The Science Behind Why That Blue Dress Looks White

the-science-behind-why-that-blue-dress-looks-white“What color is this dress?”

It’s the question that divided the Internet on Thursday — after a woman posted her photo on Tumblr earlier this week — and for such a black-and-white issue, there’s a surprising amount of gray.

To be clear, the dress is blue, with black stripes. But it didn’t always look that way to me. In fact, when I first saw an image of the dress on Thursday night, it seemed obviously white-and-gold.

But a few minutes later, as the image of the dress sat on my iPhone, the color pattern suddenly shifted to blue-and-black — it was such a jarring change that I thought someone was playing a prank and had sent me a very, very slow-transforming gif. (The image has since settled in as a hybrid, blue-and-gold patterned dress for me.)

That queasy shifting is part of what’s made the dress image contagiously viral; it’s a unique effect, on par with the rabbit-duck illusion and other memorable visual tricks. But the dress image was arguably more polarizing because so many people could only see one color pattern — and couldn’t agree on what it was.

Social media site Yik Yak ran a poll on Thursday night, asking its users, “WHAT COLOR IS THAT DRESS?!?” Tens of thousands of votes later, “white and gold” had narrowly edged out “blue and black” by a vote of 53% to 47%.

So the real question: Why does the color of the dress keep changing for some of us, and why can’t folks agree?

At least in part, it’s because some of our brains are confused by the original image’s washed-out, bluish lighting. Our visual systems intuitively know to filter out normal backgrounds and lighting in order to see the “true” color of an object — a concept called color constancy — but the bluish tint to the photo is throwing off that ability for many people.

Tweaking the brightness of the original photo (bottom left) makes the blue-black pattern more clear to everyone (bottom right).

“What’s happening here is your visual system is looking at this thing, and you’re trying to discount the chromatic bias of the daylight axis,” neuroscientist Bevil Conway told Wired‘s Adam Rogers. “So people either discount the blue side, in which case they end up seeing white and gold, or discount the gold …

Share