The job of a data visualiser is to discover the most important information hidden in a dataset and communicate it clearly and effectively to the user. This can oftentimes be difficult to do as the data is rich, the stories are complex, and hence the information is overwhelming. Thus, to help overcome these challenges, it is invaluable for a data visualiser to become familiar with pre-attentive attributes.
What are Pre-Attentive Attributes?
When the brain is first presented with visual information, it is immediately processed by the subconscious. The brain will then direct it’s conscious attention to whatever of this information it finds most salient, making sense of context and extra detail to understand on a deeper level. Pre-attentive attributes are the features of a visual that the subconscious finds most salient and hence most deserving of attention. Being mindful of the impact of pre-attentive attributes helps the data visualiser in effectively directing the user’s focus towards the most insightful information.
The pre-attentive attributes may be classified according to their form: spatial (length, width, orientation, position, grouping, enclosure), colour (colour hue, colour intensity), size and shape. Alternatively, we could classify the attributes according to what type of differences in data they most naturally emphasise: differences in quantity (length, width, colour intensity, size) and differences in quality (orientation, position, grouping, enclosure, colour hue, shape).
For instance, it may be most appropriate to represent the difference between sales figures with a bar chart utilising length, or a geographic map utilising colour intensity.
Furthermore, demonstrating a particular region’s outstanding performance compared to others may be effectively shown with a scatter plot utilising grouping, or with a line graph utilising colour hue.
Combining Attributes
It can be rather effective to combine attributes together to either provide more detailed distinctions or to emphasise simpler ones. A company’s profit figures per day throughout a month may be most easily visualised with a calendar map using a diverging colour palette, ranging from solid orange (largest loss) to white (no profit or loss) to solid blue (largest profit). This would enable the user to quickly identify profits vs. losses by focusing on colour hue, as well as the most significant profits/losses by distinguishing colour intensity.
On the other hand, it may be helpful to show profit vs. quantity sold for each product sub-category with a scatterplot that uses a different colour and shape to represent the category that each product sub-category falls into, reinforcing the difference with two independent attributes.
Further Considerations
Over recent years, data visualisers have become increasingly aware of the need to cater for users who may have visual disability. Two of the most common conditions include colour blindness (particularly red-green colour blindness) and visual impairment. As a result, visualisers should consider avoiding the traditional ‘green for good, red for bad’ and instead opt for a blue-orange colour palette that is more easily distinguishable for all possible users. Additionally, it is best to avoid using similar shapes or sizes as the differences may go unnoticed.
In addition to taking advantage of pre-attentive attributes, it is just as important to avoid types of visualisations that obfuscate them. While a stacked bar chart may show the general gist of a change over time, it can become hard to accurately compare the middle bars over time as they do not have a uniform starting or ending position, hence making it more troublesome to compare the differences in length.
Keeping in mind the importance of pre-attentive attributes in data visualisations will ensure that the user’s experience is straightforward and streamlined. Happy visualising!
Cover image by Gerd Altmann from Pixabay