Boo to a Goose
Is it wrong to do amoral things in computer games? Is it ok to bully people as a goose in Untitled Goose Game?
Things mentioned in this podcast:
For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email podcast@alephinsights.com
This podcast uses the following third-party services for analysis:
Podtrac - https://analytics.podtrac.com/privacy-policy-gdrp
Transcript
Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Chris Wragg, Peter Coghill and Nick Hare of Aleph Insights and this week we're discussing whether it's okay for geese to bully people in computer games. Nick, I have no bloody idea what this is about. Tell me, what's going on? Well a recent sensation on the internet, which I know you're not very familiar with, was the game Untitled Goose Game, is what it's called, by House House Software and it's an absolutely fantastic little game where you... Can I just say it's lovely to see your little eyes lighting up as you talk about this. I haven't seen you so excited in so long. I mean, when you think you've played every genre of computer game, along comes Untitled Goose Game and the aim of the game is you're basically a goose living in a little sleepy sort of middle England village and your job is to go round and basically you have a sort of checklist of horrible things that you need to do. You need to sort of steal people's lunch, you have to try and make a child fall over and then you have to give him the wrong glasses to pick up, you've got to trap him in a phone box at some point, you've got to try and smash this woman's vase and so there's all these missions you have to try and complete and it's a kind of a puzzle game. You're wandering around, it's very adorable, this goose. It just walks around, it goes to a pub, it goes to the high street and goes into shops and stuff and so you're basically going around being very mean to people. The aim of this game is to be mean and I can't say I felt particularly bad about bullying this kid because he is only a little virtual person but it did make me think about what are the ethics of how you behave in a computer game and perhaps more widely games in general. What is allowed? If a game encourages you to do bad things like making a child fall over, is that okay? And then more widely looking at multiplayer games, what kinds of behavior are acceptable there and so yeah, I mean I just think it's quite an interesting question.
Speaker B:Peter, Chris, weigh in. Yeah, well I mean, I've also played Untitled Goose Game and similarly didn't really think about the, didn't feel immoral conducting the acts. But then you don't in real life. But that's true, I'm always making children fall over and putting the wrong glasses on. So yeah, in that way, yeah, it mimics reality. But no, I did think about the idea of whether or not you can be bad to agents, you know, kind of computer-created agents within games. And in this game, the sort of suspicion, they wander around but they don't appear particularly intelligent, right? You know, they have quite predictable patterns of behavior and so they don't seem intelligent and they don't particularly, I mean they do exhibit emotions in that they sort of might run away from you or something. But generally, they're not very sophisticated so they don't feel like real people that you're doing this to. But there are other games, you know, where you've got much more sophisticated agents and ones that, you know, the game producers have talked about as being intelligent, being sentient, having emotional states. So an example is Mario Lives where the characters, the, you know, computer-generated characters that you interact with have emotional states like greed and hunger and so on. And the question is, under those circumstances, you know, can you start to mistreat those entities? Now, other people in games is another matter but just on the issue of computer-created agents. I think there's an interesting
Speaker A:question, I'll come to Peter in a second, interesting question here about what we've talked about before about something being sentient or sapient, okay, and I guess getting into the world of AI and all that sort of side of things. But also, I'm wondering if there's an effect, positive or negative, of how one behaves in a game like that and how then one behaves out in the real world. And that's, as you say, before we even start talking about other real-life
Speaker C:players out there. Peter? Yeah, well, I think we'll come on to that. But yeah, assuming these are dumb agents and they aren't living beings in any kind of meaningful sense, where does the harm come from? Where's the wrong in the being bad? You know, if there's no actual material wrong, no harm,
Speaker A:no foul, right? There's no one being harmed. Let's imagine that you found out that someone you knew had a penchant for buying teddy bears, taking them home and torturing them, right? The person or the teddy bears? They were torturing the teddy bears. Okay. Ultimately, fatally, so that the teddy bears would end up, you know, as just with all their stuffing out and they had to be thrown away and a new teddy bear would have to be brought. Like Sid from Toy Story. Yeah, is that okay? I mean, part of me thinks no. I'd say I'd think there might be something to worry about if I found out that that was someone's predilection. And so I think it's not just consequentialism and we can talk about that because I think there's something interesting there.
Speaker B:Well, is there something wrong with it? I mean, is it a symptom of, is the reason there's something wrong with it because it's indicative of the fact that they might harbour a desire to do that to things that can feel their stuffing being pulled out and that they might be more likely to actually do it as well?
Speaker A:Well, this is where I've not heard enough from. I'd like to hear more from Peter.
Speaker C:Yeah, I don't think being more likely to do something is enough. They have to actually do it, right? They're not guilty until they've done something wrong. But is this torturing teddy bears, maybe not a nice, safe release that they can have in their privacy in their own home, which might actually prevent them from doing things because they can get their stress out on
Speaker A:these teddy bears. And then yeah, I mean, I know this is not exactly the same thing. But I remember years ago, there was this sort of, probably here in North London, where during break times, boys weren't were discouraged from sort of playing things like British Bulldog and beating each other up. And what happened was then in class, they were a complete nightmare. Because they need that time to sort of, you know, to get it out of them. So similarly, like when you're sort of go home to have a nice, quiet evening pulling the stuffing out of your teddy bear, maybe that's good, because it means you're not pulling the stuffing out of any well. Yeah. I think I think the evidence is a bit, the evidence generally says that, you know, playing violent computer games and stuff doesn't make you violent. But, you know, but it's not. At the same time, there are bits of evidence which suggests, you know, there are impacts that it has on certain kinds of, you know, cognitive things and on behaviour. So for example, people playing a driving game makes people more likely to drive more dangerously. I mean, I think these effects are very temporary,
Speaker C: ey there was a famous I think: Speaker A:violence in the real world. And whether or not that was that causative and which direction that went in and stuff. No, I mean, but I think this comes down to a question of sort of virtue ethics versus consequentialism. And I think both get things right a bit about sort of ethics, where virtue ethics is essentially you are the judgments about good and bad or about an agent and about their kind of intentions and about the kinds of ways that they behave and not so much about the things that they do, the consequences of what they do. Consequentialism is very much the opposite and just saying, well, it's really only about the consequences of your actions is how we judge whether those actions are good or bad. Now, I can't, I think virtue, parts of us are all instinctively a bit virtue ethics-y. And I think that it is sort of, I mean, when I play like role-playing games on the computer, I cannot help, I always sometimes think, you know, this time I'm going to be really mean. I'm going to become a kind of, you know, scumbag who goes and kills innocent people and does all the missions where you have to, you know, rob people and stuff. And I can't, I just can't do, I cannot be anything other than a nice bloke in RPGs, even if it's totally consequence-free. Just try that in real life. So I'm not behaving, I'm not behaving as though consequentialism is everything, you know, because I sort of think there's a bit of my identity, which is, you know, which is about not being really horrible to people. You're just too nice.
Speaker B:But is that, do you think that is, do you think that is because you see the lack of virtue in what you're doing in that, in those circumstances? Or do you think that's because the simulation effect makes you feel like you're doing it to a real person? So if it were obvious that that thing was not a real person, you hadn't taken the imaginative leap, the suspension of disbelief to imagine that it was, do you think you could, for example, could you smash a door to pieces?
Speaker A:Uh, I don't think I could even do that physically. Well, okay. But, but, you know, if you were,
Speaker B:if you were, um, looking for catharsis from anger, could you hit an abstractly inanimate object?
Speaker A:Definitely smash a door up. In fact, I quite enjoy knocking shelves off the wall.
Speaker B:Okay. So, so, so there's something about the, um, uh, the, um, anthropomorphisation of an object like a teddy bear, which is no different to say a pot plant or something. Um, the, uh, that, you know, it's the staring eyes that make you imagine that it's an actual.
Speaker A:I think, I mean, so I think it's plausible to say that our, our kind of intuitive sense of what's right or wrong is probably as evolved in a pro in a way to approximate a kind of consequentialism so that you judge with a, someone is good or bad by whether or not their actions in general tend to produce goodness or badness. So fund is driven by ultimately by consequentialism, but actually the way that it feels is much more about, Oh, you know, is this person just a nice person or a nasty person? And, you know, being mean to people is something we're reversed to doing, even if we know they're not real people because, you know, because it operates on a more fundamental level and, you don't be horrible to things with big eyes, fluffy things with big eyes. Um, the reason we've evolved to do that is for some consequential reason, but in fact, the way that we experience it is, and that's, that's, see, that's why, that's why I don't think it's as straightforward as, well,
Speaker C:this is not a real person, so it doesn't matter. Yeah. I think it's worth moving on a little bit and addressing this, the, the do violent computer games make you more violent in real life? And don't they a bit more because it is controversial topic is often cited by the, uh, the gun lobby in America as being a major driver for youth violence and youth use of, uh, use of weapons. Yeah. Um, uh, so there, there, there were some count counter, uh, factual studies for, against the APA study, which it shows there's no, no correlation between a national level national stats between video game spending per capita and any kind of violence. If it were the case, then places like Japan would have as much gun violence as America does. Um, and it, uh, and the critics of, of this idea that violent computer games cause, cause increased violence often see, often, um, see that suggests that computer, the computer game industry is an easy target for the gun lobby as like a scapegoat for all blame, blame those diversions. Yeah. It's convenient to, to, to, for the American gun lobby to shift the blame to virtual guns and to distract from the fact that it's actually probably a major contributors, the prevalence of real guns. Um, but yeah, I mean, I don't think there's lots of studies that suggest that computer games can actually be a quite a useful release of tension and a useful way of, uh, that cathartic, uh, the, the cathartic violence in the virtual world is a safe place to do it. Um, which, which, uh,
Speaker B:I think that begs the question, you know, if you are merrily running around shooting, um, in a computer, in a computer game, yes, yes. Uh, uh, virtual agents, you know, um, you know, that might, that might either, you know, have some effect in the real world, but regardless of that is intrinsically what you are doing in the game wrong. If you're shooting things that are becoming more and more sophisticated, there's a company called, um, Wingman AI who specialize in developing virtual, um, uh, AI, um, kind of co-players for you in games, like particularly in shoot shoot-em-ups, but any, any games, that's what they do so that you can sort of, you know, if you, if you haven't got any real friends, you can play with, uh, this AI and, and, you know, they've found some quite interesting, uh, things, for example, um, in, in some of these shoot-em-up games, they, um, they have found their initial kind of attempts to create an AI meant that this, this thing would just leave the human player in the lurch because it would have made assumptions about the capability of the human player, assuming it could, you know, take out the whole floor of a building or something on its own because the AI... Imagine being abandoned by even your virtual best mate. Exactly. The, the dent in your self-esteem at that point. Yeah. Well, I see now, sorry, go on. Well, I was just going to say, so, you know, in a sense there, there is an AI, there we're projecting like, hang on a minute, you've, you've stitched me up here. You're, you're kind of, um, projecting that, um, ethical decision on the part of the AI, but, you know, obviously the flip side is if you're killing AIs, you know, is, is that, is that wrong?
Speaker A:Well, now you might be tempted to say, what the hell are you talking about? It's just computer program. And I think, you know, that's a kind of adequate response, but utilitarianism essentially says that, you know, we judge the goodness or badness of a state of affairs by the effect it has on people's utility. So in other words, what people prefer, right? So, you know, if, if there's a situation which broadly, you know, a majority of people would prefer to happen than, than, than not, then that's better than where we are now, more or less. So, you know, and now if you, machines at least beyond a certain level of sophistication, artificial intelligences have utility functions, we have to give them utility functions if we want them to learn stuff, uh, because, or at least if we want them to learn stuff for a reason, because they're trying to achieve something, uh, broadly. So non, if, if AIs that are rule-based, which, you know, a lot of them are in computer games, this doesn't apply to, but AIs which have utility functions where you say, right, you want to achieve this. We're not really going to tell you how the world works. You've got to find that out for yourself. Um, but this is what you are trying to achieve. That, that's a utility function. And if we're thwarting it, um, then in, you know, a sense we are lowering its utility. Yeah. Doesn't sound like though in a moral or ethical sense, which I think is what you're talking about. That is the question is, is at what point do, um, online entities become, you know, become worthy of moral consideration, become worthy, you know, become people, I suppose. And I think if you take, you know, uh,
Speaker B:Nick there is sort of, you know, looking at a functionalist view of, um, of whether or not something has a, uh, has an emotional state. So, you know, if, if you set something up to, um, exhibit the, uh, the properties of being hungry, for example, you know, it, it has to seek food in order to satisfy its utility function. Uh, and you deny it food, you know, is that the same as starving, starving a human? So, you know, that functionalist approach, but also from a, you know, from a perspective of behaviorism, if, you know, it's kind of this notion that you can't actually, you can't see, I don't know whether you're suffering or not. I can only ever tell from what I can actually observe. And if what you're torturing exhibits the behavior of something being tortured, you know, it's, it's exhibiting displeasure, you know, by whatever we, we sort of, um, uh, perceive, uh, then, you know, from those two perspectives, actually you are causing this thing some harm is, is a moral.
Speaker C:Yeah. And I think, I think that we're, we're, we're sort of rife with inconsistencies all the time. So, uh, there are, there are religious faiths that seek to do as little harm, minimize harm on the environment or particularly animals, because they all were considered to have some kind of soul. But we, but we, and we ourselves, we, we, we, uh, you know, we wouldn't want, we wouldn't cause unarmed, undue harm to a creature, an animal, maybe, maybe even insects and things that, that, uh, have some sort of agency about themselves, but we still eat meat. Most of us, I think all of us here are eating meat. Uh, we, we still, uh, we still put bleach down the toilet,
Speaker A:killing all those little bacterias, you know, but only 99% of known ones, only 99%. Yeah.
Speaker C:Yeah. I'm fine. But how, but how do we, yeah. But where do you, so where, where we draw a line, so would you pour, would you pour bleach on ants? I would possibly wouldn't because it was quite a horrid thing to do, but if it were, I had to get rid of some ants, I'd probably seek a sort of humane way of, of dispatching them. Yeah. Set up some jam somewhere else. Yeah. But yeah,
Speaker A:I've got no qualms against cockroaches. Those things remind me of the alien from the alien
Speaker C:films. And I I'd happily murder them. But my point is that there's some sort of threshold of like consciousness that beyond which we think we can't be violent and horrible. Well, or is it a sliding
Speaker A:scale and we moderate how moral, how, how kind of, how much we care morally by where it is on that scale. Um, but the question is whether or not it's to do with, uh, intelligence per se, or whether that's really proxying for sentience. And I think they're, they're different things. You could imagine something being very intelligent, but not at all sentient. And I wouldn't feel bad about, about, you know, doing anything to something which wasn't sentient. So I wonder if we're, we we're talking about intelligence, but what we mean is sentience. We definitely mean sentience. Yeah. Um, I want to round us off. There's a question I want to ask before I do any, anything you want
Speaker C:to round off on what? Yeah. Just the computer game world is this, um, the, the, the, the people in the middle are suggesting when I actually give you a game industry, they're already kind of tackling this and, um, lots of game, more and more games are actually, um, trying to, uh, just get away from pure violence as being the main theme of the game. Uh, example would be hit, hit man. Uh, you're actually rewarded for avoiding violence in that game. So if you complete missions without actually killing anybody, then you get more points. So there's a way, there's a way of exploring
Speaker B:violence in computer games, which is, but also, I mean, actually, if you, if you look at, um, if you look at the, the, uh, kind of analogy of sport, okay, well, sport is a sort of a virtual kind of environment, uh, where you try and, um, kind of undermine your opponent's utility function. You know, that's, that's the whole purpose and you sometimes do that violently. And yet that is seen as the, a kind of morally upstanding activity to, to engage in generally within, within society. But I mean, we didn't really touch on the issue of, um, whether or not it is wrong to inflict damage on other people, you know, other people's avatars within a, within a virtual
Speaker A:environment. I think that might be, that might make a good, another podcast actually. I agree. Yeah. Yeah. Yeah. And look, I just want to round things off. Um, I hesitate to ask this and it's quite a predictable question, but I hesitate to ask this cause you're such a bunch of geeks, but, um, what's, um, what is your favorite ever, uh, computer game?
Speaker C:Oh, that's easy. Uh, Kerbal Space Program. As an engineer, it's literally the most amazing game
Speaker A:ever made. I remember you showed me that once and I went off and had a look at it and I think I lost about five minutes and I was like, oof, it's all a bit. Oh, it's just incredible. I can't wait
Speaker C:for the sequel. Hopefully it's going to be half as good as that. Okay. Very good. Uh, Nick?
Speaker A:I think measured by, uh, I mean, I've got very fond memories of games like from the nineties, like Tomb Raider and Civilization, but the one that I think I played the most by an order of magnitude is Minecraft because I play it with my kids. So we all are in the same Minecraft world, you know, adventuring together and building stuff together. And, uh, uh, it is brilliant. I mean, it's great because it does very little for you other than give you a world and some things you can do to that world and everything else is up to you. Uh, it's a genuinely amazing landmark achievement in computer games, Minecraft. My kids like it. Maybe I need to get into it.
Speaker B:You should. It's brilliant. Chris? Well, something happened to me about the age of sort of 15 or 16 where I just... Girls. Yeah. Well, maybe. Um, but I literally, I could not play computer games after that time in a, in a way that really engaged me. I mean, I, you know, my kids sometimes play them. I go into like Untitled Goose Game. I go and have a play and after 15 minutes, I just feel cognitively drained, not in a, not in a good way. And I, for me, like, I think it's two things. The was less, there was less imagination in it. And, um, also I think I started working on a computer and associated a computer with work as opposed to fun. Um, so yeah. So my memory really is, is probably a game called Elite, which was old vector graphics style. You don't need to explain
Speaker A:what Elite is. No, you're right. But the point, the point is, it was too old for Elite. Yeah.
Speaker B:Okay. Yeah. Uh, it was basically a, um, a game where you went around the galaxy, uh, trading in different improvements. One of the most influential games. Actually I think I know
Speaker A:of this because I, again, it was via you guys recommended. I had a look and you know, never had it on the Atari 800 XL, unfortunately. So I never got to play Elite. Um, I mean, I think possibly we agree with Chris a little bit, or I'm the same, which I can't engage with computer games like I used to, you know, um, I just don't find them as interesting anymore. Um, I mean, well, you say computer games, but you guys also aren't mad on the old board games either. Focus in a kind of artificial environment with problems to solve. I don't care if it's a computer game or a board game. No, I disagree. I just love games. No, I do really love board games. Actually. I really enjoy them. I do get into them. It's just, but no, that was an awful game. Um, just not compared to you. Cause you're just such a big sort of gamer on these things, but actually I do like a board game. Um, no, my favorite Jet Set Willie, probably one of the first games I ever played. Loved that. I remember ZX Spectrum. I did. Yeah. Yeah. Yeah. And, and, but then actually really, I played a lot in the early nineties when I was at university, um, uh, on Sega was Echo the Dolphin. I don't know if you remember that. It was really cool. It was just a bit different from the usual kind of, it was a little bit, but it had this dolphin that did flips and stuff. So I quite like that. Um, okay. We'll wrap up there. Um, thank you very much indeed for listening to the Cognitive Venturing Podcast. I'm Fraser McGruer. Um, we've been here with Peter Coghill, Nick Hare and Chris Wragg of Aleph Insights. Until next time. Goodbye.
