BRAINBITES
WEEK OF MAY 15–21, 2026 · ISSUE 12
BRAINBITES — Your weekly health, fitness, and longevity digest. Five minutes that cover the highest-signal podcasts and science before they hit the mainstream.
This week: social psychologist Nick Epley walked Huberman through the research on perspective-taking errors — why you consistently underestimate how much people like you and overestimate how harshly they judge you. And the Essentials vault dropped the language-learning episode with the critical-period data that isn't as closed as you think.
From: Huberman Lab · Huberman Lab Essentials
🔢 The liking gap — the systematic error in how you think others perceive you
🎙️ Nick Epley on mind-reading failures and the liking gap
💡 Deep Dive: social anxiety is a calibration error, not a character flaw
📰 News Bits: the liking gap in conversations · language learning after the critical period · compliments as an underused intervention
GM.
The most practically useful episode this week is Nick Epley on Huberman. Epley is a behavioural scientist at Chicago Booth who studies how people think about other people's thinking — and how badly they do it. The headline finding: after a conversation, people consistently rate their partner's impression of them lower than the partner actually rated it. The gap is large, persistent, and almost entirely in one direction.
⏱ ~5 min read
🧠 THE LIKING GAP
The systematic error your brain makes about how others see you.
Two findings landed in the same week. One: after a first conversation, people believe their conversation partner liked them less than the partner actually did — a bias so consistent it has a name: the liking gap. Two: the 'critical period' for language learning isn't a cliff — it's a slope, and adults can still achieve high proficiency with the right methods.
The thread this week: the constraints you assume are hard limits are often calibration errors — in how you read others and in what you believe you can still learn.
🔢 THE VITAL SIGN · SOCIAL PSYCHOLOGY
The Gap
The 'liking gap' — the systematic difference between how much people like you and how much you think they like you.
Epley presented data from studies where strangers had conversations, then separately rated (a) how much they liked their partner and (b) how much they believed their partner liked them. The result: people consistently underestimated how much their partner liked them. The gap was large and appeared across age groups, conversation lengths, and cultural contexts.
The mechanism, Epley argued, is asymmetric access to information. You have full access to your own inner monologue during a conversation — every awkward pause you noticed, every word you second-guessed. Your partner doesn't hear any of that. They see your outward behaviour, which is typically warmer, more engaged, and more likeable than your internal experience suggests. You judge yourself on the blooper reel. They judge you on the broadcast version.
The liking gap has downstream consequences. People who underestimate how much others like them are less likely to initiate future contact, less likely to share personal information, and less likely to deepen relationships. The bias doesn't just make you feel worse — it changes your behaviour in ways that prevent the social connection your brain actually wants.
The person you talked to yesterday liked you more than you think. The tragedy is that your brain's miscalibration might prevent you from talking to them again.
Dr. Nick Epley on Huberman Lab · May 2026
🎙️ THE EPISODE · SOCIAL PSYCHOLOGY / RELATIONSHIPS
You're bad at reading minds — and that's costing you relationships
Nick Epley on the liking gap, the limits of perspective-taking, and why asking beats guessing every time.
Dr. Nick Epley · Huberman Lab · May 2026
Epley is a precise, data-driven researcher who avoids the self-help register entirely. The conversation with Huberman is grounded in published experiments with specific effect sizes and clearly stated limitations. It's the rare psychology episode that gives you a usable mental model without overpromising.
You think you're transparent. You're not.
Epley presented the 'illusion of transparency' — the belief that your internal states (nervousness, boredom, attraction) are more visible to others than they actually are. In experiments, people who were instructed to hide disgust while tasting a bad drink believed observers could tell they were disgusted. Observers couldn't. Your inner experience leaks far less than you assume. The practical implication: the nervousness you feel during a presentation is largely invisible to your audience.
🎙️ Skip to ~22:00 for the transparency illusion experiments.
Asking beats guessing — and it's not even close.
Epley's strongest finding: the best way to know what someone thinks is to ask them directly. This sounds obvious, but people overwhelmingly prefer to guess — using facial expressions, body language, and theory of mind — all of which are systematically worse than simply asking. In studies, direct questions produced dramatically more accurate assessments of others' preferences, feelings, and opinions than any amount of observation. The barrier isn't capability. It's social norm — people feel it's awkward to ask directly. The data says it's far more awkward to guess wrong.
Compliments are massively underused — the 'undervaluation of positive actions.'
Epley's lab found that people consistently overestimate the awkwardness of giving a compliment and underestimate its positive impact on the recipient. In experiments, compliment-givers predicted mild appreciation. Recipients reported significantly higher boosts to mood, self-esteem, and warmth toward the giver than predicted. The gap between expected and actual impact was large. Epley's argument: positive social actions (compliments, expressions of gratitude, random acts of connection) are systematically underperformed because people miscalculate the cost-benefit ratio.
🎙️ Listen → Huberman Lab, Dr. Nick Epley, May 2026
💡 DEEP DIVE · SOCIAL CALIBRATION
Epley gives you the social miscalibration. The language episode gives you the learning miscalibration. The pattern is the same: assumed limits that aren't there.
The language-learning Essentials episode Huberman released the same week lands on a parallel error. The conventional wisdom — that there's a critical period for language learning that closes around puberty — is overstated. A 2018 study of 670,000 English learners found that the ability to achieve native-like grammar proficiency declines after age 17, but doesn't vanish. Adults can still reach high proficiency; it just takes more deliberate practice and different methods than children use.
The parallel with the liking gap is structural. In both cases, people assume a hard constraint that's actually a soft one. 'People don't like me as much as I hope' is a belief, not a fact — and the data says it's wrong. 'I can't learn a language after 30' is a belief, not a fact — and the data says it's overstated. Both beliefs limit behaviour: the liking gap prevents social connection; the critical-period myth prevents adults from attempting new skills.
Epley's broader research programme connects these calibration errors to a single cognitive tendency: people are bad at predicting others' reactions and their own capabilities because they rely on intuition where data would serve better. The fix in both domains is the same: test the assumption. Start the conversation. Try the language. Measure the actual outcome instead of trusting the prediction.
The loneliness epidemic Galloway described in Issue 9 looks different through Epley's lens. If the liking gap is universal, then a significant portion of social withdrawal isn't caused by actual rejection — it's caused by predicted rejection that never happened. The barrier to connection isn't that people don't want to connect. It's that both sides think the other isn't interested.
Most of the limits you feel aren't walls. They're miscalibrated predictions. Test them.
📰 NEWS BITS
The liking gap persists across contexts — from first conversations to long-term colleagues.
Epley's research shows the liking gap isn't just a first-impression phenomenon. In longitudinal studies, college roommates underestimated how much their roommates liked them even after months of living together. The gap narrowed over time but never fully closed. The implication: the social anxiety most people attribute to unfamiliarity isn't just a novelty effect. It's a persistent bias in how the brain models other minds.
— Huberman Lab, Dr. Nick Epley, May 2026
Language learning after the critical period: the cliff is actually a slope.
Huberman's language Essentials episode reviewed a 2018 study of 670,000 learners showing that grammar-learning ability declines gradually after age 17, not abruptly. Adults who used immersive methods (not just classroom instruction) achieved near-native proficiency well into their 30s and 40s. The critical period is real for accent acquisition but overstated for grammar and vocabulary. The practical message: adult language learning is harder but far from impossible — and the methods matter more than the age.
— Huberman Lab Essentials, Languages episode, May 2026
Compliments produce larger positive effects than givers predict — the undervaluation bias.
Epley's lab ran experiments where participants gave genuine compliments to strangers and friends, then predicted how the recipient felt. Recipients consistently reported larger mood improvements, greater warmth, and more positive feelings about the interaction than givers predicted. The givers overestimated the awkwardness and underestimated the impact. Epley's recommendation: treat compliments like an underpriced asset. The cost is lower than you think and the return is higher.
— Huberman Lab, Dr. Nick Epley, May 2026
P.S. Give one genuine compliment to someone today — not a generic one, a specific one. Notice what happens. Hit reply and tell us if the reaction surprised you.
What did you think of today's BrainBite?
🧠🧠🧠🧠🧠 The liking gap just explained 15 years of unnecessary social anxiety
🧠🧠🧠 The compliment data is the sleeper hit. The language-learning parallel is smart.
🧠 Less psychology, more biology. Where's the exercise science?
We listen to every health podcast so you don't have to. Weekly. No noise.
They liked you more than you thought. Now you know. See you next week.
