Skip to main content
Stuff You Should Know

SYSK Gets Weird Playlist: Doppelgängers: Hey you, you're me!

47 min episode · 2 min read
·

Episode

47 min

Read time

2 min

Topics

Fundraising & VC, Artificial Intelligence, Psychology & Behavior

AI-Generated Summary

Key Takeaways

  • Genetic basis of lookalikes: A Spanish researcher at the Carreras Leukemia Research Institute analyzed 32 doppelgänger pairs from photographer Francois Brunel's 20-year project. Facial recognition AI classified 16 pairs as identical twins. Genome analysis confirmed these unrelated people share significant single nucleotide polymorphisms (SNPs) in regions governing facial structure, despite being as genetically unrelated as random strangers overall.
  • Epigenetics explains the differences: While doppelgängers share key facial-construction genes, their epigenomes and microbiomes show no closer similarity than two random people. This mirrors findings in identical twins, where shared DNA doesn't fully explain appearance differences — environment, behavior, and lifestyle modify gene expression, producing visible divergence even between people with nearly identical underlying genetic blueprints.
  • Statistical probability of lookalikes: A 2015 Australian study measuring eight facial dimensions found a one-in-a-trillion chance of two people being identical across all measures. With eight billion people on Earth, that translates to roughly a one-in-125 chance that two people currently alive share identical measurements — meaning human perception of "doppelgängers" operates on looser similarity thresholds than mathematical precision requires.
  • Fusiform gyrus and face recognition: The brain contains a dedicated region — the fusiform gyrus — exclusively for processing faces. Studies show this region activates even in people blind from birth when they touch 3D face models, but not when touching other shapes. Mild electrical stimulation of this region causes people to perceive faces in front of them as visually transforming, demonstrating hardwired face-detection circuitry.
  • Facial recognition AI bias: Current facial recognition systems produce false positives at dramatically unequal rates across demographics. A Black woman's face is 40 times more likely to generate a false match than a light-skinned man's face. This bias stems from training datasets overrepresenting white male faces. A 2019 New Jersey case saw Najir Parks spend 20 days in jail and $5,000 in legal fees due to a misidentification.

What It Covers

Josh and Chuck explore the science and folklore of doppelgängers — unrelated people who look nearly identical. The episode covers the German origins of the word, historical accounts, a 2022 genetic study using Francois Brunel's photography project, facial recognition implications, and how the brain processes faces via the fusiform gyrus.

Key Questions Answered

  • Genetic basis of lookalikes: A Spanish researcher at the Carreras Leukemia Research Institute analyzed 32 doppelgänger pairs from photographer Francois Brunel's 20-year project. Facial recognition AI classified 16 pairs as identical twins. Genome analysis confirmed these unrelated people share significant single nucleotide polymorphisms (SNPs) in regions governing facial structure, despite being as genetically unrelated as random strangers overall.
  • Epigenetics explains the differences: While doppelgängers share key facial-construction genes, their epigenomes and microbiomes show no closer similarity than two random people. This mirrors findings in identical twins, where shared DNA doesn't fully explain appearance differences — environment, behavior, and lifestyle modify gene expression, producing visible divergence even between people with nearly identical underlying genetic blueprints.
  • Statistical probability of lookalikes: A 2015 Australian study measuring eight facial dimensions found a one-in-a-trillion chance of two people being identical across all measures. With eight billion people on Earth, that translates to roughly a one-in-125 chance that two people currently alive share identical measurements — meaning human perception of "doppelgängers" operates on looser similarity thresholds than mathematical precision requires.
  • Fusiform gyrus and face recognition: The brain contains a dedicated region — the fusiform gyrus — exclusively for processing faces. Studies show this region activates even in people blind from birth when they touch 3D face models, but not when touching other shapes. Mild electrical stimulation of this region causes people to perceive faces in front of them as visually transforming, demonstrating hardwired face-detection circuitry.
  • Facial recognition AI bias: Current facial recognition systems produce false positives at dramatically unequal rates across demographics. A Black woman's face is 40 times more likely to generate a false match than a light-skinned man's face. This bias stems from training datasets overrepresenting white male faces. A 2019 New Jersey case saw Najir Parks spend 20 days in jail and $5,000 in legal fees due to a misidentification.

Notable Moment

When researchers gave electrically stimulated mild current directly to participants' fusiform gyrus during a face-to-face interaction, subjects reported watching the other person's face physically transform before their eyes — while height, clothing, and skin color remained unchanged — revealing how localized and modifiable face perception actually is.

Know someone who'd find this useful?

Episode Transcript

00:00:02 Speaker 1: I think I even say in the episode that I remember learning about doppelgangers from those time-life books about the paranormal back when I was a kid. Well, fast forward to when I'm a grown-up, a.k.a. now, I was really glad to learn that they're an actual thing. There are people walking around who have a twin that they're in no way related to, and the stories behind them tend to be very interesting. I'm out. This is the last one for me, so let me say thanks for listening to our playlist. 00:00:35 Speaker 2: Welcome to Stuff You Should Know, a production of iHeartRadio. 00:00:45 Speaker 1: Hey, and welcome to the podcast. I'm Josh, and there's Chuck, and Jerry's here, too. And this is Stuff You Should Know, the day after Labor Day edition, which Chuck loves. 00:00:57 Speaker 2: Yeah, you know, Emily showed me that article that informed part of this episode the other day. Yeah. And we'll get to it, but I started looking at this photographer's doppelganger photo series. 00:01:12 Speaker 1: Yeah. 00:01:13 Speaker 2: And boy, some of those are crazy similar. 00:01:18 Speaker 1: Yeah, what about the two bears with their shirts off? 00:01:21 Speaker 2: Yeah. 00:01:22 Speaker 1: Those guys struck me as extremely close. 00:01:26 Speaker 2: The two women with the bangs and the black tank tops looked like identical twins to me. I could not tell them apart. 00:01:33 Speaker 1: Yeah, I feel like we shouldn't tease anybody anymore. There's a guy, and like you said, we'll get to him in a little bit, or he'll figure in in a little bit, I should say. But he's a Montreal, I guess a Quebecois photographer, Francois Brunel. And for 20 years, he's been assembling... pictures of doppelgangers, people who already knew they had a doppelganger and wrote in doppelgangers he's looked for and found on the street. And he brings them together into his studio and photographs them together. But he does some like really cool stuff with it. It's not just like, here, you stand here and you stand next to him. The way that he poses them and dresses them and everything, like it creates like a sense of intimacy between them. It makes you think you're looking at identical twin photos. 00:02:15 Speaker 2: Yeah. I mean, you know, some of them people favor one another. And then when you have them styled to look alike and posed alike, that informs what you're seeing. But a handful of them, it was they look so much like identical twins that it's just it's weird to think that you could walk down the street one day and meet someone who looks exactly like you. 00:02:35 Speaker 1: Yeah. And I mean, like that is got to be pretty eerie. Luckily, one of the great things about social media is that people can post their random encounters. You got a phone …

Get the full transcript (9,120 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Stuff You Should Know transcripts →

You just read a 3-minute summary of a 44-minute episode.

Get Stuff You Should Know summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links.

other

More from Stuff You Should Know

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best Science Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into Stuff You Should Know.

Every Monday, we deliver AI summaries of the latest episodes from Stuff You Should Know and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime