user modelling in user-centred system design (ucsd) chapter 3

22
User modelling in user- centred system design (UCSD) Chapter 3

Upload: julie-parker

Post on 15-Jan-2016

291 views

Category:

Documents


7 download

TRANSCRIPT

Page 1: User modelling in user-centred system design (UCSD) Chapter 3

User modelling in user-centred system design (UCSD)

Chapter 3

Page 2: User modelling in user-centred system design (UCSD) Chapter 3

By the end of this chapter you should be able to...

• to define a user model• develop a user model• describe various ways of creating user models • explain the Model Human Processor and Simplistic

Theory• explain the Simplex One model• explain and apply Nielsen’s heuristics

Page 3: User modelling in user-centred system design (UCSD) Chapter 3

3.1 Introduction

• A user model is a psychological way of depicting the people who will use the system– Considering users’ needs and preferences– Present psychological processes, individual

differences, social context, culture, lifestyle, task objectives

Page 4: User modelling in user-centred system design (UCSD) Chapter 3

3.2 Types of user models

1.Psychological theories as user models– Human cognition is very complex– These theories expect a system analyst to be a

psychologist

2.Task-based analysis for user models– Evaluation of core tasks could lead to

consideration of how users will undertake these tasks

– Behaviouristic approach which has been overtaken by cognitive science

Page 5: User modelling in user-centred system design (UCSD) Chapter 3

3. Cut-down psychological theories as user models– Example is Model Human Processor theory (MHP)

• Simpler, without irrelevant complexities of full-fledged theory• Developed directly to solve IT design problems

– Assumes that information flows from user to system but not back

4. Simplistic psychological theories as user models– Conceptual framework for theories and research findings– To be used by both designers and computer scientists as guide

without becoming psychologists Simplex One as a simplistic theory (more details later)• Five modules of cognition – partially independent of each other• Each module has memory and local processing capacity

Page 6: User modelling in user-centred system design (UCSD) Chapter 3

3.3 User models and evaluation

HCI has many principles or heuristics for design and evaluation

Four kinds in order of ascending specificity:• Guidelines (they contain design principles)

– General & not prescriptive– Describe usability concepts to assist in judging designs– Cover a wide range of interface features and concepts– Examples

• Know the user population• Reduce cognitive load• Engineer for errors• Maintain consistency and clarity

• Style guides– Developed by companies for internal use– Enhance usability through consistency, i.e. look and feel across

product lines are maintained

3.3.1 Design principles (heuristics) for evaluation

Page 7: User modelling in user-centred system design (UCSD) Chapter 3

• Rules– Precise & prescriptive recommendations about design

aspects– Examples

• never use red text on a blue background• always position the waste bin in the bottom right hand

corner• always issue a warning before the user deletes a file

• Standards– Imposed by regulatory bodies, e.g. ISO

• Sometimes mandatory, i.e. legal requirement – could be sued if not kept

• Part of contract – won’t be paid if not kept– Specific type of system in a particular industry

Page 8: User modelling in user-centred system design (UCSD) Chapter 3

Dix’s design principles

• Learnability– Humans learn through understanding and

generalizing (i.e. have mental model) of the system as a whole

• Predictability– User’s knowledge must be sufficient to work

out what actions are possible next and what their consequences will be

• Consistency– Essential and most widely applied design

principle

3.3 User models and evaluation

Page 9: User modelling in user-centred system design (UCSD) Chapter 3

• Flexibility– Extends the way in which user and system can

exchange information

• Recoverability– Humans learn by exploration: need a way to undo

mistakes

• Responsiveness– Rate of communication between user and system

NB: Guidelines must be clear and unambiguous• Beware of “expert knowledge fallacy”

Page 10: User modelling in user-centred system design (UCSD) Chapter 3

• Aim: Simplex One facilitates the design process – it assists in understanding key design issues in terms of the psychology of the user

• Designed in two levels– Framework to understand research findings– Guide practical problem solving

• Five cognitive modules that can take input from any other module via executive function

• Theory can be used to generate simple questions on either– the strengths and weaknesses of users (3.3.3)– proposed design (3.3.4)

3.3.2 Simplex One

3.3 User models and evaluation

Page 11: User modelling in user-centred system design (UCSD) Chapter 3

Simplex One Cognitive modules

Page 12: User modelling in user-centred system design (UCSD) Chapter 3

• Perception (input)– Take in new information from the senses– Analyse & Store

• Response (output)– Select, organise, time and implement

appropriate responses

• Abstract working memory– Take, hold, and process task-relevant

memories– Limited in capacity, time and processing

rates

Simplex One Cognitive modules: more detail

3.3 User models and evaluation

Page 13: User modelling in user-centred system design (UCSD) Chapter 3

• Long-term memory– Stores occurrence of key events and symbols– Freer of constraints– Builds over time as experience grows– Episodic memory: Remember episode in context– Semantic memory: Knowledge without context

• Executive functions– Information is switched between foregoing four modules– Sequences of transfers are organised– Functions of different modules are mediated– Task requirements are organised and monitored

Page 14: User modelling in user-centred system design (UCSD) Chapter 3

• Typical questions about users (refer to text book as well):

• Perception: What are the perceptual skills of the user groups likely to be?– “How many users could have visual impairments?”– “Do the symbols chosen make sense for the user groups?”

• Responses: What are user responses likely to be like?– Are there users with problems in responding, e.g. verbally or psycho

motoric?– E.g. elderly people might have problems in controlling hand position.

• Abstract working memory: What memory capabilities do the users have?– Taking into account limitations of human memory (capacity, time and

processing rate), what can be done to make task execution easier for users?

– Do some users have significantly different memory resources than the average user? E.g. users with learning difficulties or older users

3.3.3 Evaluating Users with Simplex One

3.3 User models and evaluation

Page 15: User modelling in user-centred system design (UCSD) Chapter 3

• Long-term memory (LTM): How old are the average user? – Use metaphors to link prior knowledge with system elements– (Young users have less experience and may be less open to

metaphors but more open to new ways of organising)

• Executive functions: Might the users vary in executive functioning?– Some are better at learning than others and sometimes more

and better training might be necessary– Some might need more system feedback than others– Some users might be unaware of problems while thinking that

they are performing well while it is not the case.

Page 16: User modelling in user-centred system design (UCSD) Chapter 3

• Typical questions about the design (refer to text book as well):• Perception: Does the design allow information to be delivered in ways

that will respect users’ sensory capabilities and perceptual skills?– E.g. “Does the system cater for users with age-related or specific

sensory and perceptual deficits?

• Responses: What response options are available for users?– Can users select, organise, time and implement appropriate

responses?– Range of response media adequate?– Adequate support for users with specific problems with verbal and

psychomotor responses?

• Abstract working memory: Does the system respect limitations of human memory?

3.3.4 Evaluating Designs with Simplex One

3.3 User models and evaluation

Page 17: User modelling in user-centred system design (UCSD) Chapter 3

• Long-term memory (LTM):– Does the design rely on unjustified assumptions about users’

prior knowledge?

– Did the designer assume that his/her own prior knowledge is representative of that of the users?

• Executive functions: What demands do the design place on executive functions of users?– E.g. “What new tasks do the design expect users to learn?– Will users be expected to perform familiar tasks in unfamiliar

or contradictory ways?– Do the users require special features to enable them to

perform well?

Page 18: User modelling in user-centred system design (UCSD) Chapter 3

• Heuristic: Search strategy• Not absolute

– Involves interpretations and judgements– Checklist: Reminds the designer to be on the look-out for

something– Example: There is an undo facility

• Nielsen’s heuristics1. Visibility of system status 2. Match between system and the real world 3. User control and freedom 4. Consistency and standards 5. Error prevention 6. Recognition rather than recall 7. Flexibility and efficiency of use 8. Aesthetic and minimalist design 9. Help users recognise, diagnose, and recover from errors 10.Help and documentation

3.4 Heuristic evaluation

Page 19: User modelling in user-centred system design (UCSD) Chapter 3

1. Visibility of system status– Keep the user informed about what is going on through

providing appropriate feedback in a reasonable time– Must be timely, visible and meaningful– Message boxes, progress bars, hour glass, pop up hints,

animations, clipboard, depressed buttons, borders around selected items, etc.

2. Match between system and the real world– Speak the user’s language (phrases and concepts from the

user’s world rather than system-oriented terms)• Avoid: “cache”, “database”, “compile”, etc.

– System should look and behave in ways that are familiar to the user

• Use metaphors

3. User control and freedom– Emergency exits: Provide a way to escape from places where

users unexpectedly find themselves– Users must feel confident to explore

3.4 Nielsen’s Heuristics

Page 20: User modelling in user-centred system design (UCSD) Chapter 3

4. Consistency and standards– Don’t use the same word for different meanings or describe the same thing

with different words– Support users’ expectations– Experience with a small part of the system should facilitate easier learning of

the rest of the system– Users have good spatial memory

• Screen arrangement should be consistent• Positioning of features should be consistent, e.g. links, index, search, etc.

5. Error prevention– If possible, prevent errors in the first place– System errors

• E.g. run-time errors, erroneous processing, etc.• Unforgivable

– HCI-errors• Misleading terminology, e.g. user selecting wrong menu option is not necessarily

user error• Lack of feedback cause user to reselect a submit payment option

– User errors• E.g. typing errors, poor mouse handling, weak understanding of the system• Enforce constraints, e.g. limit typo’s by providing dropdown list as far as possible,

masks for dates and currency

3.4 Nielsen’s Heuristics (continued)

Page 21: User modelling in user-centred system design (UCSD) Chapter 3

6. Helping users recognise, diagnose and recover from errors– Use plain language to describe problem– Suggest way to solve– E.g. submitting a form with not all fields completed:

• NO: “Form not accepted”• YES: Present the form again with forgiving message and indication of

missing info

7. Recognition rather than recall– Make objects, actions, and options visible– Example: Use radio-buttons and lists rather than expecting a user to

type commands– Use clear metaphors, e.g. shopping cart– Objects should be labelled well

3.4 Nielsen’s Heuristics (continued)

Page 22: User modelling in user-centred system design (UCSD) Chapter 3

8. Flexibility and efficiency of use– Accommodate novice users while providing features for power users to work

more efficiently and productively

9. Aesthetic and minimalist design– Avoid information that is irrelevant or rarely needed– Avoid eye-catching graphics or vivid colours – it clutters the screen– Banner ads, Flash, animations, etc are waste of space and confuse users

10. Help and documentation– Easily searchable– Stepwise– Context sensitive– Terminology easily understandable for users

3.4 Nielsen’s Heuristics (continued)