
The Flux by Epistemix
Welcome to The Flux - where we talk data, decisions, and stories of people asking the what-if questions to create an intentional impact on the future.
The Flux by Epistemix
From Curiosity to Complexity: Paul Amoruso’s Journey into Agent-Based Modeling
In this episode of The Flux, host John Cordier, CEO of Epistemix, sits down with Paul Amoruso to explore his unexpected journey into the world of agent-based modeling (ABM). What started as a course recommendation turned into a passion for understanding complex systems and using data to drive better decisions.
Paul shares insights from his academic path, the value of interdisciplinary collaboration, and why ABMs foster explainability in research. He also discusses how modeling could reshape fields like education and improve knowledge diffusion. Plus, in our What the Flux moment, Paul reflects on how we might have transformed traditional learning methods if we had embraced active learning sooner.
Tune in for an engaging conversation on curiosity, computational social science, and the power of modeling to inform the future.
Welcome to The Flux, where we hear stories from people who have asked what if questions to better understand the world and talk about how data can help tell stories that impact decisions and create an intentional impact on the future. This is your host, John Cordier, CEO at Epistemix. In a world where the flux capacitor from back to the future does not yet exist, people have to make difficult decisions without always knowing how the future will play out.
Our guests are people who've taken risks, made decisions when uncertainty was high, and who have assisted decision makers by using data and models. We hope you can turn lessons from our podcast into foresight, so you or your organization can make better decisions and create an intentional impact for others.
Hey there, welcome to the next episode of The Flux.
We're here with Paul Amoruso. Paul, thanks for jumping on.
Paul: Thank you. Thank you for having me.
John: Yeah, so we're here in Santa Fe at the Complex Social Systems Society Conference, talking all things agent-based modeling, systems dynamics modeling, and complex social science. What got you into the field in the first place?
Paul: Well, it's kind of a funny, long story. But since it's a podcast, I guess we have time.
John: We definitely have time.
Paul: Alright. So, I needed to take some additional courses for my PhD program. I had already taken most of the machine learning classes that applied to what I was studying, and the other available courses didn’t really align with my interests. So, I asked my advisor if he had any recommendations. He told me to check out the industrial engineering department and see what courses they had.
In doing that, I came across a professor with the last name Garibay. I told my advisor, "Hey, there's this human-computer interaction course taught by someone named Garibay." And he said, "Oh! I was on his dissertation committee—great guy! Go ahead and take that class."
So, I took the course, but when I got there, I realized it wasn’t a guy—it was a woman. My advisor had either mixed up the name or the gender. About a quarter of the way into the semester, I went to clarify with my professor. I said, "My advisor keeps referring to you as ‘he’ and calling you Ivan, but… you’re not Ivan." She laughed and said, "Oh, that's my husband!"
That moment was hilarious because, for the first few weeks of class, I was completely confused. Same last name, same department—just not the person my advisor thought.
Anyway, I enjoyed her class, did a research paper with her, and got into human-computer interaction, user interface design, and so on. Then, the next semester, I thought, "Let me try taking Ivan Garibay’s course." But that semester, he wasn’t teaching. So, instead, I ended up taking a course taught by his advisor, which was on complex adaptive systems.
That’s when I started learning about NetLogo, modeling behaviors, behavior space, agents, and environments. I was getting familiar with the process, but at that point, I was still a bit removed from agent-based modeling and computational social science. I did a research project in that class, but it wasn’t strictly agent-based modeling—it was more along the lines of cellular automata.
Then, in my final PhD semester, I had the opportunity to take Advanced Agent-Based Modeling. That’s what really set me on this path. For that class, we had to do a project, and I was brainstorming ideas. Eventually, I landed on a project that I actually presented last night as a poster presentation.
At that point, I wanted to find a conference to present my work. I didn’t just want to get a grade—I wanted to get the most out of my project. Ivan Garibay suggested this conference, and I’m really grateful I decided to come. I've met you, I've met so many other people from different backgrounds—anthropology, geography, computer science, economics.
John: Yeah, I mean, every other conference I’ve been to, you see some variation in backgrounds, but this one really feels like a melting pot.
Paul: Exactly. The conversations are amazing. You learn so much. You’d think that with such a wide range of disciplines, there might be a disconnect—like people struggling to find common ground—but actually, it’s the opposite. It feels like an idea-generation hub. Every time I talk to someone, I walk away with new ideas.
John: Let’s go into that a little more. In a lot of fields, even within the same discipline, it can be hard for people to communicate—especially when discussing advanced topics. But in complex social systems, you just listed five different disciplines, yet everyone seems to be having meaningful conversations. What do you think drives that? Is it a common tool like NetLogo? Or is it a shared way of approaching problems?
Paul: I was actually thinking about this earlier today. I don’t think it’s NetLogo specifically, because not everyone here uses it. But we can still understand each other’s work and have meaningful discussions. I think the key is explainability.
Everyone at this conference, in my opinion, is focused on making their research explainable. We talk about how traditional AI methods—like linear models or neural networks—can be black boxes. But agent-based modeling and computational social science aim to create a transparent box where we can understand and trace the reasoning behind the outcomes. That shared commitment to explainability carries over into our conversations.
John: That makes sense. We’ve heard that before on this podcast—that agent-based modeling is really about explainability. It’s not just about predicting the future at scale; it’s about understanding why things happen so we can make better decisions.
Paul: Exactly. It’s all about curiosity and learning. People here genuinely want to understand how things work and share that knowledge with others. There’s no gatekeeping. If anything, people are eager to help each other refine their models—like, "Hey, if you add an information diffusion aspect, it might clarify your results." That openness is really unique.
John: Yeah, it sounds like a great environment. So, we have time for one or two more questions. One thing I always like to ask—since you kind of stumbled into agent-based modeling through a series of course recommendations—if someone wanted to be more intentional about getting into the field, where should they start? And what can the agent-based modeling community do to make the field more accessible?
Paul: My advice is the same thing I tell students learning programming. There are two approaches:
First, start by building a foundation—watch podcasts like this, check out YouTube videos, read online documentation. Get familiar with the basics.
Second, look for examples of agent-based models that relate to something you're interested in. If you can find a model that somewhat aligns with what you want to study, start by modifying that. Hands-on learning is key—build something small, then expand on it.
As for making the field more accessible, I think it comes down to providing more educational resources—clear documentation, well-explained YouTube tutorials. When you teach others, you also refine your own understanding. So, researchers should put more of their work out there in an explainable way.
John: That makes a lot of sense. Final question—this podcast is called The Flux, named after the flux capacitor from Back to the Future. If you could rewind history and simulate an event to show we should have made a different decision—whether it's an economic policy, social policy, or even a business decision—what would your What the Flux moment be?
Paul: This is fresh on my mind because I was just discussing it last week—education.
Traditional lecturing, where one instructor stands at the front of the class and talks at students, doesn’t make much sense when you look at student comprehension rates. Studies show that active learning—where students engage in discussions and exchange ideas—leads to much better outcomes.
If I could go back in time and model information diffusion in classrooms, I’d show just how much more effective active learning is. And honestly, I’m sure models like that already exist. But if we could visualize it in a compelling way, maybe we’d finally move toward more interactive, student-centered learning.
John: Yeah, now it’s just about making that change.
Paul: Exactly.
John: Paul, thanks so much for being on the podcast.
Paul: Thank you for having me.