Join Bill Shander for an in-depth discussion in this video Visual perception, part of Data Visualization for Data Analysts.
- If you think about the beginning of language from the earliest days, the written language was all made of visuals, it was symbology. The pictures represented the actual concepts being communicated. So as language evolved, as communications needs evolved, text was required so we could communicate more nuanced things and not take up quite so much space. That was one of the big issues. But it really hasn't helped us when communicating data, numbers. You can actually parse and understand data, numbers, more quickly and easily in visual compared to text form.
I don't know if you know this, but 30 to 50% of your brain is devoted to visual processing. Human beings are wired to be visual. Seventy percent of your sensory receptors are in your eyes. It takes a tenth of a second to make sense of a visual scene. This is all brain science, okay? This is how all humans are. It's also been proven that we have 323% better performance on tasks when we learn those things with images compared to without images. There's also something called the picture superiority effect.
You'll remember 60% of something when you learn it with imagery verses 6% without it. So ten x the results and that effect increases with age. So I'm going to go through some scientific principles behind how we process visual information. It really helps to understand these principles when you think about how to visualize data. There are really seven key concepts. They're called Gestalt Principles and they come from Gestalt psychology. We're going to go through them one by one. So the first one is called Figure/Ground.
And it's used very frequently in logo design. So the idea of this one is that human beings will see something as being in the foreground and we will then also, simultaneously, see the the thing that is in the background. So in the case of the FedEx logo, we all see the purple and orange letters, right? That says FedEx, you know, Federal Express, etc. But we also, instinctively, you know, deep down in our psychology and brain science, we all see the arrow between the "E" and the "X." And whether you've every recognized it before, whether you see this in every single time you look at every single logo, of course not.
But it is wired into us to see these types of things. So this effect, figure/ground, as it's called, really plays into data visualization in one way, primarily, and it's not hugely important amongst all these Gestalt Principles, but it is interesting. And so the key thing to understand here, is that for some reason, humans will always perceive the thing on the bottom as being the thing in the foreground and the thing on the top as being the thing in the background. So here we have two area charts and regardless of color, whether we're looking at the white chart on a black background on the left, or the black chart on the white background on the right, you're going to perceive the bottom part as being the chart and the top part as being the background.
So why does this matter? Well, it matters because I've seen people try to draw upside down area charts, for some sort of effect. And they can work, I'm not saying you can never break the bounds of these particular rules, but the fact is most humans are automatically going to perceive it this way. So just keep that in mind. The next one is known as Proximity. And this, like many of these, is pretty obvious. Human beings, we all by default, see things that are near each other as being together. So those three columns of dots on the left go together, the two columns on the right go together and the two sets of columns don't go together, right? They're different from each other.
Proximity. Makes sense. So for instance, when you look at a scatter plot like this, we see patterns, right? We see the dots that are all clustered near each other in the top right hand corner as having something to do with each other. This is just how we recognize patterns. Next one is called Similarity, also an obvious one. We see things that look similar and assume that they go together. So this one seems to override proximity, right? So the black dots go together and the white dots go together, but the black and the white don't go together.
So we'll sort of categorize things in this way. So once again, the similarity of the objects will make us categorize, right? Whether similarity is in shape, or in size, or in color. The next principle is known as Parallelism. And so the idea here is that we will see parallel lines and assume that they go together or that they have something going on that's similar and the things that aren't parallel, are different, right? And again, this is deep brain science. This is how we all perceive these things. So for instance, if we look at these two area charts, in visualization, sometimes these parallel lines, even if they're not drawn into the visualization, but we might see it, we might see, subconsciously, that line from one dip in the chart to one rise in the chart.
And those parallel forms formed by those two charts that are near each other, will help us see that there' some sort of correlation, some sort of pattern going on, again, deep in our subconscious sometimes. This next one is called Common Fate. I just love the title of this one, it's just fun. So the idea here is that you might have a bunch of things that have no discernible pattern, right? So that proximity is not playing role here, parallelism, etc., but once they move, they are perceived as having a common fate. So I'm just going to go back and forth a couple times here.
Those dots that are moving share a common fate. That's where the term comes from, right? So I perceive them as going together, as belonging to each other. So you'll see this one, obviously, in animated visualizations, particular or interactive experiences where maybe I hover over something and the state changes and lets me see which things go together. The next principle is called Closure and there's another one called Continuity. They pretty much go together, they're very similar. They're less important when it comes to visualization, once again, but they're still interesting.
So the idea here, is that the brain sees what isn't there, right? The brain completes the picture, so you don't just see the three pies here, the three Pacman symbols, you see the triangle that is in the negative space between them. This is closure. Continuity allows you to see the continuous shape. So we see the letter "S" here, we don't see two random curvy shapes. Culture can definitely play a role in this, right? You might see a worm if you're a farmer, you might see a snake if you live in the desert, you might see Loch Ness if you live in Scotland. So this isn't necessarily that specific to visualization, but it's good to be aware of this because as your visualizations create shapes, as you create shapes on the screen or on page, they can be misperceived, or seen as something that's unintended.
So, you know, you have to make sure that you're, sort of, thinking with a clear eye towards these principles when you're creating your visualizations, 'cause it could affect how your data is understood by your audience. Gestalt Principles are really key to how people perceive things visually. There are lots of techniques to use to draw the eye to show changes or patterns in data. And really it just comes down to a few key ideas, right? You can change the orientation or the length or the thickness or the shape of something, or add little hash marks or cross hatches, etc., but really it all comes down to those principles that we just discussed.
You don't need to understand cognitive science or this, you know, be able to read this brain scan, or understand every Gestalt Principle to understand how to visualize data of course. But it is nice to know about it. It's nice to have that as a background thought and you'll be surprised at how many times you think back to this when you're working on visualization projects. I know it helps me think when I'm designing how to trigger the brain, instantly, to maximize effect.
- Why visual communications matter, and how they work
- Communicating via story
- Communicating with color
- Using legends and sources
- Sketching and wireframing
- Rethinking slides, charts, and diagrams
- Rethinking your templates and brand guidelines