baboon

An olive baboon at the Seneca Park Zoo that took place in the Primate Portal experiments. Photo courtesy of the Rochester Institute of Technology.

The Human Mind’s Geometric ‘Superpower’ May Not Be So Human After All

In A Nutshell

  • A new study finds monkeys rely on the same basic mix of geometric reasoning as humans, not just simple visual pattern-matching as long assumed
  • Eight monkeys, dozens of preschoolers, and adults from an indigenous Bolivian community were tested on matching geometric shapes
  • When shapes were rotated, monkeys leaned on abstract, rule-like reasoning even more than preschoolers did on the easier version of the task
  • The results suggest human geometric intuition grew from abilities shared across primates rather than a uniquely human mental ingredient

For decades, scientists believed humans exclusively held the mental gift of encoding geometric shapes through abstract, rule-based thinking. Only humans, the theory held, could represent a square as discrete properties: equal sides, right angles, perfect symmetry. Monkeys were thought to rely mainly on raw visual pattern-matching instead. A new study in PNAS challenges that assumption.

Researchers tested monkeys, preschoolers, and adults from an indigenous community on a shape-matching task and found monkeys’ choices were best explained by the same basic mix of geometric representations seen in humans. Rather than a sharp divide, the study reveals a smooth, overlapping continuum. Monkeys aren’t as sharp as adults, but their geometric thinking is far more sophisticated than credited, overlapping considerably with young children’s.

Most provocative: on tasks requiring recognition of tilted shapes, some monkeys leaned on abstract geometric reasoning even more than preschoolers did on the easier version of the same task. That’s a comparison across different conditions, not proof the monkeys are better at geometry overall, but it’s still a striking reversal of what scientists expected.

A 2021 Study First Suggested Humans Alone Grasp Geometric Rules

Scientists called the leading explanation the “symbolic singularity” view: humans have a unique mental system for encoding geometric rules almost like a checklist, parallel sides, right angles, equal lengths. This checklist lets humans instantly recognize a square no matter how it’s drawn. Other primates, lacking this system, were thought to depend only on continuous perceptual features rather than discrete rules.

A widely cited 2021 study backed this up, showing baboons performing worse than humans at picking the “odd one out” among geometric shapes. Baboons showed no advantage when the odd shape was highly regular; humans did, and researchers read this as evidence that symbolic geometric thinking was uniquely human. This new study takes direct aim at that interpretation.

primate portal
Jessica Cantlon (Carnegie Mellon University) and Caroline DeLong (Rochester Institute of Technology) watch as an olive baboon named Olivella interacts with the Primate Portal. Another baboon who frequently participates, named Kalamata, stands nearby. (Credit: Rochester Institute of Technology)

Monkeys, Kids, and Adults Faced the Same Shape-Matching Test

Monkeys, preschoolers ages 3 to 6, and adults completed a matching task: view a sample shape, then pick which of two options matched. Correct answers earned a food pellet for monkeys, a star sticker for kids, or verbal confirmation for adults.

Eight adult monkeys participated: four rhesus macaques at Carnegie Mellon University and four olive baboons at the Seneca Park Zoo in Rochester, New York. Fifty-eight American preschoolers were recruited through Carnegie Mellon, along with 79 adults from Tsimane’ communities near San Borja, Bolivia, a farming and foraging society with little formal schooling. That last group mattered because it let researchers test whether geometric thinking depends on schooling, or runs deeper than that.

Rather than tallying right and wrong answers, researchers used statistics to determine how each group mentally represented shapes, comparing behavior against three models: raw pixel similarity, holistic shape recognition, and a symbolic model encoding properties like symmetry and right angles.

Across every group, behavior was best explained by the two sophisticated models, not the primitive one. Monkeys were not limited to basic visual matching, as the old theory predicted; they showed reliable sensitivity to the symbolic model too, though less strongly than adults. Monkeys and preschoolers, in fact, landed in largely overlapping territory, with no meaningful gap between the species.

“Our findings reveal that monkeys share geometric representations with humans, suggesting the origin of geometric intuition is shared among primates,” the authors write.

When Rotating Shapes Turned the Findings Upside Down

Its most arresting result came from rotated shapes: participants saw options tilted clockwise or counterclockwise and had to identify which still matched the sample, testing whether a shape reads as the same once turned on its side.

A subset of five monkeys tripled their reliance on symbolic-style reasoning once shapes were rotated, and preschoolers showed a similar jump. The monkeys’ jump was even bigger, suggesting it was the harder task, not the species, driving the shift.

This led researchers to question what “symbolic” thinking actually is. It turns out the symbolic model behaves almost identically to a simpler idea: recognizing a shape as itself no matter which way it’s turned. What looked like a uniquely human rulebook for geometry may really just be a sturdier version of that same basic skill, one monkeys have too, just less developed.

“Rotation invariance, rather than biological species, induced categorical restructuring of representational space,” the authors conclude.

baboon zoo
A little girl and an olive baboon interact with each other through the glass at the Seneca Park Zoo, where the Primate Portal is housed. (Credit: Rochester Institute of Technology)

The Roots of Geometric Intuition May Predate Humans

In these tasks, researchers found no clean break between the groups, only differences of degree that shifted depending on the task. Adult humans still showed the sharpest use of these skills overall, and the exact mental code behind it remains an open question. But the basic ingredients, it turns out, are not uniquely human, an ability shared far enough back in the primate family tree that geometric intuition looks less like a human invention and more like an inheritance.


Paper Notes

Limitations

The study used a relatively small number of nonhuman primate participants, eight total across two species, which limits how broadly the findings generalize across the animal kingdom. The symbolic model and the rotation-invariant visual model are highly correlated, making it difficult to cleanly separate the two explanations for behavior. The precise nature of the abstract, rotation-independent mental representations underlying performance in both species remains open, and the paper notes several alternative models have not yet been formally tested against primate data. Differences in prior task experience between species (monkeys had touchscreen experience but not geometry-matching specifically) could also influence performance comparisons.

Funding and Disclosures

This work was supported by NSF grants DRL2026416 and NSF 2148343, as well as NIH grants R01HD107715 and R01HD085996. The authors declare no competing interests. Previously published data from Sablé-Meyer et al. (2021) were used in parts of the analysis.

Publication Details

Paper Title: Continuity in Geometric Intuition between Humans and Monkeys | Authors: Jialin Li, Isabelle Boni, Logan R. Sandwick, Emily M. Sanford, Caroline M. DeLong, Li Wenjie, Margaret M. Henderson, Steven T. Piantadosi, and Jessica F. Cantlon | Affiliations: Department of Psychology, University of California, Berkeley; Department of Psychology, Carnegie Mellon University; Department of Psychology, Rochester Institute of Technology | Journal: Proceedings of the National Academy of Sciences (PNAS), 2026, Vol. 123, No. 30 | DOI: 10.1073/pnas.2532934123 | Published: July 20, 2026

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor

Leave a Comment