Collective Behavior (Discussion) — Course home

Hello and welcome. It's August 16th, 2023. We're in the collective behavior discussion section of the active inference for the social sciences course in 2023. So perhaps let us all go around and give an opening introduction and or reflection and see who joins and what happens from there. But just so that everybody has said hello. Hi, everyone. I'm Brun. 9pm in South Africa. And yeah, just excited to be and see what happens. Hi, I'm Andrew. I'm a graduate student in Indiana. And I guess my field of interest now is the knowledge management. And I think that figuring out how we collectively learn and can share information. It's a really interesting topic. And I'd like to see how this would apply. Hi, I'm Regina. I'm from Guadalajara, but I'm living in Spain, studying and doing experiments related to this topic and trying to learn a little bit of the theory that's behind it. So we can, you know, advance better. I'm excited. Hello, my name is Sasha. I'm calling in from California. I'm also a graduate student. Yeah, just really enjoyed the spirited presentation last week. I'm looking forward to a discussion. All right. Well, how about anyone can give a first reflection on collective behavior? What was their thought or context before and where are they at as we kind of begin? Yeah, Regina? Okay, so I just want to make sure that I understand because I took like two bigger lessons from your lesson, right? And the first one is that, you know, this, this famous model of active inference that has the agent and the environment and the mark of blanket in the middle, right? And just interacts with everything, right? So I think what I got from what you discussed, maybe I didn't have a good definition of an agent. But first, but you can have that agent not be just an individual, but also your definition of an individual changes, depending on what you want to study in the behavior, right? So an agent can be an ant, it can be an ecosystem, it can be an eye looking at something, or it can be as a society, for example. And I think that's a big take for me because it changes a lot the concept of how to study things and use active inference, just by changing how big the agent can be. So I don't know if I got that right, right? So first that clarification. So what is an agent? Yeah, so I don't know if there's like a formal definition of an agent in active inference, right? But if we, if we want to study whatever we want to study, because I'm not going to study because I have to do experiments. That's like my goal, whatever I'm doing. So I want to study whatever I have to choose what's going to be my agent and the environment it has. So choosing the agent, you can do it to create a model or to make an experiment, whatever. But the choice you make, it will depend on your, what you understand as an individual or as an agent, if I think so, right? Awesome. Anyone have a thought on that? No. If I continue with just to add something else and this will be my, my, um, and then I'm thinking is that, um, also a big problem that we have, or I think I have is that since I am part of the environment, um, you know, you know, the perspective of whoever's studying whatever it is, it's also going to be an effect in it. So maybe, maybe scientists already decided, but I don't know that we can never have an objective way of studying something if you're part of the environment of the agent. So I don't know, we already forgot it or we already decided that's it. We cannot have objective observations of behavior in general, but you know, you always think about like the agent environment and you studying it as something else, but there's something else is also affecting it. So I see that as a problem or maybe something that has to be said, like we are not objective because we're part of what it's affecting our agents in the behavior collectively too. So those are the thoughts I have to share. Uh, Regina, I'm, uh, I'm also new to, to all of this. So my preliminary…