Through the conversation.
22 sections010:01 – 3:33Introductory Discussion on Mathematical Models and Evolutionary Science
Hosts Dave Pietraszewski and David Pinsof introduce the podcast episode featuring guest Paul Smaldino, discussing his work on mathematical models, institutional selection criteria, bad science, and the nuances of covert signaling and punk rock.
The main thing is like why they're why math models models are stupid, we need more of them
“why math models models are stupid, we need more of them” 2:20
what matters is the selection criteria of the institutions.
“what matters is the selection criteria of the institutions” 2:41
023:34 – 12:03The Functional Role of Identity and Social Categorization
Paul Smaldino discusses the functional role of identity from an evolutionary and cultural perspective, explaining how human identity acts as a social category used for adaptive decision-making, cooperation, and strategic interaction.
Paul Smaldino explains that identity functions as a category used for adaptive decision-making such as finding cooperation partners.
“We use that for all kinds of adaptive decision making. Like how do we find partners for cooperation” 5:49
Paul Smaldino notes that identity is a multifaceted social category that involves strategic negotiation and presentation across different contexts.
“we often sort of put on, we try on roles, we try on identities in different contexts” 10:56
Host Dave Pietraszewski prompts guest Paul Smaldino to discuss his methodological approach to studying social identity and other topics. Smaldino explains that the vast majority of his work relies on mathematical and computational modeling rather than just traditional experiments or fieldwork. He describes how building simplified artificial systems helps researchers tackle complex dynamics, uncover unexpected consequences, and run numerous variations efficiently.
I mostly do modeling i i the the bulk of my work i would say 80 to 90 percent of the work i do is building mathematical or computational models.
“I mostly do modeling i i the the bulk of my work i would say 80 to 90 percent of the work i do is building mathematical or computational models.” 13:28
When you deal with a very complex system, you have to ignore a lot of it in order to get any tractability and do science.
“When you deal with a very complex system, you have to ignore a lot of it in order to get any tractability and do science” 15:55
Paul Smaldino and the hosts discuss the practical advantages of using computer simulations over traditional lab experiments or long-term field studies. Smaldino notes that while designing a valid model is the hardest part, running simulations allows researchers to test thousands of variations rapidly compared to the immense time required for developmental psychology or field anthropology.
A passage from this discussion.
“and it takes like a really long time to set up a lab or a field site and I know this you” 20:35
0524:10 – 27:36Identity Signaling and Social Diversity in Human Groups
Paul Smaldino discusses the anthropological and sociological history of identity markers, explaining how individuals signal their identity through fashion, language, and behavior to navigate group affiliations. Host Dave Pietraszewski and David Pinsof join in with a lighthearted comparison between human urban commuting and chimpanzee social behavior.
if we live in a world where people are different from each other and they meet strangers, it would be good to have a way sort of from a distance
“if we live in a world where people are different from each other and they meet strangers, it would be good to have a way sort of from a distance” 24:44
0627:37 – 36:11Covert Signaling, Implicature, and Mathematical Modeling
Paul Smaldino elaborates on covert signaling and pragmatics, explaining how humans use ambiguous language, humor, and subtext to signal their identities selectively in diverse societies without risking backlash. He then describes how researchers translate these verbal theories into formal mathematical models to test predictions.
we actually aren't just trying to honestly signal about ourselves um it's actually a lot of the time we we hide information
“we actually aren't just trying to honestly signal about ourselves” 28:49
When this is used strategically for uh identity signaling, this is what we uh call covert signaling.
“when this is used strategically for uh identity signaling, this is what we uh call covert signaling.” 32:10
Paul Smaldino discusses with Dave Pietraszewski and David Pinsof how mathematical models and artificial populations are constructed to study overt and covert signaling between similar and dissimilar individuals. Smaldino explains the trade-offs in communication fidelity, how assumptions can be tested or extended, and the benefits of agent-based modeling for investigating complex social interactions.
And you could do this if you think that's a bad assumption, you could do it and you want you should do it another way.
“you could do it and you want you should do it another way.” 37:12
Paul Smaldino explains how successful covert signals eventually become recognized by out-groups, leading to a natural lifespan for signals like dog whistles. Smaldino, Dave Pietraszewski, and David Pinsof examine how Twitter data and behavioral experiments around political polarization helped validate these theoretical and computational models.
We call out the dog whistles, usually often years after they were effective.
“We call out the dog whistles, usually often years after they were effective.” 42:31
And we found that there were a number of tweets that reliably people on the same side of the spectrum as the the tweeter on one side of the spectrum were like, Yeah, yeah, yeah, I know what this is.
“people on the same side of the spectrum as the the tweeter on one side of the spectrum were like, Yeah, yeah, yeah, I know what this is.” 47:19
0948:12 – 57:33Analyzing Political Signaling on Twitter and Behavioral Experiments
Host Dave Pietraszewski, David Pinsof, and guest Paul Smaldino discuss how political actors use overt versus covert signals on Twitter depending on audience composition. Smaldino explains observational findings from follower networks and describes a behavioral experiment where participants selected tweets for audiences with varying political distributions.
The people whose tweets were more covert tended to have more followers from both sides.
“The people whose tweets were more covert tended to have more followers from both” 50:39
When participants knew that the majority of their audience was different from them, they selected fewer tweets overall and a higher percentage of covert ones.
“they were even more careful in two ways. They just selected way fewer tweets overall. And of the ones they selected, a higher percentage of them were covert.” 56:47
1057:34 – 1:00:17Costly Signals, Bridge-Burning, and the Rise of Extreme Politics
The speakers explore theoretical extensions regarding overt messaging as a costly signal of political loyalty that burns bridges with moderates. Smaldino connects these insights back to his modeling and the noticeable surge in extreme political movements around 2015 and 2016.
Sending an overt signal acts as a costly signal of loyalty because it burns bridges with moderates and political opponents.
“if I'm sending an overt signal, I'm sending a costly signal of my loyalty to uh to leftists because I'm I am burning bridges with uh moderates and with right wing people.” 58:12
Host Dave Pietraszewski, David Pinsof, and guest Paul Smaldino discuss how the internet and political shifts embolden individuals to move from covert to overt signaling. They explore perceptions of group formidability, why minority or low-status dissidents take high-risk actions when they have nothing to lose, and examples like punk rockers in East Germany and outlaw groups.
Overtly signaling groups tend to be perceived as more formidable or as a greater political threat.
“And you might expect that groups that are signaling overtly will be perceived as more” 1:02:01
Individuals who fall below a certain threshold and have nothing to lose are incentivized to take high risks.
“Like if they keep going on the way they've always been going, they've got nothing anyway.” 1:05:43
The hosts transition to discussing Paul Smaldino's work on meta-science, specifically the natural selection of bad science with Richard McElreath. They highlight how social systems and selection pressures often produce unintended consequences that diverge entirely from individual intentions, emphasizing the necessity of formal modeling over naive individual-level analyses.
Social dynamics and selection pressures dictate outcomes regardless of individual intentions.
“is to say, look, selection, what you select for” 1:10:03
Interactions within a cultural and social context over time generate collective behaviors that cannot be understood by studying isolated individuals.
“cultural and social context over time produces behaviors that are direct” 1:11:48
Host Dave Pietraszewski, David Pinsof, and guest Paul Smaldino discuss the replication crisis in science, focusing on controversial studies like Daryl Bem's ESP paper, questionable statistical practices, and how social science often ignores cumulative foundational knowledge.
And it all like the the one of the main catalysts, it wasn't the only one, but one of the main catalysts that really kind of launched this investigation was this publication in this top journal in in social psychology by uh um by Daryl Bem.
“one of the main catalysts that really kind of launched this investigation was this publication” 1:13:02
It like it's it's just ludicrously terrible methodology, just the thinking and I got really upset uh for another reason too about it
“it's just ludicrously terrible methodology, just the thinking” 1:14:03
141:17:45 – 1:24:19An Evolutionary Model of Science and Publication Selection Pressures
Paul Smaldino explains why scientists sometimes engage in poor methodology, describing an agent-based evolutionary model that simulates how publication quantity and professional prestige create selection pressures that can disfavor careful, rigorous science.
And so we built a model because the truth is they vary in lots of ways, and they're selected in lots of ways. But we wanted to focus on the quality or rigor of the methodology.
“we wanted to focus on the quality or rigor of the methodology” 1:20:09
And we also allowed for replication. So we said, all right, if somebody else replicates your study and it doesn't replicate, that's bad for you and that hurts your prestige. But otherwise, the more the more papers you publish, the more prestige you have.
“the more papers you publish, the more prestige you have” 1:23:42
Paul Smaldino discusses with the hosts how scientific fields fail to naturally select for better methods and truth over time, drawing parallels to life history theory trade-offs between quality and quantity. They examine how career incentives in academia reward fast, catchy publications over rigorous, slow science, leading to systemic outcomes driven by institutional incentives rather than individual intentions.
you can actually get selection for misunderstanding.
“you can actually get selection for misunderstanding” 1:26:42
the science that gets done overall gets worse.
“the science that gets done overall Um gets worse.” 1:29:56
The speakers explore why appealing to individual conscience and morality fails to fix systemic issues in science, politics, and media. Smaldino and the hosts argue that institutional incentive structures naturally select against people susceptible to moral appeals, meaning lasting change requires altering the actual selection mechanisms rather than just scolding bad actors.
the people who are most susceptible to those appeals to conscience are going to be selected against
“the people who are most susceptible to those appeals to conscience are going to be selected against and under the current incentive incentive structure.” 1:33:32
selection is what matters and not intention.
“selection is what matters and not intention.” 1:32:15
Dave Pietraszewski, David Pinsof, and guest Paul Smaldino discuss how incentive structures shape behavior and institutional norms, applying these ideas to the replication crisis in psychology and the social sciences. They examine how status games, professional stakeholders, and shifting methodological standards influence scientific progress.
the main point of progress I've seen is that you can now get more status for a failed replication attempt
“The the main point of progress I've seen is that you can now get more status for a failed replication attempt than you could have in the old era.” 1:40:32
Paul Smaldino and David Pinsof discuss the role of theory and risk-taking in scientific research. They explore the balance between pursuing counterintuitive hypotheses and building on established knowledge, viewing science through the lens of collective problem-solving, variation, and evolutionary dynamics.
evolution works on two things, right? Selection and variance.
“evolution works on two things, right? Selection and variance.” 1:46:11
the speed of evolution is proportional to the variance in the population
“Is that the speed of evolution is proportional to the variance in the population.” 1:46:22
Dave Pietraszewski, David Pinsof, and guest Paul Smaldino discuss how academic publishing norms and journal requirements often select against good, purely theoretical work by demanding immediate empirical data, which can inadvertently promote superficial theories.
asking for data concurrent with theory selects for bad theory
“asking for data concurrent with theory selects for bad theory” 1:50:12
The hosts and Paul Smaldino explore his scientific identity, the role of provocative modeling stances like using extremely unrealistic models, and how expressing unconventional views helps researchers feel less isolated in academia.
Learning About Social Evolution with Extremely Unrealistic Models
“Learning About Social Evolution with Extremely Unrealistic Models” 1:53:13
models are stupid, we should have more of them
“models are stupid, we should have more of them” 1:57:42
Hosts Dave Pietraszewski and David Pinsof engage in lighthearted banter with guest Paul Smaldino about music, a Lego technic sun and moon model, and corporate sponsorships, before Paul shares a final, humorous closing word.
And we were talking about it earlier. I think it's really neat.
“And we were talking about it earlier. I think it's really neat.” 2:01:17
Dave Pietraszewski and David Pinsof wrap up the episode, expressing their enjoyment of the conversation with Paul Smaldino, encouraging listeners to spread the word, leave reviews, and support the show.
All, and if you did enjoy it, please help us by spreading the word.
“if you did enjoy it, please help us by spreading the word” 2:02:49