The scene is deceptively simple. Two knots lie before you, their loops and tangles familiar to anyone who’s ever tied a shoelace or secured a boat. One knot is reliable and strong. The other? A treacherous trick ready to unravel under the slightest strain. But can you tell which is which?

Most people can’t, according to new research. In a study, Sholei Croom and Chaz Firestone from Johns Hopkins University reveal a perplexing blind spot in human intuition: our inability to judge the strength of knots. Despite knots being woven into the fabric of daily life, our minds stumble when tasked with predicting which ones will hold and which will slip.

“People are terrible at this,” said Firestone, a researcher who has been studying human perception for years. “Humanity has been using knots for thousands of years. Yet you can show people real pictures of knots and ask them for any judgment about how the knot will behave, and they have no clue.”

The Knotty Problem of Intuitive Physics

Intuitive physics is the everyday mental ability to predict how things will behave — like knowing that a tipped glass will spill or that a stack of blocks will topple if pushed. For decades, psychologists have marveled at both the strengths and failures of this skill. While early research highlighted glaring errors — people often predicted a ball dropped from an airplane would fall straight down instead of along a curved trajectory — more recent studies have shown that we are remarkably adept at judging physical scenes, especially when the scenarios are realistic.

In most cases, our intuitive physics is pretty good. We know what happens when we knock over a tower of blocks or pour water into a cup. But this new study uncovers a new and peculiar exception: knots.

The idea for the study came from Croom, a Ph.D. student in Firestone’s lab. An avid embroiderer, Croom once looked at the messy knot of threads on the back of their own needlework and couldn’t make sense of it. The confusion sparked a hypothesis: Could knots represent a unique weakness in our intuitive understanding of physics?

To test this, Croom and Firestone designed a series of five experiments. They presented participants with pairs of knots and asked them to pick the stronger one. The knots looked similar, but they had very different strengths. For instance, one type is the reef knot, one of the strongest basic knots. On the opposite end, there’s the aptly named grief knot, which can unravel with just a gentle pull. They also worked with granny knots and thief knots.

Participants consistently picked the wrong knot, believing the weakest knots were the strongest. And for many among the few who guessed correctly, their reasoning was off the mark. They focused on visual features that had nothing to do with the knot’s actual strength.

The researchers even tried to make the task easier. They showed participants rotating videos of the knots. They displayed diagrams of how each knot was tied. But no matter the approach, people’s judgments didn’t improve.

“We tried to give people the best chance we could,” said Croom. “It didn’t help at all — if anything, people’s responses were even more all over the place.”

A Universal Blind Spot

Knots are ancient tools, appearing in artwork from Ancient Egypt, Greece, and China. Evidence suggests that humans have been tying knots since before they harnessed fire or invented the wheel. Neanderthals, too, are thought to have made cordage.

And knots matter. The difference between a secure knot and a faulty one can mean the success or failure of a climb, the safety of a boat, or the security of a load.

The study’s findings challenge the notion that all intuitive physics can be explained by a general-purpose mental “simulation engine” — a model of the world that we use to predict physical outcomes. This “engine” works surprisingly well for many scenarios, such as predicting the fall of a block tower or the splash of a liquid. But knots seem to short-circuit it.

One possible explanation is that our brains are better suited to reasoning about rigid objects — blocks, balls, and beams — than soft, pliable ones like ropes and strings. Simulating the behavior of flexible materials is more computationally demanding, even for modern computer graphics software. Perhaps our mental models prioritize speed and efficiency over perfect accuracy, making knots a bridge too far for our intuitive physics.

Could experience make a difference? After all, sailors and rock climbers rely on their knowledge of knots daily. The study didn’t test experts, but the researchers suspect that training could improve knot intuition.

If experience helps, that would suggest knots are genuinely counterintuitive, but that’s something for another study to answer. And if so, knots might offer a new way to study how we learn physical concepts that defy our initial understanding. They could become a testbed for “intuitive physics training,” revealing how we can expand our mental models through practice.

For now, the research leaves us with an unsettling realization: the world is more tangled than we think.

“It’s a nice case study into how many open questions still remain in our ability to reason about the environment,” Croom said.

Think you could do better? Take the same tests the study participants took on the study’s website.

The findings appeared in the journal Open Mind.