Sign up for the Slatest to get the most insightful analysis, criticism, and advice out there, delivered to your inbox daily.
It’s one of the deepest mysteries of the human mind: Do we all see the world the same way? In 1956, three social scientists set out to answer this question. From their offices at Northwestern University, Donald Campbell and Melville Herskovits teamed up with Marshall Segall, of Syracuse University, to coordinate an ambitious new investigation. They sent researchers on a mission to societies near and far, urban and rural: A gold mine in Johannesburg; a community of foragers in the Kalahari Desert; the Philippine island of Mindoro; and even their own college campus in Evanston, Illinois.
Tucked into each of their suitcases was a booklet of drawings, including 12 examples of a prominent figure called the Müller-Lyer illusion. You may have seen it before: When two identical horizontal lines are capped with arrowheads pointing either inward or outward, the line with inward-facing arrowheads looks longer, even though it’s not.
At least, that’s how the illusion works here in the United States. But what about elsewhere? When the study was completed in 1961, the results shocked the scientific community: Not everyone was susceptible to this seemingly obvious illusion. While students in Illinois tended to report the top line as longer, Zulu pastoralists in South Africa had a much weaker response, barely experiencing the illusory effects. And the San foragers of the Kalahari seemed not to see anything remarkable at all, just two lines of equal length, as if the illusion simply wasn’t an illusion for them. This wasn’t like finding different shapes in a complicated abstract drawing, or even having different interpretations of a novel. It was as though something about human vision was fundamentally different from culture to culture. How was this possible?
Before we unravel the mystery, let’s first explain why this matters to psychology researchers like us. Psychology aims to capture enduring truths about the human mind. But in the vast majority of cases, psychological studies explore narrow subject pools, such as college students taking introductory courses. The reason is mostly convenience; such test subjects are available in large numbers on college campuses (where most researchers are), and they are happy to give a little bit of their time in exchange for a bit of compensation, such as course credit. But there is good reason to worry about doing research this way. Who is to say that attitudes toward nationalism, or the prevalence of ADHD, or the best tricks for sticking to a new diet—or really, anything else psychologists study—generalize from one very specific group to the whole of humanity? Psychology has been around for a while, but it wasn’t until relatively recently that the field began to reckon with these concerns more seriously and systematically, under the banner of a clever acronym proclaiming most psychology research subjects are “WEIRD”: Western, Educated, Industrialized, Rich, and Democratic. The idea has gained steam over the past 15 years, and researchers are trying to do better.
But which psychological principles should we expect to vary across groups? Which findings are cultural creations limited to WEIRD research subjects, and which reflect our common humanity—true for us and everyone else? Here’s where the illusion study becomes so consequential. It’s one thing to suppose that political attitudes or dietary practices differ around the world; any tourist can attest to this. But perception itself? Could it really be that our very eyes tell us something different about the world depending on where we grew up?
Segall and colleagues thought so, and even went a step further. They proposed that Americans see the illusion only because of their overexposure to carpentry—straight lines and sharp angles that are present in urban environments and the Müller-Lyer figure, but are less prevalent in the Kalahari. Raise someone in an environment without boxy structures or rectangular windows and doors, the idea goes, and the illusion won’t exist for them. Contemporary anthropologists have further popularized this view, arguing that “the Müller-Lyer illusion is a kind of culturally evolved by-product.” Call it the Cultural Byproduct Hypothesis.
This result and the theoretical apparatus built around it are now essentially part of the psychological canon. They are often taught to Psych 101 students as both a fascinating discovery about visual processing and a cautionary tale about unwarranted assumptions of universality. We shouldn’t assume that others experience the world like we do—that much seems true and even uncontroversial. And what better piece of evidence for this lesson than discovering that we literally see the world differently depending on where we grow up.
We weren’t so sure about all this. In a new paper, we revisited over 100 years of research on perception and came to nearly the opposite conclusion: This particular visual illusion, and many other aspects of our perceptual system, arise from deep within us, are likely common to humans across the globe, and certainly aren’t mere cultural creations. As impressive as the cross-cultural studies seem (more on that in a moment), there are powerful clues suggesting that the Cultural Byproduct Hypothesis can’t really be true.
Why not? For one thing, lots of other animals see the Müller-Lyer illusion. If you train a guppy to swim toward longer lines (yes, a real thing that can be done!), and then you show it the Müller-Lyer figure, it will swim towards the top line—suggesting that the guppy sees that line as longer. This is true for a veritable zoo of nonhuman creatures, including horses, parakeets, monkeys, and lizards, who all see the illusion as well. Did the guppies’ culture create the illusion for them too? Seems unlikely.
Another clue: The illusion doesn’t even need to be made of straight lines in order to work. There are versions of the Müller-Lyer illusion composed entirely out of curves, or just groups of dots; there’s even a version that uses people’s faces. That observation calls into question the purported link with carpentry, since the whole idea was that the illusion relies on features such as straight lines present in precisely constructed environments.
Perhaps the strongest clue of all is also the most remarkable. A humanitarian and scientific project called Prakash recently offered free corrective surgery to children in North India who were born with congenital cataracts—cloudy lenses that prevent light from entering the eyes, blinding them since birth. With new, clear, artificial lenses, these children were now able to see for the first time in their lives. Astoundingly, when shown the Müller-Lyer illusion—mere hours after recovering from their operations—they reported the top line as longer than the bottom line. Not only had these children never seen carpentry, they had never seen anything.
And yet they still experienced the illusory effects of the figure.
All this and more suggests that the illusion really is a result of who we are, not the buildings we happen to grow up next to. Despite our differences, we really do see the world similarly, sharing something in common with humans across the globe and throughout history.
But wait: If the evidence against the Cultural Byproduct Hypothesis is so overwhelming, then why did those three social scientists find different results in their cross-cultural study, with some groups seeming not to see the illusion at all?
First, the cross-cultural studies were never all that consistent with one another. When we looked even deeper into the published record, it was surprisingly easy to find contradictory results: A study from the early 20th century found that a jungle-dwelling population in India showed a stronger illusion than a pastoralist community in the same country, and another study from 1970 similarly found stronger effects in a rural population indigenous to South Africa than in a nearby urban community. Even Segall, Campbell, and Herskovits’ more famous study contained contradictions within itself. For example, one of the samples showing the weakest illusion of all was a group of mineworkers. Mines, of course, are highly constructed, carpentered environments—exactly the kind of environment that should produce a large illusion, according to the theory.
Second, studies of this sort are open to bias. For starters, you have to translate the task instructions into a local dialect, which is not always easy; many of these researchers even worried about this difficulty, writing that they “were not completely sure of exactly what was communicated to the respondents at all times.” There are also biases introduced by experimenters who know something about the research hypotheses and might—consciously or not—tilt the results accordingly. In another telling passage, one experimenter wrote that he “developed very strong expectations of what answer the respondents should give to a given item, and if a respondent gave the other answer, there was the impulse to correct the respondent to ask him to reconsider.” There is even evidence that some of the cross-cultural data were excluded if the reported illusion was too strong, and that this masked some findings that would have challenged the overall narrative. It may well be, then, that the illusion was present in these diverse populations, and the experimenters simply failed to fully capture it in their measurements.
What does that mean for us today? Expanding psychological research to capture the diversity of human experience is a tide that lifts all boats, and is a project we wholeheartedly support (and engage in ourselves). But some experiences may well be universal.
