What Your Online Self Reveals About You

Summary

Hidden Brain: What Your Online Self Reveals About You Hosted by Shankar Vedantam | Released on December 16, 2024

---

Introduction

In the episode titled "What Your Online Self Reveals About You," Shankar Vedantam delves into the intricate relationship between our digital footprints and our true selves. Drawing on insights from Columbia University psychologist Sandra Matz, the discussion explores how the data we inadvertently generate online can offer profound insights into our personalities, behaviors, and even predictions about our future actions.

Knowing Ourselves: The Illusion of Self-Knowledge

[00:00 - 04:59]

Shankar begins by challenging the common belief that we know ourselves well, highlighting the biases and self-deceptions that cloud our self-perception. He references studies showing that individuals often overestimate traits such as intelligence and ethics. This segment sets the stage for understanding how our actions, rather than our self-reported preferences, provide a more accurate depiction of who we are.

Shankar Vedantam: "How well do you know yourself? Chances are you'll tell me you know yourself very well."

From Village Surveillance to Digital Footprints

[05:01 - 18:54]

Sandra Matz shares her upbringing in a small German village where community surveillance was the norm. This experience parallels today's digital surveillance, where our online activities are monitored by anonymous entities rather than familiar neighbors. Matz explains how the subtle traces we leave online—similar to behavioral cues in a physical space—can be pieced together to form a comprehensive profile of an individual.

Sandra Matz: "If you look at someone's bedroom, you can combine intentional identity claims with unintentional behavioral residues to get a pretty good sense of who the person is."

Predicting Personality Through Digital Data

[19:10 - 25:01]

Matz discusses groundbreaking research demonstrating the ability to predict personal attributes, such as income, based solely on Facebook activity. Her studies reveal that while high-income individuals frequently discuss luxury items and future plans, lower-income individuals tend to focus more on the present and themselves. This distinction underscores how nuanced digital behavior can reflect deeper psychological states.

Sandra Matz: "Lower income people talk more about themselves and the present, possibly because they are focused on making the present work."

The Power of Minimal Data: Facebook Likes

[23:39 - 25:40]

Referencing Yoyo Wu's study, Matz illustrates how as few as ten Facebook likes can allow algorithms to predict a user's personality more accurately than their coworkers. With increasing data points, the accuracy surpasses that of friends and family, showcasing the formidable power of machine learning in understanding human behavior.

Shankar Vedantam: "After 10 likes, the model could judge personality better than work colleagues. After 120 likes, better than family members."

Behavioral Residues: The Clues We Leave Behind

[26:00 - 32:24]

The conversation shifts to "behavioral residues"—the subtle, often unconscious traces of our actions online. Matz explains how integrating various data sources, such as credit card transactions and smartphone GPS data, can construct a detailed and accurate profile of an individual. This comprehensive data amalgamation allows for insights that are both intentional and unintentional, revealing aspects of our lives we might not openly share.

Sandra Matz: "Behavioral residues are all of the traces that we inadvertently leave as we go about our life, like your GPS records or your credit card purchases."

Digital Footprints and Societal Insights

[32:24 - 38:54]

Matz highlights studies linking online behavior with broader societal trends. For instance, regions with higher rates of racist online searches correlated with lower support for Barack Obama during elections. Additionally, negative emotions expressed on social media were found to predict a rise in populist voting patterns. These findings demonstrate how individual digital behaviors collectively mirror and influence societal dynamics.

Shankar Vedantam: "Negative affect on social media is linked to higher likelihood of voting for populist candidates."

Enhancing Lives Through Data: Saving Money and Mental Health

[36:38 - 42:13]

Transitioning from observation to intervention, Matz discusses how tailored digital messaging can encourage behaviors like saving money. Collaborating with the fintech company Save a Life, her team crafted personalized messages based on personality traits, resulting in a 60% improvement over existing methods in encouraging low-income individuals to save.

Furthermore, Matz explores the potential of using location data to monitor mental health. By analyzing GPS records, it's possible to identify changes in behavior indicative of depression, offering a proactive approach to mental health support.

Sandra Matz: "Targeted messages tailored to personality traits were 60% more effective in encouraging savings than previous methods."

Predicting and Preventing Dropouts in Education

[42:26 - 45:36]

Matz shares her work on predicting college student dropouts through digital footprints. By analyzing communication patterns and social interactions, her models could identify students at risk of leaving university. This predictive capability allows for personalized interventions, addressing the specific challenges faced by each student, thereby increasing retention rates.

Sandra Matz: "By understanding why a student is predicted to drop out, we can tailor interventions to address their specific needs."

Addressing Polarization with Digital Tools

[45:46 - 52:45]

Shifting focus to societal polarization, Matz proposes innovative uses of digital data to bridge ideological divides. She envisions an "Explorer Mode" where individuals can experience the digital echo chambers of others, fostering empathy and understanding. Although acknowledging the comfort people find in their own information bubbles, Matz emphasizes the potential for these tools to provide broader perspectives.

Sandra Matz: "An Explorer Mode could allow us to step into someone else's echo chamber, helping to broaden our worldviews."

Intervening with Psychological Targeting

[52:45 - 54:27]

Matz touches upon the ethical considerations of digital surveillance, advocating for responsible use of data. She stresses that while digital tools hold immense potential for positive interventions—such as mental health support and educational retention—they must be implemented thoughtfully to avoid misuse and ensure they genuinely benefit individuals.

Sandra Matz: "These tools can be incredibly helpful if used to support people in meaningful ways, like mental health or education."

Conclusion

[54:27 - End]

Shankar wraps up the episode by emphasizing the dual nature of digital footprints—how they can both reveal deep truths about us and be harnessed to improve our lives. From financial savings to mental health interventions and educational support, the episode underscores the transformative potential of understanding and leveraging our online behaviors.

---

Notable Quotes

• Shankar Vedantam: "How well do you know yourself? Chances are you'll tell me you know yourself very well." [00:00]

• Sandra Matz: "Behavioral residues are all of the traces that we inadvertently leave as we go about our life." [22:49]

• Shankar Vedantam: "Negative affect on social media is linked to higher likelihood of voting for populist candidates." [31:47]

• Sandra Matz: "These tools can be incredibly helpful if used to support people in meaningful ways." [53:05]

Key Takeaways

1. Digital Footprints as Psychological Indicators: Our online activities offer a window into our personalities, preferences, and emotional states, often more accurately than self-reported data.

2. Predictive Power of Minimal Data: Even limited data points, such as Facebook likes, can enable algorithms to predict personal attributes with surprising accuracy.

3. Behavioral Residues Reveal Unseen Aspects: The subtle traces we leave online unconsciously reflect broader aspects of our lives, including socioeconomic status and mental health.

4. Potential for Positive Interventions: By harnessing digital data responsibly, it's possible to create personalized interventions that support financial well-being, mental health, and educational retention.

5. Ethical Considerations and Responsible Use: While digital surveillance holds promise, it necessitates ethical implementation to ensure it serves beneficial purposes without infringing on privacy or fostering further polarization.

Final Thoughts

"What Your Online Self Reveals About You" offers a compelling exploration of the intersection between our digital behaviors and our true selves. Through Sandra Matz's research, listeners gain insight into how the data we generate daily can not only predict but also enhance our lives when used thoughtfully. This episode serves as a crucial reminder of the profound impact our online actions have on our personal and societal landscapes.

Transcript

[Shankar Vedantam]

This is Hidden Brain. I'm Shankar Vedantam. I have a question for you. How well do you know yourself? Chances are you'll tell me you know yourself very well. All of us like to believe this. We feel like we know ourselves better than anyone else does. Every day we make choices based on this knowledge we have of ourselves. We decide how to spend our money, who to vote for, where to go for dinner based on what we know of our predilections and preferences. But our knowledge of ourselves is not always accurate. A host of biases and self deceptions keep us from seeing ourselves clearly. When you ask people how smart they are or how ethical they are, or how good looking they are, for example, majorities say they are above average, which of course is mathematically impossible. But it isn't just about vanity. How many times have you gone to a restaurant you've been to before and ordered the same dish you ordered last time, only to remember after you started eating it that you didn't like it the last time? Or think about your last romantic entanglement that ended in disaster by the time it ended? Did you wonder how your past self could have gotten involved with someone so unsuitable? Over the last few decades, researchers in a variety of disciplines have discovered there is a much better way to understand people than to ask them questions. When you ask people what books they like to read, people will tell you about the novels and biographies they think they ought to like. If you ask them what movies they want to watch, they will tell you about the movies they aspirationally want to watch. But if instead you look at the books that people actually read or the movies they actually watch, it usually paints a different picture of their preferences. This week on Hidden Brain. How understanding what we do instead of listening to what we say can help us make better financial choices, improve our physical and mental health, and maybe even bridge our political divides. Support for Hidden Brain comes from jobs Ohio in partnership with the State of Ohio what would make someone move from California to Ohio? Maybe it's the comfortable, convenient lifestyle or the low cost of living. Or maybe it's the job, that next big opportunity you've been wanting for a while now. If you're a tech pro considering your next move, Ohio deserves to be on your list. See More [e-mail] Support for Hidden Brain comes from Discover. This time of year, you feel automatically reinvigorated with the holiday spirit, moods are boosted and that becomes infectious. Here's another automatic mood booster for you. Discover will automatically double all the cash back you've earned on your credit card at the end of your first year. Get rewarded no matter who you are or how much you spend with Discover. See [e-mail] CreditCard support for HiddenPrint comes from Lingo. You know how sometimes a meal leaves you energized and other times sluggish, foggy or even hungrier? That may be your glucose talking. Foods that are sugar and carb heavy cause your body's glucose levels to spike. Lingo by Abbott tracks your glucose 247 so you can see how your body responds to food. And over time, Lingo helps you learn to develop habits to steady your glucose which can support your everyday well being, your metabolism and your long term health. Lingo starts at $49 for a two week plan. For a limited time you can save 10% on your first order with code hidden@hello lingo.com the Lingo Glucose System is for users 18 years and older not on insulin. It is not intended for diagnosis of diseases, including diabetes. For more information, please visit hello lingo.com philosophers tell us the highest wisdom is to know ourselves. They say this precisely because knowing ourselves is difficult, not easy. It requires self reflection, self awareness, and a healthy dose of humility. At Columbia University, psychologist Sandra Matz studies how one aspect of our behavior can reveal surprising truths about who we are. Sandra Matz, welcome to Hidden Brain.

[Sandra Matz]

Thank you so much for having me.

[Shankar Vedantam]

Sandra, you grew up in a small village in Germany which had two restaurants and no shops. Can you paint me a picture of the place where you grew up?

[Sandra Matz]

Yes, happily. It's a village of 500 people in the very southwest corner of Germany. As you said, there's two restaurants, no shops, one church. I should say that's very important to the people living there. And it was really like a small community.

[Shankar Vedantam]

So one day I understand that your doorbell rang and it was a neighbor reporting a missing rabbit.

[Sandra Matz]

Yeah, I must have been, I think, eight, nine years old, and I had this pet rabbit or bunny called Schnuffel, and he was living outside. We had built him this house outside, my dad and I, and one day the neighbor comes and says that they found him in their garden feasting on their vegetables and salad. And so they tried to catch him already, but unsuccessfully. And so they were trying to get more manpower. So now my entire family is up. Their entire family is up. We're trying to get him. And I don't know if you've ever had a bunny or let alone try to catch one. They're really fast so they zigzag around, and it's almost impossible to catch them. So we must have looked like clowns running around. And that certainly didn't go unnoticed. So very soon into our hunt, I think the entire street really was involved. So we had. Someone was managing the traffic because the bunny would just kind of run from one side of the street to the next. And then we had, like. It really felt like a command center. So people were strategizing about whether we should set up a trap for the rabbit or whether we should just try to lure him in with a treat. Eventually we caught him, but it was certainly an adventure for the entire neighborhood.

[Shankar Vedantam]

How was the rabbit eventually recaptured? Was it a dramatic moment?

[Sandra Matz]

It was a dramatic moment. So it was actually one of my neighbors who leaped and caught him on the back leg. And I just remember the rabbit screaming. I didn't even know that rabbits could scream that loudly. And then all the kids were crying because the rabbit was screaming. And so I think the adults were just happy that we got him. But it was certainly a dramatic capture.

[Shankar Vedantam]

I'm getting a sense that this was a village where everyone knew everyone's business.

[Sandra Matz]

Very much so. Very much so.

[Shankar Vedantam]

So when you were 15, Sandra, you loved riding around the village with your boyfriend on a motorcycle. What was this bike like? And where would you go?

[Sandra Matz]

Yeah, so it was. I still remember it. It was like a red Suzuki Bandit. And it was. I thought it was beautiful. I was 15, he was a bit older. And we would just take it from one village to the next through the. Through the hills and up the up and down the serpentines. And I loved riding it, but I was usually in the back. So at some point, I think I got really tired of being in the back. And I knew that I would have to wait for another three years, because in Germany, you get your license at 18 at the earliest. So I kind of tried to sneakily convince him to let me just try. I found this abandoned airfield, and I just told him, let me just kind of ride for a few meters. You're going to get the bike back. It's all going to be good. He was a bit skeptical at first, but then agreed to let me try. So we start, and I don't know exactly what happened. I think we must have kind of moved to the grass. And I was trying to pull the bike back, but suddenly I think I just turn on the gas, and the bike rises. My boyfriend falls off the back, and I just speed away. So I have no idea what I'm doing now. Suddenly, I'm alone on the bike without any sense of, like, how to handle it. I'm kind of going left and right and left and right, and at some point, I essentially crash on the side. Luckily, it was still going. Going slowly. But there was my first experience riding the bike myself.

[Shankar Vedantam]

Sandra and her boyfriend weren't hurt, but Sandra had to spend a year's worth of tutoring money to get the bike repaired.

[Sandra Matz]

But in a way, that wasn't the worst part for me. The worst part was that really, I would say the minute we dropped the bike at the shop, everybody knew. So people knew about me asking my boyfriend to drive the bike. Me crashing the bike, which was even worse otherwise, could have potentially been a cool story, but certainly wasn't. So everybody knew what had happened, and just I was punished for weeks after with people asking me about it.

[Shankar Vedantam]

When you say you were punished, how so what. What was the reaction of your neighbors and friends?

[Sandra Matz]

Very. I mean, very different. So some neighbors actually just went back to their own childhood, and they were like, oh, that's like, such a brave and fun thing to do, and they just recounted their own childhood offenses. Others were like, how could you ever do this? We thought that you were actually one of the good kids on the block. So. But it was just one conversation after the other that was all about me crashing the bike.

[Shankar Vedantam]

Was there anything good that came from all the surveillance, Sandra?

[Sandra Matz]

Certainly. I mean, I think that maybe not the bike surveillance. I think this one was a hard one. But generally speaking, the fact that my neighbors knew everything about my business also meant that there was a community that I felt safe in. Right. So it was a community of people who knew me, who tried to help when I was looking for advice. And I've never quite experienced anything like it ever since.

[Shankar Vedantam]

Was there a time when you, in fact, got very useful advice from these people because, in fact, they knew you quite well?

[Sandra Matz]

I think so. I mean, some of them were just trying to interfere with my life, but some of the advice I got was incredibly helpful. So one of the ones that I still remember is that when I was finishing high school, I was thinking about doing a gap year. It sounded like a dream to me. I was like, well, you can travel the world. You can take a year off. Don't rush into university. But I remember being quite torn because not many of my friends were considering it. And I was very ambitious, I would say, at the time. So I was like, well, maybe it's just a waste of my time to spend a year traveling. I could start university, get a job, and luckily, a lot of my neighbors told me like, look, you seem like someone who's been always craving to leave this village to see some, some of the world. Why don't you do it? Like you can work for many, many years after you should really consider it. They helped me even find a job to scrape together the money. So I think that was one of the times that I felt very much supported.

[Shankar Vedantam]

And in some ways it sounds like they knew you almost better than you knew yourself.

[Sandra Matz]

In some ways I would say so because I think it was a slightly less biased and maybe self critical version of myself.

[Shankar Vedantam]

Sandra's experience with the nosy neighbours in her village is what life has been like for most humans. Through most of human history, we've typically lived in small groups and people in those groups have known everything there is to know about us. Today, many of us live in a different kind of village. It's a global village where anonymous entities rather than our actual neighbors have eyes on us. Not all of them have our best interests at heart. When we come back, what our digital footprints reveal about us and how this information can be used both to help us and to harm us. You're listening to Hidden Brain. I'm Shankar Vedantam. This podcast is supported by the Sierra Club, a nationwide force fighting for the planet for over 130 years. Under a second Trump administration, there will be unprecedented environmental challenges. Pristine landscapes will be handed over to fossil fuel giants. Clean energy jobs will be under attack. But Sierra Club's grassroots organizers and legal experts have the power to fight back. You can power Sierra Club's work to protect our lands and communities. Visit Sierra Club.org podcast to donate. Today the most innovative companies are going further with T Mobile for Business like Delta Air Lines. Together with T Mobile for Business, Delta Air lines is putting 5G into the hands of ground staff so they can better assist on the go travelers with real time information. From the Delta Sky Club to the Jetbridge, this is elevating customer experience and Tractor supply trusts 5G solutions from T Mobile for business to connect over 2,200 stores around the country with 5G business Internet plus T Mobile is powering AI solutions so tractor Supply team members can match shoppers with the products they need faster. This is enriching customer experience. Take your business [e-mail] now. This is hidden Brain. I'm Shankar Vedantam. Every day as we go about our lives, we reveal aspects of ourselves to the world. If you visit a local bakery a lot, it's probably because you like pastries and baked Goods. If you spend time in parks, it's because you value nature and recreation. Someone who rarely ventures outside their home except to go to work might be introverted. At Columbia University, computational social scientist Sandra Matz studies how the things we say and do reveal things about our thoughts, preferences, and personalities. Sandra, I want to talk about the clues we unintentionally leave behind us as we go about our lives. Let's start in the physical world. Many years ago, you were on a date and things were going well, and you ended up at your date's apartment. Tell me what you did as soon as you got to his place.

[Sandra Matz]

Yeah, that's a funny story because I remember entering the apartment and, you know, you meet someone for the first time, you have no idea who they are. And I entered the apartment and it's pristine. It's. First of all, it has this huge library, which I loved. Had books in Hebrew and English and French. So I was like, oh, man, this is a bookworm. I love it already walked to the kitchen. It's sparkling clean, which I cannot say of my own kitchen. So I was highly impressed in the sense that everything had its place. The knives were perfectly organized. I got some glasses for us and they were perfect. Like, no marks, any watermarks anywhere. And I just kind of started building this image of who the person living in this apartment, the guy I was dating at the time, who he was, and it just felt like he was this curious, book loving person with almost like an OCD sense of order.

[Shankar Vedantam]

So you weren't wearing a hat and carrying around a magnifying glass, but it feels vaguely Sherlock Holmesian to me. What you were doing in that apartment.

[Sandra Matz]

It does feel like Sherlock Holmes. And I think we do this all the time, right? We meet someone new, we look for clues of who that person is. Could be their apartment, could be what they're wearing, could be what they're saying. It's really. We're kind of trying to piece the puzzle pieces together in a way.

[Shankar Vedantam]

How did things go with this date?

[Sandra Matz]

Well, it worked out really nicely. He's now my husband and we have a. Have a 10 month old.

[Shankar Vedantam]

I'm wondering whether your impressions of your husband, your first impressions of him when you were dating, did they turn out to be accurate, Sandra?

[Sandra Matz]

Very accurate. So I think he. I would say he's probably the most curious person I know. He loves podcasts, loves reading, loves to learn everything about the world. And he also, I have to say, is a little bit OCD in a good way.

[Shankar Vedantam]

So the psychologist Sam Gosling has shown that people, in fact, are remarkably accurate at judging the personality of strangers when given the chance to snoop around their offices or bedrooms. Tell me about this work, Sandra.

[Sandra Matz]

Yeah, so this is actually work that inspired all of my research in the digital space. So Sam Gosling was one of the first people to try and figure out, like, how good are strangers at judging our personality if they just take a look at our bedrooms and our offices? And he distinguished between these two types of cues that you can find. So he said, well, some of the cues that you find in someone's office or bedroom are the intentional identity claims that we put there. Right. So we put up a poster of Lady Gaga because that's the signal that we want to send to the world of, well, we're into music, and this is the type of music that we like. But then there's also all of these other cues that we don't really think about. Right. So the socks are disorganized. The bed isn't made. It's the opposite of my husband. It's just, like, probably a little bit more disorganized. So what Sam Gosling really showed is that if you combine all of these things, you get a pretty good sense of who the person living in these places is.

[Shankar Vedantam]

So you say there are parallels between what happened in your village or your behavior when you visited your date's place and what happens to us online. It's as if your village neighbors now have access to your Facebook messages and credit card purchases.

[Sandra Matz]

Yeah. So in a way. Right. So I could take a look at your office or your bedroom, or I could see what my neighbors are doing. But on some level, I think we all now live in this. What I think of as a. As a digital village. And so we all leave these traces, these digital traces all the time that could be anything from the stuff that you post on social media. So, again, relatively explicit identity claims to this. The. The data that is captured by your smartphone. So GPS records, where do you go, your credit card, what do you buy? In the same way that we could put the pieces together from someone's bedroom, we can also do that in someone's digital space.

[Shankar Vedantam]

You say it takes shockingly little information to get an extremely granular picture about people, Even in a big town like New York City. Now, there are millions of transactions that take place every day in New York. Finding any one person might seem like you're looking for a needle in a haystack.

[Sandra Matz]

Yeah, it's actually one of my favorite studies that was coming out of mit. And what they showed is that it's very easy to identify someone based on your spending records or your GPS records. So you can imagine, as you said, like, there's millions of people in New York and even if we, say, got access to all of their credit card spending, anonymize it. So we don't have names, we don't have any personal identifiers, it's very easy to reverse engineer data. You can imagine that let's say you go and get a Matcha latte at Starbucks on 72nd street in New York at 7:20am Then you have lunch in a certain place and maybe you take a cab downtown at night. There's at some point only so many people who have exactly that same signature. So you can almost think of it as a fingerprint that is made up of your data.

[Shankar Vedantam]

I want to talk about some of the ways you and others have found that our digital footprints can reveal deep truths about our lives. In 2019, you ran a study that predicted people's income based on an extremely unlikely source. Tell me what you found, Sandra.

[Sandra Matz]

Yeah, so that's, in a way, the most interesting part of this entire field of research is like, yeah, we can identify you as a person. We can know that it's Shankar based on your data. But for me, the more interesting part is actually that we can dive into your psychology so we can take a look at what's going on inside your mind. And so the study that we did when we tried to predict someone's income was essentially relying on their Facebook data. So what is it that people talk about and post on social media? And I think there were some really interesting, sometimes quite uncomfortable truth that we discovered. But overall, the bottom line was that just by looking at what you talk about on Facebook, we can have a pretty good sense of your socioeconomic status.

[Shankar Vedantam]

I'm puzzled by how that would be the case. I mean, what does my posting about a movie that I've watched or a vacation that I've taken? How do you tell income is based on those postings, Sandra?

[Sandra Matz]

Yeah, so when you, when you start opening the black box, what you see is some of them are like, some of the cues are relatively obvious. So you can imagine that people with a lot of money, they talk about the vacations that they're going to take. They talk about expensive luxury brands a lot more often than people who are struggling to make ends meet. But there's also these more subtle cues that I found even more interesting, which is, for example, that lower income people, people. And they talk more about themselves. And they talk more about the present than higher income people. And in the beginning you might be wondering, like, why might this be the case? And I think it's just that it's really damn hard to think of anything else other than how you make the present work if you're struggling to make enough money to put food on the table. So those are all these little, I think, like secrets about what's going on inside our mind that we can uncover in the data.

[Shankar Vedantam]

What's fascinating about that, of course, is that most of us are not thinking, are my posts describing something that's happening in the present or something that is about the future, for example. But that difference in fact can reveal something about us.

[Sandra Matz]

Yeah, And I think that's the distinction between identity claims and behavioral residue that I think is so interesting. Right. So again, you might post about this luxury vacation and it's a very clear signal to the world that you're having a great time and you can afford going on this vacation. But then all of these more subtle ones where you talk about yourself, you're more focused on the present. That's certainly something that we don't necessarily intend to reveal.

[Shankar Vedantam]

You used an interesting phrase just now, behavioral residue. What do you mean by that, Sandra?

[Sandra Matz]

Yeah, so behavioral residue are all of the traces that we essentially inadvertently leave as we go about our life. In the offline context, you can imagine, again, that's like the bin overflowing. That's your socks not being organized, that's the bed not being made. And in the digital world, it's all of the traces that we generate without really thinking about it. So that could be your smartphone, for example, captures your GPS records pretty much continuously 24 7. And you're not intentionally sitting down to create a record of where you went and what you did there. But still those traces exist.

[Shankar Vedantam]

Let's look at some of the ways in which these behavioral residues can tell us important things about our lives and the lives of other people. The researcher Yoyo Wu once looked at what you could learn about a person from their Facebook likes.

[Sandra Matz]

Yeah, that was really so the research by Yo Yu Wu, I would say, was one of the pivotal studies in this field because it showed just how accurate the predictions that we can make about someone's psychology really are based on relatively little data. So she was studying the Facebook pages that people followed. So let's say CNN has a Facebook page. You can like it. And what she showed is that just by looking at your Facebook pa, an algorithm can actually predict your personality more accurately than our Coworkers could, than our friends could, than our family members could. And mind you, those are people who know you pretty well. Right? Those are your parents, those are your siblings, those are your kids. They've spent a substantial amount of time with you. And it was slightly inferior to the judgments and the predictions of your significant other. Now, this was a study that was done in 2015. It was only based on Facebook, like, so you could imagine that if we get access to all of your digital traces and apply slightly more sophisticated machine learning, that we could probably outperform even your significant other.

[Shankar Vedantam]

So the study found that after observing just 10 likes from someone's Facebook profile, the model was able to judge a user's personality better than their work colleagues. After 65 likes, it knew users better than someone's friends. And after 120 likes, better than family members. I mean, that's astonishing, Sandra.

[Sandra Matz]

Yeah, And I think what, what is astonishing to me, and I think a point that is important. Those models aren't perfect. Right. So I think any prediction always has a certain amount of error. And what we're talking about are averages on averages. These models are really accurate, as you just said, with a, with a comparison, however, we still make mistakes and at the individual level. So one of the things when we kind of make these comparisons and predictions that I want to highlight is that don't take it as a truth. Right. It's a prediction, it's a probability. It's pretty damn accurate on average. But we're still going to make mistakes at the individual level.

[Shankar Vedantam]

I'm also assuming that when you have intersecting lines of evidence. So this study was looking at Facebook likes. But if you were able to combine that, for example, with people's credit card purchases, if you were able to combine that with their Twitter feeds, if you're able to combine that with what they're saying about themselves, you're gradually producing a more and more accurate profile of who the person is.

[Sandra Matz]

Yeah, I think of it as like this puzzle that we're putting together of a person. So you get a piece here that's their social media, and then you get another piece that's their credit card spending and another piece that's their smartphone sensing data. And gradually you kind of see this person behind the data emerge. And what I think is fascinating about this, combining data sources is essentially that a lot of people always say when I talk about social media that, well, isn't it just like this curated identity of who we are? It's just like who we want to be. We all like Machalates and amazing vacations. We're never sad. So it's just like the self idealized version of who we really are. That's true for some of these identity claims. Right. Social media. But if you wanted to, let's say you wanted to pretend that you're more organized and conscientious than you do really are, maybe you can do this on Facebook for a couple of weeks. It's really, really difficult to do this across all data sources and across like months and months and months.

[Shankar Vedantam]

I'm wondering if you can talk a moment about how these sort of, in some ways mindless algorithms are painting a picture of us that's more accurate than our friends and neighbors and co workers. And whether some of that is because our friends and neighbors and co workers are bringing, you know, their own perceptions and their own biases to the equation as they're evaluating us.

[Sandra Matz]

Yeah. So I think part of it might be bias. Right. One of the things that we're limited by as humans is we have only a sliver of experience and we have our own perspective on the world and that's influencing every judgment that we make about other people. Now we also have a lot less data to work with. Right. If you look at the predictive models that we build, they are looking at millions and millions and millions of data points and all integrating them at the same time. There's just no way that we have access to millions of millions of friendships that allow us to then judge someone's personality based on their behavior.

[Shankar Vedantam]

Yeah. I mean, in some ways this is like Sherlock Holmes on steroids is what these machines are doing. Right. Because they're actually picking up huge amounts of data, far more than most of us are actually able to observe in the physical world.

[Sandra Matz]

Yeah. Let's imagine it's like Sherlock Holmes with a million Watsons.

[Shankar Vedantam]

So even our search history, what we're looking for online, can say a lot about us. Talk about this, that in some ways what we search for online can paint a very powerful picture of who we are.

[Sandra Matz]

Yeah. So Google searches, if you, if you think about that it's. Google is probably a more like a closest, the closest confidant that we, that we have. Right. We ask Google questions that we don't even dare to ask our closest friends, our partners. So on some level it's not surprising that whatever we search for on Google actually reveals a lot of what's going on inside. And that could be anything from mental health to truth about society that we might not want to see. So one of a close friend Seth Stevens Davidovich. And what he did is he looked at search data. So all of the searches that people make, and he was trying to uncover some of the relationships between what we search for and really kind of truths about society. So that could be anything from what do people search for when they search for sex? Do people look for abortions more often than we actually see in the official data? Do people search for racist jokes more often than people would admit in public? So I think Google is really this source that captures, like, what's going on inside our mind and stuff that we don't want to share with anyone else.

[Shankar Vedantam]

So we had set Stevens Davidovitz on Hidden Brain some years ago, and one of the things he mentioned was that there was this negative correlation between racist searches on the Internet and the likelihood that people would vote for Barack Obama. So in both the 2008 and 2012 presidential elections, places with higher rates of Google searches using racist terms were less likely to vote for Barack Obama.

[Sandra Matz]

Yeah, that's right. And I think again, like, if you look at the official polls, nobody wants to admit that. So those are correlations that you don't necessarily see showing up in survey data, but you do see them show up in these more hidden queues.

[Shankar Vedantam]

In another study, Sandra, you looked at the relationship between social media updates and voting, but you were not looking at explicit data like people saying they were going to vote for a particular politician. What were you looking for and what did you find?

[Sandra Matz]

Yeah, so this was a study that we did where we looked at what's driving populist voting. And what we were particularly interested in is affect. So to what extent is this negative affect and not just like the more aggressive negative affect, like anger, which you oftentimes see talked about in the media, but also the more subtle ones like sadness and depression. To what extent are those emotions as they show up on social media linked to people voting for populist candidates? So one of the elections that we looked into was Brexit in the uk, so people voting to leave the European union or the 2016 U.S. presidential election. And what you find consistently is that in areas where there's a lot of this negative affect showing up on social media, people are also more likely to to vote for these populist candidates and causes.

[Shankar Vedantam]

And again, what's interesting here is that it's like the mismatched socks in the drawer, Right? It's not a signal that people are actually thinking will say something about their political preferences. If I'm feeling upset or sad or my affect in general is Negative. I don't think it's going to reveal something about my political preferences, but in fact, it does.

[Sandra Matz]

Yeah. And you have all of these predictive models. Right. So you have all of these predictive models trying to project what is the outcome of an election. None of them really consider tone or emotional valence based on social media.

[Shankar Vedantam]

You know, I'm reminded of that analysis that found in the 2016 presidential election that Donald Trump won three quarters of all counties that had a cracker barrel restaurant, but only 22% of counties that had a Whole Foods store. Now, most people are not thinking about politics when they're shopping for groceries or dining out, but it turns out that our shopping and dining habits can reveal powerful things about us.

[Sandra Matz]

Yeah. So sometimes I think it's like the behaviors that you show. Right. Shopping and Whole Foods is probably a proxy for some of the more psychological variables. That could be anything from openness, which we know is associated with being liberal, could be associated with socioeconomic status. And the same way that negative emotions, for example, is associated oftentimes with a desire for change. And in that sense, it's not necessarily surprising that those people who feel currently bad about themselves want to vote for a candidate that promises change.

[Shankar Vedantam]

Most of us spend a great deal of time every day in front of various devices. We scroll and tap and like and listen. We search for answers to our most personal questions and post updates to our social media feeds. When we come back, how all this data can help us improve our lives. You're listening to Hidden Brain. I'm Shankar Vedanta. Support for Hidden Brain comes from Lumen. Lumen is the world's first handheld metabolic coach. It's a device that measures your metabolism through your breath. Your metabolism is your body's engine. It's how your body turns the food you eat into fuel that keeps you going. Because your metabolism is at the center of everything your body does. Optimal metabolic health translates to a bunch of benefits, including easier weight management, improved energy levels, better fitness results, better sleep, etc. Lumen gives you recommendations to improve your metabolic health. All you have to do is breathe into your lumen first thing in the morning and you'll know what's going on with your metabolism, whether you're burning mostly fats or carbs. So if you want to stay on track with your health this holiday season, go to Lumen Me Brain to get 15% off your lumen. That's L u m e n me brain for 15% off your purchase. Lumen makes a great gift, too. Support for Hidden Brain comes from SimpliSafe. If you ever worry about the safety of your home and family, there's no better time to act Right now. Simplisafe is extending its massive Black Friday deal for hidden brain listeners. Get 50% off a new SimpliSafe security system. Old school systems only take action once someone is already inside your home. That's too late. Simplisafe's active guard outdoor protection changes the game by preventing crime before it even happens. Simplisafe is extending its massive Black Friday deal for hidden brain listeners. This week only you can get 50% off any new system with a select professional monitoring plan. This is your last chance to claim their best offer of the year. Head to simplisafe.com brain that's simplisafe.com brain there's no safe like Simplisafe. This is Hidden Brain. I'm Shankar Vedantam. You wake up in the morning and reach for your phone. You open Instagram and leave a comment on a friend's vacation pictures. You sneeze and run a Google search about allergies. On the way to work, you buy a muffin at a local cafe using your credit card. Every day, we leave dozens of tiny traces of ourselves in the digital world. At Columbia University, Sandra Matz calls the accumulation of these traces our digital footprints. She is the author of the Data Driven Science of Predicting and Changing Human Behavior. Sandra, you say that the traces we leave online not only paint a picture of who we are, they show marketers and political campaigns how to influence us. Now, we've all heard a lot about the problems of digital surveillance, but fewer people know how these tools can be used for good. Let's start with the work you've done showing how psychological targeting can help people save more money.

[Sandra Matz]

So the idea here was that if we could make saving more appealing to people, to make it more personally relevant, could we help them put the extra money to the side? So we teamed up with Save a Life, which is a fintech company in the US they're trying to help low income families save for a rainy day. So the people that we work with were people with very low levels of savings, so less than $100. And our goal was to get them to add an additional $100 to their savings account over the course of four weeks. So we teamed up with the creative team of Save a Life and we essentially asked them, well, come up with saving messages that try to encourage, say, people who are very agreeable. So people who care about other, people who care about their Social relationships, and maybe tell them that if you manage to put some money to the side right now, this is a great way of making sure that your loved ones are protected. Now, if you're talking to someone who is much more competitive and critical, which is the other side of the same personality trait, maybe you want to highlight how just putting this money to the side gets them ahead of the game. Right. So we kind of came up with this different type of messaging for all of the big five personality traits. And then we just sent out the messages over the course of four weeks, and we looked at how many people eventually managed to save an additional $100.

[Shankar Vedantam]

What did you find?

[Sandra Matz]

So what we find is that essentially, if we target people with the messages that were tailored to their personality, about 11% of the entire sample managed to put $100 to decide. Now, that's certainly far from perfect, but ideally, we'd want this to be closer to 100%. But if you think about it, this means that someone is doubling their savings over the course of four weeks. And what's more important is that it was also much better than the existing messaging that Save a Life had been using up to this point. So they had been trying to perfect their messaging over a couple of years, and we were still 60% better than the gold standard that they were using at the time.

[Shankar Vedantam]

So just to underscore the principle here, what you're doing is you're basically saying we can tell what people's personalities are by the digital footprints they're leaving behind. And if we can tailor messages in some ways to match people's personalities, those messages are far more likely to break through.

[Sandra Matz]

Exactly. And you can think of it as essentially, this is what we do all the time in our offline relationships. Pretty much any type of conversation that you have is, to some extent, tailored. You don't talk about the same things or in the same way to a friend or to your kid or to your boss. So we're trying to replicate this at scale and just say, okay, what is it that you might care about? And how can we make saving more appealing to you?

[Shankar Vedantam]

Our digital footprints can also reveal insights about our mental health. You and a colleague have studied whether there's a connection between depression and a person's location data.

[Sandra Matz]

Yeah. So this is essentially research that we did with GPS records. So again, your phone tracks your GPS records pretty much 24 7. And what we were interested in is whether we could tell whether someone might be suffering from depression or not, just based on these GPS records alone. Now, if you look at the content of some of these traces that we observed, they actually make a lot of sense. So what we found, for example, is that if you don't leave your house as much anymore as you typically did, or there's much less physical activity, you don't travel to as many places as you used to. Those are all small indicators that there might be something going on. Right. It's certainly not a diagnostic tool, but it means that maybe we could be raising a red flag and say, hey, might be nothing, but why don't you check in with some support?

[Shankar Vedantam]

And I suppose there's always going to be noise in the data. So someone may have lost their phone inside their sofa cushions. And so the phone basically sits at home for three weeks. It doesn't mean that they are depressed and they haven't left their home in three weeks. It just means that the phone was lost. But I think what you're really saying, Sandra, is that in aggregate, this data in fact are telling us valuable things and at a minimum, they're basically raising a flag that warrants further investigation.

[Sandra Matz]

Yeah. So I don't think it's a deterministic diagnostic tool, but it could be incredibly helpful for people, for example, who know already that they're suffering from depression. Right. So it's like one of these mental health challenges that just pop up time and again and it's really difficult to find your way out of the valley. So once you enter the full fledged depression, it's really hard to come back. And so if we can get these early indicators of, well, maybe it's nothing, but here's like a warning system that might alert you to, well, again, there's like these changes in your behavior. You're deviating from your typical routine. Why don't you reach out to someone and see if there's something to it?

[Shankar Vedantam]

I mean, this is really no different than basically saying, let me measure your resting heart rate or your cholesterol levels. And over time, if I have enough data, it might paint me a picture of saying, you know, you're heading down a bad path, you might want to change your lifestyle.

[Sandra Matz]

Yeah. And you can do this in real time. And technically, what you could, what you could also do, if you're really thinking about this as a support system for the person, is not just alert the user, but maybe I can give you the opportunity to name two people, loved ones, someone that you want to know that you're having a hard time even if you're not in a position to tell them.

[Shankar Vedantam]

So our digital footprints not only reveal things about our past. They can also predict things we might do in the future. You once tried to predict dropout rates among college students by studying their digital footprints. How did you do this, Sandra?

[Sandra Matz]

Yeah, so this is actually one of the projects that I personally care a lot about because there's still so many students dropping out with enormous debt that they never recover. So what we were trying to do is to see if we could predict early on, once people joined university in the first semester, whether we could see if they might be struggling, integrating into the system. Maybe they're not finding the information that they should be finding. Maybe they're not embedded in the cohort as much as other people and they're somewhat on the fringes, not really connecting to the community as much. We again teamed up with a company called Radio Education. They had a sense of what are the activities that students attending? Are they talking to other students? Are they part of groups? Are they sending messages? Are they receiving messages? So we looked at all of these data traces and again, once you combine all of them, you actually have a relatively decent sense of whether someone might be struggling and whether they might drop out at the end of the semester.

[Shankar Vedantam]

And of course, when you put it this way, it seems to make sense. Now, if I know, for example, that a student doesn't have many friends and is not exchanging messages and in fact is a little bit isolated and is not spending time hanging out with other students, you know, it's not unreasonable now to say maybe the student doesn't feel like he or she belongs at university and is at higher risk of dropping out.

[Sandra Matz]

Yeah. And for me, what I love most about this is essentially it creates a path to help students. And at the very bare minimum, what it allows administrators to do is identify at risk students. Right. So if you see that there's some students who have a higher likelihood of dropping out, maybe you allocate more resources to helping them. Now, for me, the even more interesting part is that we also get a sense of what is predicting dropout for each individual student. So it could be that I, for example, when I started university as a first generation student, my problem was that I simply didn't know where all of the information was sitting. I didn't know how to get the literature, I didn't know where to search for information. And so for me, if that was the prediction that the algorithm had made, administrators could have gone in and said, here, here's the information that you need. You can pop it up on my app, you can send it in my email Just make sure that I see what I need to see. Now, there could be other people who know exactly. I know that most of my friends when I started knew exactly what they were looking for, but some of them probably had a harder time integrating with the community and finding the friends and making these connections. So for those people, if we see that that's what's happening based on the algorithm, it's a totally different intervention. Right. So then we're trying to see if we can get you involved in events more. Is there a way to ask other people to connect? So the moment that you understand why someone is predicted to be a dropout, you can also adjust the approaches that you use to help them.

[Shankar Vedantam]

In other words, instead of a one size fits all approach, now you can actually say the individual person gets his or her own approach.

[Sandra Matz]

Exactly. It's the same as targeted advertising. Right. So we kind of try and figure out what each person needs at a given point in time. Same for student dropbound.

[Shankar Vedantam]

Sandra, you say that these digital tracking tools are increasingly being used not just to identify health issues, but to actually intervene. How so?

[Sandra Matz]

Yeah, so there's really two things that the data has to offer, and I think of it as tracking and treating. So on some level, just all of the data that we generate says a lot about our physical activity, our physical health, but also about our mental health. Right. Again, we talked about GPS records that say something about whether you might be suffering from depression. There's a lot that we can learn about your mental health from what you post on social media. So this is the tracking part. But then what I think is really interesting and it's currently being developed, so I think we're really early stages, is more of the treatment part. So can I use your footprints to not only surface, let's say, the most relevant interventions to you, the same way that Amazon recommends products and the same way that Netflix recommends movies, can actually an algorithm who knows you, based on your data, recommend the best treatment for you suffering from depression.

[Shankar Vedantam]

You tell the story of a woman named Chukora Ali who was in a car accident that left her severely injured. She spiraled into depression. Tell me her story and what happened to her.

[Sandra Matz]

Yeah, so this is a really tragic story of a woman who got into an accident, got severely injured, lost. The bakery that she was running shows she was self employed, which also meant that she couldn't afford a car anymore, couldn't really provide for her family. And you can imagine that all of this takes a pretty big toll on someone's mental health. Now, with no car, no money. There's no way that you can either find a therapist, let alone drive to a therapist for like your weekly session. So what she did is she started using an app that's called Wiza, which is really trying to interact with you, give you advice, asks questions about how you're feeling, gives these little prompts and little challenges. Maybe you go out to nature and maybe you try and meditate for a little bit. And I think the way that she tells the story is that it was very weird in the beginning talking to a bot about your mental health struggles, but at some point you adjust, right, and you get used to it. And from using it once in a while, I think she started using it multiple times a day.

[Shankar Vedantam]

What was the effect of using this bot on her mental health center?

[Sandra Matz]

So in her case, I think it significantly improved her mental health. It certainly didn't fix all of the problems. Right. And still a lot of effort that you have to put in as a human being. But it felt like there was a support system that she otherwise couldn't have afforded.

[Shankar Vedantam]

And again, I don't think you're necessarily suggesting that a bot is necessarily an ideal replacement for a human therapist, but you're saying in a situation like this where in fact the person cannot afford or cannot get to a human therapist, this would be a potential solution.

[Sandra Matz]

Yeah. So I think if you have access to a human being, blood and flesh, who can be your therapist, that's probably preferable. However, there's this huge gap in terms of how many therapists there are and how many people are seeking therapy. So then there is a really huge need for people to get at least some support in cases where they can't get hold of a human being.

[Shankar Vedantam]

Many people are worried that digital tracking has increased polarization. The moment you click on one video with a political theme, the algorithms quickly paint a picture of you as liberal or conservative and start feeding you more and more of the same content. In other words, digital tracking and psychological targeting can quickly leave you inside an echo chamber. You say it's at least theoretically possible to use these same tools to reduce polarization?

[Sandra Matz]

Oh, yeah, it's one of my favorite applications. But the idea here is that it actually offers this, what I think of as like a magical echo chamber swap machine. Because it's really difficult for me to figure out, well, what is the reality of, let's say a 50 year old guy in the middle of Ohio? I just don't have direct access. Right. It's really difficult for me to step into their shoes. And see what is their day to day look like. Same for, let's say, a single mom in the suburbs of Chicago. But Google knows, right? Google knows exactly what those people see every day when they search for something specific. Facebook knows exactly what their newsfeed looks like every day. So instead of keeping me in my own echo chamber and just feeding more of the stuff that I, that I already know, they could actually allow me to hop into the echo chambers of other people.

[Shankar Vedantam]

In other words, if I know that you are basically self selecting into one echo chamber, you're saying, what if these platforms in some ways can encourage us to basically visit other echo chambers and in some ways broaden our worldviews?

[Sandra Matz]

Yeah. So it could be an explorer mode. Right. And the explorer mode at the very basic level could be, well, just do an echo chamber swap with someone. So maybe someone is happy to let you access their Facebook feed and you give them access to yours. At a more sophisticated level, they could build an engine that allows you to specify exactly which echo chamber you want to hop into. Right. I can say, here's the demographics of the person, here's the preferences, here's the age, gender, whatever you want to see, and then you can hop into the echo chamber. Now, I don't think we're going to use it all too often the argument by Google is nobody would use it because it's so comfortable in our own echo chamber. And I think that is largely true. Most of the time. We probably love to not have to go to page two of Google because we find what we want to see on page one. But I @ least want to have the option see, well, what is the search result for like immigration that someone with a totally different political ideology than me in a totally different part of the country sees that I would never otherwise get to see?

[Shankar Vedantam]

When I'm thinking about the concerns that major platforms might have in serving up this kind of information, I'm struck by the fact that in some ways, I think, Sandra, what you're talking about is the difference between the information we want and the information that we need. So the information that I want might be information that basically confirms that my preexisting views are correct. The information that I need might in fact tell me, hey, take a look at what's happening on the other side.

[Sandra Matz]

I think that's absolutely true. And in all fairness, some of it is human nature. So the reason for why these algorithms work and the reason for why companies craft them in their effort to make profits is because we love to see stuff that we believe in anyway. It's very comforting, it's very reassuring to see stuff that is aligned with our worldview. So that's why I feel like this Explorer mode is just one option that allows us to at least get some collective oversight right? So even if we're not using it as much, it still means that we have an option to see what's happening on the other side.

[Shankar Vedantam]

In our companion episode on Hidden Brain plus, we look at the downsides of digital surveillance. We take a closer look at the harms of tracking technologies and why the most popular intervention to protect people, giving them control over whether they are tracked online and whether their children are tracked online, may not be the best approach.

[Sandra Matz]

It feels much more of a burden and a responsibility that we're not really equipped to take on.

[Shankar Vedantam]

To listen, please look for the episode titled how to Protect Yourself online on Hidden Brain Plus. If you're not yet signed up, please visit support.hiddenbrain.org if you're using an Apple device, please go to apple.co hiddenbrain. Sandra Matz is the author of Mind Masters the Data Driven Science of Predicting and Changing Human Behavior. Sandra, thank you so much for joining me today on Hidden Brain.

[Sandra Matz]

Thank you so much.

[Shankar Vedantam]

Hidden Brain is produced by Hidden Brain Media. Our audio production team includes Annie Murphy, Paul, Kristin Wong, Laura Quirrell Ryan, Katrina, Autumn Barnes, Andrew Chadwick, and Nick Woodbury. Tara Boyle is our executive producer. I'm Hidden Brain's executive editor. We end today with a story from our sister show, My Unsung Hero. This My Unsung Hero segment is brought to you by T Mobile for Business. Today's story comes from Stephanie Cole. When Stephanie was a teenager, she got her very first job. It was around the winter holidays at a department store in Los Angeles.

[C]

There I was in my black skirt, my white blouse and ready to go the first day. And I had been trained, but very, very quickly. And as is true, in a department store during Christmas, it was just bustling. You know how it is at Christmas when everybody's out shopping and everybody's in a hurry and all these people around. This woman comes up to me with, I think a Christmas tree ornament she wanted to buy. And I freeze. I just freeze. All of a sudden I can't remember anything. I can't remember how to run the cash register. I can't remember anything about the transactions. I am just absolutely frozen and probably very close to tears. Just, I so wanted this to go right and it was going so wrong. She looked at me and paused. And with such a kind expression on her face said, it's all right, take your time. I'm not in a hurry. And that was the release. All of a sudden, I could breathe, I could wait till somebody else could help me. It was going to be okay. It made such an impression that all these years later, not only do I still remember it, but I find myself those words coming out of my mouth on numerous, many, many occasions over the years. You know, you encounter somebody whose first day on the job or they're just having a bad day and things are really, you can tell they're in a bad place and you can say, it's okay, I'm not in a hurry. Take your time. And it always makes the situation better. Always, always. And so this woman, I can't really remember her face, and certainly she's probably dead by now, given how old I was and how old she was. But she gave me that gift without knowing she gave me that gift. And it's lasted all these years.

[Shankar Vedantam]

Stephanie Cole is from Bainbridge Island, Washington. This segment of My Unsung Hero was brought to you by T Mobile for Business. You can find more stories like this on the My Unsung Hero podcast or on our website, hiddenbrain.org I'm Shankar Vedantam. See you soon.