues as they like, and while those clues may inform their decision, there’s no ultimate ‘solution’ to be uncovered. It doesn’t help that some of the NPCs are probably lying, and that there are a few of the ‘rules’ governing the natural/supernatural world that aren’t concretely laid out. In the end, we’re left knowing that we don’t know everything, and we have to make a judgement call. This also has the potential to make us interested in what other people chose, and why.
A staple of what we might call ‘discussable’ art is that it features some kind of intractable problem, and the only real difference in games is that the player gets to make the all-important choice. This choice therefore matters because, on the one hand, a narrative unknown tends to nag at us like a missing tooth (this is the driving force behind the whole genre of detective fiction, after all). On the other hand, we’re left in a position where our choice won’t be entirely dictated by what’s been given to us by the game (since that’s incomplete). We necessarily insert our own prejudices, morals, and patterns of behaviour into the situation, and therefore our choice inevitably reflects a part of ourselves.
Trolley Problem – Solved
The types of choice that I’ve been discussing here – ‘loadout’ and ‘problem’ – are probably the two main ways games make choices matter without affecting the main story. Hybrids are also fairly common (where a problem choice is also a loadout choice), and although I’ve mostly focused on RPGs, this isn’t the only genre that features them (take resource-management strategy games like Frostpunk (2018) and The Banner Saga (2014), for instance). Of course, I wouldn’t be doing due diligence without pointing out that some of the games I’ve mentioned do actually have some branching pathways in their main stories as well, although that gives them all the more reason to also use loadout and problem choices to keep things manageable.
To come back full circle, there is a sense in which, although they seem to matter, the loadout and problem choices do still share a kinship with the Quantum Ogre. Although – unlike the Quantum Ogre – they are choices that lead to different consequences, those consequences leave the prime narrative largely untouched, ultimately putting the player back on the path that has been laid out for them. Yet when they’re done well, it’s entirely possible that these problem choices can end up seeming far more important than the main story – and thus it will be these smaller, self-contained narratives that the player actually remembers when the game is over.
Good Reasons to Give Players Bad Choices
Polygonal Morpheus has a choice for you…
About 20 minutes into The Matrix – Path of Neo, you’re presented with the same choice as Neo is given in the movie: take the red pill and begin your journey to self-discovery and bullet-dodging, or take the blue pill and go back to being an office drone with a side-line in grimy noughties clubbing. It’s the classic call to adventure moment, and since by this point Neo’s already done the ‘refusal of the call’ and all the standard hero drama, it’s something of a no-brainer. Still, the game actually gives you the choice to refuse once and for all, and you can take the blue pill to kill the adventure before it even gets started. After being treated to Morpheus’ disappointment and a short animation where you wake up back at your computer, it’s game over (or load from the autosave).
There are plenty of other examples of this kind of thing. In Batman: Arkham City, there’s a brief section where you play as Catwoman and can choose to leave Batman to die. It actually stands out as one of the few choices in the game that can affect the plot, and unsurprisingly, it leads to a narrative cul-de-sac (what with letting the main character die). The game then drags you back to the corridor outside Batman’s cell and offers you the same choice again. In Black Mirror: Bandersnatch, you can choose to jump off a multi-storey balcony instead of Will Poulter’s character, and plummet to your death – before being given the option of going back and letting the other guy take the fall. I could go on and on (the old Sierra games and choose-your-own-adventure books are lousy with this kind of thing), but hopefully that’s enough to illustrate what I’m talking about here. These are bad choices – not necessarily ‘bad’ in a moral sense, but choices that obviously go against the goals that the game has put in front of you. Stupid choices, in other words. Often they lead to a gameover before giving you the option to go back and choose differently (you idiot), and usually they don’t have any effect on the course of the game as a whole. The reality in which you continue is the one in which you chose correctly (even if you happened to catch a glimpse of the one in you which you chose poorly).
It’s not as if the game is going to just kill Batman offscreen, right? Right?
So what is the point of having these choices in a game? After all, if you actually want to stop playing, there’s a menu for that – no need to literally kill off the protagonist. They’re barely choices at all, since once you’ve made them you’ll be forced to unmake them if you’re going to keep playing. And it’s not fair to dismiss them as bad game design, either – they’re clearly serving some sort of purpose, since with all of the narrative tools at the designer’s disposal it’s far easier to just avoid including these choices altogether. For example, The Matrix – Path of Neo, which otherwise liberally uses footage from the films for its cutscenes, goes out of its way to animate the red pill/blue pill sequence. Time and money were spent on having digital Lawrence Fishburne badly lip-sync the lines that real Lawrence Fishburne already articulated perfectly for the camera, all so the player could make the choice to short-circuit the game they’ve just started playing.
If we’re thinking about what functions these choices are serving, a couple of explanations might spring to mind right off the bat. These are the functions I want to get these out of the way quickly, not because they’re ‘wrong’, but just because I don’t think they’re as interesting. The first is thematic: The Matrix movie is all about choice and free will, and it’s not exactly subtle about it. Even though the game strips out most of the philosophy to really double down on the slow-motion violence, it’s keeping this most important of choices as a nod to its origins. The same goes for Black Mirror: Bandersnatch: choice, free will, possible worlds, video games within a video game, so on and so forth. But what about Batman: Arkham City, and all the other games where the player can make an obviously stupid choice to kill off the protagonist? Not every game is making some kind of point about free will, and even when they are, it’s not as if theme explains everything that’s going on.
Another possible explanation for the ‘bad choice’ is: because it’s fun. There’s something compelling about self-destruction, which is perhaps why people can so often be relied upon to position the metaphorical foot under the metaphorical barrel and gleefully pull the trigger. The bad choice allows the player to safely indulge this impulse by proxy, a function which is often held up as the function of art in general (let’s not get into all that here, though). Sometimes this will also be part of establishing the game’s tone, and the choice’s consequence will lean hard into the absurd or the grotesque for humorous effect. And sometimes the impulse that drives the player to pick it more a case of gratifying curiosity than masochism (‘Will the game really let me screw up so badly?’).
Indeed they do, Michael. Indeed they do…
Yet although this idea may have some purchase, when it’s applied across the board it starts to seem a bit… gratuitous. It becomes the equivalent of an optional sex scene in a violent game (or a violence scene in a sex game) – rather besides the point, but there just in case the player fancies a bit of self-destructive titillation. Include it, don’t include it, it doesn’t really matter, right? Look, here’s a free one for you:
DO NOT PRESS
There is a third explanation for the ‘bad choice’, and it’s one that doesn’t either pump it full of meaning or totally trivialise it. This explanation considers the ‘bad choice’ as serving to heighten player agency – or more specifically, the player’s sense of agency. In a nutshell, the sense of agency is the feeling we have when we’re controlling an action. Philosophical and psychological research demonstrates that this sense/feeling can be quite easily misattributed (i.e. people can be entirely wrong about what they feel they’re controlling), and can also be disrupted (for instance, in conditions like ‘alien hand syndrome’ where people feel like a part of their body is moving entirely outside of their control). It’s an incredibly complex feeling, with various different conscious and non-conscious parts that contribute to it (things like perceptual feedback, awareness of intentions, a sense of effort, etc.).
According to some theories, one of the things that contributes to the sense of agency is an awareness of the potential for inhibition – a sense that we could always stop an action in progress. This accounts for why we still feel like we’re in control of actions we take automatically, like walking while talking, or picking up a cup of coffee. We don’t need to think about these actions, and we certainly don’t need to pay attention to all the little adjustments of muscles and limbs that we’re making, but we also certainly don’t feel like we’re lacking agency. The same goes for actions we’re compelled to take by external or internal forces (e.g. handing over money at gunpoint, succumbing to the demands of an addiction). We don’t ‘have a choice’ in the sense that it seems like choosing differently would be foolish or morally unconscionable or overwhelmingly difficult or whatever – but that’s not the same thing as saying we don’t have a sense of agency for the action. (Of course, people sometimes do dissociate in such situations, but that’s a part of what I’m getting at – feeling like we’re not in control of our own body is pretty much a condition of dissociation.)
An awareness of the potential for inhibition at least partly explains how we have a sense of agency for these sorts of automatic and compelled actions without recourse to things like ‘conscious decision-making’ (or even ‘bodily feedback’ if we’re going to include thoughts). Even when some of the other things that might contribute to the sense of agency aren’t online, we’re still aware that we could always inhibit/stop/interrupt the action (or at the very least, try to). We might not want to, and we certainly don’t need to think about our potential for inhibition – but it’s something that’s there in the background.
To go back to choices in video games, it seems like the inclusion of the ‘bad choice’ is mimicking part of the basic structure of agency. It’s there to make the ‘good choice’ actually feel like a choice, even when the player knows it’s the only choice that will keep the game moving forwards. The same kind of function is being performed by ‘bad endings’ in sprawling RPGs like The Witcher 3 or Mass Effect. It doesn’t particularly matter whether the player already knows which choices lead to which endings (which is why a second playthrough or checking a walkthrough doesn’t necessarily detract from the pleasure of the game or the feeling of making a choice). After all, making an informed vs. an uninformed choice doesn’t make the choice feel any less agentive – quite the reverse, in fact.
Turns out you really can make Geralt sad after all, you monster.
These three functions of the ‘bad choice’ aren’t mutually exclusive, and there may be more that I haven’t covered here. Of course, it’s always possible to have ‘bad choices’ that don’t lead to a game over, or to present the player with two obviously undesirable choices and so drop them into a moral quagmire. Since this post is already getting quite long, though, I’ll have to deal with that another time.