Tessa Rusch is a postdoc working on computational modelling of social interactions at Caltech in the labs of Ralph Adolphs and John O'Doherty. She is also part of COVID-Dynamic project, a large-scale longitudinal study on the psychological effects of the COVID pandemic.

In this conversation, we talk about Tessa's experiences of being part of such a large project, about her move to the US just before the pandemic, and about her review on computational models and bevioural tasks of Theory of Mind.

BJKS Podcast is a podcast about neuroscience, psychology, and anything vaguely related, hosted by Benjamin James Kuper-Smith. New conversations every other Friday. You can find the podcast on all podcasting platforms (e.g., Spotify, Apple/Google Podcasts, etc.).

Timestamps
0:00:04: Before Tessa's PhD
0:04:07: Tessa's first year in the US, during the pandemic
0:16:51: Tessa's original plan for her postdoc with Ralph Adolphs and John O'Doherty
0:24:22: How COVID-Dynamic got started
0:32:42: The practicalities of running a large collaborative study
0:43:37: Social changes during an extremely variable time
0:55:03: Working with complex data sets
1:14:02: Doing COVID research while working on other projects
1:20:48: Discussing Tessa's review article about Theory of Mind from Neuropsychologia
1:47:27: Tessa's final words of wisdom

Podcast links

Website: https://bjks.buzzsprout.com/
Twitter: https://twitter.com/BjksPodcast

Tessa's links

Website: https://www.hss.caltech.edu/people/tessa-rusch
Google Scholar: https://scholar.google.de/citations?user=OzT7c-oAAAAJ
Twitter: https://twitter.com/tessa_rusch

Ben's links

Website: www.bjks.page/
Google Scholar: https://scholar.google.co.uk/citations?user=-nWNfvcAAAAJ

References
Kuper-Smith, B. J., Doppelhofer, L. M., Oganian, Y., Rosenblau, G., & Korn, C. (2020). Optimistic beliefs about the personal impact of COVID-19. PsyArXiv.
Post, T., Van den Assem, M. J., Baltussen, G., & Thaler, R. H. (2008). Deal or no deal? decision making under risk in a large-payoff game show. American Economic Review.
Rusch, T., Han, Y., Liang, D., Hopkins, A., Lawrence, C., Maoz, U., ... & Stanley, D. (2021). COVID-Dynamic: A large-scale multifaceted longitudinal study of socioemotional and behavioral change across the pandemic. PsyArXiv.
Rusch, T., Steixner-Kumar, S., Doshi, P., Spezio, M., & Gläscher, J. (2020). Theory of mind and decision science: towards a typology of tasks and computational models. Neuropsychologia.
van Baar, J. M., Chang, L. J., & Sanfey, A. G. (2019). The computational and neural substrates of moral strategies in social decision-making. Nature communications.
Van den Assem, M. J., Van Dolder, D., & Thaler, R. H. (2012). Split or steal? Cooperative behavior when the stakes are large. Management Science.

[This is an automated transcript with many errors]
Benjamin James Kuper-Smith: [00:00:00] Wait, you in Berlin?

Tessa Rusch: Yeah, I did my master thesis in Berlin.

Benjamin James Kuper-Smith: At which one? The mind and brain. Or Mind? And Brain. Oh, I can't remember you doing that. Oh, okay. Who did you do your thesis with?

Tessa Rusch: Nico Bush. I mean, I wasn't, I didn't do my master's in Berlin. I did my master's in Munich, but then, um, I went like, um, for half a year I went to Berlin and did my master thesis with him.

Benjamin James Kuper-Smith: Oh, okay. I didn't know that.

Tessa Rusch: Yeah, it's the time when I was still studying vision perception processes.

Benjamin James Kuper-Smith: That stopped though, once you did that project or?

Tessa Rusch: Um, I don't actually know. Yeah, it stopped. I actually, um, my, um, my bachelor's thesis was in visual perception. My master thesis was in visual perception. I worked as a he in a visual perception lab.

All I did my entire. Pre PhD academic training was, um, focused on visual perception, [00:01:00] um, and alpha oscillations. And then I kind of stopped and moved to Hamburg and worked with Jan. I mean, Jan just needed someone who knew EEG, kind of, and I wanted to switch to more computational, um, approach,

Benjamin James Kuper-Smith: which doesn't exist in vision

Tessa Rusch: Own, I mean, of course it does.

No, I was just like, to be honest, when I started my PhD, it was kind of, um, I wasn't really sure whether I wanted to stay in academia. Um, it all sounded very stressful. Uh, and so I took a break for a while and I worked at the, uh, at the, I visited in Dar. On like a, like I was just kind of like a long-term hebe person and did, um, program their experiments and like ran a hundred of hours of, um, recordings with kids.

Um, and then I figured, oh, I'm doing this. I might as well do, get a PhD,

Benjamin James Kuper-Smith: get a degree for this. Yeah,

Tessa Rusch: yeah. Don't, don't send that on the radio. I don't want people to [00:02:00] know that I wasn't actually, no. Um, so yeah, no, I kind of, I kind of switched fields completely. Um, I thought it sounded interesting was, um, what Jan put in this ad, which actually ended up to be lace thesis.

So when I applied, um, the position was already taken and so, um, I was actually quite, it was stressful and it was fun because Jan said, yeah, sure, you can come and you can pretty much do whatever you want, which sounded great, but then on the other hand, you are 20 something and you have no clue and you're completely new to the field.

So, um, can also be kind of daunting,

Benjamin James Kuper-Smith: but yeah. Okay. I didn't know that you. Division the entire time. But I guess you, you said that academia center stressful, I guess that turned out not to be the case at all, right? It's very relaxing.

Tessa Rusch: Oh yeah. It's

Benjamin James Kuper-Smith: like at a spa every day.

Tessa Rusch: So walk in the park. Um, yeah, no, I mean, it is stressful and we all know that and every, like, all of us are struggling with that, I [00:03:00] think, and everyone at some point in our career think about quitting maybe.

I mean, there are certain things that are great about it and there are other things that are really hard and, um, yeah, difficult to deal with and I think. From what I've heard and what I feel myself, it's like people go through this, these oscillations in a way. Like sometimes they're really excited about it and then everything is just really tough and hard and you don't see any perspective.

And it's, but I mean, I think for me, the biggest advantage of academia is it's like, I kind of like to hang out with these people. They're all slightly weird and have weird interests and very specific interests. And it's usually, it's very hard to find, um, a researcher that isn't fun to talk to,

Benjamin James Kuper-Smith: but you know, you can have the social interactions without having to work with them.

Right.

Tessa Rusch: That's true. You can just

Benjamin James Kuper-Smith: be friends with them. You don't need, is this just one big socializing program for you?

Tessa Rusch: Yeah. That's true. You spend a lot of time at work, don't you? And then, um, imagine being in an office and you have all these colleagues that you don't get [00:04:00] along with or you don't find interesting, and then you have to spend all your day with them.

Benjamin James Kuper-Smith: Yeah. By the way, have you actually, so you've been in the US for like. A year now, right? More or less. Exactly. Or,

Tessa Rusch: um, yeah, I started working here May, uh, March 1st, 2020.

Benjamin James Kuper-Smith: Okay. So, yeah. Now, so we're recording this on the 22nd.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Actually, what was the, uh, so a few questions about that. First is, have you actually seen anything of the US yet, or have you just been at home for the last year completely?

Well, I've

Tessa Rusch: been at home a lot like everyone else. I mean, we, we got here and when we actually, um, we got here two weeks before the travel ban, um, was instantiated. So we got here two weeks before we actually would've not been able to come here. And the funny thing though is at the time there was like very few people and like maybe two, three people on the plane weren't wearing masks.

And there was still this thing in China and officially there wasn't a [00:05:00] single case. I think, uh, in the us I think also they had, like, the tests weren't working in the beginning or something. Um, but so we got here and COVID wasn't a thing yet, and so it was just normal life. LA is very much defined by traffic and it was very, very dense traffic.

Um, it was very hard to get to places. And so we actually, like, we managed to buy some furniture, which was nice because, um, and we already had a place, place when we moved here, so that was good. But, um, then I think I went to the office for about a week and a half, and then it started being like, okay, um, it's this weird thing going on.

We don't really know when we'll be able to come back, so try to get oil your belongings and clean out the fridge. And it's, it was, it started to become this, this really scary, daunting thing. And, um, I, and then they said, you can stay home if you want to. And I kind of decided to stay home because I was taking public transportation for about an hour one way.

And, um, I wasn't, and, and people started to wear [00:06:00] like garbage bags and all kinds of stuff to protect them. What was really weird.

Benjamin James Kuper-Smith: Over the face or just as,

Tessa Rusch: uh, yeah. Of their entire body, essentially. Like they were wearing, wearing homemade hazmat suits. And then, um, pretty soon. Yeah, of course. Yeah. Um, so I'm still wearing one as you obviously.

Benjamin James Kuper-Smith: Yeah, that's true. Um,

Tessa Rusch: yeah.

Benjamin James Kuper-Smith: Tessa is, is wearing just rubbish bags, bin bags right now.

Tessa Rusch: Yeah. Um, and, uh, no. So I, uh, that's when I started to stay home. And then three days later, um, actually on a Friday night, we got an email. It's like, go to the office, get everything you need. You will not be able to enter campus, campus from Monday on.

So we all, like, we went, like, made the schedule, and then we all went in and we grabbed our screens and our keyboards and emptied the fridge. And we all met in a kind of hurry and we wouldn't, like, we didn't really know what was going on and we packed everything and we had a rental car at the time, and so we packed everything in the car.

[00:07:00] Jonathan picked me up and. So we drove home and then, um, it started raining, which is a very weird thing in Germany, uh, in, in Southern California. And it started raining for about two weeks. Very heavy rain. It was a pandemic. We were on lockdown, everything was closed. And that was just at the beginning. And we sat an hour, like one bedroom apartment.

Um, and it was, uh, yeah, it was pretty scary to be honest. In the beginning, we didn't really know what was going on and everything was closed and the shops were empty. The shelves were completely empty, not just toilet paper, but anything that you could eat was gone from the shelves. Um, and these gigantic lines in front of the grocery stores.

And so it was just really, really weird. Yeah.

Benjamin James Kuper-Smith: And you didn't know anyone there, or,

Tessa Rusch: I didn't know. I mean, I, I knew the two people that used to live in my apartment and I, I know, but the majority of my colleagues, I hadn't met. I have met like three or four people and the majority of my colleagues [00:08:00] I met over Zoom and we started having these social zooms with the lab and we had very frequent lab meetings.

And Ralph, um, so I'm with Ralph AOLs and Don already. And, um, Ralph started giving, like, we had to give presentations in our lab meeting about COVID, about the virus, about mitigation strategies, how it's gonna affect different parts of the world. So we actually went through a really scientific progress, uh, process of processing it, um, which was interesting.

So they, not as a

Benjamin James Kuper-Smith: project, just

Tessa Rusch: No, no, just in the lab meeting. Just the lab. And we kind of split up the topics and it was just. To get ourselves informed and know what's going on. And we were looking at the different predictions at the different models. And actually I remember back at the time when we saw that, it was like, okay, we're gonna have probably, we're gonna have a spike in spring here, and then if we go on lockdown and everyone stays home, we're gonna flatten the curve, which was this big thing on Twitter.

Right. Hashtag flatten the curve. Yeah. And then what would happen is there's gonna be this [00:09:00] bigger, gigantic peak in November, OC October or something like that. I don't remember the exact date, but looking at this graph, it was like, oh, actually it's not gonna get any better anytime soon. And that was, these were the models in March, and so we already knew.

We could have all already known that it's gotten to be really bad in, in, in the fall. And that's what happened. It was just, it was exactly how it was predicted. And then, um, of course it was also a strange time because, um, it was still the, um. Prior administration and it was, um, there were a lot of questions about visas and they started questioning, um, whether students that are all online actually are gonna lose their student visas because they don't actually have to be here.

There was this travel ban. Um, it was all these, um, these blockages of J one visa, which, um, all the scholars are on, um, the majority of the scholars are on. Um, and so we really just made it to the US because right after we came here, [00:10:00] um, we wouldn't have been able to enter the country anymore. And a couple of people got stuck in Germany, um, or in China.

And so we had lab meetings. It was really hard to find a time for lab meetings sometimes because we had to cover China at all, all the way to Hawaii, which is a pretty big range. And that was sometimes hard. Um, yeah.

Benjamin James Kuper-Smith: But did you want to come to the Yes then, or were you a bit hesitant about doing it?

Tessa Rusch: Well, when we left, it wasn't a thing.

It was this, I mean, it was a thing, but it wasn't a thing in Europe and the Western world, I would say. And it was this. It was similar to, um, when SARS or the swine flu was around and we knew that there was something going on or Ebola, which I kind of, I mean, this was a really eye-opening time, I think to some degree because there have been several pandemics over the last decades.

It was just we as Europeans or as US Americans or the western, northern western world have not been, um, that strongly affected by it. And [00:11:00] that was just. It was weird how all of like, I mean, you empathized a little bit with China and with, oh, they're all on lockdown and they're sitting in their homes and it's, they can't leave the city and that, oh, actually we're affected too, and this is not something that can happen here.

And I think it was just, I think for our generation, it's probably the first time we really experienced something that dramatic. And from traumatic, maybe, I mean, we haven't experienced any war in Europe. We haven't experienced, um, major political crises. Um, there have been economic crises, but it was never, I think for us, um, that was kind of the first time that we experienced something like that.

I, I, I wonder whether our parents, maybe with pure novel, it was similar when they weren't able to leave or eat fresh vegetables or something like that. But even with, um, the, no, I'm blanking. Oshima. No.

Benjamin James Kuper-Smith: Oh, the, the Japanese, the, the,

Tessa Rusch: yeah, what's the name? The salami? Earthquake.

Benjamin James Kuper-Smith: Yeah. That was Fukushima I think.

Tessa Rusch: Fukushima, [00:12:00] yeah. Even that didn't really affect us, right. Even though it was a gigantic. Um, so we have been sitting in this very comfortable nest and everything was fine and we were one ring and boring about very everyday things. And so it was, this is why I think this will be interesting for us as a generation or like the generations around us, because we've never really experienced anything like that before.

Benjamin James Kuper-Smith: Yeah. I found it really hard

to,

Benjamin James Kuper-Smith: oh, sorry. Continue.

Tessa Rusch: No, I was just, um, I mean, we've been in the US for a year now and we couldn't, we weren't able to travel back to Germany. Right. So it was. My mom said it at some point when I talked to her on the phone, it was, oh, it's kind of like during these times of war when it was just these letters from the front and people couldn't come home.

So we literally couldn't come home. I mean, it wasn't, it wasn't, we were fine. We have a job, we have a home. We're very safe. Um, it was just an interesting [00:13:00] experience not being able to go see a family or anyone.

Benjamin James Kuper-Smith: Yeah, yeah. Yeah. I mean, what I find weird is that, you know, I mean, if I think in some sense you could say also quite affected by this, not only because I did a research project on it and spent a year of my life, uh, doing this thing and also moved during a pandemic to a different city, which in this case wasn't that much of a problem, but it's still weird being in a new city and everything's closed the entire time.

But in a way, like, I don't think I've really been affected, like, in a way it's weird that like this huge thing's happening kind of around you. But I mean, number one, I'm not. I'm kind of introverted, so I'm not someone who needs to go out to town every day or something. Right. Like I, my life is pretty similar compared to the way it normally is.

And second, also, like in terms of just like job security, like, this happened at the second year of my PhD. Right. So, and I had four, I have four years funding, so it's, it's pretty [00:14:00] much guaranteed that it's gonna be over by the time I, I have to look for a job basically. Right. So in a way, like it's, I've been really lucky in just like how it fit into my life.

Like if this had happened two years earlier, I would've, God knows, like that was not a fun time in general. And then to have this on top of it would've, yeah. Would've made things pretty rough. But, so in a way, yeah, I don't even know whether it's gonna make, like, in terms of like memory, when I would've think back, like, let's say in 20 years about like my life.

I'm not sure it's gonna be a huge thing. But for me personally, well, because again, like I've, I'm, I'm what I, I do what I do anyway. I'm at home and I read books. It, it's not, I mean, of course it has like an effect and it wasn't always easy, but I think in a way I've been lucky that it, even this big thing hasn't really affected me that much.

But again, I think it's more due to circumstance rather than anything else.

Tessa Rusch: Yeah, I'm, no, I mean, I've, I have to admit it wasn't for me. It was really hard [00:15:00]psychologically. Especially in the beginning when we went on to do this lockdown extremely quickly, from zero to 100 essentially, and everything was closed from one day to the other.

And we were just, I mean, the US or California closed really early compared to other US states. And then all these, um, news from New York were coming in and bergamo Italy. And, um, yeah, I was, I, I remember checking all numbers every morning, different newspapers, different statistics. Mm-hmm. Every day. And I like reading different models, reading different predictions.

Um,

Benjamin James Kuper-Smith: yeah, I had that too, but it lasted for like a week or two. Like it wasn't, it was pretty much, I, it was basically when we were collecting the data for our study, which was mid-March to, well, we finished in mid May, but basically the main work was done from like mid-March to, let's say three weeks into early April.

Then there was the same thing, but like [00:16:00] outside of that.

Tessa Rusch: It's, I think to be honest, it's still the first thing I do in the morning. I look at the R value.

Benjamin James Kuper-Smith: Oh, really? Okay.

Tessa Rusch: In the us in Germany, in the region where my parents live, um, and then in California and I, I always know whether it's above or below one.

And

Benjamin James Kuper-Smith: so what's it now? What's it today?

Tessa Rusch: It's a little bit over one. In Germany it's like, yeah, it's slightly over one, but it has been for a couple of days and, and the, the, the trajectory that Germany is taking is really bad. I'm really worried.

Benjamin James Kuper-Smith: Yeah. I was about say it's in a way that's not surprising because the numbers are going up for the last few weeks, so

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Almost. Of course it's above one, but yeah, we'll see. I just booked train tickets to visit my family over Easter. We'll see whether I can use them. Um, maybe I should have waited until tomorrow because they're making some sort of decision today. But. Uh, we'll see. By the way, what did you, so you mentioned you are working in Ralph AOL's AOL's Lab.

What did you, I mean, we'll get to that, you spent your [00:17:00] time doing COVID mainly this year, but what did you actually want to do? Like why did you, what was the initial plan for going there?

Tessa Rusch: I mean, the initial plan, um, as I said, I'm with, um, John had Doty and Ralph ados. Um, and so it's, it's two labs. Um, and John, a Doty's focus is decision making.

Benjamin James Kuper-Smith: Uh, so just briefly, is it like equal, like half, half or is it like mainly one with a bit of methods help from the other or how exactly?

Tessa Rusch: It's, I mean, I have, my, my office space is in the ADOS lab. Um, but that was just because there was more room, so it was more of a logistical decision. And also we moved to this new building and I have a beautiful view of the St.

Gabriel Mountains, and I'm very happy about that.

Benjamin James Kuper-Smith: Tessa showed me, it's a very nice view.

Tessa Rusch: Um, and so, no, I mean, um, I applied with both of them last, no, no, last 2019. Um, and so it is essentially half, half right now. I spent more work on, uh, in Ralph Eros lab because the COVID study is [00:18:00] run with the ADOS lab.

But, uh, I, I'm in both labs lab meetings and participate in discussions. And, um, so, and essentially that was also, um, the goal because they, their shared expertise is actually what is, what is great because, um, John is a, is an expert. Modeling of decision making and, um, has decades of work on, um, different forms of decision making, habit formation, um, all these kind of things.

And then, um, Ralph on the other hand, um, is investigating social percept, uh, social processes, um, in healthy individuals and individuals with autism. And it's just, um, the combination was just great. And they have worked together a lot before. So it's, Caltech is a really small school, um, and they have collaborated a lot before and they've always had shared projects.

Then there's this Conti Brain Imaging center here, which they also, um, choir together. So it's, it's not hard to collaborate here. It's, I mean, you can easily talk to [00:19:00] everyone and, um. That's actually, I think a pretty great thing about Caltech is that people really collaborate a lot and there's a lot of interaction between the different labs.

And I mean, Ralph just, um, just a couple of years ago, maybe 2018, 19, I'm not sure, he published this book with David Anderson, which is actually comparing or is covering emotion processing from mice to human. Um, so it's, um, it's a huge field and it's just, um, yeah. I'm, I'm really happy about the collaborations.

You also get in contact with a lot of people that had been at Caltech before or that just collaborated with them on other projects, and so you get a much bigger network of people that are interested in similar things or you just, you hear about a lot more things than I'm used to, um, from Germany. Yeah, and especially, I mean, that was one upside of the pandemic is that you can now, I mean the, the COVID project is, uh, there are seven different institutions working on that, and it's [00:20:00] not a problem because we always meet on Zoom.

Right. So it doesn't really matter where you are and you can just work from anywhere and always everyone can join. And also you have, we have a lot more lectures and talks from outside. Um, you can easily invite people to your lab meeting to give a talk. I mean, that actually I kinda like, and it works much better than it did before, I feel.

Benjamin James Kuper-Smith: Yeah. I think that's is something that's. Been handy for the podcast that, I mean, it's kind of coincidence, but I think it probably helped in asking people that now everyone is completely familiar with all of these things and doing it regularly. There's no, I think like if I'd asked maybe two years ago, I mean obviously people used Skype and all that kind of stuff, right?

Or Google Hangouts I think was two years ago. But anyway, like people were using all these things, right? But you know, you'd use it like once in a while. It wasn't this thing where you go like, oh yeah, I do this like, video thing over the internet with people every day. So I think it probably also helped there and just [00:21:00]making it more accessible for people to say, sure, I'll do the interview.

Um, but did you have a specific research project within or was it just, Tessa, you're great, do what you want to do, we're here to help you.

Tessa Rusch: I didn't have, um, a specific research project to be honest. So, um, they're very flexible. And the first day I got here, we went for lunch and Ralph said, so what do you wanna do?

Which is a pretty big question. And I honestly sat there and I was like, um, that's

Benjamin James Kuper-Smith: too Extens question for the first meeting.

Tessa Rusch: And, um, but he, I think he honestly meant it. He was like, so what? You're interested in what you wanna do? And I was terrified. I mean, I was terrified in general because there are a lot of impressive people here and you don't always feel like you fit in, in the beginning or even now.

Um, so, um. I mean, I wanted to continue the work that I did in my PhD to some degree. So what I did in my PhD was looking [00:22:00] at, um, theory of mind processes and taking a more computational approach to social inference processes. So that was still generally, um, the plan, but it wasn't all worked out yet what I was gonna do.

So I pretty much actually started by reading all the grants that they had currently, um, like they had received in the, in, in, in the years and months before I got here. Um, and looking at how it fits together. And then Ralph encouraged me and John encouraged me to just talk to lab members, hear what they do, um, see where I can collaborate with them.

And it was, it was a really interesting experience in the first two weeks because I could actually meet people. Um, yeah. And then it got a little harder, but, um, and it. Took me a while to find, um, the project. That is gonna be my project now. And then I'm actually, I'm gonna work on with a colleague, um, that started a postdoc here as well.

And so, um, but it was, it was an interesting process and it was actually quite fun to bounce ideas around. Um, and, but it was similar to the beginning of my PhD where there was not a [00:23:00] clear, I mean, I wasn't hired to work on a project, it was more, okay, so let's think of, of something, which as I said, it's great because you can be, um, very explorative and creative, but also it is extremely difficult and confusing in the beginning because there's so much you can do and then it has to be, um, exciting and novel, but also it has to be manageable and it has to be, I mean, it has to cover all these different things.

And so, um, it was, it was, to be honest, it was quite a challenge, but now I'm really happy with what we came up with.

Benjamin James Kuper-Smith: It also sounds like your PhD was good preparation then in terms of the like big picture and like figure out your own ideas. Right.

Tessa Rusch: Yeah. And um, I mean that wasn't just my ideas, right? It was with, um, Jan and Michael and it was also all of us bouncing ideas around and, and y um, Martine, Bert.

And so, um, no, it was, but yeah, it was a good preparation in the sense of, okay, you, there is a lot going on, but still, I feel like it's still extremely [00:24:00] difficult and I, yeah, it's the most fun and the most difficult part of academia, I think.

Benjamin James Kuper-Smith: Yeah, I mean that's the thing about doing something new, right? You can't just, yeah.

People can't just tell you what to do. Yeah. Uh, yeah, I mean, we'll I think we'll talk more maybe than maybe about like what you did to your PhD and about that kinda stuff. Maybe in, maybe a bit more later now. So as we mentioned, you, you spent the first year basically of your first doc doing something completely different.

Um, I mean, you already mentioned like that you were already, that you had these lab meetings where people were present stuff. Did it start then from there that. People said like, I dunno, we've already done all this research into this. Or was it, how did that start?

Tessa Rusch: Um, actually it started with Uri Moss who used to be a postdoc with Ralph and is now, um, a Chapman University in Orange County.

Um, I think they had a conversation just, huh, this is an interesting time and [00:25:00] it's probably gonna affect, um, biases a lot, especially, um, in the US or just generally anti-Asian biases. And um, so they just started chatting I think, and then Ralph sent an email, um, over the lab mailing this and hey, um, I was just chatting with Uri and.

We wanna do something. I mean, it seems like it's a good time to do something who's interested. And then a couple of people responded, it was Ying and myself in the beginning. Um, and, um, Damien Stanley, who's in New York and who also was a postdoc here, and, um, because Damien has done a lot of work on implicit and explicit biases.

And that's, um, because, so Ralph was like, oh, if you're doing something on biases, we should definitely call Damien. And so that's how they got Damien. And then it was like, okay, so, um, who else is interested? And it was yanking and myself. And then we started chatting essentially. So we just had the Zoom calls, like, okay, so what are we gonna do?

Um, we have to be quick because it's, it's already happening. It [00:26:00] was, I think mid of March and then we started brainstorming and we had a couple more people join. Um, and it was a very, um, fast and exhausting period of time. So we launched this COVID dynamic study on April. Fourth. And so we decided, so we decided we wanted to do something longitudinal, something really long longitudinal because as I said, we, we looked at these projections and it was kind of clear that this thing was not gonna go away soon.

Um, and a lot of things are gonna happen and this is gonna be period of extreme change and all kinds of things. And, um, so we were sitting there and discussing, so what would potentially be, um, relevant and what was relevant at the time was that toilet paper and, um, people staying at home. And so, um, but it was, okay, so, so things are gonna change.

And it was, and, and we were just sitting there and trying to. Think about and predict what would happen in the next, uh, couple of months, um, over the [00:27:00] next year. And so we, we focused on designing a study that would take almost a year. Um, and of course in the beginning, the massive problem was funding as well, because actually running a study with a thousand people over a year in the beginning, weekly, then biweekly, and then later we went to a monthly schedule, is actually quite a, um, expensive endeavor.

Benjamin James Kuper-Smith: Yeah, that's what I, when I presented in the lab, I was like, guys, they paid them like $10 an hour. They must have paid like a hundred thousand dollars for this study or something like

Tessa Rusch: that. We did pay a lot. I mean, um, yeah, we did pay a

Benjamin James Kuper-Smith: lot of money. Yeah. Yeah, exactly. You don't have to say the exact figures, but I was just, look, just looking at like the data you report, it was like, this is an expensive study.

Tessa Rusch: It was an expensive study, but we also wanted, um, because I mean, it is a longitudinal study, right? And it was conducted over the internet. And so we also wanted people to be happy, right? We wanted them to come back. Um, and also, I mean, it was a time of, it was a gigantic economic crisis, right? So, um, a lot of people actually started doing these [00:28:00] online studies to, um, earn a little extra money and you wanted to pay them fairly, right?

So, um, they got 10 hours, uh, $10 an hour, and each 10 hours per dollar each, um, session lasted about an hour. And we just, um, yeah, I mean they were helping us with a complex project and we really wanted them to be fine and to be happy. And, and later clinical psychologists join, joined the project as well, and people that are studying substance abuse, um, and trauma.

And so they were like, oh, people, you have to provide some resources. I mean. These people are really depressed, as you can see from their scores. And you guys have to, we, we have to provide resources on, um, for psychological help or food banks or these kind of things because people were really struggling and people sometimes really wrote very heartbreaking messages.

Um, our participants essentially, and so. Yeah, we, we started this thing and we thought, okay, let's start, let's run it on weekends, because probably the new cycle is gonna pick up, um, on Monday again. And so [00:29:00] we wanted this period between Friday and Monday when things were a little slower and new, um, policies had already been Ed and then before the new stuff happened on Monday, so that we'd be able to kind of capture things in time.

Um, but then things got really crazy in the US over the last year and it really didn't matter whether it was the weekend or not, I feel, but that was the plan. So this is why, um, kind of the, the structure of the study was to record every weekend, um, from, we gave them, um, 48 hours to, uh, complete the study from Saturday morning till Monday morning.

And they could do it whenever they had time, but we just wanted to, um, have it in a fixed period of time. We tried to sample all over the us We tried to sample more, um, like a larger age group. But of course it's, there is the sampling bias. There is a massive sampling bias, um, in and running a study over the internet.

And um,

Benjamin James Kuper-Smith: did you consider, I mean, so you did it bio prolific, did you consider doing it as a representative study or was that just gonna completely ruin the budget? [00:30:00] 'cause it cost like 30% more or something on there, right? To do representative sample?

Tessa Rusch: Um, no, we did sample. I mean, actually you can, you can choose some criteria, right?

So we tried to make it so we, we actually launched it in different batches over the US so that we could cover the east, the middle and the west. Um, so we try to get the same amount of participants in each, um, in each area. And then we also launched it in different, um, age brackets, um, so that we could cover, um, uh, the, a higher age bracket, uh, more representatively, but it's, for example, it's extremely difficult to sample a politically representative sample.

Even over prolific, and I mean, the majority of prolific users are more left-leaning Democrats, uh, in the, in the us. Um, they're less conservative voters. Um, and also it's a very, it's a, it's a very dynamic characteristic of a person, right? Especially in a year of an election. So we couldn't, we couldn't [00:31:00] really representatively sample for that.

Um, and also, um, other like, um, race and ethnicity was, um, hard to cover. So that's what I mean. And also it had to be really quick so that we, it was kind of like a trade off, okay, we either launch it today or we wait another week, get this, figured that out. So it was really just a push and pull between what can we actually do because we were three people in the beginning running this thing.

Benjamin James Kuper-Smith: Okay. No, like, what I meant was that on prolific, you have the option to say that you want a representative sample according to age gen and ethnicity in the country. But that then doesn't include geographic, I don't think. And political, definitely not.

Tessa Rusch: Yeah,

Benjamin James Kuper-Smith: so, but the thing is also that cost then, like basically prolific then has to really organize and get the people who are like, uh, between 65 and 70-year-old Asian woman or whatever, right.

Like, they have to get those people so it costs like 30% more or something.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: But I guess you kind of did that in a more like manual way then.

Tessa Rusch: Yeah, we try to, um, account [00:32:00] for geographic location and um, age, and it did work to some degree. We got participants from all 50 states and I think our oldest participant was 83.

Benjamin James Kuper-Smith: Yeah, no, it looks pretty good. I mean, as you mentioned, the, I mean, you have the C right? About like the political stuff? Yeah. Like half your people identify as Democrat in your study, whereas what is it like a third usually do it in the us So it is biased in that sense. But no, I mean like, I mean the thing is also like even if you have, even if you have like a fully representative sample according to six criteria, that's only gonna be for the first data collection, right?

And as soon as you have dropouts, it's just going to deteriorate anyway.

Tessa Rusch: Yeah, exactly.

Benjamin James Kuper-Smith: Okay. So that actually that's, so one question I had like in general is like just, or not one question, but like one area I'd like to cover is the practicalities of doing this research because it's, I mean, you said it, you started with three people, but I mean when you look at the paper, there's like six people, seven people on here [00:33:00] named, and then like another 10 who aren't even named.

And Ralph ados is one of the people who isn't even named, he's like part of the COVID dynamic team. So how did. Did you, did just more people find out what you were doing or did you reach out to people? Or how did kind of this team grow and develop or change?

Tessa Rusch: It was very dynamic. It was a mixture of both.

Um, I mean, there are, as I said, there's people from seven different, um, institutions involved in this project, and it ranges, uh, I mean it covers cognitive sci, uh, neuroscientists, um, psychologists, political scientists like Mike, Mike Alvarez for example, joined who is, I mean, he does a lot of research on elections and political research here.

Um, and he, like, he did a huge amount of work on the, on the election in, um, 2019. Um, and so for example, it was great to get his, um, insight and advice on the study because he actually was the first one who was like, okay, yeah, I know your psychologist really don't [00:34:00] care about what your sample looks like.

Essentially, you just take whoever you can get, but. No one else really accepts that. So you should really think about this, that, that it's mostly young, well-educated, um, college students essentially, that you're testing usually. Um, then there were, there was joining from Yale Law School who got another, like put another perspective onto this, um, more about the policies and the, um, for example, um, his student, Tony Lawrence manually went through all the, um.

All the policies in each state and collected it on like every wave. She went through this whole thing and was like, okay, what are the, what's the GA max number of people that can gather? Is there, um, is mask, is there a mask mandate? Is there a stay at home order? And she collected all of this manually because at the time it wasn't, there was no like freely available, uh, website that was collecting all that information.

And so it was a huge amount of work by a huge amount of people [00:35:00] collecting, like getting funding is, was a huge task. And, and Ralph spent a lot of time just trying to find the money to run this thing. Um, and then, yeah, we had to code up all the tasks and the, the questionnaires and then, um, actually monitor the study because it had to run in these, um, 48 hours.

Right. And if people encountered any technical difficulties, we had to help them. So there was. We monitored it from 7:00 AM Eastern to 10:00 PM like, uh, west Coast time. So there was always someone, um, monitoring the prolific messaging board and seeing whether people had any kind of problems. And, um, we had like this elaborate, uh, Damian created this really beautiful figure of how we solve technical issues.

It's like this gigantic flow chart. Um, and so, um, so

Benjamin James Kuper-Smith: did you have many technical issues? Because I mean, we basically the same thing with let's say half the participants you had, or I mean, not even that we Yeah, and I never, I mean, a few people messaged me here and there, but it was like. Three, four

Tessa Rusch: something.

Um, [00:36:00] it's not as much. So the questionnaires weren't a huge issue, but we also ran these tasks. Um, and then for example, um, I don't wanna bash any browsers here, but some, um, didn't work. Um, and we had to find all of these out, uh, things out in the beginning, right? And, um, loading, um, all the images sometimes didn't work and some people had just like connection issues or they were worried because there was this short period of time where they could complete it and then they, um, didn't have time or they, like, they were interrupted and they were very worried about, um, not being able to finish it.

Um, and so they contacted us or like all kinds of other things. So just they wanted to clarify something or, um, a link wouldn't work. So it's, um, I mean, there are a ton of different things and when a thousand, like 1000 people at a time do it, you have to be quick to respond to these messages and keep track of them.

So we came, we became this like a customer service thing for a while.

Benjamin James Kuper-Smith: Yeah, I mean that is always a, I did have this [00:37:00] huge, I mean like I have it in general with online studies when you, once you put it live, 'cause it was like, oh God, I hope I didn't mess up. Otherwise I'll get like 200 pointless responses now.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: But it's not a problem because I mean, you know, whatever. It's like, especially online, also often a short study that you, I did before the pandemic. Uh, so, you know, worst case scenario I waste a bit of money, which is silly, but, you know, not that big of a deal. But with the COVID study, it really was like, if I get this wrong,

Tessa Rusch: I'm gonna waste,

Benjamin James Kuper-Smith: I ruined our day to connection basically.

Yeah,

Tessa Rusch: yeah.

Benjamin James Kuper-Smith: It's like I better not get this wrong.

Tessa Rusch: No, and I mean, of course we got things wrong, right? We had technical problems and then we, like, we discovered bugs every once in a while. So we had to adapt it every once in a while. And, and of course there, I mean, we're only human, right? And there. It was, as I said, it was a very big time pressure, so of course we got things wrong, but yeah.

Benjamin James Kuper-Smith: No, no fatal errors.

Tessa Rusch: No fatal errors. But, um,

Benjamin James Kuper-Smith: like you forgot to record [00:38:00] the participant ID or something, or

Tessa Rusch: that did not happen. But, um, other things happened.

Benjamin James Kuper-Smith: Do you wanna disclose them or is that,

Tessa Rusch: um, um, I mean, it was like, uh, forgetting to make responses for us, or forgetting, um, a question there and here or something like this.

You know, it was like this gigantic amount of questionnaires, um, and tasks or, um, we had this, uh, one task that was actually a real interactive. Game where, but so, because it was hard to run it, um, the time, so they got feedback, the next wave they returned and then it was just, um, difficult to make sure that they all got the correct feedback.

Um, and I, to be honest, had not, I've never run an online study before the COVID study, so I was kind of learning as I was going. So, um, for example, some, that, that was one of the big technical issues is, uh, that, um, this feedback, for example, wouldn't load and, um, or for this participant wasn't there, or something like [00:39:00] that.

And we just had to make sure that everything's there.

Benjamin James Kuper-Smith: Yeah, I mean, I made one really annoying error. It's just so stupid. It, it's not a problem in the end, but it was this thing where, so we had like, you know, when we created the server, we had all these different questions and sometimes we'll make new versions of the questions, um, right.

Just to improve it. So then when we went to the, that was the first data collection, that was the second data collection. I just copied the projects over, right. And we added some stuff and that kind of thing. And anyway, I was like, oh, well, you know, I can just delete some of these old questions we're not using, uh, because, you know, we're not using them.

And then I could confuse and accidentally put the old questions in that kinda stuff. I turned out not to be the greatest idea that I ever had, because it turns out the way the data was recorded is that every question you have is a column, which means that even the questions you don't use are a column in your dataset, which means that once you take some out, you now have the fun task of finding out, which, like you have to rearrange all the columns you have, [00:40:00] which in a way is not that big of a problem because it all shifts systematically.

But that was an unnecessary amount of work we had just because I for once tried to be tidy.

Tessa Rusch: Yeah, no, I, I totally get it. It is. So, um, Andy and Lynn have created these. Ginormous tables because also we, yeah, we changed things around as we went along and like the variables changed name, and so there's like a huge table that just aligns the variable names and all the different waves and all the different participants and uh, it just like, so that all have, like, all columns have the same and that, I mean, that was a, we're still working on that.

We're still working on cleaning and sorting out the data. It was,

Benjamin James Kuper-Smith: yeah. Do you also find that the most fun task of the entire project, sorting data, making, putting a lot of time into not messing up majorly?

Tessa Rusch: Yeah, I mean that's, um, I mean for me this was the first time working in such a big group of people.

I mean, as, as researchers, you usually, you like have one or two [00:41:00] collaborators that you work with on a project, right? But you don't work with, um, 10, 20 people at a time. Um, and that was an interesting experience. I mean, we had to change our coding habits. Um, we had to make sure that everything's well documented.

But, um, and I learned a lot from, from, for example, from Lynn, who's. Great at organizing these kind of things and keeping structure in these things and, and documenting everything very perfectly and neatly. Um, and I mean, of course it's a very tedious job, but it was actually great to learn from other people how they do this, um, and also develop like, uh, different strategies to how to deal with these things.

Um, and now as I said, we're still cleaning and sorting and organizing the data. And as we go back and forth between people, um, you have more control. So like someone, someone creates the table and then someone else starts using it and realizes, oh, there's an issue here, or this doesn't make sense. And so there's, I think it's actually much more, um, controlled than when you do it by yourself, right?

Because [00:42:00] usually you, you code it like an analysis, for example. Um, and no one ever goes through your code. And I'm pretty sure there are mistakes here and there. I mean. It's, it's very unlikely, um, that it's not. And so that was, I mean, I, to be honest, I really enjoyed working in this, um, big team, especially in this time of lockdown because it was, it became like, um, a very, uh, tight connection because it was, it was always, we were always in a rush.

It was always time pressure. Um, and, um, I think I have never said goodnight to as many colleagues, like as, as often as I did now over Slack. Um, and um, I think it was actually, I learned a lot from this. And I think it's, um, and as I said also from having people of different expertise in there and for example, having a political scientist tell you a little bit about how other fields and like social science do these kind of things, which I don't know how psychology and cognitive science got around [00:43:00] not taking care of these things.

Benjamin James Kuper-Smith: Like, like which things,

Tessa Rusch: for example, not sampling representatively or not, um, weighting your data properly so that your data, like your sample is, um, more representative. Um, just not looking at that. Um,

Benjamin James Kuper-Smith: but isn't that just, it seems to be like a lot of those things are more relevant to this research than to other research though, right?

Like if it's something like this where your perceptions, that kind stuff really depend on your political affiliations, that kind stuff, and then that makes a huge difference. But I wonder like to what extent it really makes that much of a difference in most research.

Tessa Rusch: I mean, when you look for example at social decision making, um, it is something that is heavily influenced by your culture and how you grew up and how you've been socialized and educated, right?

Yeah. I mean there are a lot of studies on cultural differences. For example, um, how some societies are more individualistic and others are more collective, and I'm. Pretty sure that [00:44:00] um, there are very different views even between Germany and the US and how self-dependent or independent you have to be or um, how much you rely, um, on the society and on the group.

And I mean, that was actually something that I think was really interesting in the beginning of COVID because, um, for example, this mask, like wearing a mask and not going out is a gigantic public goods game, right? It's not fun. And um, especially for young people, it was this, okay, we am probably not gonna get hurt as badly.

And it is a cost for myself staying at home and not meeting other people. But if everyone behaves in the same way. Like, if everyone doesn't stay at home and doesn't wear a mask and still meets people, then we're all gonna be heavily affected because then the, the hospitals are gonna be overwhelmed and then, um, nothing works anymore.

And then when I get into a car accident or something, I'm also, they can't treat me because the, the hospital is completely full. So it is this, it really is this question [00:45:00] of, um. Do I work with everyone else? Hoping or expecting that everyone else is also following that strategy or that practice? Or am I just, um, looking out for myself here?

Am I looking for my own needs? And I think, I mean, that is something really interesting. I think about this pandemic is how we as a society collectively behave. And also that we have these collective experiences. Right now, we have at the same time have very similar experiences. I mean, it's very different for each and every one of us because as you said, you're like, you're more of an introvert.

You don't care. It's fine for you to sit at home great and read books and yeah, it's, it's. Perfect. Don't bother me.

Benjamin James Kuper-Smith: I haven't had to make any excuse for not wanting to go to party in months. That's amazing.

Tessa Rusch: And then there are other people who were like really dependent on social context and social interaction or, um, as you said, we both kept our jobs.

We, we were able to work from home, um, and other people lost their jobs or they still had to go to work and, um, be exposed to [00:46:00] very, um, dangerous situations. And so it's, um, there is a similarity in how we experience this time, but there's also each individual person's experience, which is very different.

But there is like this base fluctuation underlying it, which is very similar, um, at least within a nation maybe, or in a, in, in an area. And even though we were in this period of, um, being so separated, like socially distant, I, I mean it's still a huge, uh, social psychology. Question an experiment, essentially.

It's like, how are we gonna behave? It's this, uh, I mean this crazy thing about buying toilet paper, why on earth was toilet paper sold out so quickly and it was this, I mean, and it was this massive, um, social feedback loop. Everyone else was buying it. I need to buy.

Benjamin James Kuper-Smith: Yeah. That was the one thing in Germany we had, like, there was nothing with food ever that was apart from like maybe one or two days when like certain food items were gone.

That, but like toilet paper was a thing where, I mean, like, [00:47:00] even my, my mom said that she, you know, she lives in like a, a countryside basically, where it's very relaxed because it's like everyone has their own house and all that kind. Like it's just, you know, you don't have these like, dense, um, and also like the supermarkets.

Um, and she said, you know, like, well, I like always just like in the cell has like some, some stored goods, right? Just, uh, like in a storage room, she's like, you know, I'm on my own. I had like toilet paper for two months or whatever. But then after like f 4, 5, 6 weeks of just not seeing any toilet paper in the supermarket, she did go like, uh, I'd better buy someone when I get it.

Like when, when they have the opportunity. She started asking like, the, the people in the supermarket, like, when are, when are you guys going get more toilet paper? Because I'm starting to get worried here too.

Tessa Rusch: Yeah. Yeah. They started selling single rolls, um, at the checkout here. So you could only take two rolls at a time.

It was also, I mean, you could only take two of the same items, for example, at the same time. [00:48:00] Like two loaves of bread, for example,

Benjamin James Kuper-Smith: milk and like that kinda stuff. Yeah.

Tessa Rusch: Yeah. I mean it's um. And I mean, that's a massive social phenomenon, right? It's like a massive social feedback globe that was going on there.

Benjamin James Kuper-Smith: But one thing that really surprised me, um, when we were talking about like differences between groups and that kinda stuff, one thing that really surprised me when, or maybe I should say, so I lived in Hamburg from, well until end of November of 2020, and now in Heidelberg since December. And in Heidelberg people, you know, like I was in, I was in Hamburg November, and then two days later I was in, in Heidelberg and Heidelberg.

People wear masks more. It's, I find it quite noticeable. You just go there. There was this like, and I think it wasn't, you didn't have to at the time. There's this big like shopping road basically where it's like a pedestrian center and it's like, you know, shops everywhere and pretty much everyone was wearing a mask there, which definitely wasn't the case in Hamburg.[00:49:00]

And what surprised me most almost is like I had it again this morning. You see these like young kids, like young lads, like 15 year olds or whatever, outside, chilling, wearing a mask. Like, it really surprised me that like those were the last kind of people I thought who would be sticking to the restriction when no one is watching when they're just amongst themselves, but they're completely sticking to it.

It's fantastic.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Um, not an handbook, I can tell you that.

Tessa Rusch: It's interesting, right? I mean, it's not that far apart. I mean here you can also see it different, um, parts of LA County, um, or when you drive 40 minutes, it's a completely different picture. Here. Everyone goes and people go running with a mask.

People go hiking with a mask and they're always wearing a mask essentially. And it's become this kind of, um, grading thing that when you wear, when you walk in with your mask down, and then you pass someone, you automatically put it up. It's like, um, and or other people don't put it up or don't even have it with [00:50:00]them.

So it's kind of becoming this political statement in a way. To wear a mask or not wear a mask. Um, when you're outside in

Benjamin James Kuper-Smith: the us Yeah,

Tessa Rusch: in the us yeah. But even within California, which is considered to be very,

Benjamin James Kuper-Smith: right. Sorry. Yeah, of course. California is like 300% Democrat or something.

Tessa Rusch: Actually not. So there is, um, there is variability here.

Um, but it's, um, no, it's just, yeah. It's interesting. I think,

Benjamin James Kuper-Smith: I mean, is it also interesting just how the norms change in terms of, you know, there's this, I think this was, I definitely had the sort, and I saw, I think I saw a tweet about this as someone like, you know, whenever you see someone outside w.

Walking around outside with a mask on, you kind of wanna keep distance from them 'cause like, oh, they've probably got something gross. But now since the pandemic, it's like, ah, they're safe. I, I can like always walk closer to them than to someone else like that. Because like in Germany, no one, like, you couldn't even buy masks, right?

Like, because it, it just wasn't a thing at all. Um, I mean I heard like, you know, some Asian countries, you have it where it was very common [00:51:00] and usually people were out running around with a mask. In Germany would be Asian people. Right. Would be like, you'd very occasionally see someone with a mask and it would almost always be someone from Asia 'cause it's just a norm thing.

But yeah, that's changed. That's a norm. That's just completely flipped.

Tessa Rusch: Yeah, I mean that's something that is interesting to look at. I mean, that's also part of the study to some degree. Like, one part of it is, um, especially Uri and, um, Gideon looking at this, um, changes and norms and, and behaviors. Um, but for example, when you were just mentioning it with, um, mostly Asian people.

It's part of, and when you're sick, you're wearing a mask. You won't get onto a subway if you have a cold without a mask kind of thing. Um, as I said, like part of the study is also looking at, um, implicit and explicit racial biases. I mean, especially right now, for example, there have been a lot of reports about, um, anti-Asian, um, crimes.

Um, oh, that was

Benjamin James Kuper-Smith: last week, right? [00:52:00] Or something.

Tessa Rusch: Yeah, last week there was a shooting in Georgia with the majority of victims being Asian women. And, um, it's being discussed whether it is a hate crime or. I mean, there are all kinds of questions about the motivations of the shooter, um, but also small. I mean, it's not just shootings, but it's also smaller events, people being insulted on the street and these kind of things.

And I mean, we all have biases. And I noticed, I mean, I noticed in my bias that at some point whenever we, like when restaurants open up again, for example, when you could sit outside, I always felt like we should go to an Asian restaurant because I trust Asian people way more to behave responsibly with respect to the pandemic because, um, they know what they're dealing with.

Asia is doing a much better job at handling the pandemic than most Western countries. And I mean, it's also a bias, it's a super strong bias and, um, but it can,

Benjamin James Kuper-Smith: half the people are probably born in the US

Tessa Rusch: [00:53:00] Yeah. But it feels, um, safer and more organized and there's hand sanitizer on every table and, um, it's more, more structured where I feel, but it's, it's, it's a super strong, um, bias.

And so. I think it is interesting how that changes over time. And then also with the killing of George Floyd, that added to the dynamics of biases, I think, especially in the us. So I think when that happened, we kind of thought, okay, this is not the COVID dynamic study anymore. This is just,

Benjamin James Kuper-Smith: yeah. I mean you've, it is also, you've got so many different questions in there and so many things you are asking about.

You are just, if we forget about COVID for a second, you probably are just getting a lot of data just about how all these variables change over time and how these perceptions and norms, whatever, just change Yeah. Just because of the stuff that's happening that's, let's say, largely unrelated to COVID.

Tessa Rusch: [00:54:00] Yeah. Um, I mean we had, um, these questions about implicit and explicit biases towards, um, Asian Americans. Um, and, uh, we kind of included the question about, um, biases towards. Black people, um, as a control condition in a way, as a comparison of also like we had like, um, as a, as a control con condition to compare it to anti-Asian biases.

Um, and then, um, this social civil rights movement

Benjamin James Kuper-Smith: ruined your control.

Tessa Rusch: I mean, really. No, I mean, I, I, I think, I mean, it was impressive. It was impressive how long people went to the streets and how, um, I mean yeah, how strongly, um, that influenced, um, the world in, in the US and in US politics and, but it was just an extremely variable time and um,

Benjamin James Kuper-Smith: and you've got data for everything.

Tessa Rusch: We've got data for everything. [00:55:00] Yeah. Well, I mean, yeah, we'll see.

Benjamin James Kuper-Smith: Uh, speaking about having data on everything, so one thing I really struggled with in our study is. And our study is obviously much, much smaller than yours. Uh, we have three time points for data. We have, I dunno, let's say there's like 20 to 30 variables or something that are like the main ones.

I found it really hard to like, I mean to be fair, we also tried to kind of put it in one paper and white one paper outta this. But that was a real struggle to kind of, that's basically what I'm finishing. Well, I would be finishing right now if we weren't talking, um, to like really just find that, get, put that into one framework so you have like one paper where we could describe what we're doing in a coherent way.

How on earth are you gonna do that? I mean, I guess you're doing lots of studies out this in a way that's the answer. But how do you actually, for a binary question, have you actually already started analyzing data or is it still like organizing, [00:56:00] sorting through.

Tessa Rusch: Um, I mean, it's a mixture of both. So, as I said, this is a huge team effort, and different people contributed different questions and different ideas and different focuses on, on, on different topics, right?

And so, um, and also, um, the study contains about well established questionnaires, like psychological questionnaires or political questionnaires and, and, and tasks. And then there are more experimental measures that people develop just for, for this study, kind of. Um, and so, um.

Benjamin James Kuper-Smith: And then you have the external stuff,

Tessa Rusch: and then we have the external stuff on top of that.

Yeah. So, um, it's, um, there are a bunch of people working on different, uh, projects and we meet once, once a week. Um, we have this, um, kind of, um, data structure system where you have to, um, you have to present to the group, um, what you wanna do with the data. You have to request the data, um, like all the variables you want.

Um, and, and this is not, um, I mean this is kind of like to make sure that everyone knows who's working on what, and there are more [00:57:00] people joining now that the data is there, um, working on different projects and we get help from Ians, for example. Um, but, um, this, the, the idea behind that was to make sure that no one is just p hacking and just.

Looking around in the data and exploring it as long so until they find something, but that there is a discussion beforehand and people present it and we make sure that everything's sorted out. But there are a ton of different projects, um, going on. And some have started ana analysis, yet some are still, um, in the exploration phase.

Yeah, I mean, um, one, I think like one paper was published on that and that was on mask norms. And that was actually done by Carolyn Lawrence, um, the person that also collected all this data on policies, and she's a, um, law student, so she was particularly interested in how these, like the fact that these mask law laws, because before COVID you weren't allowed to wear a mask.

Right. Um, so, um, yeah, there are, um. As I said, people from Rutger is looking at [00:58:00] substance abuse. There are, um, is a huge study that Ralph and um, um, one of Ralph's, um, grad students are working on, which isn't particularly related to COVID. It was just looking at, um, change over time, looking at how states and traits change over time within an individual.

Um, and, and, um, doing like the running, um, uh, simulations on that. Um, we're looking at more as a, yeah, I'm mostly focused on the, on the social questions and the bias questions. Um, and we're looking at, um, we're starting to analyze things there, but it's also still in the process of making sure that the data is all high quality data.

Um, that we make sure that there are no weird data points in there and everything works and everything's correct. So we're currently transitioning from this period of, um, sorting everything to actually looking at it.

Benjamin James Kuper-Smith: Yeah. Uh, one question I had, which is. Kind of random one. Um, so you have all these, so I'm looking right [00:59:00] now at your table one where you'd say like, what you collected when.

Tessa Rusch: Mm-hmm.

Benjamin James Kuper-Smith: And again, I'm immediately contrasting this, like what we did. So we basically had like some questions other, then each time we collected more data, we added questions, but we maintained the initial thing. Whereas in your case it seems questions came in and out almost of the survey. And there's like some that are, I mean, let's just take the humanitarian egalitarianism, uh, questionnaire I guess, or whatever.

Mm-hmm. Um, you collected that on the first two data collections, then on the fifth and then on seventh.

Tessa Rusch: Mm-hmm.

Benjamin James Kuper-Smith: But not on anyone on the eighth or any in the middle. Um, like how did you. There, there's quite a few of these, which are kind of, have a, seem to have a, a regular rhythm in terms of data collection, like number one.

Why, um, like how did you, why was that, and what, how did you decide when to collect what? Yeah, maybe that's, [01:00:00] let's just leave that question then I'll ask the rest later so you don't have to remember five questions.

Tessa Rusch: Um, so why, why we selected different questions at different time points was, um, as I said, there were a lot of people involved, a lot of questions being asked and it just, I want one issue was just, um, there wasn't enough time to do everything, every time because, um, we felt like more than an hour is just too hard, too difficult on the participants, especially with all these questionnaires.

It's a lot of questions, it's a lot of reading, and you don't want to tire them out too much and make it too hard on people. Um, so that was why we didn't collect everything every time. Um, then another issue is a lot of these questionnaires, for example, um, ask about in the last 30 days, for example, or in the last six months.

Um, and so, um, if you collect data every week, you don't need to repeat this questionnaire because you're asking about the last 30 days. Um, then, um, as I said, there's this, um, one project that is [01:01:00]specifically focusing on, um, how it's. Like what is considered to be a state and what is considered to be a trait of a person.

So what is something that stays stable over time, relatively stable over time, and what discriminates people from one another? Um, and the other thing is like the state, what is, um, that is more dynamic, it changes more over time and was more situational. Um, and so we, um, selected for example these questionnaires, which are considered to be more, um, trade specific, looking more at trade measures.

Um, we repeatedly sampled them, but not every time. But there there's more the basic question of do these things change? Um, is that something that is actually stable in a person? When you look, for example, at the state trade, um, anxiety questionnaire where you have like, where you ask about states, um, anxious states and anxious traits, and then the assumption will be you ha don't have to repeat a trade questionnaire because this is something that is stable.

Right? But, um, especially right now. It could be that, um, actually your [01:02:00] trade anxiety changes over time and it's, I mean, it's a period of continuous, um, stress over months and months, um, at a time. And so there is the potential for things to vary, but we thought we don't have to sample it every time. Um, and then other things,

Benjamin James Kuper-Smith: so, and then you just went, let's say you don't have sampler every time, then you just say, do we have time for this, this collection?

Tessa Rusch: Well know we have, um, we have a planning sheet where we have, um, we, we mark for each, um, each questionnaire or each part, each task, um, what the desired frequency would be. So monthly or bi-monthly or just once, for example, like the basic big, um, demographic questionnaire. In the beginning it was fine to do it only once.

Um, and then, um, we just. Or mostly Lynn, um, planned, um, what was gonna be collected when, and then some things were being discussed and we had, um, trial runs before we started. Um, every wave, like a bunch of people went through the thing and we looked at how long does it take? Um, and then we had to adapt maybe [01:03:00] sometimes a little bit.

Um, and then, I mean, that stopped after maybe 10 waves or something. But we did. Really check it a lot in the beginning. And then also we added things as, um, events unfolded, right? So for example, before the election we added questions about, um, political views and who you trust most. And then during the protests, I mean during, during this, um, big Black Lives That Matter movement, there were a lot of protests all over the us.

Um, and so we added questions about that when there were the gigantic wildfires on the west coast and we have a pretty big sample on the West coast. We added questions of, have you been affected? Had you, have you been evacuated? Have you lost anything? Do you know anyone who lost a home or something like that.

Um, and now we added questions about the vaccine rollout, right? Um, and whether they would be willing to, um, get the vaccine or whether they had the option to get the vaccine and these kind of things. So, um, as some things were fixed and stable over time because, um, those were planned to be repeated and then we adapted it also as things, as [01:04:00] events unfolded over 2020 and 2021.

Benjamin James Kuper-Smith: Um, God, there was one point, what did I wanna say? Gone. Um, I don't know what I wanted to say.

Tessa Rusch: Maybe we'll come back to you

Benjamin James Kuper-Smith: probably like at a point when we're talking about something completely different and it's no longer relevant. Yeah. We're talking about like competition, modeling of social interaction.

Like by the way, I, yeah. Um, okay. Yeah, I mean, I, I, I guess we kind of figured it would be something like that, that it would be a kind of some, I mean, like one, you also have one question about life time trauma or something like that, right? Like, gonna questions that you don't need to ask every time. Yeah.

But then, so how do you, again, like just, I mean, I'm just curious now from a, from a just practical perspective because I found this is like ours much, much smaller project. It's like nothing [01:05:00] I've done before and I don't think I'll do again, but like, okay, so you have this. As you described in the Preprint, you have this protocol of how you can get to analyze the data.

Um, is it, is it basically anyone who has something they want to do just comes to that meeting and says, I wanna do this? Or is it kind of more clearly defined where we say almost like, this is like your area, this is your area. Or it just seems like, you know, it could, this is just another like thing that has to be organized, otherwise everyone's just running around at the same time and Yeah,

Tessa Rusch: I mean, it does take time and it is complex and it's also a different, like an interesting social you.

Interaction and experience. Um, but um, no. So as I said, there are some measures which are more experimental, for example, that people specifically designed, um, for this and put a lot of work into. And it's kind of clear that this is the people that design that have the right to use it first, essentially.

But then, um, when [01:06:00] you, um, when you wanna look at a certain question and you think, oh, this could be interesting too. You, you present it and you, you ask the people that develop the tool or something like, do you wanna collaborate on that? Should we work on that together? Um, and um, yeah, this is how it work.

And then some of these questions, for example, these well-established psychological questionnaires, it is, we just put them in there. None of us have any kind of intellectual rights. Regarding these questionnaires, right? So, um, people can, can join it and use it. But so right now the group is, um, so I mean the people that are involved and the people that are listed as collaborators on, on this project and as, as, as, yeah, as people working on this project.

And then the idea is once we got a chance to look at the data ourselves in a year to make it simply publicly available. Um, and before that, if you wanna use the data, um, you can always contact us and ask to collaborate on it. And then you just also have to present in this meeting. And, um, we, we just, yeah, we just wanna [01:07:00] see what your plans are, what you wanna do with it.

Um, but I think it makes a lot of sense to join forces with other studies, um, that looked at similar questions, um, over this year. I mean, there are a ton of studies looking at psychological change. Over 2020. Um, and it would be interesting to compare different samples. It would be interesting to compare different, um, communities, um, and also just, um, to increase the power, for example, or see whether we can replicate our findings, um, in a different sample.

So I think I just

Benjamin James Kuper-Smith: remember what I wanted to say.

Tessa Rusch: Yeah. Okay. Go ahead.

Benjamin James Kuper-Smith: Um, it was about the, how long the questionnaire can be because that's, I mean, I already found an hour is, um, quite demanding. Like, I think like one of the big things I remember doing repeatedly is like, you know, every, and of course we had five people in our paper and, um, everyone's like, why don't we collect this?

Why don't we ask that thing? I'm like, no, it's already 25 minutes long. [01:08:00] Leave me alone. We're not putting anything more into this.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Um. Yeah, but it's a really tricky to, but did you notice any, like data quality reduction throughout the experiment or something like that?

Tessa Rusch: Um, I mean, so the way that we did it is we had a first set of questionnaires in the beginning.

Um, then they did these, um, decision tasks or like these more classic trial by trial tasks. Um, and then they repeated the second batch. Oh. Or they finished the second batch of questionnaire, so we kind of split it in two, so it wouldn't be the same thing for an hour. So they had a little bit of a break in the beginning, uh, in the middle where they did something else, um, and where they had to like, yeah, make decisions or rate something or something like that.

Do look at, we have a bunch of different quality measures that we went through, so we don't see a strong decline, um, within the questionnaires, I think. But, um, also we haven't looked at that in detail because, um, we also changed the order of the questionnaires. [01:09:00] So, um, we haven't, we haven't analyzed that in detail.

We have the order and we will have to look at this. Um, but for now we only look at, at the quality overall. Um, and we de do see a decline over the first couple of waves. Um, and we also see a. Quite a bit of attrition over the first couple of waves. And we, um, so when you look at, um, the first wave, we start with I think 1700 something people, and then it, it declines a little bit over time.

And we also, for the first five waves, we actually excluded people. So we have like a bunch of attention questions in there. We look at the, um, duration, if they're too fast. Um, we look, and we also have this one question. It's just, how likely are you to stick with this study? This is a longitudinal thing.

How likely are you to come back? And if they said, I'm not. And we didn't invite them again, for example. Um, and so, um, that was, that those were the criteria we used to select our sample in the beginning, and then we had this pretty stable group of [01:10:00] people that always returned and that had a pretty high quality, like had high quality data.

Um, and we look at, I mean, we look at a bunch of things, for example, free response questions. Is there, is there other reasonable responses in there? Are there words in there? How many words are in there? How long are the response strings, like in these long questionnaires that it is just all the same? I mean, you can't really detect random choice behavior, but as long as it's, um, repetitive and you can see it.

Um, and so we include all of these things. It's obviously, it's not a perfect measure. Um, and it is demanding and long, but, um, I think, yeah, I mean, it's a trade off as you said.

Benjamin James Kuper-Smith: Yeah. But wait, you just said that the data quality decreased throughout the waves, right? Is that what

Tessa Rusch: No, it was just, um, no, no, it didn't decrease.

It was just the, um. In the beginning we excluded a couple of people, so the proportion of people that had not so great quality was larger in the beginning. And so the proportion of that people decreased. Yeah. So, okay. That

Benjamin James Kuper-Smith: makes misunderstood because yeah, I thought [01:11:00] like, wait, but your table

Tessa Rusch: is the opposite?

No, the quality. Okay, good. Yeah, I mean, it's fewer people because we have attrition, but, and also we excluded a bunch of people, but, um, the proportion of people with high quality data Inc increased over time. Yeah. Okay.

Benjamin James Kuper-Smith: I think I wanna slowly wrap up the COVID stuff, but the, actually, so, so one, one of the last questions I had was about the Preprint.

Just what exactly was the purpose of the Preprint? Is it to kind of just like, give an overview and say like, this is what we're doing so you don't have to do it in the individual papers, or,

Tessa Rusch: um, '

Benjamin James Kuper-Smith: cause in a way, like, I, you know, I presented in an app and I always, at the end, I felt like. Okay. Like I've presented a study with no data right now.

It was a bit weird.

Tessa Rusch: I mean, there, it, it is, it is kind of like a, um, a data release paper in a way. Right? It is presenting the data. And the potential of the data, essentially. So there are no results in the paper. Um, but I mean, it's a, it is a common thing to present data and we just [01:12:00] wanted to make sure, um, that the data is really well defined and that we pro, I mean, we provide all these additional measures and, um, these quality checks and, um, the weights and everything.

Um, and it's just to, um, it's more of a methodological paper in the sense saying, this data is here and this is how the data was created, and this is what the data can do, or what you can do with the data. Um, but it's, it's not a classic, uh, research article in the sense of like, we collect the data and then here are the results.

There are no results in there, essentially besides, um, presenting the data in itself and showing that it has, um, good quality and that, um, there's variance in it. Um, and that there's difference between individuals and differences within the group. We just wanted to make sure, um, that, that this is good, reasonable data and, um, we just wanted for people to be able to see and, um, explore that for themselves essentially.

Benjamin James Kuper-Smith: Mm-hmm.

Tessa Rusch: So it is a, it is a data release paper

Benjamin James Kuper-Smith: without releasing the data right now. [01:13:00]

Tessa Rusch: Yeah. Well, because, um, it's only, um, yeah, I mean it's, it's, it's gonna be released, uh, once we're finished, um, with everything. Um, so yeah, it is a data presentation paper, essentially. Yeah.

Benjamin James Kuper-Smith: Yeah. I mean, I think it also, like, it does make, makes complete sense to, to do that from the perspective of just, I think also for the authors, just to kind of have like one thing to organize like what the data is and Yeah.

It makes it, I was just, yeah. Curious what exactly the, yeah. So it's basically presenting the data for when it comes out in, in a year.

Tessa Rusch: Yeah. And also when it's being used. Um, what is actually the data that you're working with. Um, when, when it's being used to, um, explore specific questions, it is just there and it is reported on what we have and how it looks like, because I think it is important, especially with such a big sample, it's important that what kind of sample are you working with and, and how does it compare [01:14:00] to the general population.

Benjamin James Kuper-Smith: Um, so now moving towards your, what we could call your main research, even though it isn't right now. Um, kinda the last question about the COVID thing is then like, so how does that fit in with your postdoc? Like, how long do you plan on spending on the COVID research? How long is your postdoc position?

Like, how does, like, what are the practicalities of that?

Tessa Rusch: So, uh. I'm, I'm working on, um, both my project, um, and, and the COVID study at the same time. And then there are some other things I have to work on. Um,

Benjamin James Kuper-Smith: some old projects. Maybe

Tessa Rusch: some old projects. Yes, we all do. Um, and no, it's, it's a lot, it's a lot of work, but, so I'm thinking, I mean, funding is always a question, but I'm at least gonna stay here another year, maybe two years.

We'll see funding wise how things go. [01:15:00] Um, and, uh, how does it fit in? I mean, part of, I mean, for example, we have this altruism task in there. We have this public goods game in there. Um, these are things, as I said, I think, um, in the beginning, the, or just the entire COVID pandemic is a bit of a gigantic public goods game.

Um, and so it is interesting to, to model essentially, um, anonymous, um, large scale, um, group interactions. Um, it's also, for example, yeah, that's, um, this altruism task is more, I mean, it's kind of like a, um, dictator game. Um, but, um, classic

Benjamin James Kuper-Smith: altruism, the

Tessa Rusch: dictators

Benjamin James Kuper-Smith: of the

Tessa Rusch: world. Yeah. I mean, yeah, the question, well, what is altruism altogether?

Right? Um, but, um, uh, I mean it's, um, it's, it's just for, I mean, my. My general research interest is social behavior, right? How do we as humans interact with other humans, um, and how do does the behavior and, and, and the thought process of other people influence our [01:16:00] own behavior? And this is, this is a huge part of it.

It's for example, who has the, the option for altruism, essentially. I mean, you can look at altruism, but then, um, if you are really, if you are on your last dime and you, you can't pay your rent and you can't pay for your food, and then it's a very different question. It's like, am I willing to give 50 cents to another participant as when you are, I don't know.

Yeah. You're living in a huge mansion and you really don't care about the money and you just do it for fun and then, and you have a lot more opportunity, for example, for altruism. And, and, and so, um, it is,

Benjamin James Kuper-Smith: have you heard of Christoph? Kwan? I think he has a similar study on Yeah,

Tessa Rusch: that's that's an interesting person who, who's that?

Benjamin James Kuper-Smith: I don't know, but, uh, some person who seems to have some sort of research project going on for the last few years on whether you can or cannot cooperate in a social dilemma.

Tessa Rusch: Yeah. I mean, yeah. Um, and it is very much dependent on the situation you're in [01:17:00]and, and the opportunities you have. So that's how that fits with my, my research and also like, um, for example, biases.

Um, they do influence our behavior towards people. Um, and I haven't really worked on that a lot, but it's, um, it's not that. It isn't part of social cognition, right? It is a huge part of social cognition, how we perceive people in the first moment, how we see them, um, these intrinsic biases that we have, that we have learned, that we, um, were taught how do they influence our behavior.

And it's just, um, expanding essentially this, um, this question of social cognition to a more realistic environment, a more real life environment that is not just so focused on what's going on in the lab. And I think it's, yeah, it's been a great experience and I think it will help to get new ideas and, and, and continue this kind of research.

Benjamin James Kuper-Smith: But is the, the public goods scheme that your, the data that you collected as part of the COVID study, is that as [01:18:00] like standard public good game, or. Is it like, did you change something about it to make it more applicable or

Tessa Rusch: No, it is, um, pretty standard public goods game. It's just that it's essentially anonymous because the groups are randomly selected on each wave, so they can't really learn anything about their partners because it's, you kind of learn more about the entire group, essentially the group behavior, but it is.

Classic center public goods game. And they know the other players are the other participants. Yeah. And we actually, there's, I got an email once from a participant saying, why are people not investing? This is so stupid. And then calculating, like giving me all the, the calculations about it. And it's like, why are, is, are these actually real people?

Because this is very irrational behavior. And that was, I mean, that was a great message.

Benjamin James Kuper-Smith: I mean, that's in the, in the very first prison cinema dilemma, right? Yeah. There's this by flood where he has at the end the comments and the second person that keeps saying like, come on. I'm like, doesn't he [01:19:00] realize that this is the, the more rational thing to do?

I'll cate a few times, so he'll get it. Yeah. And they're like, oh, he really doesn't get it. Like, what's wrong with it?

Tessa Rusch: Exactly.

Benjamin James Kuper-Smith: Yeah. I mean, I guess it also makes it easier if you then like didn't have a predefined project per se. That you like definitely wanted to work through at CarTech or something, right?

Like if it's, if you're interested in like social interactions and dynamics, then you are in a sense doing that. Um, it's just somehow for me, it seemed like you, maybe just because of what you did in handbook, that's just not the direction you planned to go when you arrived.

Tessa Rusch: No, but I mean, I'm, I mean, I'm kind of expanding the direction a little bit and also with the project that I'm working on now is, um, I mean what I did during my PhD was a very specific work, right?

It was very specific on how do we, how can we model very rational theory of mind processes or what's rational is, is is a bad word. I think it's a, [01:20:00] a very structured and strategic theory of mind processes essentially. So it's a very specific part of theory of mind and social interaction that we're looking at in these economic games or in these, in these very structured interactions when we, when we explore that.

And so, um. It's, I think it's actually, it follows from that to expand it a little more into what, what, what are the alternatives that, I mean, it is, it is modeling and it is capturing a part of social behavior, a very specific part. But it's, um, social interaction, social behavior is very broad and very big.

And they're very different options to do it and engage in it. And, um, if we wanna predict and explain human behavior, we have to, to look at more aspects.

Benjamin James Kuper-Smith: Okay. Should we, should we look at the old aspects though?

Tessa Rusch: Let's look at the old aspects. That

Benjamin James Kuper-Smith: was good.

Tessa Rusch: I mean, I'm not saying that's important. It's not, I'm not saying that it's not important and that I don't enjoy this work. Um, [01:21:00] I'm just saying that it's part of a, of a broader approach.

Benjamin James Kuper-Smith: Yeah, of course.

It's, I mean, it's one approach to studying a type of behavior or phenomenon or whatever. Uh, it's not the only approach. Although that's a pretty good approach. Thank

Tessa Rusch: you.

Benjamin James Kuper-Smith: Yeah. Maybe if you, if you remember, can you give a like one to two minute summary of the paper,

Tessa Rusch: of the review paper

Benjamin James Kuper-Smith: of the review of your neuropsychology paper, theory of mind and Decision Science towards the departure of tasks and computational models by au

Tessa Rusch: It is, um, it is a paper reviewing different kinds of computational cognitive models for social decision making.

And as I said before, it's um, looking at very structured social interactions that are often studied in at the, like, interaction between economics and cognitive neuroscience that are, um. Inspired a lot by economic [01:22:00]research. Um, some are more from psychology, some are from, uh, from more from economics. And we're looking at this inter intersection essentially, where these two things come together, which study similar questions with very, very different approaches.

Um, and it has grown together more and more over the last years, I think, um, that, um, methods have been shared and, and approaches have been shared. And so, um, as I said, it looks at different computational cognitive models. So formula, formula, formula. Well, formalization for formalization, I can't

Benjamin James Kuper-Smith: formalization,

Tessa Rusch: formalization.

Thank you. Um, of, um, of these kind of, um, interactions, how can we quantify social interactions or social strategic interactions and, um, uh, theory of mine generally is referred. Through the process of inferring other people's mental states. Um, although even this term is very differently used by different people in different fields.

[01:23:00] Um, but it's, it's a, it's, it's a term from social psychology, developmental psychology. The question is how to what degree can these different computational models comput capture the theory of mind processes and also in which, um, lab tasks in which lab environments and settings are such, um, inference processes about other people's mental states actually elicited, which is not a trivial thing, I think.

Benjamin James Kuper-Smith: Yeah, that's something I have a few questions about later. Um, and then you also have the, but first, uh, you also have these two. Dimensions according to which you,

Tessa Rusch: uh, yeah. We're trying to characterize, um, tasks and computational models according to two dimensions, as you said. Um, one is uncertainty and, um, distribution of information among individuals.

And the other one is kind of, um, interactivity, uh, to some degree. So how, um, relevant is the [01:24:00]behavior of other agents, as it is called in decision neuroscience, um, or in, in decision science, um, affected by your own behavior. How much are you affected by the other person's behavior? So how, how interactive is this essentially, and how, so for the tasks, it's um, like what is this task?

Does it have this component or this characteristic? And for the models it's more, can it capture, um, this component or characteristic?

Benjamin James Kuper-Smith: What exactly was the motivation behind this? Sorting it or like having these two dimensions and using that to kind of structure tasks and models.

Tessa Rusch: That is, I mean, this is act actually from when we developed the task for my, um, PhD and, um, because we, so we were starting out with, okay, we wanna look at theory of mind processes in an interactive setting.

Um, and we wanna look at how um, actually the neural signal between two interacting partners [01:25:00]interaction influences is influenced. And, um, so really look at interactive theory of mind processes. And then we sat there and we looked at different tasks, these classic economic games, um, and, and other tasks.

And we realized, um, that we weren't quite happy with it because we thought, um, it doesn't really, I mean, there's a huge. Possibility that it doesn't really elicit these kind of mental processes because there are other far easier options to solve these tasks. And so, um, actually that was, um, an email that I still have saved on my desktop as a note, um, from Martin, hi, who at the time was a postdoc with Jan.

And he, he, he wrote me this email and was like, okay, here are these, um, criteria for which I think what elicits theory of mine doesn't. I think you have to have, uh, it has to be, I mean obviously it has to be interactive. Otherwise, um, this is something that we realize very honest. It's like if there's no, um, behavioral relevance, then yeah, what the [01:26:00] other person does is irrelevant to you.

Um, and so, um, it has to be interactive. And then the second part was your belief or your state of representation of the world has to be different from the other person's representation, um, or of the state of the environment, because otherwise you can't discriminate them experimentally. So, um, if they're identical, you might as well think about what the other person's mental state is.

But if it is identical to your own mental state, then you as an experimenter can't discriminate between these two things because they're identical. So how would you discriminate between the two? Um, and that is kind of how it started out. Um, and, um, then we started thinking about which environment and which situation would actually create such a scenario.

Or in the real world, where is this the case? Where do you have, um, for example, asymmetric information or you have uncertainty? Um, there's, there's, um, there are multiple possibilities for what you are currently thinking about the environment or what kind of information you have about the environment, how you, what it is that is driving your behavior.

Um, and, um, [01:27:00] where you're interacting with another person. And that's how these two dimensions essentially came along because we think, um, uh, the more interactive, the more behavioral relevance there is, the stronger you start engaging in these thinking about what does the other person think and potentially how am I being perceived by the other person?

How does my behavior influence the other person? And then if you were, um, very good at this kind of recursive reasoning and could go even further in the sense of, I think that you think that, I think there's actually a funny, uh, friends scene where, um, it's about two of them having an affair and the others don't know.

And there's this scene, it's like they know that we know and it's like, but they don't know that we know that they know. And then it goes on and on and on like this. So, um, yeah. Um, and um, so that's what

Benjamin James Kuper-Smith: you got your inspiration for.

Tessa Rusch: From friends. Um,

Benjamin James Kuper-Smith: okay.

Tessa Rusch: Yeah. To

Benjamin James Kuper-Smith: solve that problem for that question,

Tessa Rusch: yeah.

It's a very intellectual [01:28:00] journey. Um, um, yeah, so that was the motivation for that. And so that's how we developed this and how we looked at these different computational models to think about how we can capture that and quantify that. And, um, we came to this conclusion that you have to have these, um, certain criteria to make sure it's not that there's no theory of mind elicited in these other situations.

I have no idea whether it is, it could be, and it could be different for different people, but just the, the likelihood or the probability that it happens is larger in such an environment because there is relevance of it. The other person's mental state has relevance for you and the other person's behavior has relevance for you.

Benjamin James Kuper-Smith: Okay. That's slightly changed. Yeah. It changes a little bit how I understood then. Okay. So it is also a part like, as an experiment of what can you actually get out of this task in that sense, not just from a what does a person do, but like what can I infer from what they do?

Tessa Rusch: It is, um, it is an assumption of ours that this is stronger [01:29:00] in these situations.

I mean, if for us it makes more sense that people would engage in this kind of thinking in certain situations, but in the end, this is an imp empirical question, right? I mean, we can make all kinds of assumptions about how humans reason about things. Um, and we can from introspection, we get certain informations, but this is just ourselves, right?

So we're assuming that certain scenarios are more likely to elicit these kind of reasoning processes. Um, but as I said in the end, it is an empirical question.

Benjamin James Kuper-Smith: Yeah. So if you don't mind, can I be slightly critical of some things in the paper?

Tessa Rusch: Sure, go ahead.

Benjamin James Kuper-Smith: So, one thing, as you said earlier, that you need interactivity for to have three minor on these tasks and.

You talked about behavioral relevant, like because of behavioral relevance. Basically, if it doesn't matter to me, then what? But to me those are just not the same thing. Right? Like in, so for example, you have the, the two, like two of the examples you give, uh, one is the false belief task and the other are, uh, D games like the prisoner dilemma or stack [01:30:00] hunter, whatever.

And it seems to me that according to figure four, uh, where you basically, you list them, right? And I think the idea is that kind of the more towards the right and up they are, the more likely the tasks, sorry. The more they are, are, the more uncertain they are. And the more interactive they are, the more likely they are to elicit theory of mind and the prism.

But it seems to me that, for example, a prisons cinema can very easily lead to a lot less theory of mind than a false belief task, even though it's more interactive because, so in the false belief task, it seems to me that. Or rather, it seems to me a lot about the number of options, like to the almost action space or belief space of someone.

So like if you have a false belief task and the other person can have like, there's like six different things they could believe, right? Then I have to figure out, okay, is it that option or not? Why would it be this optional? Not all that kinda stuff. Whereas in a pre dilemma, it's a binary decision. It's much, much simpler in that sense.

Mm-hmm. [01:31:00] Uh, the complexity of the situation is much reduced. So it seems to me that those examples, even the examples you give to me, almost point towards the opposite of what you're trying to do, um, in terms of. So rather that I would almost think that they are reversed in terms of how much theory of mind they elicit, if that makes sense.

Tessa Rusch: So, um, first it, it's, um, not necessarily that it elicits more theory of mind. So this, um, I think it's the y axis is more about interactivity, so about this, um, recursive reasoning essentially. Um, and so I, I, I would agree that, I mean, what you're actually saying is action uncertainty. You don't know what the other person's gonna do and there is a, yeah, a lot of uncertainty about what the other person's going to do.

So there's not much state uncertainty, but there's a lot of action uncertainty potentially. Or maybe there's even more, um, state uncertainty. And I would agree that if there is more uncertainty, um, it would elicit, um, it, it has the potential to elicit more reasoning about the other person. But on the other hand, if it's completely irrelevant to you.[01:32:00]

Then why would you think about it?

Benjamin James Kuper-Smith: But you could make a false belief task relevant right? By saying you get a euro every time you get it correct. Or whatever. Right? And then

Tessa Rusch: for example, yeah, that's, that's true. But then you would still, um, your prediction of the other person would be completely inconsequential of the pers uh, to the person that you're predicting.

So, um, this is kind of what, um, we thought with this recursive process is if my behavior, if my prediction of the person that has the false belief, for example, is correct or incorrect, that is completely irrelevant to the person engaging in, in this task, making the decision essentially, because they don't know what kind of prediction we're making.

On the other hand, if you would say that this is your opponent or this is your partner or something, and if. He or she predicts correctly or incorrectly what you're doing. Um, then you get, you get a price or, or something like that. Or you will find each other in, um, in the museum where you're looking for each other and you know where the other person's going or something like that.

Um, or only if you, if you reach the goal together, you can open the [01:33:00] gigantic chess box and get the price and the pot of gold or something. Then there's this interactive behavioral relevance essentially. On the other hand, if you have a lot of uncertainty and the other person's behavior is completely irrelevant to you, let's say you're in a supermarket and there's another person going shopping, there are thousands of choices, but you usually don't walk, or at least, I don't know, maybe you do.

I usually don't walk behind people and be like, I think that person's gonna grab that apple and that you

Benjamin James Kuper-Smith: should do it, Tess. That's amazing.

Tessa Rusch: Yeah. I mean it's, it's, especially

Benjamin James Kuper-Smith: during COVID working closely behind people.

Tessa Rusch: Oh yeah. That's not creepy at all. Um, nope. Uh, yeah. So, um, I mean there are a lot of o options for social interaction and a lot of options for social inference.

But I think, yeah, if if the other person's behavior is not relevant to you, then it's just an e energetic question essentially. Why should you think about it?

Benjamin James Kuper-Smith: Yeah, I mean, maybe I agree then with the, with what you just said and the overall framing, but maybe not precisely the [01:34:00] examples given or I dunno, the false belief task enough.

I don't know.

Tessa Rusch: I mean, the false belief task essentially as it is classically as you, you see a story, you read a story about someone doing something and then, then you have to predict, um, what that person's gonna do. And it's essentially the person left the room and something changed. And so the person has a false belief and now you have to predict their person's behavior based on that other person's belief essentially.

Um, but it's a, it's a completely passive task.

Benjamin James Kuper-Smith: Maybe it's, to me that it just seems more open-ended. In terms of like, where are they looking? And okay, if you, if it's like very, very minimalistic, you only have two options, then it's binary also. But it just seems to me like the way they're looking again, that just elicits way more than if I'm doing a prison dilemma and the other person affected for the first time in a row.

Tessa Rusch: Yeah, no, um, I, I totally agree. And, and that's why I think, for example, the prisoner's dilemma doesn't have a lot of uncertainty. Um, it's pretty clear, depending on how you frame it or like who you interact with. Let's say you interact with very different [01:35:00] people, um, then you again have uncertainty about the motivation of the other person.

Uh, but if you, if you interact, if you play a prisoner's dilemma a thousand times with the same person, I'm pretty sure that you're gonna develop some sort of heuristic. It's just gonna, we're gonna press the same button over and over again and there's learn a strong learning effect. Yeah. Maybe, um, maybe the figure isn't very good at making that clear.

It's, it was a bit of a difficult, um, characterization of putting it in this two dimensional space essentially. But, um, no, I agree. We should have probably made that clearer.

Benjamin James Kuper-Smith: Yeah, I mean, like about, so especially because, you know, I've been doing basically for the first year I, my PhD, I thought a lot about the dilemma and use it.

So maybe I'm focusing a bit too much on that, but the, but I, but I will.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Um, uh, but for there especially, it seemed to me that, so there were like a few things. I mean, like you said, like when you have lots of iterations with one person, it tends to become fairly obvious what the next person's gonna do.

And in some [01:36:00] of the iterative or iterative prison dilemmas we had, I mean, the first thing is reciprocity just super high. It's a very strong correlation. People pretty much will cooperate as much as the other person will. Um, and you often have people who, you know, basically reach a stable state of mutual cooperation, which would affection.

It's pretty, I mean. There's some variance, but it's, it's pretty common. But the other thing I was really thinking about here with theory of mind in the prison dilemma is that, and this is something I feel like maybe I missed it, but it seems to me that it's missing from the discussion in the paper is just in general the idea of heuristics.

Because it seems to me, I mean this is also the general thing for life. Uh, one can completely agonize over what someone might be thinking all the time. Or you go, well, let's do that thing and see what happens. And it's in the prison dilemma, it's much stronger that Right. Re like tit for tat. Basically you just do whatever the other person did in the long run.

No one's gonna really do better than you. And you don't have to think at all about what the other person's thinking. [01:37:00] Right. It's just, it's almost like a non theory of mine task to me. Um. Yeah, I dunno whether I have a specific question here other than wanting to have I, I felt the need to say that.

Tessa Rusch: No, I'm, I'm, I'm, I agree.

Um, and there is a huge amount of research on that, right? Um, as you were saying, and a lot of that research includes your own work,

Benjamin James Kuper-Smith: not that big a proportion. You made it sound like I've been doing a lot of research here relative to the rest of humanity.

Tessa Rusch: Um, great. Benjamin Cooper Smith and the President Dilemma.

Um, it's pretty much not yet the Cooper Smith dilemma. Um, no. So, um, yeah, I agree. And so for example, some of these computational models include that to some degree. For example, a fictitious play model essentially counts the choices the other person made previously. So it's just. It's just taking the, the frequency of the, of the other person's action and predicts the, the [01:38:00] next action based on the frequency of previous actions.

And there's absolutely no inference process there. Um, the only thing is like, what are you gonna do if the other person's gonna do this? And the but for the prediction part, you make no inference about the other person. And yeah, tit for that definitely should have been in there. Um, once they lose shift, for example, all these things.

But we, um, we didn't focus on these models there because I mean, they, they never claim to involve any kind of theory of mind process, right. They're not claiming to, to capture,

Benjamin James Kuper-Smith: I mean the, a cognitive theories, right?

Tessa Rusch: They are cognitive the theories, but they're not, they're not focusing

Benjamin James Kuper-Smith: on theory. No. I mean, they're a cognitive

Tessa Rusch: Oh

Benjamin James Kuper-Smith: yeah.

Like if you have evolutionary simulations, it's about something that doesn't even think. Right. They just have a certain reflex. You could almost say. Yeah.

Tessa Rusch: Yeah. So, um, yeah. So this is why we, um, the, the, for example, that's what we included and, uh. The fictitious play model, for example, because it has been used a lot, but it doesn't make any assumption about what drives the other [01:39:00] person's decision making process.

Benjamin James Kuper-Smith: Yeah, exactly. And I mean, that's one reason that I am, like, in my own stuff, I'm, I'm wondering like with how much I'm gonna use these two by two games, um, because in a way the information you get out of it is very, uh, very

Tessa Rusch: limited. Yeah.

Benjamin James Kuper-Smith: Yeah. I mean, I still find it interesting as it, because it's binary, um, that has a certain, uh, neatness to it rather than if you have like complex data.

But yeah, it is kind of limiting in terms of what you can actually really get out of it or.

Tessa Rusch: Well, I mean, it's, it's always a trade off, right? If you have more freely interacting people, you probably get richer social interactions and the, the, the inference processes are much richer. But then on the other hand is how do you quantify that?

It's, after all, it's this private cognitive process that you have no way of accessing unless the person tells you what they were thinking. And then you're making the assumption that the person themselves, um, themselves has, um, [01:40:00] this access to their own cognitive process, and they're able to vocalize it and, and, which is.

It captures a part of this cognitive process, but it doesn't capture all of it. Um, probably. Um, and so, um, I mean this is the, this is I guess where these, um, these decision models are handy is because it gives you some sort of quantification. It makes it, um, it makes 'em measurable comparable, but of course it is very limited.

And especially as you get into larger state spaces or larger action spaces, they become very hard to compute and it's becoming very messy. And then the question again is, is that what humans are doing? Are we following this complex basin, um, learning process and updating process, or is it something else?

And is this just approximating it? Um. Yeah.

Benjamin James Kuper-Smith: Yeah. I mean, that's one thing I found kind of curious. While reading your paper, I can't remember where I wrote it, but somewhere I wrote in the margins. Like this is almost like assuming the homo economicals, but from a [01:41:00] psychological perspective, like all this inferential process is going on rather than someone saying, ah, whatever.

I'll just do that.

Tessa Rusch: Yeah,

Benjamin James Kuper-Smith: we'll see

Tessa Rusch: what happens. And I mean, even just introspectively, like think about it, when are you really thinking about another person's mental state?

Benjamin James Kuper-Smith: Not enough or, yeah, so I'm told

Tessa Rusch: no, but in the mo like most times you make very heuristic and fixed decisions. Like, um, I mean, I, I find always this example when you're passing someone in the hallway and usually everyone goes to the right, right?

And then, but sometimes there's this weird situation where everyone goes back and forth and this is a clear sign of, okay, the, the norm somehow didn't work here and we're stuck in this weird situation where you don't think, where you go and you don't think about it, will the other person take the left or the right?

He will just do something.

Benjamin James Kuper-Smith: Yeah. Yeah. I mean, I feel like also in like introspective real life, like even if you think about someone about what someone's doing, you have like three or four factors you consider, right? Like that's the most you go [01:42:00] like, okay, were they like in a general good mood or whatever, how do they usually, you know, whatever.

Like it's not like this super complex reasoning that really goes on. I mean, I guess maybe most, some people say that that's kind of what emotions are, this kind of integration of lots and lots of different, uh, influences, but at least on the cognitive level or the conscious level. Yeah, I think it's heuristic, heuristics all the way.

Tessa Rusch: Yeah, it's emotion and memory. I mean, you have a lot of, for the most people, you have interacted with them a couple of times and you had previous experiences that probably weighted into different degrees. Um, some were very influential of your image of that person. And so you, and you have met a lot of other people and you can probably, you, maybe you can make some kind of comparison between.

People and like likely behavior that these people are gonna exhibit. Yeah. But, but as I said, I mean the social interactions are so diverse and so different. Um, and it's, we're probably not using the same process in every [01:43:00] situation. But when you think about, so like really these really economic games, uh, like when you're on the stock market or when you're really trying playing poker with someone and you're thinking about whether they're actually have good cards or not, and whether they're just bluffing, it's like then you're really trying maybe to think about this, but it's probably very specific situations where the stakes are high and, and where you have a strong drive to do that.

Benjamin James Kuper-Smith: Yeah. That's another thing I wrote down. Your third dimension could be just stake size.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: Because that's really like even in a business dilemma, right? Like if you're doing lots of tasks and you earn like a year or 10 or whatever, people will probably not think about it as much as if you say like, there's this, I mean the the like, uh, Richard Thala has like some of these papers where they look at game shows, right?

Like those people are probably thinking about this quite a bit when it's like $300,000 or whatever.

Tessa Rusch: Yeah. I mean we probably would get very different results if you'd be like, okay, now you're playing for $50,000, what is the other person going to do? [01:44:00]

Benjamin James Kuper-Smith: Well, I mean, cognitively you'll probably get different things.

Yeah. But I dunno, in terms of behavior, I mean, there are some studies that look at stake size and it seems to be, I think people are a bit more idealistic.

Tessa Rusch: Mm-hmm.

Benjamin James Kuper-Smith: Uh, no, sorry, that, sorry, wrong. That was hypothetical stuff. Um, I think people behave a little more traditionally economically rational. The, the, the more money is at stake, but not that much, I think.

Tessa Rusch: Well, but, um, I mean,

Benjamin James Kuper-Smith: but the thought process going into it might be quite different.

Tessa Rusch: Yeah. And it could be again, um, where are you, what's your action space in the sense of what are, how much money do you have? What's your life situation in a way? Can you afford to be, to be, to share or can you not? Or, um, are you used to not sharing and all these kind of things?

Um, I mean, people are, make decisions in very different situations. [01:45:00] Um, and I think the context in which a decision is made is not, um, irrelevant.

Benjamin James Kuper-Smith: I mean, it's basically everything there is, right? Even if you have personality stuff, it. Yeah, I think like when you look, when like the, I've been thinking about that a bit.

Like when you look about like what influences how someone makes a decision, you have the environmental factors and the kind of more personal factors. But it seems to me in many cases the environmental factors are just, just outnumber the personal quite a bit. And maybe the, the personal ones are, it's, unless like anything's too extreme, like unless like it's about huge amounts of money or the differences are huge between options or whatever.

I think like if, if everything's kind of similar, then the personal stuff is probably more important, like piece by piece compared to environmental stuff maybe. But I think there's just so many environmental factors that like accumulation of it just, yeah.

Tessa Rusch: Yeah. I think there's this one paper, um, that looks, for example, ed, um, an [01:46:00] equity aversion and guilt aversion or something.

So where you have the option to the other person knowing that you didn't share. Or the other person doesn't know. And then, I mean, it's a very different incentive in a way. The other person thinks you're fair, but you're actually keeping quite a lot of the money or um,

Benjamin James Kuper-Smith: do you remember what that paper is?

Tessa Rusch: Um, I think it's, um, nature Communications and One Bar at all, I think.

Benjamin James Kuper-Smith: Okay. I have a look. Yeah, I might actually talk to someone soon who does inequality aversion of animals. Uh, don't wanna name the name because we haven't scheduled anything yet, but a person said yes, but that would be, that would be fun.

Tessa Rusch: Oh yeah, it would be cool. Sounds

Benjamin James Kuper-Smith: interesting. I dunno whether animals have guilt aversion. I dunno. That's a good question. Yeah, it's just weird. Like in general, like when I think about, I think I wrote this to you in my email when I asked you whether you wanna do this. Like, I've basically not been thinking about Theory of mind at all.

And I've been working on, you know, princess Dilemma, that kind [01:47:00] stuff for a year now. And somehow it's this weird thing that. In some sense seems super important. And you know, as, as you said, like one of how we spend a large part of our lives, right? Thinking about what other people are thinking. But in a way also, you can also manage without it pretty, pretty well in many situations.

Tessa Rusch: Yeah.

Benjamin James Kuper-Smith: I think I've, I've run through my points.

Tessa Rusch: Okay.

Benjamin James Kuper-Smith: And I think you also might want a slight break before your next meeting.

Tessa Rusch: It's okay. That's what we do nowadays. Sit in front of our screen, talk to people.

Benjamin James Kuper-Smith: Okay. I'll just stop recording. Yeah. Or do you have anything you wanna say? Any declaration to make any i I, uh,

Tessa Rusch: final words of wisdom?

Benjamin James Kuper-Smith: Yeah, exactly. Something you, what you've learned and that you think people should know. Oh, [01:48:00]

Tessa Rusch: no, no. I'm not gonna claim to have any kind of knowledge that everyone should know.

Benjamin James Kuper-Smith: Okay. Is that your knowledge then? Everyone has their own piece of knowledge or,

Tessa Rusch: um, yeah, no, sorry. There's no grand statement.

Benjamin James Kuper-Smith: No wisdom of tesa.

Okay, then I'll stop recording.

14. Tessa Rusch: COVID-Dynamic, an extremely variable year, and theory of mind

Featured Posts

Read Full Transcript