If you have been on a social media site in the last few days then I’m sure you have probably seen the image above before. The biggest question on the internet has been, which do you seen, black-and-blue or white-and-gold? The picture showed up randomly on Tumblr from a dress maker in India and since then it has taken the internet by storm. This entire issue is all thanks to our top-down processing.
We must first realize that color is a subject perception. Meaning that what I would call red, doesn’t necessarily mean that another would call red (or whatever color). This is the same with all sensory information; for example, I am a super taster, which means that I have more bitter receptors on my tongue than most other individuals and what others may taste as only slightly bitter, I’ll be in the trash can trying to get the remnants off with a fork. The difference with vision is that we take in additional contextual information such as shadows.
Shadows play a huge part in how we perceive many things visually such as motion and color. The video shown above shows a few examples of just how much our visual perception relies on shadows. This is purely top-down processing. We see the shadow and expect that the color is actually lighter than it is. This is referred to as color constancy. Color constancy is a Gestalt principle which basically suggests that the context in which we view an object influences our perception of that object.
With this in mind we must then understand what top-down processing is. To put it simply, top-down processing is how we form expectations of the world around us and how magicians, for example, are able to fool us. Now, look at the picture below, do you see anything? This is a good example of top-down versus bottom-up processing. Bottom-up processing just see’s the image as a bunch of dots whereas once I tell you that there is a Dalmatian in the picture with its nose to the ground that it then becomes a top-down issue. If you look close enough, it’s as though you can actually see the lines which make up the dog, but look again, they’re not there. It’s all your brain connecting the dots and creating the object out of the minimal amount of information given.
So this goes to a larger question of how we perceive our world. How could a simple dress fool the way in which we see? I think the biggest thing we must realize is that color isn’t a physical thing that actually exists. Color is our brains interpretation from data it receives from our eyes which is simply waves of radiation bouncing off of the objects that we are looking at. Simply put, our brain creates color. A fact that I have found interesting from my very first Psychology class is that since vision is simply light waves bouncing off of an object, when we think of color, the object is every color EXCEPT the color we see. For example, I have a green apple sitting on my stand right now. As light waves hit the apples skin the light is absorbed by the skin of the apple, except the green waves are bouncing off and traveling to my eye which then my brain interprets into “green”. The apple is not actually green; green is the only color it is NOT. With this in mind we can see how easily our brains can be fooled into seeing something that doesn’t actually exist, especially when we through shadows into the mix as well.
So then back to the original dress debate. The answer is that the dress is really blue and black. The reason 3/4ths of those who look at it see white and gold is because of the light behind the dress. It’s tricking your brain into thinking that the dress is in shadow so it then must be lighter than what are eyes are seeing. The question I have now though, is how the other 25% of people (I’m included) didn’t see white and gold. Even after I knew the answer and knew why it was this way, I still fail to see white and gold. Perhaps that is a question for a later post. I leave you with this ASAP Video on the matter.