Russian Blues: Can Words Unlock Our Brains?

In 2007, Winawer and colleagues published evidence that Russians do not see the world as the rest of us do.

Feel free to add your own joke here.

Specifically, Winawer’s study showed differences between the colour-perceptual capabilities of Russian and English participants.

More simply: Russians could see more colours.

Just as interesting as the discovery was the explanation: that the difference between Russian- and English-speaking participants could be a function of their native tongue.

Composition 6 (Kandinsky, 1913)

It is difficult to describe a world without language.

A truism, sure. However, the relationship between language, perception and cognition might be more complicated than you expect.

The Sapir-Whorf hypothesis (also known as linguistic relativity) holds that the structure of a language has an effect on its speakers’ cognition. As such, the way in which we perceive the world is altered according to the language that we speak.

There are two broad schools of thought that describe strong (language determines and constrains thought) and weak (language influences thought) versions of the Sapir-Whorf hypothesis.

The strong version of the Sapir-Whorf hypothesis was originally proposed by Benjamin Lee Whorf himself.

Whorf focused his research on the Hopi language spoken by the Native American Hopi tribe. Notably, he claimed that the Hopi language did not have a word for time — and that its speakers lived in a timeless manner.

Critics have argued that Whorf’s claims were based solely on linguistic data. As such, the conclusions he drew about general aspects of cognition were likely overstepping the mark.

Further, later analysis suggested that Whorf was incorrect anyway — Hopi speakers do have a concept of temporality. Although this analysis further pointed out that the concept of the passage of time was, in fact, also present in the language.

Whorf (right, year unknown) was a student of Edward Sapir (left, c. 1910) — the linguist who initially outlined the idea of linguistic relativity. The two never formally proposed a hypothesis together, and the term bearing their names was proposed by another of Sapir’s students, Harry Hoijer.

Even beyond this inaccuracy, Whorf’s strong version of the hypothesis has fallen out of favour. However, that does not mean that the Sapir-Whorf hypothesis has been rejected entirely.

Instead, the debate has focused on its weak version — in which thought is influenced by language (though not constrained by it).

A particularly compelling debate has centred on the possible influence of language on colour perception.

Several studies have found that speakers of different languages show differences in their capacity to recall colours. If a language uses the same term to recall two different colours, its speakers are more likely to confuse those colours in a memory task than do participants whose language does separate the colours.

These experiments have themselves been criticised by those that hold a ‘universalist’ view. Steven Pinker argues that such differences in memory can be attributed to engaging in two different memory systems — that support recall of non-verbal visual images and verbal labels.

He argues that better memory would be expected in speakers with more verbal labels because “presumably two types of memory… are better than one.”

Pinker further argues that other tests are confounded by language biases that affect a participant's decision-making. As such it is not so much that a person is physically unable to perceive a difference between two stimuli, but that their choice to group or organise them might be influenced by different variables.

Universalists make another quite reasonable claim. Humans across all cultures and ethnicities share a biological colour perceptual system — beginning with the three types of cone cell in the retina. Given that this shared physiology exists, there is no particular reason to believe that humans will physically perceive the same stimuli differently.

There are people that have abnormalities in their cone cells. Most commonly, this emerges as a developmental problem in one or more of the three types of cone cell. This results in the person being unable to recognise a difference between certain colours. In rare cases in which the person has no cone cells at all, they are simply unable to perceive colour. Collectively, these conditions are known as colour blindness.

As a final brief aside, scientists have also identified a condition known as tetrachromacy — used to describe people that possess a fourth type of cone cell. In very, very rare cases this can lead to superhuman colour perception capabilities.

Pinker suggests instead that all human brains share a common language — Mentalese. He argues that spoken language is simply that — a translation of Mentalese that makes possible everyday communication.

Steven Pinker (2011)

And that leads us quite neatly back to the Winawer study — which implemented an objective perceptual task in an elegant solution to the problems identified by Pinker and others.

The experiment presented participants with a triad of coloured squares. Two squares were the same colour whereas the third square was a slightly different shade.

A figure is reproduced below from Winawer (2007) that illustrates the experimental set up. In short, the squares were arranged on the tips of a triangle. Participants were asked to identify which of the bottom two squares matched the top square (i.e. they had to pick the matching target and ignore the distractor).

The task was relatively challenging. We are not talking about two greens and one orange. Instead, the squares were varying shades of blue.

In English, we separate different shades of blue: royal, navy, sky, baby, and so on. However, these are not “different colours” in the same way that “green” and “orange” are different. Instead, they are varied forms of the singular concept “blue”.

However, this is not the case in Russian. What English speakers describe as light and dark blue, Russians instead refer to as goluboy and siniy. These terms are used behaviourally as descriptions of different colours (rather than shades). Importantly, there is no single word in Russian that can be used to describe all shades of ‘blue’ (as demonstrated in the spectrum below, reproduced from Winawer’s study).

The closest analogy I can provide (in the constraints of the language I use) is to consider orange and red (or you might prefer, orange and yellow or green and yellow). Although these colours can have similar shades, there is a definite boundary where one becomes the other. In a clearer manner than the transition between light and dark blue.

In line with this, the boundaries between goluboy/siniy and light/dark blue were calibrated to participants individually. The test then compared two conditions — firstly, comparisons of two shades within a category (e.g. light blue vs light blue, goluboy vs goluboy, etc); and secondly, comparisons of two shades across the established boundary (light blue vs dark blue or goluboy vs siniy).

Winawer found that English-speaking participants did not differ in their performance within or across boundaries. That is, they did not have an advantage when the distracting square lay on the opposite side of the light/dark blue boundary.

However, Russian speakers performed better when the target and distractor were on either side of the goluboy/siniy boundary than when both squares lay within either category.

Winawer concluded that linguistic representations “meddle in even surprisingly simple objective perceptual decisions”.

Humans do share a biological colour perceptual system, however, it is part of a complex neurocognitive system. So although we do see the same world, we do not necessarily see the world the same way.

Reproduced with permission from figure 1 of Winawer (2007). Shows a spectrum of 'blue' (or goluboy/siniy in Russian). The testing format (with baseline, target and distractor squares) is also displayed. Copyright (2007) National Academy of Sciences, U.S.A.

The Sapir-Whorf hypothesis is a fascinating example of an interdisciplinary issue that still enjoys debate. At first glance, it is a fascinating insight into the ways in which we differ.

My suggestion is that these differences need not be divisive. After all, they show that our brains share a common mechanism that allows them to develop as they do.

In this light, our differences are simply reflections of the details of our experiences.

---

If you are interested in learning more about cognition, check out this interactive article in the Experimental Psychology 101 series. See how many illusions fool your brain!

---

References

Kihlstrom, J. F., & Park, L. (2016). Cognitive psychology: Overview. Reference Module in Neuroscience and Biobehavioral Psychology, Elsevier

Pinker, S. (1995). The language instinct: The new science of language and mind (Vol. 7529). Penguin UK.

Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A. R., & Boroditsky, L. (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences, 104(19), 7780-7785.

---

Image references

Benjamin Lee Whorf (Year Unknown) [Photograph]. https://en.wikipedia.org/wiki/File:Benjamin_Lee_Whorf.jpg

Kandinsky, W. Composition 6 (1913) [Oil on canvas].

https://commons.wikimedia.org/wiki/File:Vassily_Kandinsky,_1913_-_Composition_6.jpg

Edward Sapir (c.1910) [Photograph].

https://commons.wikimedia.org/wiki/File:Edward_Sapir.jpg

Lincoln, R. Steven Pinker (2011) [Photograph].

https://commons.wikimedia.org/wiki/File:102111_Pinker_344.jpg

Martini, J. Use Your Brain (2020) [Photograph].

https://unsplash.com/photos/Iod3vdjKE1E (Cover image)

Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A. R., & Boroditsky, L. Figure 1 (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences, 104(19), 7780-7785. Copyright (2007) National Academy of Sciences, U.S.A.