Interview

How we perceive colour depends on our culture and language

Interview with AcademiaNet scientist Anna Franklin, professor of visual perception and cognition

14. 12. 2016 | Our ability to see a colour is limited by the words we have to describe it, and understanding more about colour categorisation could help improve how colour-blind children learn and develop, according to Anna Franklin, professor of visual perception and cognition at the University of Sussex in the UK, who is studying the relationship between language development and colour perception.
Bild vergrößern
(© fotalia / Anna Subbotina)



Prof. Franklin, how many colours are there?

Prof. Anna Franklin: Colour is a spectrum and research has shown that we are capable of seeing millions of them: if you show people two very close shades of green they can tell that they are not the same. But rather than name all of these as distinct colours, we tend to categorise them into groups like blue, green and so on.

Is there a fixed number of colour groups?

Different languages break up colour in different ways. For example, some languages have only four basic words for colour whereas others have 11 or 12. One of the big questions in the field has been whether the way in which cultures carve up the colour spectrum is completely arbitrary.

You're investigating the link between language and colour as part of your CATEGORIES project, funded by the EU's European Research Council. Does categorisation develop as we learn language or is it innate?

One part of our project aims to look at what age we begin to categorise colour: when do we turn millions of colours into families or groups. We have shown that four-month-old infants know that two greens are the same kind of colour rather than belonging to separate categories.

How do you quiz infants about colour?

Dr. Anna Franklin
Bild vergrößern
Dr. Anna Franklin
We give them habituation tasks: we show the infant a shade of green and get them used to it. Then we show them a new colour – if the new colour is blue, they respond to it as though it is something novel. If it is another shade of green, we can see they are already used to it because they are not as excited by it.

In terms of how different the colours are, the new green and the new blue could be the same distance from the original green but the infant can tell whether the new colour belongs to the same category as the original. So it's not that they are more different from the original, it's that they are categorically different.

What we want to know now is how many colour categories infants start out with and how this aligns with language. Do infants carve up colour in the same way as languages do it?

What do you expect to find?

Our theory is that there are natural fault lines that infants have; that we have a basic template for categorising colour and then you personalise that depending on what distinctions are important in your culture.

How else might language influence colour perception?

Another element of our work is looking at the relationship between naming colours and colour perception. For example, Russian has two words for blue – it distinguishes the darker and lighter blues into separate categories. Russian speakers, because of this fundamental distinction, are more sensitive to colours in that region of the spectrum. We can test this by measuring the electrical activity of the brain from the scalp in Russian speakers compared to others when they are asked to differentiate between two shades of blue.

What we found is that there are differences but they come relatively late – a couple of hundred milliseconds after seeing the colours. This suggests that two people with different colour lexicons see the colours the same way but they think about colour differently: the difference is cognitive rather than visual. Language hasn't fundamentally altered how colours are seen, but it has changed what we do with the information.

What brain regions are involved in naming colours?

We've identified a region of the brain responsible for categories of colour. It's called the middle frontal gyrus. For example, if you are just passively viewing colours and it changes from green to blue, the middle frontal gyrus is sensitive to this.

What about people who are colour blind?

Some people with colour vision deficiency see colours differently because they are lacking a certain type of cone – colour-sensitive cells – at the back of the retina. For normal colour vision you need three types of cones which respond to long, medium and short wavelengths.

If you are missing one type of cone it can affect your perception but also influences whether you like certain colours. We all like colours we can name quickly, so the way in which we categorise colour affects our aesthetic experience of colour. If it's a struggle to identify and name a colour we tend not to like it and this is an issue for people with colour vision deficiencies.

A photo of a dress went viral last year because people disagreed about what colour it was. Why did it cause such confusion and excitement?

This is not an area where I am working, but the dress has been the subject of much debate and psychological research. Some have suggested that the dress phenomenon may relate to what's called colour constancy. We are all capable of keeping the perception of colour constant in our minds even if the light changes. For example, a yellow banana under blue light still looks yellow to us because our brains compensate for the blue light.

We like colours to be stable as it makes the world more predictable. One argument made by some is that the illumination of the dress makes for an ambiguous and unstable colour.

We have conducted a study looking at colour constancy in the context of how children learn. If you can keep a colour constant and compensate for lighting, it is actually easier to learn the word for that colour.

How could all of this be applied?

We are just about to start a new project which will design a test that diagnoses colour vision deficiency in young toddlers. At the moment most tests are designed for children older than four years of age, but a lot of development has already taken place by that age – and a great deal of it is linked to colours. Just think how important colour is in nursery education – for coding things, grouping things – a lot of educational material is heavily reliant on colour.

These materials can be more difficult for people with colour vision deficiencies to access, potentially putting them at a disadvantage. Some studies suggest that children with colour deficiencies are mistakenly presumed to have a learning disability because they cannot understand what they are being shown.

We hope that by diagnosing colour vision deficiencies early we could level the playing field for those children.


Courtesy of Horizon, The EU Research and Innovation Magazine, by the EU Commission
  (© Horizon Magazine)

More information

Testimonials

  1. Read what our members say about AcademiaNet.

No more excuses!

  1. Please download the brochure "No more excuses" and read more about female experts in Europe, and about AcademiaNet.

News

  1. New study reveals ancient galaxies' frenzied starmaking

    AcademiaNet member Prof Karina Caputi and her team have discovered that rapid star development is more widespread in early galaxies than previously thought.

  2. Find and further the outstanding female scientists

    Female scientists are not only underrepresented in academia, but also in terms of articles they publish in scientific journals. Nature journals' Editor-in-Chief Philip Campbell talkes about how to change this.

  3. Society's Problem with Parity

    This week we celebrate Marie Curie's 150th birthday. Since her time as a scientist, the situation for women in science has changed a lot. Nevertheless, only a minority of science professors today is female. AcademiaNet spoke with Professor Polly Arnold who is the producer of "A Chemical Imbalance" and a strong fighter for women in science.

  4. A new role for exosomes in type 2 diabetes

    In healthy people, a type of exosomes – tiny structures secreted by cells to allow intercellular communication – prevent clumping of a protein that leads to type 2 diabetes. In patients with the disease, these vesicles don’t have the same ability. These are the results of a new study by AcademiaNet member Professor Pernilla Wittung Stafshede.

  5. "Female academics tend to be too passive"

    Marlies Knipper, Professor of Molecular Physiology at the University of Tübingen, is determined to do her part to bring about gender equality. She has organised a Club of AcademiaNet scientists who meet regularly to discuss challenges they face and to further their knowledge on topics that are relevant today. We talked to Prof. Knipper about what motivates her to stand up for women, and which actions she would like to see more of from fellow academics.