A summary of a seminal book in data visualisation theory

Why Visualisation?

Visualisations can serve two purposes:

• To communicate an idea (‘a picture is worth ten thousand words’)
• To use graphical means to create or discover the idea itself

The second of these is the focus of this book, 'Using Vision to Think'.

Information Visualisation is defined as The use of computer-supported, interactive, visual representations of abstract data to amplify cognition (i.e. the acquisition of knowledge).

How does visualisation amplify cognition?

• By increasing the memory and processing resources available to the user
• By reducing the search for information
• By using visual representations to enhance the detection of patterns
• By enabling perceptual inferences
• By using perceptual attention mechanisms for monitoring
• By encoding information in a manipulable medium

Human Perception

Information Visualisation is dependent on the properties of human perception, so its necessary to have a basic understanding of our perception system, in other words how the eyes work!

There are two levels.

The first level of the visual system is the retina. The retina is good at detecting movement or other changes in the visual environment and in maintaining a rough representation of the location of shapes previously examined. The second level of the visual system is the foveola, a higher resolution field that is focused automatically on stimuli like movement, strong colour, intensity or contrast. The system as a whole maintains a surveillance of the entire visual field with the retina, and at the same time positions the foveola to sample areas of interest.

Due to the ways our eyes are structured, visual information can be processed in two different ways. Controlled processing, like reading, uses the fovea and is slow and thorough. Automatic processing is superficial and fast, occurring unconsciously. Techniques to aid pattern detection and search should use automatically processable features, colour and size being good examples.

The Building Blocks of a Viz

A visualisation is made from: spatial substrate, marks and the marks graphical properties.

• The spatial substrate is the use of space in the viz. The most fundamental aspect of a viz is its use of space. Space is perceptually dominant, meaning we notice it the strongly! Spatial position is such a good encoder of data the first decision of visualisation should be which variable gets spatially encoded over others.
• The marks are the visual things that occur in the spatial substrate.
• The mark’s geographical properties show structure. For example a line between two marks shows a connection.

Interactivity Theory

Interactively modifying and augmenting visual structures exploits time to extract more information from the visualisation, compared to if it were static.

There are three levels of interaction: 0.1 seconds, 1 second and 10 seconds:

• 0.1s (Psychological ovement) - A time frame that makes changes fuse into a single precept, creating the direct experience of cause and effect.
• 1s (Unprepared response) - A time frame that create a smooth interaction with the user, allowing new information to be conveyed without slowing the user down.
• 10s (Cognitive Skill) - A time frame that is the pace of routine cognitive skill, i.e. thinking about something.

Interactivity Techniques

One particularly useful interaction technique for handling large information interactively is overview + detail. In this technique two displays are coordinated, one giving an overview and the other giving detail.

An interesting variant of this is the focus + context, where the detail area is embedded in the overview.

Based on these techniques, A good mantra to use when designing any visualisation is: “Overview first, zoom and filter, then details on demand”