Maybe you’ve been fascinated by Taylor Swift’s nuptials at Madison Square Garden and hoping all month for some good photo leaks. Or perhaps you’ve been following Lionel Messi’s record-setting World Cup run and eagerly seeking details about the man behind the magic. Or maybe you’ve simply wondered why *other* people care so much about celebrities and their private lives. This month’s Q&A is with Princeton psychologist Anuj Shah on the science of so-called parasocial relationships — the one-sided “relationships” we form with celebrities who, naturally, don’t know us at all. We’ll dig into the misunderstandings, and somewhat surprising opportunities that these asymmetrical relationships create.
But first, a little good news: Choiceology will be back with fresh episodes and a brand new season on August 24th! We’ll have an amazing lineup of stories and guests to share very soon; I can’t wait for you to hear what we’ve created. In the meantime, here are some recommended listens and reads…
Recommended Listens and Reads
If you’re obsessing about cyclospora (who wants explosive diarrhea!? Ick!), then it might be a good time to revisit a classic episode of Choiceology on understanding the way we over-weight the risks of rare events with UCLA Prof. Craig Fox. You could also cut to the chase and read an excerpt from my interview with Craig in this back-edition of Milkman Delivers.
The hidden costs of remote work deserve attention: In this thoughtful New York Times op-ed, a team of careful economists highlights social and emotional costs of the remote work revolution that you may not have considered.
Two terrific new books penned by friends in the behavioral science community have come out since my last newsletter. In HUMAN RAISED: Nurturing Connection, Curiosity & Lifelong Learning in the Age of AI, Dana Suskind offers thoughtful guidance for caregivers on using AI-powered technologies to support their children’s development without sacrificing human connection. And in LEAVE THE LIGHTS ON: How Joyful Decisions Can Save Our Species, Elizabeth Dunn and Jiaying Zhao explain how you can make a positive impact on climate change by taking steps that bring you joy.
Q&A: Understanding Parasocial Relationships
This month’s Q&A from Choiceology is with Princeton psychology Professor Anuj Shah, who shares insights from his research on parasocial relationships and why our minds struggle to appreciate the asymmetry inherent in our connections to celebrities.
Me: One of the most interesting things about our “relationships” with celebrities or “parasocial relationships” is how confused we are by the peculiar fact that we can know so much about a person who doesn’t know us at all. Can you describe how this biases our judgment?
Anuj: We generally expect there to be some symmetry or reciprocity in our relationships. So, if I see you as a friend, then there’s a good chance that you see me as a friend as well. If I see somebody as a stranger, they probably see me the same way. But that’s not always true, right? There are times when I know more about somebody than they know about me, but, in general, people might see symmetry where it doesn’t exist.
So, when I know a lot about a celebrity, I might feel like, “Oh, they must have some knowledge about or familiarity with me as well,” when in fact there’s a deep asymmetry there.
Me: I have to admit, as a professor with 150 students, I always find it a little scary how my students expect me to know things about them just because they’ve gotten to know me a bit from listening to my lectures all semester. Of course, you have the same experience.
Anuj: Absolutely.
Me: So, it happens with real celebrities and not real celebrities. Why do you think knowing things about someone makes us feel like they know us too? You talked about the fact that we expect symmetry, but where is that expectation coming from?
Anuj: Well, I think there’s two things. First, there generally is reciprocity or symmetry in our relationships, so it’s usually true. But I also think that when we’re trying to get inside the minds of others or think about how other people see things, we usually start with ourselves and adjust away from that.
So, it’s both generalizing from past experience and just focusing on ourselves and our own sense of familiarity when trying to get a sense of what other people know about us.
Me: Could you describe some of your favorite research on this topic?
Anuj: We have online experiments where we tell participants that they’re going to be interacting with another online participant, but they’re otherwise going to be anonymous. So, they think they’ll be interacting with a stranger, but there’s a bit of deception here in that there is no other person. Instead, we simulate the “partner’s” responses, so we have total control over what participants are seeing. And in all of these studies, we ask participants to answer a few basic icebreaker questions. For example, “what is your employment status?” “What kind of area do you live in?” “What is your marital or family status?” And there’s multiple choice responses.
And then we ask, “If you were to meet this person, how well do you feel like they would know you?” Another way we do it is by asking participants to write down four things that are true and one lie about themselves. And what we find is that when participants have more information about their partners, they think their partners are more likely to know which statement is a lie.
Me: That’s fascinating. Those experiments are so clever. But I also love the field study you ran that shows this phenomenon can really matter in important policy contexts.
Anuj: Yeah, we did that experiment in partnership with the New York Police Department. Before we had done any of the lab experiments, we were doing qualitative interviews with folks who had been previously incarcerated. And one of the questions we asked was, “There were probably times when you thought about going forward with a crime or offense and you stopped. What might have made you stop?”
It wasn’t the most common response, but a few times people talked about knowing the officers in their neighborhood. One kid, for example, talked about an officer that everybody in the neighborhood knew, nicknamed Birdman. And they said, “Whenever Birdman was around that would get us to step back or chill.” It’s interesting because he wasn’t saying that he was worried about how this officer would see him, he was saying “Oh, because I know this officer, he must be paying more attention to me.”
From there, we started wondering if understanding this psychological phenomenon could change how we do community policing. Part of the goal of community policing is for people to know their officers better and for officers to know the community better. But officers often want some level of anonymity because they worry that anything people know about them could come back to hurt them.
In our intervention, we worked with neighborhood coordination officers (NCOs) and residents in public housing developments throughout New York City. We either randomized housing developments into a condition where it was business as usual, or we sent mailers that included three facts about their NCOs, like what their favorite sports team is, how long they’ve been an officer, or why they wanted to become an officer.
We found that residents who had gotten the mailers were more likely to say, “If I did something illegal, I think this officer would be more likely to know about it.” We found modest, short-lived crime reduction of 5-7% for about three months after the intervention. Those effects fade, but it was promising evidence that when residents learn more about their neighborhood officers, they think the officers are more attuned to their actions.
Me: It’s really a remarkable study. I know the effects are a little wobbly, but even the hint that this can matter in such an important setting is exciting. Do you worry that this bias could lead us to make any mistakes?
Anuj: Sure. For example, if you become familiar with your doctor, you might assume that they would be more familiar with anything that’s going on in your life, any changes in your health or ailments. And when you have a short amount of time with your physician, you might leave some things unsaid, when in fact it would be helpful for your doctor to know. So, I worry about the potential for under-sharing. When you know something about somebody else, you might think that they already know the things that are actually just in your mind.
Me: That’s a great point. How are you applying what you’ve learned from your research to your own life?
Anuj: Well, like you mentioned, I’m more aware of it when I teach. The fact that I’m sharing stories with my students and they’re going to have a level of familiarity with me that I don’t have with them. So, I’m more intentional when I have interactions with my students to make sure I’m asking them about themselves. That way, we try to balance some of the asymmetry. I’m also more aware when I know more about others than they know about me because they’re more public than I am, or because I spent time stalking them on social media. And I try to share more about myself if I’m talking to them.
Me: Thank you, Anuj. I really appreciate you taking the time to share these insights.
This interview has been edited for clarity and length.
To learn more about Anuj’s work, listen to the episode of Choiceology where we dig into his terrific research on the psychology of parasocial relationships.
Wishing you a happy end of July. I’ll be taking August off to relax and recharge, but you can look for a fresh new edition of Milkman Delivers in your inbox come September.
Katy Milkman, PhD
Professor at Wharton, Host of Choiceology, an original podcast from Charles Schwab, and International Bestselling Author of How to Change
P.S. Join my community of over 100,000 followers on social media, where I share ideas, research, and more: LinkedIn / X / Instagram / BlueSky / Threads
Small nitpick on the NYC police study: "5-7% crime reduction" sounded interesting (I was going to tweet about the study haha) but it felt too good to be true, so I looked into it and found this article: https://datacolada.org/101
Apparently the statistically significant reduction (6.9%) is for crime within 250 ft during the first 3 months (p=.036). Crime within 65 ft (the preregistered distance!) gave p>.2 for all 9 duration choices (see Fig 2 in the linked article).
The paper tested 81 distance-duration combinations (65 ft to 750 ft, 1 to 9 months), and 16 were statistically significant, which appears promising; however, the linked article found that if the p=.036 result were a false positive, there would be a 43% chance of getting at least 16 statistically significant results across the 81 combinations.
So I think the evidence for a 5-7% crime reduction from such an intervention appears stronger than it actually is. Interesting article, though-- thanks for sharing!
Fascinating topic. Parasocial relationships may be one-sided, but their influence is not. The people we follow can shape what we notice, value, buy, and even how we see ourselves, often without us realizing it. Understanding that influence is increasingly important in a world where attention is constantly in demand.