imagindi  ·  mind & cognition

Evolutionary Biology·Neuroscience·Philosophy of Religion

The Brain We Have, Not the Brain We Imagine

If a mind this intricate were handed down by a perfect designer, why does it come with a blind spot, a spine that buckles under its own upright walk, and a birth process still dangerous enough to kill? The newest evolutionary biology suggests these aren't oversights in an otherwise flawless plan. They're the signature of a process with no plan at all.

·  ·

The human brain weighs about three pounds — 1.3 to 1.4 kilograms — and somehow that's enough to generate every poem, proof, and civilization we've ever produced. It runs on roughly 86 billion neurons, a figure established by Suzana Herculano-Houzel's team using a cell-counting method nicknamed "brain soup," though a 2025 re-analysis in Brain argues the real number is better described as a range than a fixed point.¹² Those neurons form connections numbering somewhere in the hundreds of trillions — estimates vary widely depending on method, and no one has an exact count. What's not in dispute is that nothing humans have built comes close to matching the brain's efficiency per watt.

That efficiency is organizational as much as physical. When you look at a tree, and when you close your eyes and picture one, you aren't switching to a separate imagination system — you're recruiting much of the same visual machinery twice. A widely cited 2019 review in Trends in Cognitive Sciences found that imagined and perceived stimuli produce strikingly similar activity patterns across visual, parietal, and frontal cortex, differing mainly in whether bottom-up sensory input is present.¹² A 2023 fMRI study went further, showing parietal cortex actually carries a stronger representation during imagery than early visual cortex does, reversing the usual perception-dominant pattern.¹³ And a 2025 Neuron paper mapped how the brain tells an imagined scene apart from a perceived one — largely by tracking the strength and source of that shared signal, not by routing it through separate hardware.¹⁴ The brain doesn't keep a dedicated inner-vision department in reserve. It reuses one system for two jobs, and tracks the difference by how the signal arrives.

The brain isn't a locked vault with 90% of the rooms dark. It's fully lit, all the time — just organized so that most rooms stay quiet unless a very specific key turns.

01What Savant Syndrome Actually Shows

It's tempting to read savant syndrome as proof that a genius is sleeping inside every skull. That's not quite what the evidence says, and it's worth being precise here, because the "untapped 90% of the brain" idea is a myth — brain imaging shows virtually the entire organ is active over the course of a day, just not all at once for any single task.

What savant syndrome does show is narrower and stranger. A 2023 systematic review found that exceptional savant skills — in music, calendar calculation, art, memory — appear in roughly one in ten people with autism spectrum disorder, and can also emerge after brain injury in people with no prior sign of them.³ A related line of work looks at metabolic and network-level changes in the brain that accompany these skills, suggesting they ride on unusual patterns of connectivity and inhibition rather than a simple surplus of raw material.⁴ The leading hypothesis, sometimes called paradoxical functional facilitation, is that damage or atypical development can release specific, narrow capacities that are normally kept suppressed — not that everyone is secretly capable of anything, given the right unlocking.

The perception–imagination overlap points to the same underlying picture. The brain's limits are rarely a matter of unused hardware sitting idle, waiting for a key. They're a matter of how already-active systems are gated, weighted, and coordinated moment to moment — which circuits get priority access to shared resources, and when. "More brain" isn't sitting in reserve. Better coordination of the brain that's already fully switched on is what separates ordinary cognition from its rarer, stranger edge cases.

02A Machine Full of Failure Points

Whenever something looks engineered — a jet turbine, a Mars rover — we assume an engineer. The brain is more sophisticated than anything humanity has built, so it's a short step to assuming a designer built it too.

But complexity is a question, not an answer. And if the brain were the work of a perfect, all-knowing designer, its failure modes are hard to explain. It is vulnerable to stroke, aneurysm, epilepsy, Alzheimer's and Parkinson's disease, tumors, infection, and autoimmune attack. A few minutes without blood flow can destroy millions of neurons permanently. None of this reads like flawless engineering. It reads like a system with a long, improvised history.

There's a logical snag here too. If complexity always implies a designer, the designer — being at least as complex — should need one as well. Insisting the universe can't exist unexplained while its explainer needs no explanation is an asymmetry that has to be justified, not assumed. And even if design were granted, it wouldn't say which of humanity's thousands of gods did the designing. Complexity alone can't adjudicate between competing theologies.

03What Evolution Predicts — and Where the Evidence Gets Contested

Evolution doesn't design. It edits. Small changes accumulate, useful ones persist, and the process has no destination beyond differential survival. That produces something closer to a repeatedly patched structure than a blueprint executed from scratch — which is exactly what shows up across human anatomy.

It's also worth being precise about why traits persist, because natural selection isn't the whole story. Motoo Kimura's neutral theory of molecular evolution, now a mainstream part of the field, holds that a large share of genetic variation is fixed by random genetic drift — the chance ups and downs of allele frequencies in finite populations — rather than by any selective advantage.¹⁰ A trait, or a stretch of DNA, can spread through a population simply because of who happened to survive and reproduce, not because it made anyone fitter. Founder effects and population bottlenecks work the same way, locking in whatever variation a small founding group happened to carry.¹¹ None of this displaces selection as the main force behind adaptation — eyes, immune systems, and the brain's core architecture still bear its fingerprints. But it means "evolution produced this" is not automatically the same claim as "evolution optimized this." Some anatomical quirks may be neutral leftovers that selection never bothered to edit out, rather than compromises it was forced into.

The textbook examples are real, but several are more contested than the popular version admits, and it's worth being straight about that.

Contested
The retina. It's often cited as "wired backward," with light passing through neurons before reaching the photoreceptors, creating a blind spot. That much is true. What's less often mentioned is a 2009 biophysics paper showing this inverted arrangement is actually a superior space-saving solution in small eyes, and may have helped — not hindered — the evolution of sharp vertebrate vision.⁵ The blind spot is real; "obviously bad design" is not the settled conclusion the popular version implies.

Contested
Childbirth and the pelvis. The classic "obstetrical dilemma" holds that a wider birth canal for easier delivery would compromise the narrow, efficient pelvis bipedal walking needs. Recent genetic and biomechanical work complicates this. A 2025 genome-wide study in Science found the width of the birth canal and the width of infant heads are genetically correlated — a coevolutionary solution, not a straightforward trade-off — while newer biomechanical studies suggest a wider pelvis wouldn't meaningfully hurt walking efficiency at all.⁶⁷ The danger of human childbirth is real and still costs lives; the specific evolutionary story behind it is actively being revised.

Partially contested
The spine. Bipedalism does shape spinal disease risk — a 2020 study tied the vertebral shape most associated with spondylolysis directly to the degree of an individual's bipedal adaptation.⁸ But a broader recent synthesis argues that for most people, the spine's basic design is functionally sound for upright walking, and modern back pain owes more to sedentary posture and lifestyle than to bipedalism itself.⁹ Evolution built a spine that mostly works; modern chairs are doing some of the damage evolution gets blamed for.

Wisdom teeth crowding out of undersized jaws, and the airway crossing paths with the esophagus, remain straightforward cases: neither serves any function, and both are easiest explained as leftovers of an ancestral anatomy repurposed rather than redesigned.

04A Brain Built From Layers, Not a Single Vision

The brain itself carries this same patchwork logic. Circuits handling fear and basic survival sit alongside newer cortical regions built for planning and abstract reasoning, and much of human behavior is a negotiation between the two rather than the output of one coherent design. Logic and impulsiveness, compassion and tribalism, creativity and bias — these aren't bugs layered onto a clean system. They're what you'd expect from a brain built in stages, by a process with no final draft in mind.

05What This Is Good For

If the real limit is coordination rather than locked capacity, that changes what's worth doing about it.

Education built around "unlocking hidden potential" is chasing the wrong target. What the evidence favors instead is training the coordination itself — retrieval practice, interleaved study, deliberately switching between modes of thinking — since the ceiling is how well existing circuits are marshaled, not how much spare capacity is sitting in reserve.

Creativity techniques that work — sketching before you know what you're drawing, changing physical location, altering sensory context — make more sense in this light. They aren't bypassing a locked vault. They're feeding the same shared perception-imagination system a different signal to reorganize around.

AI research gets a concrete architectural hint from this: the fact that perception and imagination in the brain diverge mainly in signal source, not in which module handles them, argues for building shared generative/predictive cores rather than bolting a separate "imagination module" onto a perception system.

Public science communication has a cheap, immediate fix — retire "we only use 10% of our brains" for good, and replace it with the less catchy but accurate version: full engagement, imperfect coordination, and gains that come from practicing that coordination rather than searching for a hidden switch.

Research funding arguably has more to gain from mapping how signals are gated and routed between regions that are already active than from the search for idle neural real estate — because the data increasingly suggest there isn't much of that left to find.

None of this makes the brain less remarkable — if anything, more. That an undirected process spanning billions of years produced something capable of asking whether it was designed at all is itself the more interesting fact. We are not the output of a perfect blueprint. We are the current draft of a very long revision — imperfect in ways that are, on close inspection, part of how a mind like this became possible at all.

References

1. Azevedo, F.A.C. et al. (2009). Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. Journal of Comparative Neurology, 513(5), 532–541. onlinelibrary.wiley.com/doi/10.1002/cne.21974
2. (2025). Eighty-six billion and counting: do we know the number of neurons in the human brain? Brain, 148(3), 689–697. academic.oup.com/brain/article/148/3/689/7909879
3. Park, H.O. (2023). Autism spectrum disorder and savant syndrome: A systematic literature review. Journal of the Korean Academy of Child and Adolescent Psychiatry, 34(2), 76–88. jkacap.org/journal/view.html?uid=799
4. Onin, I., Hanoglu, L., & Yulug, B. (2023). The savant syndrome: a gift or a disability? A deeper look into metabolic correlates of hidden cognitive capacity. Endocrine, Metabolic & Immune Disorders - Drug Targets, 23(2), 250–253. pubmed.ncbi.nlm.nih.gov/35400329
5. Kröger, R.H.H., & Biehlmaier, O. (2009). Space-saving advantage of an inverted retina. Vision Research, 49(18), 2318–2321. pubmed.ncbi.nlm.nih.gov/19591859
6. Stansfield, E. et al. (2025). The genetic architecture of and evolutionary constraints on the human pelvic form. Science, 388. science.org/doi/10.1126/science.adq1521
7. Wells, J.C. et al. (2024). There is an obstetrical dilemma: Misconceptions about the evolution of human childbirth and pelvic form. American Journal of Biological Anthropology. pmc.ncbi.nlm.nih.gov/articles/PMC10952510
8. Plomp, K.A., Collard, M., et al. (2020). Spondylolysis and spinal adaptations for bipedalism. Evolution, Medicine, and Public Health, 2020(1), 35–44. sciencedaily.com/releases/2020/03/200305132047.htm
9. (2026). Viewing low back pain through the lens of spinal evolution: Understanding the morphology and limits of the human spine. PLOS ONE. pmc.ncbi.nlm.nih.gov/articles/PMC12810782
10. Kimura, M. (1968). Evolutionary rate at the molecular level. Nature, 217, 624–626; overview at Nature Education Scitable, Neutral Theory: The Null Hypothesis of Molecular Evolution. nature.com/scitable/…neutral-theory
11. Understanding Evolution, UC Berkeley. The neutral theory. evolution.berkeley.edu/…the-neutral-theory
12. Dijkstra, N., Bosch, S.E., & van Gerven, M.A.J. (2019). Shared neural mechanisms of visual perception and imagery. Trends in Cognitive Sciences, 23(5), 423–434. sciencedirect.com/…S1364661319300592
13. (2023). Neural representations in visual and parietal cortex differentiate between imagined, perceived, and illusory experiences. Journal of Neuroscience, 43(38), 6508–6524. pmc.ncbi.nlm.nih.gov/articles/PMC10513072
14. Dijkstra, N. et al. (2025). A neural basis for distinguishing imagination from reality. Neuron. cell.com/neuron/…S0896-6273(25)00362-9

Imagindi · Mind & Cognition 14 sources