Skip to page content or Skip to Accesskey List.

Work

Main Page Content

Where What Vision Theory In Practice

Rated 4.22 (Ratings: 6)

Want more?

 
Picture of hgquinn

Heather Quinn

Member info

User since: 23 Sep 2001

Articles written: 2

Understanding the Where and What vision systems helps with better

visual design:

In my first

article about the Where and What vision systems (defined by Margaret Livingstone,

a neurobiologist, in her book Vision and Art: The Biology of Seeing (Margaret

Livingstone, David H. Hubel, ISBN 0810904063)), I wrote about the differences

between the two vision systems, and I made a strong claim: By learning about

the vision systems and their differences, you'll be able to do more effective

visual design.

This article puts that theory to work.

I've taken my own website, and analyzed it for Where/What balance. Using what

I learned from the analysis, I made some changes to my site's title and navigation

graphics. The result is a more effective page design, with a small investment

in time and effort. And the work involved becomes an addition to my experience

as a visual artist.

How can Where/What analysis help the designer?

Whether designing or redesigning, the designer works in a recursive loop: doing

work, viewing it, critiquing it, working out changes to make, then back to the

original work to apply the changes, and so on. You don't want to be stuck in

recursion mode forever, and you want your final work to be effective. The objective

nature of Where/What analysis can help you break out of the design loop, and

make sure your work is communicating with viewers the way you intend it to.

If you have a problem with an existing design, Where/What analysis often gives

you the hidden key to your design problem. It also gives you an approach

for developing a solution. And it's a tool that lets you recheck your

redeveloped design to make sure it's working. It gives you a level of

solid confidence about your design decisions that's a welcome counterbalance

to the otherwise subjective nature of design.

Finally, if you need to sell an employer or site owner on a particular design

of your own, Where/What analysis can be used to show why an old design doesn't

work and your new one does.

A real-world example: my own website

When I did a couple of bits of digital illustration recently, I featured a

large thumbnail of one on my site's main page, but the new illustration made

my site's design look "off" somehow. Also: I often feel driven by

changes in the world (including seasonal shifts) to modify my website's look.

With a design problem of my own making, and midwinter having generated an impatience

for springtime in any form, I was motivated to do a redesign.

In my redesign project, I had two goals: To resolve the visual disharmony that

a new illustration created on my site, and to shift my site's look a little,

in honor of the upcoming spring season.

Creating an inadvertent design problem:

Before I added the new illustration to my site's main page, the page looked

like the screenshot above. The page was designed to allow the eye to move

from the branding in the top left corner, diagonally down and right to the main

page content. The page elements had a good balance between Where features

(mainly high contrast and sharp edges) and What features (color, texture, details

and apparent depth) that captured a viewer's attention, and held it for a while.

The next screenshot, below, shows the page with the new illustration in place.

You'll see that adding the new illustration has messed up the page's visual

dynamic.

Comparing the two screenshots without thinking about the Where and What vision

systems, we'd say that the new illustration is too strong. It pulls the

eye down to the bottom right of the page, and keeps it there.

How can we fix this? My questions grew. Should I remove the illustration? Make

it smaller? Is the problem that it's too square? Is that big apparent curve

holding the viewer too strongly? Is it the dog's face? What can I change? What

if I like the illustration as-is? Should I change the branding area instead,

making it stronger to balance the illustration? If so, how should I make it

stronger? Wouldn't making it stronger create a new imbalance on the page?

Yow! It was time for a Where/What analysis. The analysis would tell me what

was really wrong, and help me figure out how to fix it.

Doing the Where/What analysis:

Below is a table of possible Where and What vision system features:

The Where vision system sees large

shapes and movement fast, and fatigues quickly.
The What vision system

sees things slowly, processes details, and is long-lasting and resilient.
The Where vision system is triggered by: The What vision system is triggered by:
High contrast Low contrast
Edges Soft value shifts
Large shapes Details
Angles Curves
Sharpness Textures
Apparent movement Apparent depth
Where vision system special features: What vision system special features:
Is 100% colorblind Sees color
Is used to determine who close things are,

and and how fast things are moving, thus telling people WHERE something

they're looking at is.
Is used to recognize details, including

facial recognition, thus telling people WHAT something they're looking at

is.

Taking an inventory of the new illustration's Where and What qualities, I found:

Attention-getting Where features: The illustration has large, high-contrast

shapes, and some sharp edges. These are very Where, and draw the viewer's eye

down to the new illustration right away.

Attention-holding What features: The illustration has texture, color, curves,

low-contrast areas, and a face. These are very What, and hold the viewer's gaze

long after the initial attention grab (achieved by the illustration's Where

features) has fatigued.

Looking at the Where/What analysis more deeply, to develop a solution:

The Where/What analysis of the new illustration showed it had both strong Where

and strong What features.  Overall, it had the strongest combined Where/What

features of anything on the page. Close-up snips of the branding area and the

new illustration show this more clearly. Notice that the illustration (bottom

snip) has significantly more large shapes and high-contrast, which are strong

Where features, as well as more value shifts, detail, texture, curves and that

cute little face, all strong What features. This is why it competes so strongly

with the logotype graphic. And even in the snips graphics above, even though

the snip of the branding area partially overlays the snip of the new illustration,

it's the new illustration that draws and holds attention best.

Because the new illustration engages both of the viewer's vision systems so

strongly, attention is captured by the new illustration and never released to

the rest of the page. It's hard for the casual viewer to pay any attention to

the site's branding and rest of the page content.

(In a grayscale version of the snips graphic, below, you can see each element's

Where qualities more easily.  When you view in grayscale, there's no color

to excite the What vision system, so it's easier to see Where features.)

My site's original design was done with the Where/What vision systems in mind.

The original page focus is the branding area in the top left of the page, consisting

of a logotype and an illustration of a horse (designers are workhorses, you

know!). The logotype and branding illustration have large shapes, sharp edges,

angles and high contrast. These features made it the most Where area of the

page, so it would get attention first. The color of the logotype and its background,

and the texture, detail, curves and color of the horse illustration, were What

features that were deliberately introduced into this area of the design to make

it less Where. I wanted viewers to see the logotype first, and stick around

long enough to read it, too. The Where vision system fatigues very quickly.

To hold attention on a Where element, you must add What features to counterbalance

the element's Where qualities. By creating graphic with a combination of strong

Where/What features, I ensured that the branding area would become the page's

main focus.

All this planning was undone by the intended well-balanced strengths of the

new illustration, which communicates strongly and directly with both the Where

and What vision systems.

The solution:

So what's the solution? The source of the problem is the strength of both Where

and What features in the new illustration, vis-à-vis the original branding

graphic.

I don't want to change the illustration. My only option is need to turn my

attention to the branding area, specifically the logotype and its background.

Also, I don't want to spend a lot of time fixing this problem. That means I

decide there'll be no changes to shapes, sizes or positions of the original

logotype graphic slices.

I turned back to my Where/What knowledge to assist in developing my solution.

I figured I could rebalance the page by adding stronger Where features to the

logotype, to enhance its basic attention-getting qualities. But I knew that

if I did this, I'd run a strong risk of making it so strong that people would

see it and run the other way. So, as well as giving the logotype more Where attention-getting

qualities, I'd probably need to enhance its What attention-holding qualities,

to give it a good balance so people would stay long enough to read the logo,

just as in the original.

Remember this key to good design: Too many Where elements grab attention,

then lose it. Too many What elements may never have the chance to grab

attention long enough to hold it. You need both!

Below are snips and a screenshot that show how the solution worked out.

The logotype was made higher-contrast, and a block of warm transparent color

in back of the logotype was removed. High contrast and less color create stronger

Where, attention-getting, qualities.

The increase in Where had to be balanced by reducing the logotype font size,

and adding a fine, colored texture in back of the logotype. Texture and more

color create stronger What, attention-holding qualities.

The logotype and its illustration were now rebalanced. But, when the now-stronger

branding area was seen against an already strong navigation block, it was clear

that together they were too Where. The navigation block had its font size reduced,

and the font color lightened to reduce contrast. This improved the What qualities

of the navigation block, and created a better Where/What balance for the branding

and navigation area together.

See the Where/What-developed solution as a live page.

Overall, these small changes completely rebalanced the entire page, and worked

well on other pages on the site that had not been otherwise changed. In addition,

when a warm color block was replaced with a textured green, and the size of

the logotype font was reduced and its color darkened, the site stepped into

a kind of spring dress, with cooler colors and more delicate elements. It took

about 90 minutes to do the analysis, make the changes, and publish the results.

When you first starting thinking about how the Where and What vision systems

experience your designs and how you can best communicate with them, you'll have

to do a bit of depth-thought and explicit analysis. Over time, you'll integrate

your awareness of these issues into your general design skills, and incorporate

Where/What knowledge into your active work intuitively. Whether you're just

starting to think about these two vision systems, and how humans process what

they see, or you're farther along in understanding how these vision systems

work, grappling with these issues will help you design with more effective results.

Artist turned programmer turned self-taught web designer, Heather enjoys walking the cliff's edge of trying to deliver effective visual design in a continually morphing medium.

Still a dayjob programmer and web designer at a medical software company, WindyHillDesign.com is her nightjob workplace, and her personal face is shown at WindyHillDesign.org.

The access keys for this page are: ALT (Control on a Mac) plus:

evolt.org Evolt.org is an all-volunteer resource for web developers made up of a discussion list, a browser archive, and member-submitted articles. This article is the property of its author, please do not redistribute or use elsewhere without checking with the author.