A SEMIOTIC ANALYSIS OF PERSUASIVE TECHNOLOGY

An Application to Obesity Management

Isaac Wiafe, Keiichi Nakata

Informatics Research Centre, University of Reading, Whiteknights, Reading, U.K.

Keywords: Persuasive Technology, Captology, Weight Management Systems, Obesity Management, Semiotic

Analysis.

Abstract: The semiotic approach to information systems development has proved over the years as a reliable

framework for analyses, and it has been mainly focused on the development of organizational information

systems. However, the usefulness of the approach extends beyond the organizational context to other areas

of systems development. This paper presents a semiotic analysis of persuasive technologies, leading to the

design and implementation of a system for obesity management.

1 INTRODUCTION

Persuasive technology (PT) has opened new

channels to the contribution and application of

technology in the health sector. It is the use of

computing technology for persuasive activities,

mainly designed for social benefit purposes

(Lockton et al., 2008), and its application areas

range from healthcare to environmental issues. In

healthcare, PT has been applied in areas of teenage

pregnancy, STDs, weight management, and general

wellbeing among others. Despite growing interests

in PT and its potential benefits, it still lacks

established methods to systematically analyse and

understand the problem domain that leads to its

effective and systematic development.

This paper explores the use of methods in

organisational semiotics (OS) to design a PT,

particularly for combating obesity and overweight

conditions and to design a motivation technology to

encourage overweight and obese patients on healthy

eating and lifestyles.

In the next section, a review of literature on

obesity and PTs is presented. This is followed by a

descriptive analysis of weight management PT using

the semiotic framework. The systems analysis

section follows, and it focuses on requirement

analysis including the stakeholder analysis and the

design of a demonstrator system. Implementation

issues were also discussed and finally the system

was evaluated, leading to a conclusion.

2 BACKGROUND

Obesity is a condition in which excess body fat

negatively affects one’s health or wellbeing and it is

diagnosed using the Body Mass Index (BMI) or the

Quetelex index (WHO, 2000). Figures indicate that,

currently 33.3% of the male population and almost

60% of female are unhealthily heavy, and nearly a

third of boys and girls under 11 years will be

overweight or obese in 2010.This increase is a result

of unhealthy lifestyles of families; such as eating of

unbalanced diet and engaging in less physical

activities (Campbell et al., 2001).

As the years go by, human habits on food has

been affected a great deal. This is mostly due to the

abundance of food, and also it is available in more

varieties than before. Food is cheaper, especially

processed ones (Meikle, 2001). Marketing strategies

developed by retailers encourage more consumption.

In addition, with the increase of knowledge workers,

people are made to sit by their desks mostly,

resulting in the increased sedative activity.

Genes also plays a major role in obese

conditions, as obesity tends to run in families. It is

estimated that 19.8% of children living in

households where both parents are either overweight

or obese were themselves obese compared with

6.7% of children living in households where neither

parents were overweight or obese (Jotangia et al.,

2005). Obesity related genes affect one’s eating

habits and food metabolism which results in

increased risk of obesity. However, people with

157

Nakata K. and Wiafe I.

A SEMIOTIC ANALYSIS OF PERSUASIVE TECHNOLOGY - An Application to Obesity Management.

DOI: 10.5220/0003269201570164

In Proceedings of the Twelfth International Conference on Informatics and Semiotics in Organisations (ICISO 2010), page

ISBN: 978-989-8425-26-3

obesity-related genes are not destined to be obese;

rather they have a higher risk of becoming obese

(Bourn, 2001, WHO, 2000, Jotangia et al., 2005,

Campbell et al., 2001, Kumar and Johnson, 2006)

and thus needs to increase their activity levels.

Another key cause of the sharp increase in the

adulthood obesity population is the increase in the

childhood obesity. Research has revealed that most

cases of childhood obesity develop to adulthood

(Bourn, 2001) and this creates concerns on factors

that cause childhood obesity. Parents serve as role

models and influence the development of

overweight and obesity at different stages of their

children’s development; during gestation and

infancy, toddlerhood or preschool stages and

schooling age or early adolescence (Dietz, 1994).

Hence lack of proper parenthood and lifestyle

choices makes their wards become obese. As

children become obese, they develop to adulthood

with obese and overweight conditions in majority of

cases. This situation creates a vicious chain and

according to studies it is one of the most

contributing factors of the increased in obese and

overweight conditions (Kumar and Johnson, 2006).

2.1 Persuasive Technology Intervention

As the heat to curtail the menace increases, most

interventions are geared towards providing tools,

training and support to health professionals, youth

institutions, schools and other relevant stakeholders.

The focus is on combating the obesogenic

environment, making healthier choices the most easy

ones, creation of education awareness for all by

putting in place practical, ethical and evidence based

interventions across board (Jackson, 2006).

Like human persuaders, PT brings about

constructive changes in many domains by designing

applications intentionally to change attitudes and

behaviours in a predetermined manner (this should

not be confused with side effects of the use of

technology, which are not usually intended). This is

because computers are sometimes considered to be

more credible than humans (Fogg, 1997) and hence

a better choice for persuasion. They also present two

ubiquitous trends which is converging and creating a

new opportunity in preventive healthcare (Intille,

2004). The rapid improvements in mobile computing

devices and the ability to carry these devices almost

everywhere makes it possible for technology to be

used for healthcare purposes at places where humans

are not welcome (Fogg, 1997, Intille, 2004). The

second property which is the ability to give real-time

suggestions makes conveying of “just-in-time”

motivational messages possible. As such, the future

of PTs in healthcare interventions appears to be

bright. However, design issues are limitations to its

success. As social animals, human’s attitude and

emotion are unpredictable; it may be independent or

dependent on a number of factors which are not

easily determined. This presents a critical challenge

to the development of PTs. To have a better

understanding of how they should work, it is

envisaged that semiotics can be used to analyse and

make explicit some critical factors to be consider

during analysis and design.

3 SEMIOTIC ANALYSIS

The application of semiotics presents an informative

approach for the descriptive analysis and design of a

weight management system. In general, there are

three major approaches in the study of semiotics for

information systems development, namely,

knowledge-oriented, behaviour-oriented, and

system-oriented approaches. This paper focuses on

the behavioural approach.

Our ability to transfer knowledge depends on the

use of signs which affects our behaviour (Andersen,

1990) and attitude directly. However, some research

disputes that semiotics should not be considered as

part of cognitive psychology and it puts it out of

context in the enquiry into the work of the human

mind (Jorna and Liu, 2008). That notwithstanding,

the issue of persuasion is related to our intention,

motive, behaviour and attitude, thus the

consideration of signs as artefacts used in the

transmission of knowledge and information is

critical in the development of PTs and cannot be

ignored completely. As such, an analysis based on

the semiotic ladder introduced by Stamper

(described in (Liu, 2000)) was considered at all

levels and the following observations were made.

Figure 1: The semiotic framework of PT adopted from

(Liu 2000).

Human Information SOCIAL WORLD: change in

attitude or behaviour

Functions PRAGMATICS: interaction and dialogue

for context

SEMANTICS: Persuasive messages and

interaction

The IT

Platform

SYNTACTICS: interface design, language, choice of

words, icons and patterns

EMPIRICS: media channels, synchronization traffic, device

memory, processing speed

PHYSICAL WORLD: hardware, sensors

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

158

3.1 Social, Pragmatic and Semantic

The social world deals with social norms and

impacts of signs and their uses. How society accepts

the use of a particular sign and its meaning, by

considering both the direct and indirect effects. In

the development of a PT the ability for the

technology to create real effects in the social world,

e.g., change of attitude or behaviour, is the primary

interest in this level. In the case of obesity, some of

the most important issues that affects the

persuasiveness can be identified as, privacy, motive

of design and legislature surrounding the healthcare

industries. Most patients suffering from obesity and

overweight conditions prefer to be anonymous due

to the stigma attached to this unhealthy condition,

thus the need to considered all these issues during

the design stage is crucial.

At the pragmatic level, the design intention, and

the communication methods is considered. Like

human persuaders, a persuasive technology needs to

adopt a strategy. It was observed that pragmatics

mainly focuses on the primary task support (Oinas-

Kukkonen and Harjumaa, 2008) and principles such

as reduction, tunnelling, tailoring, personalization

are some of the methods that can be applied to

achieve the desired intention for the envisaged

weight management system. Also, at this level, it

seeks to identify whether the intended meanings of

messages are understood effectively in the context of

users. Different meanings can be given to the same

sign depending on its observer. For example, a “do

not eat now” message displayed to a user may have

variable meanings depending on the time of display,

how it is displayed and whom it is displayed to.

When this message is displayed to a hungry user,

he/she may see it more of a coercive message rather

than persuasive, whereas the same user might

consider it as a persuasive message if he/she just

finished his/her lunch. By concentrating on dialogue

support systems (Oinas-Kukkonen and Harjumaa,

2008) a good rapport can be established between the

user and the technology which will enable the

delivery of messages in the appropriate time to

enhance systems credibility.

Whenever credibility is achieved, assurance is

created and audience is established. At the semantic

level, it is expected that the system conveys the

persuasive message in a manner that is seen as the

truth to the user. This can be achieved by the

demonstration of facts which can also be associated

with credibility. More important, is the relationship

between the semantic level and the pragmatics level.

In the case of the weight management systems, the

ability to establish credibility is crucial in the

persuasive approach since users need systems they

can trust. The use of convincing logics and

demonstration of principles in weight management

would enhance the credibility of the system. Design

intentions and methods should be explicit and users

should be made to understand the rationale behind

the task they perform.

3.2 Syntactic, Empirics and Physical

As a persuasive system, the approach used may vary

for each individual. As such, extra care must be

considered in the selection of “language” and

structure used. A clear and well organized formal

language structure will enable the user to acquire

and understand the persuasive message. However,

the system should take into account all categories of

users. The interface design for variable age groups

must be considered, choice of words, icons and the

way they are presented in order not to create

ambiguity and distortion of intent, is very crucial.

The interface structure, ease of navigation through

the system and complexity of icons used are some of

the few things which need attention at the design

stage. Familiarity enhances motivation (Nass et al.,

1995) since users will not need to study new task

and activities in order to follow a PT System.

Empirics deal with the study of the quantitative

and statistical properties of signs in the physical

media. What is analysed in this level is how the data

and information on which the persuasive messages

are created are effectively transmitted to the

interface. In this case of a weight management

system, consideration should be given to the various

media channels. For example, since it would be

expected that users can access the system readily,

internet and synchronization traffic should be

considered if the system will operate over the

internet. Also, mobile device have relatively less

memory and slower processing clocks, thus the need

to consider these factors at the analysis stage, as

users will not be considerate and enthused with PT

systems when they are uncomfortable to use.

The physical world concentrates on the physics

of the envisaged system. In this case, our

consideration is focused on the hardware that users

interact with, which also processes and manages

data. A critical attention is needed on the types of

hardware which is to be used for the design and

implementation. PTs are purposefully designed to

motivate and persuade for attitude and behavioural

change as such familiarity with the hardware is a key

factor to the success of persuasion. As much as

A SEMIOTIC ANALYSIS OF PERSUASIVE TECHNOLOGY - An Application to Obesity Management

159

possible, the use of a new hardware should be

avoided unless it is essential. Also, since in some

cases the ubiquitous nature of the system might have

enormous effects on its use and success, there might

be instances where there will be the need to

introduce new hardware and this should be carefully

thought through.

4 OBIMO PET DESIGN

The approach used for the requirement analysis

process was a standard specification approach which

elicits the requirements of all stakeholders involved

(Shvaiko et al., 2005) in the development and usage

of a weight management system. In this, a thorough

review of literature on some existing systems,

enabled the identification of relevant stakeholders

and a familiarization process helped in the

identification of key stakeholders whose interests are

to be met, leading to the identification of relevant

events and determination of inputs and outputs

necessary for accomplishing those events. For

simplicity, the activities were categorized into:

• Stakeholder identification, their roles and

interest

• Elicitation and definition of software

requirements

Stakeholders are individuals, organizations or group

of people actively involved in a project and their

interest can be positive or negative (Start and

Hovland, 2004). Due to constraints on time and

resources, it was not made possible to have formal

interviews with stakeholders in order to identify

their needs. However, extensive review on

literature, case studies and casual interviews were

used and the stakeholders were categorized based on

the roles they play in the project. Figure 2 is a

representation of stakeholder categorization.

Figure 2: Stakeholder onion.

4.1 Stakeholders’ Needs

Our study established that though there are various

weight management systems in operation, they are

not effectively utilized. This is because weight

control issues are behaviour - oriented (Kumar and

Johnson, 2006), however, most existing systems are

not designed to motivate or persuade whereas

stakeholders require an effective system with a

persuasive functionality which shall able to:

• provide health status in terms of weight

• provide information on healthy food recipes

• be able to suggest healthy menus

• provide assistance to workout activities

• be easily accessible and readily available

• provide information on their performance

• motivate to control weight

• respect their privacy

• be easy to use

4.2 The ObiMo Pet Solution

In line with the need of stakeholders, a weight

management system was designed to include

motivational functionalities as required. The solution

system was named ObiMo Pet, because it combines

the catering of a Pet with an obesity control system.

ObiMo Pet is a PT interface developed for PCs and

Mobile devices to motivate and encourage

individuals to follow and monitor their eating and

physical activity. Figure 3 is an overview of the

design concept. It comprises a virtual pet and the

traditional weight management system. In the

system, users are made to set their own goals and

each time a user accomplishes a goal, points are

allocated for the goal. The points can be energy

points or reward points. Reward points are allocated

for accomplishing food menu goals (calorie intake)

and energy points are allocated for exercise and

workout goals (calories burned). The user uses the

accumulated points to cater for their virtual pet and

the status of the pet in the ObiMo Pet Community

(virtual community) is directly proportional to their

performance in terms of weight management.

Figure 3: Overview of design concept.

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

160

Reward points are earned as a user set targets on

calorie intake reduction, by consuming healthy but

low calorie food. It is computed as follows:

Reward Points = ((Target calorie intake -

recommended calorie intake)/10 + (target

calorie intake/actual calorie intake)) X 100

Cash Amount = ((Target calorie intake -

Recommended calorie intake)/10) + (Target

calorie intake/ Actual Calorie intake)) 100

Where

(Target calorie intake – recommended

calorie intake) ≥ 0

This is because, users are not expected to set targets

below the recommended calorie intake in order to

lose weight, and since rapid weight loss has serious

health consequences. Dieticians recommend that,

calorie intake is based on the type of occupation and

daily activity level and these values would be

computed accordingly as a hidden feature in the

system. This means that in an ideal situation, a user

will earn 100 points for accomplishing his target as,

(Target calorie intake – recommended

calorie intake) = 0,

and

(Target calorie intake/actual calorie intake) = 1.

Also, energy points are equivalent to the total energy

burned, hence

Energy points = total energy burned

The reward and energy points are used for various

activities in the ObiMo community and this enables

the pet to gain higher status in the virtual

community. There are 3 main factors which

contribute to the status (rank) of a user’s pet. These

are the level of strength, happiness and health. The

sum of the 3 factors compared to other pets is the

rank of a user’s pet. In each activity (e.g. play, bath,

dance, shop, etc) some amount of energy is lost, and

converted to points gained for performance

(happiness, strength and health) summing up to

indicate the status of the user’s pet. In the real world,

users save real cash when they eat less food and gain

muscle strength when they engage in physical

activities. This has been replicated as reward and

energy points respectively in ObiMo.

ObiMo Pet is designed to operate on two

different devices (PC and mobile devices). However,

it is not necessary to own both devices in order to

use the system, though a combination of the two

devices is highly encouraged as it adds value to the

persuasiveness and motivational nature of the

system. The architecture of the system is illustrated

in Figure 4. The main database resides on a server

which connects to users over the internet.

Users have client databases on their devices so as

to enable them to use the system when it is offline.

The system automatically synchronizes with the

remote database whenever it connects to the internet.

In addition, the mobile device can be synchronized

with the PC to enable the user have full access to the

system even without internet, though some

functionalities such as “status in virtual community”

is only active when there is internet connectivity.

Figure 4: Architecture of ObiMo pet.

5 IMPLEMENTATION

The software was developed in a Visual Studio

environment and the codes were written in Visual

Basic. The rapid prototyping model was used and it

involved the development and construction of

prototypes, by merging various structured

techniques to accelerate the software development

(Whitten et al., 2000).

5.1 Hardware Components

The system was implemented on two devices; PC

and mobile handset. The PC interface was installed

on a system with, Windows Vista Home premium,

Intel core Duo T8100 @ 2.1GHz, 2046MB of RAM

and 250 GB HDD, WLAN. The mobile interface

was installed on a Sony Ericson Xperia X1 with,

Built-in GPS, USB support, WLAN, Microsoft

Office Mobile, Internet Explorer, Windows Media,

Exchange ActiveSync, Voice control and Utility

Applications.

5.2 Software Components and GUI

As a PT, the interface is expected to be interactive,

since it is to perform activities of a human

persuader, there is a need for a high level of human-

computer interaction. As such, the interface was

A SEMIOTIC ANALYSIS OF PERSUASIVE TECHNOLOGY - An Application to Obesity Management

161

designed to be simple but interactive. For

demonstration purposes, this paper presents a

selection of screenshots for the mobile interface.

Figure 5: My Pet.

The system has features such as profile

management, tools, pet selection, and activity

predictions among others. For instance, the profile

manager enables the system to capture useful vital

statistics such as name of the user, address, type of

occupation, gender, date of birth, vital statistics,

among others. The system uses this information to

create an account for the user. Most of the

information provided is also used for accessing the

user and calculating vital information which the

system uses for its weight management approach.

For instance, the type of occupation of the user

enables the system to determine one’s recommended

calorie intake per day which is very essential for

rewarding the person.

Figure 5 is a screen shot of the “My Pet” window

for the mobile device. It can be observed that the

system displays the status of the pet by showing the

progress for the happiness, health, strength and the

overall rank. In addition to this, a text to speech

function recites to the user, the performance of his

pet compared to other users. Users can also click to

feed, shower or play with their pets. As mentioned

earlier at the design stage, respective points are

accumulated as the user interacts with their pets.

Figure 6: My Tools.

The tools window (figure 6) provides the user

with a set of useful tools necessary in weight

management activities. These functions includes,

BMI calculator, BMR calculator, Performance

charts, Calorie burned calculator, Personal calendar,

and recipe manage, among other. For example, the

recipe manager displays a list of meals and their

corresponding recipes, hence users are provided with

guidelines for preparing new meals whenever they

are introduced.

Figure 7: GPS System.

The mobile device interface also has an activity

inference feature. This tracks a user’s speed with the

help of an in built GPS receiver and infers the

possible current activity. When an activity is

inferred, the user is made to confirm or select one of

the possible activities (see figure 7). After which the

user needs to input the expected duration of his

selected activity. The system then displays his

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

162

expected calorie consumption. This facility serves as

a self-monitoring system for the user.

6 DISCUSSIONS

As part of the objectives of the research, the

proposed solution was evaluated. In the case of a

persuasive technology, the appropriate method of

evaluation should be to make the system accessible

to a number of randomly selected people over a

defined period of time. This will enable the

identification of the efficiency and effectiveness of

the system. However, due to limitations and

constraints on resources and time, this activity could

not be performed. Instead, the approached used was

to assess the method of design and reflect on how it

addressed the intended persuasion. One key

characteristics of a persuasive technology is the

ability to measure progress.

6.1 Intentions and Methods of PT

PTs should be assessed by their intention and

methods and from a deontological viewpoint, the

intended persuasions and methods need to be in

accordance with moral principles by considering

whether the technology or the methods used are

harmful, discriminative or lead to the infringement

on privacy. Also, during the analysis stage, at the

semantic level, it was established that the intended

meaning of a persuasive system should be made

explicit.

The intention of ObiMo Pet was to motivate its

users to control their weight by following through

activities which are designed for weight

management. It is clear that there is no hidden

agenda and users of the system have full control of

their lives. However, there might be the possibility

that other unintended behaviours of a user may

change as a result of using the system

(microsuasion). For example, a user may start

keeping pets or improve upon the way he interacts

with his pet. Another aspect examined was system

abuse. The interactive and gaming nature of the

system might make a user be addictive to the

system. This might end up in two possibilities:

• The user may learn to cheat the system in order

to gain more point for the pet interaction. This

will lead to the user not achieving his goals but

rather using the system for a different interest

• The user may attempt to set unreasonable goals

which might put his health in danger.

6.2 Privacy

Privacy is a delicate issue and it does not occur only

in areas of persuasive technology. As identified at

the social level, a persuasive technology should fit

into social norms and ethics. Accordingly, the status

and competiveness of a user is transferred to a

virtual pet and by this, the user remains anonymous

to other users. In addition, users need to log into the

system each time they use the system by entering a

user name and password and this create restriction to

accounts.

For the mobile device, due to convenience and

screen size limitations, the login function was not

implemented, however it is expected that users have

full control of their mobile phones than they do for

their PCs.

6.3 Implications of Semiotic Analysis

The semiotic analysis described in Section 3

highlights the issues that should be considered in the

design of PT. In particular, since the purpose of PT

is to create positive effects in terms of changes in

attitude and/or behaviour, the analysis at the Social

World level is crucial in achieving or evaluating the

effectiveness of PT. While there are a wide range of

systems and tools for PT developed and propose so

far, these are often developed without any

systematic approach. The semiotic analysis enables

the designers to consider how the implementation of

PT would lead to the intended effect through the

analyses at each level of semiotic framework.

In the design of ObiMo Pet, the semiotic analysis

was useful in considering design features. However,

this was still implicit. There is a potential in the

development of a PT design framework based on

semiotic analysis to provide a more systematic

approach to the design of PT.

7 CONCLUSIONS

This research has presented a conceptual design for

persuasive weight management system (ObiMo Pet)

to motivate users who are suffering from overweight

and obesity to follow a weight management plan. It

explained the significance of the semiotic approach

in the analysis of the state and problem of obese and

overweight conditions, and how it can be applied in

the design of persuasive technologies.

During the requirement analysis relevant

stakeholders where identified and their roles and

impact in connection with the new system was

A SEMIOTIC ANALYSIS OF PERSUASIVE TECHNOLOGY - An Application to Obesity Management

163

established. Though the system was partially

implemented, it managed to explain the

appropriateness of the conceptual design.

It can be recalled that during evaluation, it was

established that users may be addicted to the

interactive Pet and this may tempt to the user

abusing or cheating the system. It will be of great

interest if a function can incorporated into the

system to cater for this happening, i.e., to dissuade

this behaviour. Also, the activity prediction feature

of the system on the mobile device was limited,

since GPS receivers are limited in terms of reception

(cannot receive signals in rooms and buildings).

However, different hardware can be used to detect

movement or activities and this can be incorporate

into the system to enhance self-monitoring.

Finally, there is a potential in developing a more

systematic approach to persuasive technology design

and development based on semiotic analysis. Such

an approach would be necessary for further research

into persuasive technologies.

REFERENCES

Andersen, P. 1990. A theory of computer semiotics:

semiotic approaches to construction and assessment of

computer systems. Computational Linguistics, 18,

555-562.

Bourn, J. 2001. Tackling Obesity in England. Report by

the Comptroller and Auditor General. House of

Commons, 220, 2000–2001.

Campbell, K., Waters, E., O'meara, S. & Summerbell, C.

2001. Interventions for preventing obesity in

childhood. A systematic review. Obesity Reviews, 2,

149-157.

Dietz, W. H. 1994. Critical periods in childhood for the

development of obesity. Am J Clin Nutr, 59, 955-9.

Fogg, B. J. 1997. Captology: the study of computers as

persuasive technologies, New York, ACM New York.

Intille, S. S. 2004. A new research challenge: persuasive

technology to motivate healthy aging. IEEE

Transactions on information technology in

Biomedicine, 8, 235-237.

Jackson, A. 2006. Combating the obesogenic environment.

London: Weight Management Centre.

Jorna, R. & Liu, K. 2008. ORGANIZATIONAL

SEMIOTICS Round Table Workshop ‘An

organizational semiotic view on interculturality and

globalization’at the IASS 2004 Conference. Relation,

10, 7692.

Jotangia, D., Moody, A., Stamatakis, E. & Wardle, H.

2005. Obesity among children under 11. London:

National Centre for Social Research, Department of

Epidemiology and Public Health at the Royal Free

and University College Medical School.

Kumar, H. & Johnson, P. 2006. The causes of obesity -

genes, behaviour and the environment [Online].

London: Cancer Reserach UK. Available:

http://info.cancerresearchuk.org/healthyliving/obesitya

ndweight/whatcausesobesity/ [Accessed May 15

2009].

Liu, K. 2000. Semiotics in information systems

engineering, Cambridge Univ Pr.

Lockton, D., Harrison, D. & Stanton, N. 2008. Design

with intent: Persuasive technology in a wider context.

Lecture Notes in Computer Science, 5033, 274.

Meikle, J. 2001. Action urged to cut the fat of the land.

The Guardian [Online]. Available: http://www.

guardian.co.uk/society/2001/feb/15/publichealth.ukne

ws [Accessed Febuary 15, 2009].

Nass, C., Moon, Y., Fogg, B. J., Reeves, B. & Dryer, C.

Year. Can computer personalities be human

personalities? In, 1995. ACM New York, NY, USA,

228-229.

Oinas-Kukkonen, H. & Harjumaa, M. 2008. A Systematic

Framework for Designing and Evaluating Persuasive

Systems. Lecture Notes in Computer Science, 5033,

164-176.

Shvaiko, P., Léger, A., Paulus, F., Rocuet, L., Nixon, L.,

Mochol, M., Kompatsiaris, Y., Papastathis, V. &

Dasiopoulou, S. 2005. Knowledge Processing

Requirements Analysis. KnowledgeWeb.

Start, D. & Hovland, I. 2004. Tools for policy impact: a

handbook for researchers, Overseas Development

Institute.

Whitten, J. L., Bentley, L. D. & Dittman, K. C. 2000.

Systems Analysis and Design Methods 5e, McGraw-

Hill Higher Education.

Who 2000. Obesity: preventing and managing the global

epidemic. World Health Organization Technical

Report Series.

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

164