Let’s get this out of the way right now: great apes aren’t failing at being human. They’re excelling at being apes — and that’s precisely what makes their cognition so hard for us to grasp.
For decades, comparative psychology ran on a simple premise: humans sit at the top of a ladder, and other animals climb from bottom to top by getting less wrong. Great apes? Closest rung. Then monkeys, then dogs, then rats — each step down was a little less cognitive oomph. But understanding these behaviors often requires looking at how they evolved (see: The Collaborative Apes: How Evolutionary Biology Explains Evolved Altruism and Community Trust).
Manuel Bohn and his colleagues at Leuphana University of Lüneburg hit us with the real gut punch: maybe the problem isn’t that apes get it wrong. Maybe the problem is that we keep asking questions designed only for human brains.
Their latest work — a 2026 study published in Psychological Science and building on earlier findings in Nature Ecology & Evolution — tracked 48 apes across four species (chimpanzees, bonobos, gorillas, and orangutans) through six cognitive tasks over 18 months. What they found wasn’t subtle: cognition in great apes isn’t just variable across individuals — it’s structured differently. Not as a scaled-down version of human cognition. Not as a broken mirror. As an entirely different architecture.
This isn’t just about apes failing our tests more often. It’s about the fact that high performance on one social-cognition task (say, following where someone is looking) tells you nothing about performance on another (like understanding communicative cues). In humans? Social abilities tend to cluster. Not so in apes — their cognitive worlds don’t sort themselves the same way.
So why does this matter? Because if ape cognition doesn’t map onto our mental categories, then every time we cite “ape-like” reasoning in infants or compare ourselves to chimps as though they’re a simpler version of us, we’re missing the point entirely.
Here’s what I’m saying: if you want to understand ape intelligence, stop asking how close they get to us. Ask what questions they are answering.
The Three Myths About Ape Cognition (And Why They’re Falling Apart)
Let’s dismantle a few myths before we go much further.
Myth #1: Ape intelligence is just a degraded version of human intelligence.
Nope. The data doesn’t support it, and here’s why that matters: when scientists assumed apes were running on a stripped-down human OS, they only designed tests suited to that OS — things like mirror self-recognition, button-pressing for symbolic communication, or false-belief tasks. Fail on those? You’re labeled cognitively limited. But what if your species evolved different priorities — say, spatial memory for navigating hundreds of Foraging paths in dense forest, or social strategy calibrated to fluid group compositions? Your "test failure" is just the wrong test altogether.
Myth #2: Cognition within a species is uniform.
Bohn’s team tracked each of their 48 apes across multiple tasks — not to find the average ape, but to map the distribution. What they saw? Huge individual variation in how apes approached each task. Some bonobos excelled at attention-following but flopped on spatial reasoning; orangutans reversed the pattern. Group membership, rearing history, and prior research experience were strong predictors of success. That’s not noise — that’s signal telling us cognition is deeply contextual.
Myth #3: Social and non-social reasoning should correlate.
Here’s where the real whiplash happens. Human cognition clusters: people who do well on theory-of-mind tasks usually do well on other social tasks, and the same for mechanical reasoning. Apes? The study found no correlation between performance on different social tasks (e.g., following gaze vs. interpreting communicative cues). Meanwhile, non-social reasoning tasks did cluster — suggesting a coherent executive function suite that’s independent of social tuning.
That mismatch between human and ape cognitive architecture shouldn’t surprise us if we actually stopped to ask what cognition is for in each species. For humans, social coordination was a select pressure — our minds are wired to read others as quickly as we read the world. For apes like orangutans — famously solitary — cognitive energy got routed toward ecological problem-solving instead. The structure of their intelligence reflects the ecology they evolved in, not ours.
Beyond the Mirror Test: Why Standard Tools Betray Ape Intelligence
Let’s talk about the mirror self-recognition test. You’ve seen it: a red dot painted on an animal’s forehead; if they touch it while looking in the mirror, they’ve supposedly passed the test of self-awareness. Great apes — check. Humans older than ~18 months — check. Elephants, magpies, maybe dolphins? Occasional passes.
But here’s the thing: Bohn and his colleagues didn’t run a single self-referencing task in their 2026 study. Why? Because for solitary orangutans or highly hierarchical bonobo troops, mirrors don’t capture the full picture — and more importantly, they’re a human-centric metric.
The authors put it bluntly: "We don’t have assessment tools that have been particularly built to assess the different aspects of great ape cognition." The standard battery — gaze following, communication cue interpretation, detour tasks, button-pushing for food — was built on assumptions about how intelligence should work. Not how it does work in apes.
That’s critical. For example, one task measured whether an ape could follow a human’s pointing gesture to locate hidden food. It tests social communication and cue interpretation — textbook "social cognition." But it says nothing about whether the ape can remember where it last found food, or how many different routes are optimal on a given day. Those tasks do exist (they were part of the same study), but they’re lumped under "non-social reasoning," which suggests we’ve already baked in a false dichotomy. What if spatial memory and gestural communication aren’t supposed to talk to each other at all? What if they’re different subsystems operating in parallel?
Bohn’s team found that performance on social tasks didn’t correlate across individuals — high scorers on gaze-following weren’t necessarily good at interpreting pointing. But non-social tasks did correlate, suggesting that a unified reasoning subsystem exists — it’s just not tied to our pet theory of what "social" means in an ape context.
It’s like evaluating two languages by testing how well speakers can transliterate the alphabet — you’ll conclude they’re both flawed. But neither language wants to transliterate the alphabet; their grammar, idioms, and social constraints were built for different purposes. Stop testing writing when you’re trying to compare speech.
Developmental Paths, Not Linear Climbs: Why Age Doesn’t Predict Ape Ability Like It Does Human
Here’s a curveball most people don’t see coming: the 2026 study didn’t just track performance across tasks — it tracked how each individual ape’s abilities changed over 18 months. The short answer? Stability, but not the kind we expected.
For humans, certain cognitive milestones (language, theory-of-mind emergence) follow a fairly tight developmental timeline. Deviations raise flags — early or late, they often correlate with later outcomes.
Great apes? No such lockstep. Individual variation wasn’t random noise; it was structured and stable within individuals over time. Some bonobos excelled at social tasks from day one and held that edge; others got better with exposure to research settings. Age didn’t guarantee improvement — rearing history and early social experience mattered more.
Manuel Bohn put it starkly: “We very much have these kind of developmental and individual differences perspectives for humans. And so, we thought this was clearly missing in great apes.”
That’s the key — comparative psychology didn’t lack data on ape development; it lacked a theoretical vocabulary for individual difference. The study forces us to treat apes as if they have life histories that shape cognition, not just evolutionary blueprints. An orphaned orangutan raised in a human care facility will think differently — not worse — than an orphaned chimpanzee reared by peers in a semi-wild sanctuary. Their cognitive architectures aren’t broken or incomplete; they’re different paths up the same mountain, viewed from opposite sides.
What does this mean for the future? Longitudinal studies. Lots of them. Tracking cognitive development across apes in naturalistic and semi-natural settings, alongside those in controlled research labs. Comparing not just what they can do, but how and why they choose to do it.
The implication is uncomfortable: much of what we thought was core to ape intelligence may just be their response to our unnatural testing conditions. If an ape fails a test, is it because they lack the ability — or because we asked the wrong question at the wrong time in an unnatural setting?
That’s not just inconvenient. It demands humility.
The Point of It All
Here’s the real gut punch: if great apes don’t think like us, then much of what we thought was foundational to human uniqueness — cooperation, empathy, theory-of-mind — may not be uniquely human in the way we imagined. That’s not a downgrading of human intelligence; it’s a recentering of the problem space.
For decades, cognitive scientists have been hunting for the “cognitive niche” that made us human — some magic ingredient in our brains that lets us outthink other species. But if apes operate on entirely different architectures, then the search needs a complete overhaul. We’re not just looking for our cognitive niche; we’re asking what other niches of thinking even exist, and how many we’ve overlooked because our instruments were built to only recognize ourselves.
And that’s the beauty of Bohn and his colleagues’ work. They’re not declaring victory; they’re handing us a new compass. Their paper isn’t an endpoint — it’s a call to arms for anyone willing to question the bedrock assumptions of comparative psychology.
It’s time we stopped trying to force ape cognition into human-shaped boxes and started listening to what the apes are actually telling us — in their own terms. Their cognition may not mirror ours, but that doesn’t make it less valuable. It makes it more valuable: a window into possibilities we hadn’t even dreamed of.
So let’s stop pretending apes are just less-smart humans. Let’s stop calling their behavior “failures” when they’re actually executing a different algorithm entirely. And let’s build the tools, ask the questions, and write the papers that honor their intelligence on its own terms.
The point isn’t to prove how smart apes are — it’s to prove how we haven’t yet learned how to see them clearly.
Great Apes Don’t Think Like Us — And That’s the Point © 2026 Felix Sterling. First published in Biology & Natural History, General domain.