Two Years Later, Vision Scientists Are Still Studying the Dress

Yes, that dress.

Apr 7 2017, 5:04pm

In 2015, an image of a striped dress took the internet by storm. Some people saw the stripes as gold and white, while others saw them as black and blue. The debate raged for days in the trenches of internet forums, families were divided, and lives were undoubtedly lost in defense of stupid opinions. But when the dust settled and the carnage was seen clearly by the light of day, what did the internet have to show for itself? Were we any closer to discovering the true color of the dress, or had internet reputations been torn asunder in vain?

We could borrow Situationist critique and discourse on the Spectacle of the Dress until the end of days, but according to the company who made the dress the garment is actually black and blue. But that still doesn't answer the question of why so many people saw something totally different—a question that scientists are still puzzling over.

According to new research published today in Journal of Vision (which announced that it would be compiling a special issue on vision science related to the dress) the dress discrepancy boils down to individual assumptions about how the dress was illuminated.

"The original image was overexposed, rendering the illumination source uncertain," Pascal Wallisch, a neuroscientist at New York University, said in a statement. "As a result, we make assumptions about how the dress was illuminated, which affects the colors we see."

Read More: There is the Dress and Only the Dress

Wallisch's conclusion was based on an online survey of 13,000 people, which suggested that those who thought the dress had been photographed in a shadow would see it as gold and white, and those who thought it was artificially illuminated would see it as black and blue.

While this may seem like a trivial observation, it speaks volumes about how the brain works to color correct the things we see based on context. In other words since shadows are blue and artificial light tends to be "yellowish," the brain would subtract either blue or yellow light from the image to render the 'correct' color of the dress, depending on the assumed context.

Less certain was what would make them believe it was in a shadow in the first place, but according to Wallisch, this depended on a participant's general exposure to daylight. Those who woke early and went to bed early are more likely to see objects illuminated under a blue sky than those who are more active at night, when everything is illuminated by artificial light. "This suggests that whatever kind of light one is typically exposed to influences how one perceives color," Wallisch said.

Similar hypotheses have been raised elsewhere, but the perceived color of the dress is based on the color of the background, rather than lighting. Other non-dress related research has found that physiological factors such as the sensitivity of photoreceptors (rods and cones) in the eye also influence differences in color perception.

In any case, the dress has become another significant data point in the science of color perception, where one of the main research questions is whether or not two people looking at the same thing are actually seeing the same color.

"The answer is 'not necessarily,'" Wallisch said. "If illumination conditions are unclear, your assumptions about the illumination source will matter, and those might depend on lifestyle choices, such as when you go to sleep."

Subscribe to pluspluspodcast , Motherboard's new show about the people and machines that are building our future.