Daniel sent us this clip that's been everywhere today. Trump's Energy Secretary, Chris Wright, is in the Oval Office, starts saying "A hundred and twenty years ago, Albert Einstein said that..." and Trump cuts him off immediately with "Nobody cares." Then he tells him to continue. And the thing is, you can watch this ten times and still not know if it was a joke, a power move, or both. I've literally watched it twelve times now and I'm less sure than when I started.
What's wild is how the same three seconds splits people completely. Half my group chats are convinced it was pure hostility, the other half think it's obvious deadpan. Same clip, same words, totally different read. I had one friend who said "that's just how guys talk" and another who said "I would quit on the spot if my boss spoke to me like that." They're both reasonable people.
That's the puzzle right there. Not the politics of it, but the mechanism. Why does one person hear a joke and another hears an insult? What does your answer say about how your own brain processes humor? Because it's not just a matter of opinion, it's revealing something about your internal wiring that you probably haven't examined.
Because it does say something. It's not random. Your read on that clip is basically a personality test you didn't know you were taking. And unlike those BuzzFeed quizzes, this one actually has a research literature behind it.
Let's take it properly. Let's figure out what just happened in our own heads.
The psychology literature actually has a framework for exactly this. Rod Martin and his colleagues published the Humor Styles Questionnaire back in two thousand three, and it's become the standard way researchers categorize what people are doing when they joke. Four distinct styles. And the questionnaire itself is worth describing because the questions are surprisingly revealing. They ask things like "even when I'm by myself, I'm often amused by the absurdities of life" or "if someone makes a mistake, I will often tease them about it." Your pattern of agreement across forty-some items sorts you into these four buckets.
Four ways to be funny. Or not funny, depending. And most people aren't pure types, right? You get a profile, a blend.
First, affiliative humor. This is the warm, inclusive stuff. Puns, witty banter, telling jokes to ease tension and bond with people. It says "we're in this together." Low risk, high social reward. Think of the person at a party who tells a story where everyone looks good, including themselves. They're not punching down or up, they're just creating shared laughter. Second, self-enhancing humor. That's the ability to laugh at life's absurdities even when things are going badly. It's a resilience mechanism. You're not putting anyone down, you're just maintaining perspective. The classic example is someone stuck in a terrible traffic jam who says "well, at least we'll have a story about how we missed the wedding.
The "well, that happened" of humor styles. Which, by the way, is a phrase I've come to rely on heavily. I think I might be a self-enhancing type by default.
Third, aggressive humor. Sarcasm, teasing, put-downs, ridicule. It can be genuinely funny, but it's used to critique or dominate. And this is where the nuance matters. Aggressive humor isn't necessarily hostile in intent, it's often a bonding mechanism within groups that have established trust. Roasting culture, for instance. But the structure is hierarchical. Someone's status is being adjusted, usually downward. And fourth, self-defeating humor. Making yourself the punchline to gain approval, often to an excessive degree. "I'm such an idiot, haha" but there's real self-esteem erosion underneath. The key distinction from self-enhancing is that self-enhancing humor acknowledges the absurdity without devaluing the self. Self-defeating humor makes you the target.
When two people watch that Oval Office clip and disagree, they're essentially running it through different default humor settings. Someone whose primary mode is affiliative hears Trump's interruption and thinks "that's hostile, that excludes." Someone whose mode is aggressive hears it and thinks "that's just how we banter." And they're both convinced their read is obvious.
Here's where the personality research gets interesting. These styles correlate with the Big Five traits. Affiliative and self-enhancing humor map to high openness and agreeableness. Aggressive humor correlates with low agreeableness, no surprise there. Self-defeating humor tracks with neuroticism. Your humor style isn't random, it's an expression of your underlying personality structure. Martin's team found these correlations across multiple cultures and age groups. They're robust.
Which means the Trump clip functions as a kind of Rorschach test. Your reaction reveals your own humor signature more than it reveals anything about what he actually intended. And we may never know what he actually intended. That's almost beside the point.
That's what we want to map out today. A taxonomy of personal humor signatures. What your style says about your cognition, your social processing, and in some cases, your neurology. Because once you understand the framework, you start seeing these patterns everywhere. Not just in politics, but in every awkward dinner party and misinterpreted text message. That moment where you stare at your phone and think "was that sarcasm or are they actually mad?" That's the exact same mechanism.
Text messages are the worst for this because you've stripped away every cue except the words themselves. No tone, no facial expression, no timing except the typing indicator which is its own anxiety generator.
Let's go deeper on the behavioral markers of these four styles, because they show up in ways people don't always recognize. Affiliative humorists, for example, tend to tell stories where everyone's in on the joke. They use "we" language. They'll say "remember that time we..." and the punchline is shared experience. If you pay attention to pronoun use in social settings, you can actually spot the affiliative types by how often they distribute the credit for the funny moment. Self-enhancing types, by contrast, often joke alone. They're the ones muttering wry observations to themselves during a delayed flight while everyone else is fuming. They don't necessarily need an audience, the humor is serving an internal regulation function.
The "I'm not trapped in here with you, you're trapped in here with me" energy, but directed at the universe. That's self-enhancing humor in its pure form. You're reframing a frustrating situation into something absurd and survivable.
That's actually a perfect example. And it's worth noting that self-enhancing humor is the style most strongly associated with psychological resilience. Studies of trauma survivors show that those who can find moments of dark but non-self-destructive humor during recovery tend to have better long-term outcomes. It's not about minimizing the trauma, it's about creating enough cognitive distance to keep functioning.
Which is different from self-defeating humor, where the joke is "I deserved this" or "of course this happened to me, I'm me." That's not distance, that's internalizing the negative event as identity.
Now aggressive humor, the third style, has specific markers too. Interrupting, nicknaming, framing insults as "just jokes," using sarcasm to establish hierarchy. And here's the thing. Trump's communication pattern maps onto this profile almost point for point. The nicknames, the interruption timing, the way he'll say something cutting and then wave it off as banter. Whether you find it funny or not, it's textbook aggressive humor style. The pattern is so consistent that political communication researchers have published papers specifically analyzing his use of humor as a dominance signaling mechanism.
The "just joking" deflection is the tell. It's a way of delivering the dominance message while maintaining plausible deniability. "I was just kidding, lighten up." Which puts the receiver in a double bind. If they object, they're humorless. If they laugh, they've accepted the status adjustment.
The fourth one, self-defeating, has its own tell. The person who preemptively mocks themselves before anyone else can. "I'll probably mess this up, haha." It's preemptive surrender disguised as humor. You see this a lot in workplace settings where someone's about to give a presentation and they open with "well, this is going to be a disaster, but here goes." They're trying to lower expectations so failure hurts less.
Which sometimes works socially in the short term but has real costs long term. If you keep telling people you're incompetent as a joke, eventually some percentage of them will believe you.
It correlates with neuroticism in the Big Five model. These aren't just preferences, they're personality fingerprints. Openness correlates with affiliative and self-enhancing humor. Low agreeableness, aggressive. Neuroticism, self-defeating. Your joke style is a window into your emotional architecture. And the correlations aren't small. We're talking effect sizes that would make a personality psychologist sit up straight.
Which brings us to the evolutionary question. Why be funny at all? What's the adaptive payoff? Because humor is cognitively expensive, it's socially risky, and it doesn't obviously help you find food or avoid predators. So why did it evolve?
This is where the signaling theory gets fascinating. Evolutionary psychologists argue that humor is a costly signal of cognitive fitness. To be funny in real time, you need rapid semantic processing, theory of mind, social calibration, and verbal fluency all firing simultaneously. It's a display of neural horsepower. You're essentially saying "my brain works so well I can afford to play with it." And costly signals are honest signals precisely because they're hard to fake. If your cognitive machinery isn't up to the task, you can't just decide to be witty.
Like a peacock's tail, but for cognition instead of genetics. The peacock's tail says "I'm so healthy I can grow this ridiculous thing and still survive." A well-timed joke says "my brain has so much processing capacity that I can generate novel incongruities in real time while reading the room.
And it's not just individual fitness signaling. Humor also functions as a coalitional signal. Shared laughter says "we process the world the same way, we belong in the same group." Inside jokes are basically in-group membership badges. You laugh because you recognize the shared schema, and that recognition reinforces tribal bonds. There's a reason every tight-knit group develops its own private lexicon of references that make no sense to outsiders.
Which explains why nothing bonds a team faster than a running joke that nobody outside the team understands. I've been on teams where the shared humor was literally the only thing getting us through a terrible project. And years later, I don't remember the project details but I remember the jokes.
Conversely, why nothing alienates faster than a joke that lands with half the room and leaves the other half cold. You've just drawn a bright line between in-group and out-group in real time. Everyone in the room instantly knows who belongs and who doesn't. That's why humor misfires at parties feel so viscerally uncomfortable. It's not just awkward, it's a failed coalition signal.
Deadpan humor becomes this fascinating edge case. It's a high-risk, high-reward signal because it deliberately strips away the prosodic cues that normally flag "this is a joke." No exaggerated intonation, no wink, no smile. The receiver has to infer intent purely from content and context. You're essentially saying "I trust you to get this without me holding your hand." Which, when it works, creates an incredibly strong in-group feeling. And when it fails, it fails catastrophically.
The cognitive load difference is measurable. Processing a slapstick joke is low-effort. The incongruity is visual and immediate. Someone slips on a banana peel, your brain resolves the mismatch between expectation and outcome almost instantly. Deadpan requires you to hold two competing interpretations in working memory, the literal statement and the intended meaning, while simultaneously modeling the speaker's mental state to decide which one they meant. It's a multi-step inference chain. Neuroimaging studies show increased activation in the medial prefrontal cortex and temporoparietal junction during deadpan comprehension compared to signaled humor. These are theory of mind regions working overtime.
It's basically the difference between reading a sign and reading a poem. The sign tells you exactly what it means. The poem requires you to hold multiple possible meanings in suspension and choose among them based on context and your model of the poet's intent.
This is why deadpan fails so reliably across cultural boundaries. The Davies study from nineteen ninety, replicated in twenty eighteen, found that Americans misinterpret British deadpan as hostility roughly forty percent more often than British listeners do. Same words, same delivery, completely different cultural schemas for what counts as a joke. Forty percent is huge. That's nearly half of cross-cultural deadpan attempts landing wrong.
The British have centuries of institutionalized deadpan. It's baked into the comedy tradition from Oscar Wilde through Monty Python to the present. The American default is more explicit signaling. So when a Brit says something absurd with a straight face, the American brain doesn't get the "this is play" flag it's expecting. It defaults to the literal interpretation, which often sounds insulting.
Across neurotype boundaries, the gap is even wider. If your brain processes language with a literal bias, deadpan just sounds like lying or hostility. There's no prosodic safety net to catch you. You're being asked to perform a cognitive operation that your neural architecture doesn't automatically support.
That neurotype boundary you just mentioned, that's where the research gets revealing. Francesca Happé's work in the nineties on what she called the "strange stories" test. She gave participants short vignettes where a character says one thing but means another. Sarcasm, white lies, irony. Neurotypical adults catch the intended meaning about ninety percent of the time. High-functioning autistic adults? That's a thirty-point gap on a task that seems effortless to most people.
The gap isn't about intelligence. These participants understood every word perfectly. The difference was theory of mind, the ability to model what another person believes and intends. Sarcasm requires you to hold two representations simultaneously: the literal statement and the speaker's actual belief. If your brain doesn't automatically run that second model, you just hear someone saying something false. And saying something false, in most contexts, is lying. So you attribute dishonesty or confusion to a person who was actually being ironic.
Which means deadpan is practically designed to fail across that particular neurotype boundary. No prosodic cues plus literal processing equals total ambiguity. It's not that the person can't understand humor in principle, it's that this specific delivery mode removes every feature they rely on to identify humor.
It's not just autism. There's a related but distinct condition called alexithymia, difficulty identifying and describing your own emotions. About ten percent of the general population has it, and it's massively underdiagnosed because people don't realize it's a thing. They just think they're "not emotional people." Alexithymic individuals don't just struggle to produce humor, they have reduced humor appreciation. The emotional resonance that makes a joke feel funny doesn't land. They might intellectually recognize that a joke was made, but the visceral experience of finding it funny is muted or absent.
"humorless" isn't one thing. It could be theory of mind differences, it could be emotional processing differences, it could be cultural mismatch. The idea that someone simply "has no sense of humor" as a personality trait is almost never accurate. It's almost always a specific processing difference that looks like humorlessness from the outside.
The clinical literature backs that up. True humorlessness is so rare it usually signals something neurological. Which brings us to one of the strangest phenomena in neuropsychology: Witzelsucht.
No, I know you're about to explain it, but I had to.
It's German for "joking disease." And German has these wonderfully precise compound words for clinical phenomena. Mario Mendez documented it in two thousand five. Patients with frontotemporal dementia, specifically damage to the right frontal lobe, lose the ability to stop making puns. Compulsive, inappropriate wordplay. They find everything hilarious, especially their own jokes, but they can't read a room. The inhibition mechanism that normally says "maybe don't make a pun at a funeral" is just gone. Mendez described patients who would pun continuously through medical examinations, through family arguments, through situations of obvious distress. The humor generation engine was stuck in the on position with no regulatory brake.
Humor regulation has a physical address in the brain. It's not just personality, it's circuitry. You can lose the ability to stop joking the same way you can lose the ability to stop saying certain words. It's a disinhibition syndrome, just a very specific one.
The flip side is equally localized. Patients with right hemisphere damage, same region, often lose the ability to detect sarcasm entirely. Winner and colleagues showed this in ninety-eight. They understand the words, they understand the context, but the sarcasm just doesn't register. Accuracy drops to about fifty percent, which is coin-flip territory. These patients aren't confused about language, they're selectively blind to ironic intent.
The same brain region that stops you from being the guy who won't shut up with puns is also what lets you recognize that someone is being ironic. Lose it one way, you become a compulsive joker. Lose it another way, sarcasm becomes invisible. That suggests humor production and humor comprehension share neural real estate in a really specific way.
And this is how you start to distinguish between a quirky humor style and something that warrants clinical attention. The question is always: is there a pattern, and is there functional impairment? Someone who loves puns and occasionally misreads a room is just a pun enthusiast. Someone who cannot stop making puns in contexts where it's damaging relationships, losing jobs, alienating people, that's a different category. The boundary is whether the behavior is ego-syntonic or ego-dystonic, whether the person can modulate it when the situation demands.
What about ADHD? You mentioned earlier there's a humor signature there too. I'm curious how that fits into this framework.
Canu and Carlson, two thousand three. Individuals with ADHD show elevated rates of self-deprecating and absurdist humor. The hypothesis is that it functions as a preemptive social repair mechanism. ADHD often comes with social rejection sensitivity, you get negative feedback from peers for being impulsive or missing social cues. Making yourself the joke before anyone else can is a way of controlling the narrative. You're saying "I know I'm different, I'm acknowledging it first, now we can move on.
"I'll mock myself before you get the chance." Which is self-defeating humor mapped perfectly onto a specific neurotype. And I can see how that would develop. If you've been punished socially for being impulsive your whole life, you learn to get ahead of the punishment.
The absurdist strain makes sense too. ADHD brains are highly associative, jumping between distant concepts rapidly. Absurdist humor relies on exactly that, unexpected juxtapositions that somehow cohere. It's a cognitive style that naturally produces a certain kind of comedy. The same mental mechanism that makes it hard to stay on one topic during a conversation also generates the kind of surprising connections that make absurdist humor work. It's a cognitive difference that's a liability in some contexts and an asset in others.
We've got this whole landscape now. Different humor styles, different neurotypes, different cultural schemas, and a neurological basis for the extremes. The question is what do we actually do with this information? How does knowing any of this change how you navigate a conversation?
Here's a practical question. You're in a conversation, someone says something that could be hostile or could be a joke, and you have about half a second to decide which. What's the actual decision tree? Walk me through it.
I think there are three questions worth running in that half-second. First, does this person have a pattern of this kind of humor? If your friend Steve has been deadpanning you for three years, the prior probability that this is a joke is high. If it's someone you just met, you don't have that data yet. So you're operating with a much weaker prior, which means you should be less confident in either interpretation.
Pattern recognition over snap judgment. What's the second question?
Is there a plausible alternative interpretation? Not "could I theoretically read this as a joke," but "given what I know about this person, is humor a reasonable fit for what just happened?" This is where you check for incongruity. Did they say something that contradicts an obvious fact? Did they exaggerate to the point of absurdity? These are structural features of jokes, not personality features of the joker. And third, what's the social cost of getting it wrong in each direction? If you laugh and it was serious, that can be awkward. If you take offense and it was a joke, you've just signaled that you don't speak their language. Which cost is higher in this specific relationship?
That third one is underrated. Most people default to protecting themselves from being laughed at. But misreading a joke as hostility can do more long-term damage to a relationship than the reverse. If you laugh at a serious statement, you can apologize. If you accuse someone of hostility when they were joking, you've essentially told them you don't trust them.
For the deadpan humorists listening, the ones who keep getting misinterpreted, there's actually a strategy that works. Within your first few interactions with someone new, explicitly flag your style. "Just so you know, I'm pathologically deadpan, you'll know I'm joking when I say something absurd with a straight face." It feels clunky, but you're giving them a schema they can reference later. You're essentially providing the prosodic cue verbally that your delivery style won't provide vocally.
Like handing someone a decoder ring before you start transmitting in code. It feels awkward in the moment, but it's an investment in future communication efficiency.
And for people on the other side, the ones who struggle to read humor, there's a pattern recognition approach that doesn't rely on tone of voice. Learn to spot the structural features. Exaggeration to the point of absurdity. Incongruity between what's said and the context. Timing, the pause before or after a statement that flags it as performative. These are cognitive signals, not emotional ones. You can learn them systematically. It's like learning the grammar of humor rather than trying to feel the music of it.
Instead of "do they sound like they're joking," it's "does this statement contain structural markers of humor." That's trainable. And it means humor comprehension doesn't have to be purely intuitive. You can build a compensatory system.
And the overarching heuristic for everyone is what I'd call the benefit of the doubt default. Given that humor styles vary dramatically across personality types, neurotypes, and cultures, and given that most people most of the time are not trying to be hostile, the rational prior is positive intent. Assume it's a joke until pattern evidence accumulates that it isn't.
Which is not the same as being naive. It's a Bayesian approach to social interaction. You're updating your model of the person as data comes in, but your starting assumption is calibrated to the base rate of actual hostility, which is lower than most people's anxiety suggests. Most ambiguous statements in everyday life are not attacks. They're jokes that landed weird or statements that came out wrong.
Here's a concrete thing you can try tomorrow. Next time someone says something ambiguous, count to three before reacting. Those three seconds give your brain time to run the pattern check, the alternative interpretation check, and the cost assessment. Most misinterpretations happen because we react before we've finished processing. The half-second reaction is the enemy of accurate humor comprehension.
Three seconds sounds like an eternity in conversation, but it's actually not. It's about the length of a breath. And the difference it makes in response quality is enormous. I've tried this and it works, but it takes practice because the impulse to react immediately is so strong.
Let's return to that clip we opened with. Armed with all of this, watch it again. Trump interrupts with "nobody cares," Wright pauses, half-smiles, and continues. What's actually happening there?
I think there's a third read that's more interesting than "joke or hostility." Trump uses aggressive humor as a bonding signal to his base. It's a dominance display wrapped in banter, and the audience that shares his humor style reads it as "he's one of us, he says what we're thinking." The interruption isn't just directed at Wright, it's a performance for the audience that enjoys that kind of humor. And Wright knows this. He's not the real target, he's the straight man in a bit that's playing to a different audience.
Wright's response is actually masterful if you read it through the self-enhancing lens. He doesn't get flustered, doesn't push back, doesn't apologize. He just absorbs it and keeps going. That's resilience-oriented humor in action. He treated the interruption as an absurd moment in a long day and moved on. The half-smile is the tell. It says "I recognize that was a moment, I'm not going to engage with it on the terms it was offered, and I'm going to continue doing my job.
Which means the exchange worked for both of them, in their respective styles. Trump got the aggressive humor signal, Wright demonstrated unflappability. Two completely different humor languages, both satisfied. Neither had to switch styles or accommodate the other. They just processed the same three seconds through completely different frameworks and both walked away fine.
This brings me to the open question I can't stop thinking about. As AI assistants become more embedded in social contexts, how are they supposed to handle these humor style mismatches? Right now, every major language model defaults to affiliative humor. Safe, inclusive, low-risk. Puns and gentle banter. They're trained to never be aggressive, never be self-defeating, never be deadpan unless explicitly asked.
The AI equivalent of a corporate icebreaker. "Why did the neural network cross the road? To optimize the objective function." It's humor that's been sanitized to offend no one, which also means it connects with no one in a meaningful way.
But is that the right default? If I'm an aggressive humorist and my AI assistant keeps responding with "that's an interesting perspective, let me help with that," we're speaking different languages. The assistant is failing the same way a neurotypical listener fails with deadpan. It's not understanding the social signal being sent. And over time, that mismatch erodes the sense that the assistant actually understands you.
The alternative is terrifying. Do you want your AI attempting aggressive humor? "Nobody cares about your calendar reminder, Dave." That's a disaster. The potential for harm is enormous because the AI doesn't actually understand the social context well enough to know when aggressive humor is bonding and when it's just cruelty.
It is, but it's also a real design question. Humor style matching might be as important as tone matching for social AI. And we haven't even started thinking about what it means for an AI to have a humor signature at all. Is affiliative the right default, or is it just the safest? And is "safest" actually the best design principle for a tool that's supposed to feel like it understands you?
I suspect the answer is that safe is correct for now, but incomplete long-term. The same way early GPS voices were universally calm and generic, and now we have celebrity options and regional accents. Someone's eventually going to ship a deadpan AI mode, and it'll either be brilliant or catastrophic. There's no middle ground with deadpan. It either works perfectly or it fails completely.
That's where I want to leave this. Your sense of humor isn't just a personality quirk, it's a window into your cognitive architecture. What you find funny, how you signal humor, how you read it in others, all of it maps onto how your brain builds models of other minds. Pay attention to what makes you laugh and why. It's more diagnostic than most personality tests. And it's more useful, because you can actually use the framework to navigate the world better.
Next time you watch a clip like the one that kicked this off, and you feel that instant reaction, pause. Ask yourself what your reaction says about you. Because it's saying something. It's telling you which of the four humor styles is your default, and that's information worth having.
Now: Hilbert's daily fun fact.
Hilbert: In the late Victorian period, European orientalist scholars widely believed that Tang dynasty court officials communicated using a secret whistled register of classical Chinese that only the emperor could decode. This theory persisted for decades before being abandoned as a complete fabrication. The scholars had apparently confused a description of musical notation used in court rituals with a linguistic phenomenon, and then embellished the account through several rounds of increasingly confident citation.
A secret whistling bureaucracy. And also like something someone made up at a dinner party that got out of hand.
That's going to haunt me. The image of officials whistling coded messages across the Forbidden City is now permanently lodged in my brain.
This has been My Weird Prompts. Find us at myweirdprompts dot com or wherever you get your podcasts. I'm Corn.
I'm Herman Poppleberry. Thanks for listening.
