• Blind and low vision users     Barker, Wakabayashi
• Color vision confusion     Beck, Whitty
• Cognitively disabled     Cain, Yearsley
• Deaf & hearing impaired     Coles, Young
• Mobility impaired     Ellison, Yuan
• Children     Foreman, Zhang
• Elderly     Frisbie
• Low education, low motivation     Gasca
• Non-English speakers     Glubochansky
• Non-US culture     Hisada
• Cross-language information retrieval     Koemans
• Slow speed connections     Lai
• Small screens     McCully
• Wireless     Sam
• Speech recognition     Stevens
• Textual substitutions for audio/video     Sung
• Helping novice web users     Towle
• Designing help methods     Villarta

If a discrimination decision takes the same amount of time regardless of the number of distractors present in the display, it is processed pre-attentively; that is, human perception notices before conscious focus.

Certain simple shapes and colors "pop out" from their surroundings. The theoretical mechanism underlying pop-out is called pre-attentive processing because logically it must occur prior to conscious attention.

The features that are pre-attentively processed can be organized into a number of categories based on form, color, motion, and spatial position (Ware, p.165):

Form Line orientation Line length Line width Line collinearity Size Curvature Spatial grouping Added marks Numerosity Color Hue Intensity Motion Flicker Direction of motion Spatial position 2D position Stereoscopic depth Convex/concave shape from shading

The experiments reported in this article seriously challenge the common belief that motion captures attention in a stimulus-driven fashion...Motion joins a growing list of features that can guide attention when they constitute a target's defining attribute, but do not capture attention when they are irrelevant to the observer's task...Only when subjects adopt a deliberate state of attentional readiness for attributes like motion, orientation, or color singletons do such attributes guide attention.

Together, the experiments reported here are consistent with the claim that the appearance of new perceptual objects, and little else, captures attention....New objects are automatically assigned high attentional priority. (Hillstrom and Yantis, 1994)

A concept called the useful field of view (UFOV) has been developed to define the size of the region from which we can rapidly take in information.

The UFOV varies greatly depending on the task and the information being displayed. Experiments using displays densely populated with targets reveal small UFOVs, from 1 to 4 degrees of visual angle.

Essentially, the problem is how to attract the user's attention to information outside the UFOV. For a number of reasons, the options are limited. We have a low ability to detect small targets in the periphery of the visual field. Peripheral vision is color-blind, which rules out color signals. Saccadic suppression during eye movements means that some transitory event occurring in the periphery will generally be missed if it occurs during an eye movement. Taken togtether, these facts suggest that a single abrupt change in the appearance of an icon is unlikely to be an effective signal. Common examples of this are changing the color of an icon and causing the flag on a mailbox icon to be raised. Such events are likely to be missed unless the user includes the icon in an explicit monitoring schedule.

Interestly, more recent work has suggested that it may not be motion per se that attracts attention, but rather the appearance of new object in the visual field. [See above]

Unfortunately, may Web designers generate a kind of animated chart junk: Small animations with no functional purpose are often used to "jazz up" a page. The presence of these superfluous distractions interferes with our ability to make effective use of moving interrupts. (Ware, p. 160 - 161)

The most common mistake in the design and placement of signs and other message media is the thought that the customers are going into a store. When we're talking signs, it's no longer a store. It's a three-dimensional TV commercial. It's a walk-in container for words and thoughts and messages and ideas.

People step inside this container, and it tells them things. If everything's working right, the things they are told grab their attention and induce them to look and shop and buy and maybe return another day to shop and buy some more. They are told what they might buy, and where it is kept, and why they might buy it. They're told what the merchandise can do for them and when and how it can do it.

A great big three-dimensional walk-in TV commerical.

And just as in scripting and directing a TV commercial, the job is to figure out what to say and when and how to say it.

First you have to get your audience's attention. Once you've done that, you have to present your message in a clear, logical fashion-the beginning, then the middle, then the ending. You have to deliver the information the way people absorb it, a bit at a time, a layer at a time, and in the proper sequence. If you don't get their attention first, nothing that follows will register. If you tell too much too soon, you'll overload them and they'll give up. If you confuse them, they'll ignore the message altogether.

You can see, then, how a fast-food restaurant is zoned: The deeper in you are, the longer the message can be. Two or three words at the door; a napkin filled with small type at the tables. I passed a fast-food place the other day with a perfect window sign. It bore this eloquent phrase: "Big Burger." Only when you entered the place did you come upon another sign explaining the details of the teaser. (They were selling...big burgers.) That's smart sign design-breaking the message into two or three parts, and communicating it a little at a time as the customer gets farther into the store. Thinking that every sign must stand on its own and contain an entire message is not only unimaginative, it's ignorant of how human brains operate.

• One is to grab and go. This is easiest at sites selling books and records--there's something in particular you're looking for, as a gift or for yourself, and you find it fast, buy it fast and split.
• Two is to browse when you've got a few minutes to kill and are already on-line. You may buy something or not, depending on what you find. You can always use something.
• Three is to do some info-loading, gathering specs and product reviews before you purchase an event that may take place on-line or not.
• Four is for those who don't wish to buy or even browse but need to get in touch with a company to ask a question about a product they already own, or to file a complaint, or to make contact for some other reason. These people are using the Web instead of the toll-free phone number most firms maintain, knowing that endless time on hold can be avoided by using e-mail.

Starved for Attention: As Internet traffic gets increasingly concentrated on powerhouses like AOL and Yahoo, these are the lengths to which new sites will go just to get noticed. By Austin Bunn. New York Times Magazine, 2.13.00. p. 20

Web producers call it "pop-up hell" -- you come upon a site you don't like, you hit the back button on your browser and as many as 50 additional sites pop up. It's a cheap but effective way to inflate traffic numbers, and its especially popular among porn sites. PornCity.net, a free homepage company, has used the technique, with a more restrained three to five windows at a time, and Nielsen//NetRatings recently ranked it as the Internet's most popular adult site. The drawback: it tends to alienate browsers, even pornography browers. "The surfer does not want to go through that," says Paul Strouse, C.O.O. of PornCity.net. Use of the technique among nonporn sites is limited but growing. "We have an industry policy against it," says Rich LeFurgy, chairman of the Internet Advertising Bureau. "But when the coin of the realm is how much traffic you have, people do some questionable things."

The problem was that the browser's Back button had a different effect in the search results page than it did in the data pages. (Johnson, p. 339)

The basic rule is that developers must determine what users expect the Back button to do in the important cases within a Web site and, assuming there is substantial agreement among users, try their best to make the Back button match those expectations. It doesn't matter that users' expectations may not be 'logically' correct from the developers' point of view. If users' expectations differ systematically from those of the developers, the users' expectations should take precedence. (Johnson, p. 340)

Whereas the desktop metaphor previously was themetaphor, many different metaphores are increasingly becoming a prominent part of commercially available software.

Generating Metaphors
Listen to how users understand their computer systems

Build on already existing metaphors Metaphors reflecting physical structure are often successful.

Use predecessor artifacts as metaphors Using predecessor artifacts, like the physical card files at libraries, for instance is a way for the user to benefit from his or her previous experience.

Note metaphors already implicit in the problem description For instance, links are directional (i.e., information can be sent only from the source to the destination). The link metaphor could give rise to more specific metaphors like "pipes" and "paths."

Evaluating Metaphors
Choose a metaphor with a rich structure The 'link' metaphor does not have much structure in the sense that it is rather general or abstract and does not suggest any specific aspects, like directionality or one-to-manyness. In contrast, the 'TV broadcasting' metaphor is more concrete and comes with a rich vocabulary - stations, channels, reruns, serial, receivers, VCRs - all part of the everyday experience of most people.

Choose a metaphor suitable to the audience For the metaphor to support the users' understanding of the system, the literal meaning needs to be familiar to the users. For instance, for a nontechnical user community, a 'pointer' metaphor in a computer science sense would be an inappropriate metaphor for links.

Kim H. Madsen, "A Guide to Metaphorical Design" in Visualization and Design, 1994.     Homeland security example

• Web Farms
• Navigational View Builder
• Self-organizing maps
• Star Trees
• Internet foyer
• Tree Maps
• Case Studies

However, clear differences emerged within the three visualization classes, with the zoom technique being the most beneficial to users. In this regard, the most favorable visualization tool was not theoretically motivated but instead the most familiar. The expanding outline technique showed less of a decrement in performance and resulted in higher subjective ratings. The next most favorable technique of distortion, although less familiar to participants, shares a common characteristic of the expanding outline technique in that it provides context to the user in a single display. That is, both expanding outline and distortion provide details of the immediate environment and a larger context in a single display. M. Heo & S. C. Hirtle, An Empirical Comparison of Visualization Tools to Assist Information Retrieval on the Web, JASIST, 52(8):666-675, 2001.

Hillstrom, A. P. and Yantis, S. "Visual motion and attentional capture" Perception & Psychophysics, 1994, 55 (4), 399-411.

Johnson, J. Web Bloopers Morgan Kaufmann, 2000

Underhill, P. Why We Buy: The Science of Shopping. Simon & Schuster, 1999

Visualization and Design, Communications of the ACM, December 1994, v. 37, n. 12.

Ware, Colin. Information Visualization: Perception for Design. Morgan Kaufmann, 2000.