Human Behaviors Science Still Cannot Fully Explain
From yawning to déjà vu, the human body runs on mechanisms science can measure but not fully explain. Here's what researchers actually know—and don't.
Written by AI. Mei Zhang
Photo: AI. Tomoko Hayashi
Your brain is, right now, doing at least a dozen things nobody has fully explained. Not "we're still working on the details" unexplained. Actually unexplained — as in, the mechanism is clear, the purpose is murky, and the evolutionary story keeps developing new plot holes.
I find this genuinely delightful. And also a little unsettling, which I'll get to.
Science is very good at the how. The why is where things get interesting. 🧬
The ones we thought we'd solved
Take sleep. We know a lot about what happens during it: cerebrospinal fluid flushes metabolic waste from brain tissue, neural connections reorganize, memory consolidates, hormones recalibrate. The brain doesn't go quiet — certain sleep stages produce activity patterns that look almost identical to wakefulness.
Here's the problem: none of that explains why the whole system requires unconsciousness to run. A maintenance system that needed you offline would be, from a pure survival standpoint, a strange design choice. Dolphins solved this differently — one brain hemisphere sleeps while the other stays alert. Some migratory birds catch sleep mid-flight. Humans got the version where we have to fully disengage from our environment for roughly a third of our lives, vulnerable to anything that wanders past.
We know what sleep does. We still don't fully know why it has to look like this.
Dreaming is the nested mystery inside that mystery. Most dreams happen during REM sleep, when emotional processing centers run hot and the logical-reasoning regions dial way down — which is why dream logic feels airtight until you wake up and realize you were doing calculus with a talking refrigerator. Theories range from emotional regulation (replaying charged memories in a safer internal environment) to threat simulation (rehearsing social and physical challenges) to memory consolidation. What remains genuinely open: why does the brain need to generate experience to do any of this? We can measure the neural activity. The phenomenology — the fact that it feels like something — remains its own question.
The social ones are weirder than they look
Yawning seems trivial until you learn that the oxygen-intake theory was tested and discarded. (Controlled studies showed oxygen levels don't meaningfully affect how often people yawn.) The leading current hypothesis is brain temperature regulation — the jaw stretch and blood flow changes around the skull may cool neural tissue slightly, nudging cognitive performance back up when attention starts to slip.
Then there's contagious yawning, which is where it gets interesting from a research-reliability standpoint. You've probably heard that yawning spreads because of empathy — that it's a mirror-neuron social thing. That story has had a complicated replication history. What a 2021 study published in Scientific Reports actually found is more specific, and frankly more fascinating: people who score high on psychopathic traits are less likely to yawn contagiously. That's a narrower, testable claim — and a much better story about how social sensitivity maps onto a reflex. It's also a good reminder that "linked to empathy" and "here's the precise mechanism" are not the same sentence.
Laughter has a similar gap between popular understanding and the research. The neuroscience researcher Robert Provine spent years documenting that people laugh far more frequently in social settings than alone — and that laughter is less about humor than about social signaling. The brain coordinates emotional centers, motor regions, and social processing areas all at once when we laugh. It likely evolved as a bonding mechanism: a fast, wordless way to signal that a situation is safe, that this group is okay, that you can relax. Unexpected outcomes that resolve harmlessly, tension that breaks cleanly — these seem to be the universal triggers, across cultures, even when the specific content of what's "funny" varies enormously.
And yet: no complete biological explanation for why this particular vocalization, this particular physical response, became the signal.
Blushing is the most philosophically strange one to me. It's an involuntary advertisement of vulnerability. Blood vessels in the face dilate automatically — you cannot stop it, you cannot fake it convincingly — and the result is a visible signal that communicates embarrassment, awareness of a social violation, emotional transparency. The theory is that this involuntariness is the whole point: because it can't be faked, it's a credible signal of genuine remorse or honesty, which helps defuse social conflict faster than words can. You broke a rule; your face admits it before your brain can spin a defense.
Here's what I notice, though: most blushing research has been conducted on lighter-skinned populations, because the redness is easier to measure visually. How the signal reads across populations with different skin tones, or how the social interpretation of blushing varies across cultures, is underresearched. The mechanism is universal; the science of it isn't quite yet.
The evolutionary leftovers
Goosebumps are pure vestigial reflex. In animals with thick fur, the arrector pili muscles raising each hair follicle serve real purposes: trapping insulating air, making the animal look larger to a threat. Cats do this. It works. Humans lost most of our body hair and kept the reflex anyway, which now fires during cold and during emotional experiences — powerful music, awe, nostalgia. The nervous system seems to have repurposed a temperature-and-threat circuit for emotional intensity. Why? Still an open question. The body is running ancient software on newer hardware and occasionally the processes just... fire.
Tickling is in this same category of we understand the parts but not the assembly. The brain actively predicts and suppresses the sensory response to self-generated touch — which is why you can't tickle yourself effectively. That prediction system is genuinely useful: it helps the nervous system distinguish "I touched myself" from "something external touched me," which matters a lot if the external thing might be dangerous. The laughter that comes from tickling resembles social play signals more than distress signals. Beyond that, the picture gets complicated enough that I'll stop before I overstate what the evidence supports.
The one that keeps me up at night (productively)
The inner voice.
Many people experience a continuous internal monologue — language areas of the brain activate, subvocalization happens, thoughts arrive pre-packaged in words. Researchers think this developed alongside spoken language; as humans evolved external communication, the brain started running language systems internally too. Makes sense.
Except: not everyone has this. Some people think primarily in images, in emotions, in abstract spatial concepts. No running commentary. The inner voice isn't a universal feature of human cognition — it's one cognitive style among several, and those differences appear to influence memory, learning, and creativity in ways that are still being mapped.
This is where my biotech-and-genetics brain goes alert. Because "inner voice" variation has real implications for how we understand neurodivergence. Autistic individuals, people with aphantasia, people with ADHD — cognitive diversity isn't just about behavioral outputs, it's about the actual texture of inner experience. When research assumes a verbal inner monologue as the default, it builds models of cognition that don't generalize. That's not just an academic gap. It shapes how tools get designed, how learning environments get structured, how neurological conditions get assessed.
We can measure the brain activity associated with inner speech. We cannot yet fully explain why some minds narrate constantly and others don't — or what that means for the people inside those minds.
Here's the thing that actually unsettles me: science has gotten very good at mapping the what of human experience — the neural correlates, the evolutionary pressures, the physiological mechanisms. The gap that remains isn't in the data. It's in our understanding of why subjective experience exists at all, and why it varies so dramatically between individuals who are running the same basic biological hardware.
We've measured the dream. We haven't explained why it feels like anything.
Mei Zhang covers biotechnology, genetics, and the future of medicine for Buzzrag.
More Like This
Black Hole Paradox: Are Reference Frames the Key?
Exploring how reference frames might resolve the black hole information paradox.
How Maxwell Unified Electricity and Magnetism
A compass needle twitched in 1820 and set off a chain of discoveries that now powers every wireless signal in your life. Here's the physics behind it.
Exploring Cosmic Time Delays and Dark Energy
Time delay cosmography may unveil dark energy mysteries, resolving Hubble tension with new cosmic insights.
What Color Is Monday? Inside the Brain That Sees Time
Synaesthesia isn't imagination—it's perception. Cognitive psychologist Mary Jane Spiller explains how some brains create colors, tastes, and shapes most of us never see.
Brainwaves and Communication: Beyond Words
Exploring how brainwave tech could revolutionize emotional communication.
Intelligence Beyond the Brain: Michael Levin's Radical Rethink
Biologist Michael Levin argues intelligence exists on a continuum reaching down to cells and molecules. Here's what the evidence actually supports—and where it gets murky.
What Rocks Know That Scientists Forgot
Earth scientist Anjana Khatwa argues indigenous knowledge and geology tell the same stories. Mei Zhang on why that matters—and who decides which stories count.
California's Heat Waves Are Getting Deadlier at Night
Scripps scientist Sasha Gershunov explains why California's heat waves are shifting—more humid, more nocturnal, and more deadly than ever before.
RAG·vector embedding
2026-07-11This article is indexed as a 1536-dimensional vector for semantic retrieval. Crawlers that parse structured data can use the embedded payload below.