UNIVERSITY OF BRISTOL
FACULTY OF ENGINEERING

Second Year Examination for the Degrees of
Bachelor and Master of Engineering

May/June 2008

2 Hours

HUMAN-COMPUTER INTERACTION
COMS21301

This paper contains four questions.
Answer any three questions.
All questions carry 20 marks each
The maximum for this paper is 60 marks

Q1

You are an HCI consultant who has been hired to help in the redesign of a graphical
user interface for a drawing program. There is one function in the program that is used
very frequently, and the company has decided to make this function available as a
button on the interface. The company has hired you to determine which of the
following designs for the button, D1 and D2, will result in shorter movement time for
the user:
D1: a small (1 cm diameter) button placed close (8 cm) to the main work area
D2: a larger (2 cm diameter) button placed further (32 cm) from the main work area
The company has done some preliminary testing with the interface, summarized in the
table below:
Distance
8 cm
64 cm

Target Diameter
8 cm
2 cm

Average Movement Time
0.25 sec
1.0 sec

Using the following formulation of Fitts’ Law equations, answer the questions below:
ID = log2(2A/W), where A is the amplitude, and W is the width
MT = a + b*ID, where MT is movement time, a the intercept, and b the slope
(a)

Calculate the index of difficulty for the two targeting tasks that are implied by
the two proposed designs. You may estimate log values using the table given.
(2 marks)
20 = 1 25 = 32
21 = 2 26 = 64
22 = 4 27 = 128
23 = 8 28 = 256
24 = 16 29 = 512
D1: ID = log2(2*8 / 1) = 4 (1 mark)
D2: ID = log2(2*32 / 2) = 5 (1 mark)

(b)

Determine the predictive model of movement time that can be used in this
situation. Make sure you state values for the intercept a and the slope b.
(4 marks)
0.25 = a + 4b (1 mark)
1 = a + 5b (1 mark)
a= 0.1, b=0.15 (1 mark)
MT = 0.1 + 0.15*ID (1 mark)

(c)

Using your model, predict the movement time for the two targeting tasks that
are implied by the two proposed designs
(2 marks)
D1: MT = 0.1 + 0.15 * 4 = 0.7 sec (1 mark)
D2: MT = 0.1 + 0.15 * 4 = 0.85 sec (1 mark)

(d)

The company want to know whether to design their application for a Windows
or Mac platform, and have some concerns about which operating system offers

2

the best scroll bar functionality for the drawing program. Design an experiment
to test whether horizontal application menu bars found in Microsoft Windows
are faster to access than the horizontal application menu bars found on the Mac
OS. You may assume a generic application which has menu bar items such as
File, Edit, View and Help. Note that there are many possible solutions, so
concisely motivate your design using the following headings: Participants,
Null hypothesis, Variables, Conditions, Task description (instructions, time
limits etc.) and Type of statistical analysis.
(12 marks)
Any suitable answer will gain marks, 2 available for each heading. Most obvious and simple design is one
between subjects to avoid learning effects with a mac and windows condition assigned to each group separately.
Dependent variables could be time to select a range of menu items from a range of menus and maybe distance
travelled with mouse cursor as well to measure any movement redundancy. Issues to consider should include
controlling where the mouse starts and what learning effects there are between data points as well as between
subjects (may require counterbalancing). The most likely analysis would be a comparison of means using a t-test.

Q2

(a)

Give two reasons why good interface designs can fail to become successful.
Explain both reasons and give corresponding real-world examples for both.
(8 marks)

A number of possible reasons why good designs fail were discussed in course – any two would be accepted with
up to four marks available for each. For example:
• A good design may fail to become successful because consumers purchase things based on a number of
factors, not just usability or good design. For example, a good design may fail because it is more expensive, or is
packaged unattractively, or is less well advertised. Real-world examples: Macintosh computers had a more
usable design, but were more expensive than IBM PC clones; higher-quality home telephones that are easier to
use are also more expensive.
• A good design may fail to become successful because there is already an entrenched or de facto standard, and
there would be additional costs involved with switching. Real-world examples:
Macintosh vs. Windows, Beta vs. VHS, Dvorak keyboard layouts vs. QWERTY layouts.
• A good design may fail to become successful because even when people are willing to make purchasing
decisions based on usability, it is difficult to test usability on the showroom floor. This reason generally affects
larger items with a higher purchase effort. Real-world example: home appliances (e.g. washers, stoves) that
cannot be used realistically in the store.
• A good design may fail to become successful because it is “ahead of its time.” That is, some other societal
variable is not yet in place to make people believe that they can use the device. Real-world examples: Apple
Newton, Xerox Star.
• A good design may fail to become successful because people have become suspicious of the technology after
early failures. That is, the “once bitten twice shy” rule. Real-world examples: Talking vending machines, voice
recognition in consumer appliances.
• A good design may fail to become successful because there are not enough people with that specific need to
comprise a market. Commercial products generally need a certain minimum market size in order to survive.
Real-world example: Iridium satellite phones (cell coverage was good enough for most people, so too few
people bought the technology).
• A good design may fail to become successful because different people have different definitions of
“good.” That is, there may be design pressure in directions that take away from usability. Real world example:
John Denver’s plane was a good design in terms of fire risk, but not as good from a usability perspective.

(b)

Describe in detail the problems that can face an ethnographer who wishes to
explore whether an existing design will suit a particular situation. You should
mention at least two practical and two methodological challenges.
(12 marks)

3

Examples below, up to 3 marks available for each point. Broadly practical issues involve working out the best
way of keeping the data ‘naturalistic’ because ethnography is behavioural rather than behaviourist.
Methodological issues involve balancing the demands of particular design constraints against the desire to only
describe what is observed rather than what might be sought and interpreted.
Access
Getting in
Gatekeepers can be managerial sections, admin staff, shop floor’ workers
Some areas might be regarded as off-limits to observers
Observer might be associated with vested interests
Ethnographers must gain acceptability - open and honest, showing respect
Gaining credibility may include working shifts, sharing conditions, a non-intrusive demeanour (without
being too self-effacing), sharing dress codes
Initial phase
familiarising yourself with the organisation
make notes on everything that you see and hear, on what you’re told directly and on what you hear on the
other side of the room
sketch plans of office spaces and desktops; glance at official documents and scribbled notes
tape record anything that you can
start anywhere you can
hard to work out where you should be looking
Focus of the study
Choice between the innocent ethnography and informed ethnography
How theoretically, strategically (etc.) informed should you be at the outset?
One choice is progression from one to the other
Full participation or observation only?
Duration of the Study
Effective ethnographers can grasp key aspects in a relatively short time
o some aspects of the work may not be routine but exceptional
o Knowing what problems occur, how frequently, and what their significance is, how they are dealt with
and with what degree of 'competence' can provide very useful information
There are no obvious completeness rules
o Flattening of the learning curve
o Knowing what you don’t know

Q3

(a)

Describe the two alternative strategies for assigning participants (or ‘subjects’)
to conditions in quantitative evaluation. List two benefits and two drawbacks
for each alternative.
(6 marks)

Within subjects (1 mark)
One group of testers who use both systems
Advantages: (0.5 mark each)
Much more significance for a given number of test subjects.
Need fewer subjects for significant differences when using small number of conditions.
Disadvantages: (0.5 mark each)
Users have to use both systems (two sessions).
Order and learning effects (can be minimized by experiment design).
Between subjects (1 mark)
Two groups of testers, each use 1 system
Advantages: (0.5 mark each)
Users only have to use one system (practical).
No learning effects.
Disadvantages: (0.5 mark each)
Per-user performance differences confounded with system differences:
Much harder to get significant results (many more subjects needed).
Harder to even predict how many subjects will be needed (depends on subjects).

4

(b)

A manufacturer introduces a new kind of cell phone. He wishes to claim with
95% confidence that the average time it takes to charge the phone is c minutes.
For this purpose, he tests 20 cell phones and observes the following times to
charge:
78
83

82
70

84
79

75
85

80
81

86
84

74
73

87
91

82
73

77
77

Use an independent one-sample t-test to find the optimum value of c. You may
assume that the time to charge the cell phones is normal, and the resulting t
value is 2.09. Note that the degrees of freedom measure used in this test is
equal to the sample size minus one.
(14 marks)
mean = 80.05
variance = 78-80.05^2 + 82-80.05^2 + … + 77-80.05^2 / 19 = 29.62
standard deviation = root(29.62) = 5.44
standard error = 5.44/sqrt(20) = 1.217
t = mean-testvalue/standard error
testvalue ’c’ = mean - standard error*t = 80.05 – 2.09*1.17 = 77.51

Turn Over
Q4

(a)

According to Shneiderman (1998), an icon is “an image, picture or symbol
representing a concept.” Describe the conceptual metaphors which icons
contributed to resulting in the graphical user interface revolution.
(4 marks)

Icons were first developed as a tool for making computer interfaces easier for novices to grasp in the 1970s at
the Xerox Palo Alto Research Center facility as part of their Graphical User Interface (GUI) creation (1 mark),
the desktop metaphor that is so prevalent in modern personal computing today (1 mark). No longer would the
display be simply lines of code and commands, it would be graphical with a true representation of typefaces and
images. The bitmapped GUI display helped promote the concept of WYSIWYG (what you see is what you get,
1mark) allowing people to laser print exactly what they saw on the screen. Icon-driven interfaces were then later
popularized by the Apple Macintosh and Microsoft Windows operating environments. The Macintosh, released
in 1984, was the first commercially successful product to use a GUI. A desktop metaphor was used, in which
icons were created so files looked like pieces of paper (1 mark); directories looked like file folders; there were a
set of desk accessories like a calculator, notepad, and alarm clock that the user could place around the screen as
desired; and the user could delete files and folders by dragging them to a trash can icon on the screen.

(b)

Define and give an example of an auditory icon. Why are auditory icons
useful, and how they differ from Brewster’s (1995) implementation of
‘earcons’? Include in your answer one example of each.
(8 marks)

Sighted users often supplement the graphical user interface with sounds that represent certain events carried out
by their computer, for example emptying the recycle bin sounds like paper being screwed up and thrown away
could be auditory icons associated to this event (1 mark). These sounds become familiar relatively quickly
because they tend to sound similar to the action they represent and are often accompanied by a visual prop
(warning message or dialogue box). (1 mark)
Auditory Icons are “everyday sounds mapped to computer events by analogy with everyday sound-producing
events. Auditory icons are like sound effects for computers” (1 mark). Developed by Gaver (1989).
Auditory icons rely on an analogy between the everyday world and the model world of the computer. Natural
sounds have associated semantics which can be mapped onto similar meanings in the interaction (1 mark)
- e.g. throwing something away ~ the sound of smashing glass (1 mark for example)

5

-

Additional information can also be presented:
Muffled sounds if object is obscured or action is in the background
Use of stereo allows positional information to be added
Items and actions on the desktop have associated sounds
o Folders have a papery noise
o Moving files – dragging sound
o Copying – a problem…sound of a liquid being poured into a receptacle, rising pitch indicates the
progress of the copy
- Big files have louder sound than smaller ones
Auditory icons tend to be intuitive, but not all computer operations can have associated sounds which are
representative of those computer operations. This is where earcons can be used, however they have to be “learnt”
to have meaning. Earcons are defined as “non-verbal audio messages that are used in the computer/user
interface to provide information to the user about some computer object, operation or interaction” (Blattner et
al, 1989) (1 mark). Stephen Brewster (Brewster et al, 1995) successfully implemented earcons for the
sonification of specific interface elements, such as graphical buttons and scrollbars. (1 mark)
- Synthetic sounds used to convey information (1 mark for any example)
- Structured combinations of notes (motives) represent actions and objects
- Motives combined to provide rich information
o Compound earcons
o Multiple motives combined to make one more complicated earcon
- Family earcons
o Similar types of earcons represent similar classes of action or similar objects: the family of “errors”
would contain syntax and operating system errors
- Earcons easily grouped and refined due to compositional and hierarchical nature
- Harder to associate with the interface task since there is no natural mapping

(c)

Define what a tacton is, and explain why tactons are useful, including an
example of one in your answer. Describe how crossmodal icons and how other
senses such a taste and smell may in the future have their own icons associated
with them.
(8 marks)

Tactons are structured vibrotactile messages which can be used to communicate information non-visually (1
mark). They are used for communication in situations where vision is overloaded, restricted, or unavailable, such
as navigation systems for blind people, or in mobile/wearable computers. (1 mark)
Tactons are similar to Braille in the same way that visual icons are similar to text, or earcons are similar to
synthetic speech. For example, visual icons can convey complex information in a very small amount of screen
space, much smaller than for a textual description. (1 mark) Earcons convey information in a small amount of
time as compared to synthetic speech. Tactons can convey information in a smaller amount of space and time
than Braille. An encoded tactile message may be able to communicate its information in fewer symbols. The user
feels the Tacton then recalls its meaning rather than having the meaning described in Braille (or speech or text).
(1 mark) The icon is also (in principle) universal: it means the same thing in different languages and the Tacton
would have similar universality.

6

Examples (1 mark) include putting your phone to silent – most allow a vibration mode which just vibrates if you
get a message or have a call but we can varying the vibration pulse depending on who is calling. Or copying – an
earcon might use a filling up a jug noise, for a tacton we could increase the frequency of the vibrations till it goes
silent – thus in vibrations you know the item has been copied. For tactons several elements can be altered:
- Frequency, Waveform, Duration, Rhythm, Body Location, Spatiotemporal patterns
Crossmodal icons = earcons + visual icons + tactons. (1 mark) Future research will show which form of iconic
display is most suitable for which type of information. Visual icons are good for spatial information, earcons for
temporal, tactons operate both spatially and temporally so they can complement both icons and earcons.…
smellcons… tastecons… (1 mark) As have icons, earcons and tactons associated sight, sound and touch events to
them both smell and taste could be used in the future for example maybe you receive an email from a friend who
wears a particular perfume as you read the email a device releases a small puff of their perfume or it could be
used by supermarkets to tempt people with the smell of fresh bread when they are doing their online food
shopping. Smells trigger very powerful and deep-seated emotional responses, and this additional element to the
internet could enhance users' online experience by adding that crucial third dimension, bringing an extra whiff of
realism to the internet. (1 mark)

7