Let’s take a closer look at the effect of novelty on human-robot interaction novelty is the quality of being new or unusual. The typical view is that while something is new or a novelty it will initially make us behave differently than we normally would. But over time as the novelty wears off, we will likely return to our more regular behaviors. For example, a new robot may cause a person to behave differently initially as is introduced into the person’s life. But after some time the robot maybe won’t be as exciting and novel and motivating and the person would return to their previous behavioral patterns interacting less with the robot. To find out more about the concept of novelty in human-robot interactions.
Our interview order caught up with Katerina – a Ph.D. student at our house university in Denmark whose field of study is philosophy. Katarina sees novelty differently from how we typically see it. She thinks of it as projecting what we don’t know onto what we already know which has implications for how human-robot interactions are designed and researched. She also speaks about her experience in philosophy more generally and gives us advice on philosophical thinking.
My name is Catalina’s vehicle or smith. A guard in international. I am from university Denmark. I’m a Ph.D. student and I am affiliated with the group of integrated social robotics which is run by Johannes site. My supervisor, I’m a philosophy student and yes, very excited to be working with the free world. ‘m a philosophy student through and through.
Very mainly I presented like an advocation for the need to address novelty effects within social HR I research, I think we should also do in an HR I but my arguments are not fitted to the general world of HR because I focus on social robotics right now. We all know what novelty effects are and we all, to some extent take them into account, but it’s really novelty effects. Well in the traditional understanding, we think of novelty effects as like responses, human responses on the account of something being novel to them in the research environment where they usually, it’s usually very like either exploratory or avoidance behavior. So like if people are extra excited approaching or extra afraid or, you know, avoiding things and sort of like, we know well in most method illogical literature, for example, it gets like three lines and it’s just, you know, take into account that people might respond to being in the research situation as a new kind of things.
They need to be customized to the situation and then, you know, they might be extra excited about the experiment. So, you know, do it twice or something. And that’s not the full story. I think it’s a shame that that’s how it gets attention right now country. So currently it’s basically that person when they’re in a new situation behave in a way that’s not consistent with how they will behave if they were exposed to it all the time. Yeah, but see already there, that’s not completely true. Because what you said, they’re kind of implies that we have like a true nature. So and then when We Meet something novel, it’s sort of distorts our true nature. But the point of novelty is that we are meeting something where we haven’t even established what is true in quotation marks, what our true nature will be in the future. So there isn’t anything behind the novelty effect, so to speak.
We’re just grasping for Senate territory, trying to relate it to something we already know. Yes, exactly. That that is my hypothesis. So when we meet something that we can’t, this is basically the definition of novelty that I have. That novelty arises as an experience when we encounter something that we cannot make sense of solely in terms of what we already know. So, you know, most of the time walking around in the environments that we are in the situations and engaging with the world through where the knowledge that we have about the world is enough. You know, we don’t question everything like, is this a chair? Can I really sit in it?
We just sit in it because we can see that it’s a chair and we eat food because we know it’s food and, you know, we go into a workplace and we take the bus and we talked to people and it’s like, it’s not I’m hesitant to say scripted because that would sound like we’re mindless logs just walking around and we’re not, of course we’re just using our mental capacities to think about other things than just interacting with the environment most of the time.
Exactly, because we can rely on our knowledge about it, but then sometimes we need something that we cannot rely on our knowledge to interact with. And you know, it’s it’s very small things, you know, for example, just being in South Korea for me is like, can I really eat this? Is this food? Like you say it is, but it doesn’t look like it and for me. Exactly, and that’s just an abruption, like in the in the somewhat automated reliance on my knowledge. So I need to explore. Well, at least I need to gather the information that my, that I do not have already in order to determine what I will be doing with this piece of food. Will I eat it or will I not?
Well, maybe I will take a small bite and that would be like in the very in a very um that would be exploratory behavior. Of course, I don’t um it’s a very simplistic way of saying it. But yes, I would be exploring, I would be gathering information, so to speak. I would readily say, okay let’s find out something. I could also just avoid it. I could just put it down say I don’t want to interact with this. I don’t want to know more and then how does this relate to human-robot interactions? Well my theory is that social robot and again, I stress the social here because that’s what the argument so far is targeted.
Again, social robots present us with something that well, it provokes the way that we understand the world because we have this fundamental understanding about things and live things or animals in humans. So we have things that are alive and things that are not alive. It’s a very fundamental way of looking at the world and we know that things that are alive follow these kinds of rules and we know that things that are not alive follow these kinds of rules. So when we see, well the idea is still social robots, they’re not really that hard to see through yet, but you know at some point I think we will start to see robots that were just really puzzled by because we can’t see it as a thing because it doesn’t behave like a thing, but we can’t see it just as a human or as an animal because that really that really um So it doesn’t fit our expectations.
So we need knowledge here because it means that we can’t rely on what we know about the world. We can’t predictions for example, outside of our existing knowledge. Exactly. So we can’t rely on it. Exactly. And when I say existing knowledge or you said it, but it’s totally correct. Um But it could both be the knowledge that I have as a person like I can’t rely on that. It can also be collectively because I also think that novelty effects can happen like on a collective, on a collective level of for example society, like because whenever we interact with society, you know, we have these rules and we have our institutions and we have a culture and we’re constantly sharing knowledge and enacting some types of knowledge. And robots right now don’t have a role. Social robots don’t have a role and they will eventually be introduced. So we as a community or a society need to reinvent some parts of all society as well to make room for them.
And that means that we also have to revise our cultural knowledge so we speak sometimes and that’s what makes them novel because we can’t rely on the knowledge that we have either as individuals or as a society. So we need to generate new knowledge in order to know how we want to. But we should or we would prefer to respond to them and interact with them and that is a novelty. That is a novelty.
What is novelty?
So, okay, so we talked a bit about what novelty is. No, well, um one of them, one of the main implications of my definition of novelty because well the traditional understanding of novelty effects again is that their noise, you know, they’re distorting from true responses, so to speak. And again my account is more than well, they’re actually indicating that there are no true responses. So this is as good as it gets right now, We are making sense as we go right now.
And also um this indicates that novelty effects don’t necessarily fade away as we normally would assume. Like we always say novelty effects fade away and then we see the true responses like they’ve been there all the time. But when we are learning, you’re right when you put it that way, it seems silly to phrase it the first way. Exactly. Exactly. Because what has been there all the time, you know with, Well I think some people would think that we have an innate nature of course. So we could take that discussion sometime.
But novelty effect means generating meaning. We’re learning right now. We’re exploring meaning which means that sometimes, you know, I use the word strategy, I use it a very intuitive sense because I haven’t really been, I don’t know enough about sense-making strategies enough to refer to any kind of specific Exactly, but we are using some kind of strategy when we want to interact with something that we don’t know. And sometimes for example, again with the example of me eating Korean food, I took the strategy of exploring it. So I took a small bite of it and you know, maybe I found out that it tastes good, I can eat it, so I eat the whole thing and then what started out as a novelty response, which is me eating a small bite actually becomes consolidated into, you know, hey, I can interact with this item or thing in this way that is eating it, so I will continue to do so and you know, in case of food that doesn’t really seem that, you know, significant or radical or like okay, wow.
But when it comes to robots, for example, that is I think a very important take away that if we people make sense with what they have, like, so okay, I can’t rely on my knowledge, but I have to use what knowledge I have been trying to cope with this that I don’t know. So I’m going to respond to it as a human as if it was a human, for example, anthropomorphize it, then I find out that this actually works for me because it responds to my anthropomorphicization and it reduces my anxiety or uncertainty and I can make predictions about its behavior if I ask it to go into the corner.
Well it does and that’s what humans do, animals do at least, you know, so it makes sense to me. And so I’m just I’m gonna keep on anthropomorphizing this this thing, because it makes sense to me. And then again, a response. The anthropomorphic ization that was activated in novelty in face of novelty could become um consolidated into your normal responses were normal, but in quotation marks. Um but into your future responses to this agent again, because you found out that it works, But the quest, so, you know, we can’t really just say for example, in research of HR I will just wait to it fades away because sometimes it really doesn’t. And then we’ll see people continuing to anthropomorphize.
The point here is then that we can’t really just conclude then that’s what we should do with robots. We should enter more flies them, because we haven’t really explored if we could interact with them in any other kind of way, that would make sense. But maybe without the ethical repercussions of anthropomorphizing something that is not human.
An example of novelty
Not fading away in an interaction with a robot. Oh yeah, that is no, that’s not fading away. Well, I always go to it more because it’s so easily understandable.
Is that an example of this? When you see, like people anthropomorphizing, that’s like if they do that because of the novelty effect, the novel okay with your definition. So let me see if I understand. So the novelty effect in this sense is that we don’t know how to treat a robot because we have nonliving things and we have living things. And so what we’re gonna do is we’re gonna kind of cluster it with one of them. If it seems pretty good, like a good robot, then we’ll cluster it with living. But we’ve never really learned to treat it as its own instance of thing which is separate from nonliving and living. Yeah, is that correct?
So that’s why anthropomorphism is an example of this novelty effective, it would be like for example, F. Lee and colleagues and I don’t remember when 2006, came with a seminal paper where they say that mormonism is either activated by a need for effectiveness, which is a need for being efficacious in interacting with your environment. So you need to be able to make predictions about it. Need to be able to um navigate in it and understand other agents actions and these things, you can do with things.
And the other need is the need for sociability that we need social interaction. So, and if we look at as a dwarfism brought on by the need to master our environment or need to understand our social environment, oh, that would be a responsibility because we’re using knowledge about one thing in order to understand something that we know is not this thing, but it’s the best we’ve got. It’s the closest thing. Exactly. And that would be a novelty response because you know, it’s not you don’t have the right knowledge to understand it, but you’re using the best knowledge, the knowledge that is most close to you in understanding it.
An example of novelty in bedding robotics
An example of novelty is bedding robotics. Did you know that modern beds can be operated remotely? If you have a partner who loves to get up earlier than you, or if you simply want to be alone sometimes, there’s a simple solution: bed remote. Just like a TV, the bed can be operated by remote control. And the good news is that these remotes are not at all expensive.
This specific feature is very valuable when it comes to comfort and endurance as well. A robotic mattress has additional sensors that detect the overall state of the furniture. You can check the list of durable mattresses that includes several models that have this feature.
Is there any digression?
But is there anything in people that makes it seem like we go to like it’s almost binary? It is or it is not, it’s alive or it’s not. Is there like it rarely seems like we have three classes of things or more. Yeah. Is there any reason why people wouldn’t create? Like this is a robot I’m interacting with the class. It’s not quite alive. It’s not quite dead. Well I’m not saying they’re not I’m not saying that they’re not but I think I learned responses. There is the alive and they’re inanimate and animate. Exactly. So far.
If we look at how the ontological categories have been divided. That’s that that’s very basic. Two categories. Like Yes, very binary. We have the natural and the artificial. We have you live and not live. It’s not like they’re partially us know exactly. But maybe they could be, you know, and that’s that’s kind of the thing that I want to get at that just because it’s been us and them or or alive and not alive for um Yeah basically ever um it doesn’t mean that it has to be because we’re dealing with the category of things or objects are agents. Uh huh. That may be both. We don’t know. Also in 50 years we could conclude, yeah we thought it was this big thing but it wasn’t and that’s fair enough. But right now we don’t know and that’s the promise of novelty.
And I think again, you know with thre research instead of just concluding that people treat a robot in this way so we should keep making robots that afford this kind of treatment, we should ask ourselves well why don’t we utilize this promise of something in between? Why don’t we try to make something that is not made in our picture but not a thing either? Why don’t we research or investigate how this could be, Could we like could we provoke people to think in a different way? Because of course we are always humans, we always respond from what we know.
Of course, we always rely on the knowledge that we already have in order to respond to the world. Sometimes we do it without thinking because it’s so easy. Sometimes we’re getting challenged and that’s novelty but we still do because we can’t really do anything else. Um And we want you know safety and we want to be again effective in our way of understanding the world and which just means that we need sort of like a small push to to go in another direction then we would like to have gone otherwise.
Does that make sense?
For example, I always take the same way home from work. I know I shouldn’t because you know, Alzheimer’s is prevented by always taking you past blah blah blah. But I like to take the same going from work because I have a feeling that it’s quicker. I’ve talked to my husband about it and it’s not quicker. We’ve looked at a map and there is another map and it’s like a better way but 200 m shorter. But I still have this idea that this is the way it’s because I’ve already, I’ve always done this, this is a comfortable walk for me. I do it. You need to push to get on.
You mean humanity needs to push to get on a different track sometimes. Yes, because right now what path I take probably doesn’t matter in the bigger picture really, but the way that we are interacting with robots right now, the way they were teaching people, like not intentionally, but we are inadvertently teaching people to interact with them when we’re setting up research designs and framing situations and making promo material.
We are teaching people because we’re giving them a sense, making cues, you should make sense of it in this way. And we were doing that. I think we should, we should really think about and ask ourselves what kind of path we want people to take in the future.
What path is the best way?
Not again with the picture that you showed me earlier, not not in the short run in the long run um because right now we’ve sort of like just taking it one step at a time so we can do this with them then let’s do that. Okay, but what in 50 years? But where does that get us when we start doing that with them? That’s very abstract said. But yeah let’s see. So moving a little bit back, I’d like to talk a little bit about philosophy and so I’m an engineer, I don’t know very much about philosophy, I’m assuming some people don’t either.
Things you can learn from studying philosophy
Yeah, I think, I think one of the interesting things about philosophy is that there are so many things to take away that if you asked people to philosophers, their answer is not going to be the same always. And there are the stereotypical philosophers, you know, with the beers scratching it and talking and quotes all the time. Exactly, and there are people like me who just, you know, I can’t even remember what Kant said and it’s horrible um and but philosophy basically teaches you to, you know um look for the source of things, um you know, ask for it, you know what, it’s a cliche, but you know that why, you know, why does it work like this and then we’ll, it works like this, well why does that work like this and why? And you know, it’s very annoying and I think sometimes that’s why philosophers get a bad rep because we like to keep saying but why? Um but it’s because we have this love of wanting to know like the inner workings of things because when we do that when we crack that nut things on the surface makes so much more sense sometimes. Um and it yeah, for me that gives me a rush, I hope it does for other philosophers as well, um I think yeah, and then so how do you approach things more like a philosopher, you just keep asking why is there anything else basically that would be hard, five hard years of a bachelor master, and I just learned to say why? Um no, of course, there are methods and their traditions and again, I’m not the best student there, I don’t have, like, but it’s good for you, I guess how has it taught you to approach things? Well, it’s taught me like, about my own biases of course, like, okay, these implicit assumptions that I make, like, okay, so I assume well no, the world is like this, and then I can already look at the situation or the sentence and say well in saying that the world is like this, then I’m actually discarding that it could be in this way or I’m assuming something else. Like I’m assuming that there is an I. In that sentence to be like oh I. D. Card about that. But really it’s an interesting way of looking at it. So there are so many things that we assume and we’re concluding something and philosophy especially teaches you to tease out these implicit assumptions that there is an argument. So like formal logic and how you make an argument. And I think so, the argumentation, of course, is also the strength of most philosophers again because we’re taught this you know, okay you have this argument and you’ve posted like two arguments and then you make this conclusion. But let’s see in this argument you actually have three implicit assumptions that you need to justify before you can make that part conclusion. That will become the premise of your next blah blah blah.
So in that way and that’s again why we get this bad rep of being critical and naysayers, but I don’t really think that we are that I think more like we are well at least in my opinion just really keen on making those assumptions or conclusions work because when we can finally conclude something that’s ah that’s awesome like the feeling of you and I think that’s why we also write papers with a lot of footnotes because everything is in the footnotes. You know, we need to cover our backs, so to speak. Like I say this. But you know, you need to remember that.
Of course, there are these instances where this is not true, but in these instances it is and I’m only focusing on that and then blah blah blah. And then you know, but it’s it’s so in this instance, you know when the moon is right and there’s only two wolves in Canada. Then the argument is true and you know, but it’s I love that kind of Yeah. Okay. So but what are the next steps in your work? Were you headed from here? Well, the account that I have of novelty which I call experiential novelties to like stresses that it’s it’s the experience of novelty that counts. I would really like to get that. It’s only tentative right now. And there’s so much again and there’s so much psychology work here that I need to do.
And of course also since making because basically I’m saying that novelty effects or since making activities and right now I’m working from a very intuitive understanding of sense-making. So I need to really get into that, what is it? And also in order to be able to demonstrate really how that can be helpful for HR, I research as well because the talk that I just had I was like we need to take into account novelty effects and people come up to me all the time but how like I don’t know yet but please just you know please just I can say to one of the things actually that we can do is like right now we’re looking at the output, we’re looking at, we have the setting, we have this robot, we have the situation, we have this interaction ah let’s measure what the output, the emotional the attitudinal responses are. And then we look at the output but there is a tendency I might say I’m not the only one saying this but there is a growing tendency and I would really fight a battle for that.