MSc-IT Study Material
June 2010 Edition

Computer Science Department, University of Cape Town
| MIT Notes Home | Edition Home |

Answers and Discussions

Answer to Review Question 1

Sound is not persistent, carries relatively less information, conveys a single "linear" stream in information, tends to perceivable at a greater distance than visual output, is more or less non-directional, and it is difficult for us to choose to focus on only a part of our "audio field".

This means that sound is not necessarily appropriate conveying all types of information, and not good for conveying large volumes of information. But it is particularly good for providing certain kinds of alerts, giving short term feedback that actions have been carried out, augmenting interfaces to provide additional cues to the user (e.g., Earcons).

The non-directionality and range of sound may also have implications for privacy and use in situations where sounds would create a disturbance.

Answer to Review Question 2

Main advantage of voice is that it doesn't require a special phone.

The voice message may take longer to play than the visual presentation takes to read.

If the time at which the call was made is the important thing, the need to wait until the earlier part of the audio message has played might be frustrating.

Audio interfaces tend to place a greater burden on the user's memory.

Recording contains unnecessary words and could therefore be made more efficient.

Answer to Review Question 3

Time taken may be much greater than for sighted users.

Many web sites have substantial non-text content, such as graphics, animations and so forth, which a voice synthesiser would be unable to cope with. (The problem can be alleviated, but not eliminated if web page designers use ALT tags when images are included in web pages.)

Web pages are often designed in an inherently two-dimensional way, making use of layout and organisation in significant ways. A voice synthesiser, though, reduces all pages to a one-dimensional stream of words, losing much of the structure that is otherwise visually presented.

Answer to Review Question 4

Attention is the selection of one of a number of possible perceptual and mental activities to focus on. The idea is that we have limited cognitive capabilities, and therefore have to be selective in what we perceive or think about.

Inattention can result in failing to carry out tasks (forgetting them – a "lapse") or carrying out a similar, more familiar, task in place of the desired one (a "capture error"). Focusing attention of the wrong elements of a user interface can lead to distractions, making the user's task harder, and to the user missing important stimuli (such as notification of an emergency).

Answer to Review Question 5

Visual and audio alerts – to draw the user's attention to specific and important events and conditions (e.g., emergencies, or system generated warnings).

Motion, blinking, annoyance, etc – can be a distraction and an annoyance. Use only occasionally and with care!

Colour, intensity, size, etc. – to make specific pieces of information stand out from the background.

Spatial and temporal cues – to help the user to focus on where they are currently working in a large space (e.g., a form) or what stage they are at in a process with several stages (e.g., online purchasing).

Structure and layout – structuring can help users decide what to focus attention on when a large amount of information is presented, and to stay focused.

Answer to Review Question 6

Size limits of working memory have no implications for visual menu design - the user is not expected to retain all menu items in memory, that is what the menu labels are for!

However, working memory size and decay time is highly relevant for the design of audio menus, where the user may have to remember the menu items until after the menu has completely finished playing before making a selection.

Answer to Review Question7

Knowledge or information that is in the world (i.e., in the environment or the user interface) must be perceived and recognised by the user. Knowledge in the user's head must be recalled before it is of use.

People are generally better at recognising things than recalling material from memory. Therefore, if we can build more of the information about how to use a system into its user interface, we have reduced the demands on the user's ability to recall knowledge. Therefore the mental effort of using the system should be reduced.

Discussion on Activity 2

Wide / shallow schemes are unlikely to be as effective for audio menus. The benefit of having many options in a menu doesn't necessarily apply, since the user must wait for all of the options to be spoken.

However, one problem with deep audio menu structures occurs if user gets lost, makes an error, or has to backtrack.

One way to investigate this without the need to implement a voice menu system is to carry out a "Wizard of Oz" experiment. An experimenter plays the role of the computer, and speaks the menu items in response to the user's actions. See, for example, (Dix et al., 1998) or (Preece et al., 1994) for more details about this technique. The results, in terms of time taken, errors, and so on can be compared with results from using the menu structure of an application program.

Discussion on Activity 3

Applications are many and include selling anything where being able to smell, touch, or taste a sample of the product would make the user more confident that it is what they're looking for. For instance, users could smell food products, perfumes or cosmetics before they buy.

Touch could be similarly used to allow users to feel the texture of products before buying. Some sites marketing clothing attempt to give an idea of the texture of fabric used by providing pictures. This could be greatly enhanced if the user were able to feel "for real".

Discussion on Activity 4

You probably found that it takes longer to find information in the second display than the first. A study cited in Preece et al (1994) found that on average people took 3.2 seconds to find information in the first screen and 5.5 seconds in the second. Although the two screens contain roughly the same amount on information, the structuring of information in the first makes it easier for users to focus their attention on the particular type of information they are searching for.

Discussion on Activity 7

Many examples or modes exist. Already mentioned are is the TAB key which works differently in different places. Similarly, In Word, Control-D sometimes opens the Font dialogue box, and sometimes duplicates drawing objects; a title-bar icon with an X in it sometimes closes a document and sometimes an application plus documents; and so on.

Measures to reduce the problems include: reducing the number of modes and mode-dependent commands, making modes work in a more consistent way, making it very clear what the current mode is (so that the current mode can be perceived rather than having to be remembered). The best solution is to eliminate modes entirely and make commands work the same way in all circumstances, thus reducing demands on both memory and attention.

Discussion on Activity 8

Typically, expert users have a surprising poor recall of the functioning of the systems they use. What they do tend to have a better knowledge of is how to interpret information that is available on the screen, and well as information that is not available on the screen, See (Preece et al. 1994, Chapter 5) for a fuller discussion.