Seeing Without Thinking: The Built-In Visual Cues Your Brain Already Knows
FM (Friday Morning) Reflection #17
Throughout the past week, I reflected on the communication gaps we often encounter in daily life, and realized that sometimes we need to refresh on the essentials. I wrote this in the spirit of “w hat can I offer?” and believe it’s a good start.
–Eric
Our brains are wired to perceive certain things without us having to think about them. They just register, and you can’t “not see” them even if you tried.
Knowing about these innate capabilities can be incredibly helpful, especially when it comes to data visualization.
The Eight Pre-Attentive Attributes
There are eight visual properties that we intuitively perceive:
Shape
Orientation
Size
Color
Value
Texture
Position and Order (they’re related, and for still images, expressed in planar dimensions like x, y, and z)
These are called pre-attentive attributes, which means our brains recognize them instantly, without conscious thought. They are simply obvious to us, thanks to how we process information.
Jacques Bertin and the Foundations of Data Visualization
These pre-attentive attributes were first identified by Jacques Bertin, a French cartographer. In Sémiologie Graphique (1967), Bertin articulated the first theoretical principles for the display of data and rules of visual organization.
While data visualization principles have been around since the ancient Egyptians and were improved upon by many innovators over time, Bertin was the first to propose a formal system that linked data elements with visual symbols.
Here are examples of each attribute in action:
Bertin’s system, and the identification of the variables shown in examples above, helps us to understand relationships in data visually, without the mental gymnastics of interpreting numbers in lists.
You’ll feel the difference when first trying to discern a stock price trend from a table of numbers, and then relieving the pressure on your brain by looking up a chart of the trendline. A well-designed chart makes the information—and more importantly, the takeaway—instantly apparent.
Bertin’s insights, synthesized with the work of Daniel Kahneman on our two systems of thinking (System 1 and System 2), as well as the prolific writings and appearances of Stephen Few and Edward Tufte, provided the foundation for the contemporary practice of data visualization.
Jeffrey Shaffer and I used this confluence of knowledge as inspiration to create the first Data Visualization course at the University of Cincinnati in the fall of 2012—a course that Jeff still teaches today, and that has equipped thousands of students to be better data analysts and communicators.
Using Clear Markers for Better Communication
In That's the Ticket: Shifting How We Talk to Help AI-Assistants (and People!) Succeed, I introduced the idea of using clear verbal markers to help participants and AI-assistants navigate meeting discussions, curate notes and summaries, and conclude with better takeaways. Examples include:
"Here's today's objective"
"That's a decision - next steps are..."
"To-dos: [specific action], [responsible person], [deadline]"
"This is a point to highlight in the notes"
In our world, where making sense of data and refining it into information and knowledge is one of the main ways we create value, the visual variables that Bertin identified are essential elements as well. They help us communicate effectively, whether preparing data for presentation or decoding it as a recipient.
Our tools—and by tools, I primarily mean spreadsheets—often don’t make this easy. Odds are you’ve spent a lot of time fiddling with different shapes, colors, chart types, and 3D options to create a visualization for an important presentation, only to feel it’s never quite right. Maybe you’ve even had to pre-apologize, saying, “I know this is an eye chart, but…”
Let’s agree: when we have to start by acknowledging deficiencies like this, we’re basically saying, “It’s going to be a slog. I’m sorry.”
A Quick Recipe for Effective Data Communication
Here’s a quick 1-2-3 recipe to improve the effectiveness of your visualization of data and insights communications:
Remove anything unnecessary: Extra grid lines, redundant axis labels, meaningless colors, and 3D effects—take out anything that isn’t essential.
Stick to basic chart types: Bar, line, and scatterplot charts cover the vast majority of use cases.
Use pre-attentive attributes wisely: Direct your reader’s attention by leveraging these attributes. For instance, if all the bars on your chart are black, use color to highlight the one that matters most.
Time to Experiment!
In a follow-up column, I’ll provide examples to illustrate these concepts in practice—for now, I encourage you to experiment with them. When we come back together, let’s compare notes.
Have a great weekend!
Photo Bonus
Speaking of signposts and wayfinding… For a time I lived on Church Street in the Old City neighborhood of Philadelphia. On my way to and from work every day, I always appreciated this mural in the 2nd Street MFL Subway station, and how it grounded me with a sense of the local landscape. Shot on Lomography LomoChrome Turquoise XR 100-400 Color Negative Film with my Nikon FM2.