Pre-attentive attributes are types of visual stimuli that are processed subconsciously and can be used in Data Visualization to guide the consumer to exactly where we want them to focus. So, we can add (or even layer) visual cues such as movement, shape and colour to draw attention to where we as the designers believe it should be focused.
In a previous blog I started with the image below as my initial analysis and I will use this as an example, using Oracle Analytics, we can very quickly add some subtle pre-attentive attributes to create a project that is more self-explanatory.
In the picture above, on the right, we have a bar chart of a group of customers and we have ordered these by the amount they have spent with us. Therefore Customer 98 on the far right is the highest spending customer. On the left we have a map showing the location of the customers, with the circles sized by the amount they have spent.
If I use my cursor to brush the Customer 98 bar, then this highlights the customer on the map so that we can easily identify them.
Now, imagine my job is to create a project and presentation for me to show at a company event whereby we focus on the highest spending customer. I want to have the audiences’ attention driven straight to the customer I want to focus on, but also show the other customers and their spend for context.
If I right click onto the customer I want to focus on in the bar chart, I get an impressive menu of options.
Here I can click on Color and then I’m given another list of options. What I choose to do is select the Data Point from which I have performed the right-click operation (that being the highest spending customer, Customer 98).
This shows me the current colour setting for that data point.
I want this to stand out from the uniform green of the other customers. I am going to choose a dark blue as this will visually differentiate the bar from the others.
Let us see what effect that simple change has had to the bar chart.
To me this has had a striking and immediate impact. Without any other prompting our eyes are drawn to the clear differentiation which really stands out.
Let us now have a look at our current project as a whole.
My thoughts are now, ‘wouldn’t it be great if we could do the same with the map?’. Let’s do that – and fortunately it is a consistent process.
My map data is using Latitude and Longitude so the data point descriptions shown above are those co-ordinates. As expected, we get to see the colour chart and the current assignment
Change the colour on the map for the specific customer to the same colour as we chose for the bar chart data point, so we have an immediate associative link by colour
As we expected, this customer on the map now appears in the new colour
Now when viewed as a project, are you instantly drawn to the relative images on the map and the bar chart? Do your eyes move between the blue objects in each?
Without the addition of any textual description, the pre-attentive attributes are already doing a great job.
Now, we did discuss briefly at the beginning that we can layer these choices, so we could make the blue bar wider on the bar chart (but would add much value and possibly cause the other names to be squashed?). We can certainly add another layer of attributes by changing the shape of the customer on the map to really differentiate the customer by giving it a distinctive edge.
Let’s experiment – fortunately Oracle Analytics makes this easy for us as we intuitively right-click and select Shape and the data point.
We are provided with a selection of different shape options for our specific data point.
Select the X and press OK and let’s see what happens.
We now have a lovely X, and when has an X on a map not been significant? There is indeed treasure here. With just a little bit of ‘formatting’ we are emphasizing the results of our analysis with the judicious use of some subtle pre-attentive attributes and helping to direct the consumers exactly where we want them to look.
I hope that you found this interesting and I’d certainly be interested in hearing your views on how you use these sorts of techniques in your data visualization projects.