Abstract Modern life has increasingly become intertwined with technology, and recent years have witnessed a growth in technologies that support people in, for instance, leading healthier and more sustainable lifestyles. At the centre of this growth has been persuasive systems design, which has been shown to have a positive effect on individuals’ behaviour and their use of systems. This dissertation consists of five studies, encompassing a literature review, two quantitative studies with a total of 227 respondents, and two constructive studies that address the central research question of the dissertation: How can integrating judgment and decision-making processes and persuasive software design patterns enhance the development of behaviour change support systems? The primary theoretical framework for the research is the Persuasive Systems Design model. This is a model that outlines the key requirements for developing persuasive systems, consisting of the theoretical underpinnings, persuasion context analysis, and four feature categories. In recent years, improving the design of persuasive systems to better achieve their intended objectives has been an important topic. This dissertation, in addition to examining the role of persuasive software features in influencing behaviour, also integrates behavioural economics and software design patterns into the design of persuasive systems. Additionally, the interplay between the categories and other constructs such as perceived competence is investigated through statistical analyses. Overall, results reveal that persuasive system features have an impact on the efficacy of behaviour change support systems. Additionally, integrating behavioural economics concepts that explain the reasons why individuals deviate from expected behaviour and software design patterns can help improve the development of persuasive systems and further enhance their efficacy. / Tiivistelmä Teknologia on yhä tiukemmin osa nykyelämää. Viime vuosina on tapahtunut kasvua ja kehitystä teknologioissa, jotka tukevat ihmisiä esimerkiksi elämään terveellisemmin ja ympäristöä säästäen. Tämän kasvun keskiössä on ollut vakuuttavien järjestelmien suunnittelu, jonka on osoitettu vaikuttavan positiivisesti sekä ihmisten käyttäytymiseen että järjestelmien käyttöön. Tämä väitöskirja käsittää viisi tutkimusta, sisältäen kirjallisuuskatsauksen, kaksi kvantitatiivista tutkimusta yhteensä 227 vastaajalla, ja kaksi konstruktiivista tutkimusta, jotka yhdessä vastaavat väitöskirjan päätutkimuskysymykseen: Kuinka arviointi- ja päätöksentekoprosessit sekä vakuuttavien järjestelmien suunnittelumallit yhdistämällä voidaan edistää käyttäytymisen muutosta tukevien järjestelmien kehitystä? Ensisijainen teoreettinen viitekehys tutkimukselle on vakuuttavien järjestelmien suunnittelumalli (Persuasive Systems Design model). Kyseinen malli määrittää keskeiset vaatimukset vakuuttavien järjestelmien kehittämiselle. Tärkeänä aiheena on ollut vakuuttavien järjestelmien suunnittelemisen edistäminen, jotta niillä voitaisiin paremmin saavuttaa aiotut päämäärät. Vakuuttavien järjestelmien ohjelmisto-ominaisuuksien vaikutuksesta käyttäytymiseen tutkimisen lisäksi väitöskirja yhdistää myös behavioristisen taloustieteen ja ohjelmistosuunnittelumallit vakuuttavien järjestelmien suunnitteluun. Lisäksi kategorioiden ja muiden käsitteiden, kuten koetun pätevyyden, vuorovaikutusta on tutkittu tilastollisen analyysin keinoin. Kaiken kaikkiaan tulokset paljastavat vakuuttavien järjestelmien ominaisuuksilla olevan vaikutusta käyttäytymisen muutosta tukevien järjestelmien vaikuttavuuteen. Lisäksi integroimalla behavioristisen taloustieteen konsepteja, jotka selittävät syitä, joiden vuoksi yksilöt käyttäytyvät odotetusta poikkeavasti, ohjelmistosuunnittelumalleihin, voidaan auttaa edistämään vakuuttavien järjestelmien kehittämistä ja parantaa niiden vaikuttavuutta.

Physical inactivity is a growing public health problem in the 21st century. A large proportion of the adult and adolescent population is not achieving recommended levels of physical activity which has significant consequences for both the individual and our society. With no evidence that the levels of physical inactivity will improve in the future, traditional methods of addressing this problem have evidently proven ineffective. A relatively new and promising approach to this problem that has become increasingly important in recent years is smartphone-based behavioral interventions to promote physical activity and other health-related behaviors. While several studies have shown many positive and promising results with this approach, other studies have also shown that the method is not completely understood. For instance, research shows that many of the health and fitness applications available today do not adequately adhere to behavior change strategies or health-related research, suggesting that these applications are not reaching their fullest potential. The goal of this study was to develop and evaluate an application to promote physical activity that was aligned with such research. A high-fidelity prototype based on the Behaviour Change Wheel was developed in Adobe XD and later evaluated by four participants. The outcome of the evaluation indicated that the developed prototype was well received overall, and participants believed it would trigger the desired behavior change of an increase in physical activity. However, due to the study limitations of unrepresentative participants and evaluation environment, time frame and application levels, nothing was truly proven in this regard. Further improvement of the prototype and future studies are needed to evaluate the prototype more accurately. / Fysisk inaktivitet är ett växande hälsoproblem för vårt samhälle under 2000-talet. En stor andel av vuxna och ungdomar uppnår inte rekommenderade hälsonivåer för fysiska aktivitet vilket har påtagliga konsekvenser för både individ och samhälle. Utan några forskningsbevis på att nivåerna för fysisk inaktivitet kommer att förbättras i framtiden har traditionella metoder för att motverka detta problem skenbarligen visats sig vara ineffektiva. Ett relativt nytt och lovande tillvägagångssätt för problemet som blivit alltmer relevant under de senaste åren är smartphonebaserade beteendeinterventioner för att främja fysisk aktivitet och andra hälsorelaterade beteenden. Medan ett flertal studier visat många positiva påföljder och lovande resultat med detta tillvägagångssätt, har andra studier samtidigt visat att metoden inte är helt förstådd och otillräckligt efterforskad. Bland annat visar forskning att många av dagens tillgängliga hälso- och träningsapplikationer inte optimalt följer riktlinjer för vare sig beteendeförändringsstrategier eller hälsorelaterad forskning, vilket antyder att dessa applikationer inte uppnår sin fulla potential. Målet med denna studie var att utveckla och utvärdera en applikation för att främja fysisk aktivitet som var baserad på sådan forskning. En prototyp av hög funktionsnivå baserad på The Behaviour Change Wheel utvecklades i Adobe XD och utvärderades av fyra deltagare. Resultatet från utvärderingen pekade mot att den utvecklade prototypen var väl uppskattad av deltagarna som även ansåg att prototypen skulle framkalla den önskade beteendeförändringen för ökad fysisk aktivitet. Emellertid, på grund av studiebegränsningarna med bland annat icke-representativa deltagare och utvärderingsförhållanden, tidsram samt applikationens funktionsnivå, kunde inte något absolut bevis framläggas i detta avseende. Ytterligare förbättringar av prototypen och framtida studier är nödvändiga för att utvärdera prototypen mer exakt.

This research presents the design and evaluation of an intervention that introduces modification of time perceptions as one of the solutions to promote sustainable behaviours. It is demonstrated in this thesis that unnecessary energy use is often caused by temporal tensions, defined as the relation between actions to be performed and available time. This research proposes that it is possible to deliberately reduce temporal tensions, and this can motivate people to behave more sustainably. Persuasive technology and human-computer interaction provided the tools needed to manipulate time perceptions and therefore bring about changes in the specific behaviours that result in unnecessary energy usage. Previous studies indicate that behaviours play an important role in energy consumption. From the different domains of energy use that could be examined, cooking was chosen to be the platform where the studies on behaviour change and energy use would take place. How behaviours influence energy use motivated the design of empirical studies to understand behaviours related to domestic energy use and identify what are the determinants of these behaviours. Each determinant was related to a strategy to be included on a behaviour change intervention. A wider survey was developed to understand students acceptance of a set of proposed energy saving techniques, and resulted in a vast volume of information about user preferences and intentions to perform the suggested energy saving behaviours for cooking. It emerged that participants rushed into the cooking tasks without much deliberation, consequently not following preparation procedures and thus using more energy. Information gathered during the first studies also showed that participants behaviours were partially motivated by the need to speed up the cooking process in order to reduce boredom when they were waiting for the food to cook, consequently resulting in extra energy usage. The knowledge gathered from the preceding steps and a literature review informed the design of strategies to modify the non-sustainable behaviours and promote energy saving. A user-centred design process involving an idea generation session and scenario analysis was used to provide a set of strategies to be embedded in an intervention, containing the specific methods to tackle the correspondent determinants of behaviours. The specific needs of the cooking activity indicated that an electronic intervention was an adequate platform to be implemented and tested. Two high resolution working prototypes of the electronic interventions were developed as mobile phone applications. The final study comprised the evaluation of the proposed interventions in improving aspects of the cooking activity, the acceptance of the interventions and effectiveness in promoting energy saving.

The need for a more sustainable approach to water consumption has increasingly gained attention in the last decade. The domestic sector accounts for over half of abstracted water in the UK and, as such, has become a major target for water efficiency interventions. Current research and water efficiency interventions are dominated by a positivist approach, focusing on a limited range of factors that can be quantitatively measured. This thesis questions the dominant approach and argues that a more holistic overview of water efficiency can be achieved through the consideration of socio-technical and behavioural theories. Taking a more constructivist approach, this research draws on four theories from socio-technical and behavioural fields and combines them to create a framework for the analysis of water efficiency interventions. The framework is applied to two case studies, exploring water users' perceptions of water, water supply, personal water use, and their responses to the water efficiency interventions. The case studies were selected to provide examples of current mainstream approaches to water demand management. Research methods used included semi-structured interviews and observation. The research findings support the argument that the current dominant approach to domestic water efficiency interventions is limited and, in some cases, ineffectual. Issues of trust, knowledge, motivation and the relationships between water users and water companies were raised. The thesis concludes that the use of a constructivist perspective could help to provide a more effective approach to understanding and improving water demand management.

Visualization tools are frequently used to help people understand everyday data in their lives. One such example is visualization in behaviour feedback tools. Behaviour feedback tools are used to try to help people improve their health or personal well-being or to carry out sound environmental sustainability practices. However, understanding and reasoning about personal data (e.g., pedometer counts, blood pressure readings or home electricity consumption) or gaining a deeper understanding of one's current practices and learning how to make a change can be challenging when using data alone. My literature review of this field showed that two of the main challenges in actual practice are providing a context in which to reason about the data and reducing the cost of maintenance to fit those tools into everyday life routines. Thus, I propose to integrate time-varying feedback data within a personal digital calendar. This combination of calendar and feedback data can provide contextual information to interpret data and make the data accessible in an attentionally ambient way that is suitable for maintaining awareness. I propose that the familiarity and common practice of using digital calendars can minimize the cost of learning and maintenance for people and easily fit into one's daily life routines. The viability of this approach was confirmed in my quantitative lab experiments. The results showed that visualization of feedback data integrated on a digital calendar is comprehensible, and it does not interfere with regular calendar use with proper visual encodings. After confirming the viability of my proposal, I implemented the on-calendar visualization as a web application that was synchronized with Google Calendar API and a real-time feedback data stream. To further investigate this approach in a real life situation, I deployed the application in the field for longitudinal field studies: two case studies as pilot deployment and an eight-week field study. Results showed that people liked the idea of integrating feedback data into their personal digital calendars. It required a low cost in learning and maintenance. The calendar events provided rich context for people to visualize and reason about their feedback data. The design enabled people to quickly identify and explain repeated patterns and anomalies. Meanwhile, I found that people's existing information use habits (in this case, how they use digital calendars) can highly influence the effectiveness of the feedback design. Moreover, I derived a feedback model that identifies basic components in feedback design and illustrates the role of feedback tools. With that I articulated possible design barriers that could prevent ongoing use of feedback tools. Reflecting on the effects of the on-calendar design approach, I discussed design implications inspired by this work. This work introduces a reflective approach in feedback design that can easily fit into people's existing information ecosystem (specifically, a personal digital calendar in this work). The main contributions of this thesis are: the first systematic literature review of personal visualization design used in everyday life; the design and implementation of an on-calendar design that integrates feedback data on people's personal digital calendars to provide context for reasoning and support easy access for ongoing use; the extended definition of ambience from spatial location to attentional demand; a viability study to confirm the on-calendar design approach; longitudinal studies to investigate the effects of the on-calendar design approach and the feedback model of design mechanism to inspect ongoing factors in feedback designs. / Graduate / 0984, 0384 / dhuangca@gmail.com

Communicating climate change and other long-term environmental issues to the wider public is a challenging process involving many barriers to action. Visualisations have the ability to overcome these barriers. Furthermore, the benefits of visual images over verbal information have been supported. However, there is a lack of research on how visual images can motivate behaviour. Based on Elaborated Intrusion theory, it is proposed that visual images can be internalised as mental images which can act as a ‘motivational bridge’ for pro-environmental behaviour. Six studies exposed participants to visual pro-environmental messages. Throughout all studies images were internalised as mental images, associated with intrusive thoughts. Consistently, mental images were strongly associated with pro-environmental goals and behaviour change (self-report and actual). Also, interfering with visual mental imagery, using cognitive tasks, reduced self-reported behaviour change (Study 6). Overall, two motivational roles of mental imagery emerged: mental images can trigger pro-environmental goals and can strengthen the relationship between pro-environmental goals and behaviour. The development, strength, and vividness of mental imagery depended on interacting individual and message characteristics. A vivid message was more effective when the message topic was relevant to the target individual (Study 2). Also, a message in line with existing values evoked more vivid mental imagery (Studies 2 to 6). A message could activate specific values as well if mental imagery was not interfered with (Study 5 and 6). And finally, positive and negative images were associated with different feelings, but could both be internalised and motivate behaviour (Study 3 and 6). A combined message could overcome the negative feelings associated with a fear appeal (Study 6). This thesis developed a theoretical framework, integrating approaches from social and cognitive psychology, which can help explain and predict responses to visual environmental messages. Six studies showed that the effect of a visual image on behaviour depends, among other factors, on its ability to trigger recurring mental imagery. Insights provide opportunities for designing evidence based visual pro-environmental messages. In turn this can maximise the impact visual messages have on changing people’s pro-environmental behaviour.

Regular physical activity is associated with improved physiological and psychological wellbeing, by reducing the risk of chronic illnesses such as cardiovascular disease, cancer, obesity, diabetes, osteoporosis and depression. There is a common perception that physical activity levels in the population are declining, and one of the biggest changes affecting this is occupational based activity. Since adults spend on average over 50% of their waking hours at work, work sites have the potential to be an important setting for health promotion initiatives. Cognitions and behaviours are key causal factors behind many of today s most widespread health problems and illnesses. The stage of change model has been highlighted as having intuitive appeal because it considers the dynamic nature of attitudes and behaviour change. This thesis is concerned with the application of the stage of change model to an occupational health intervention promoting physical activity. Several research studies were undertaken to explore the experiences of employees with workplace health initiatives and investigate the strategies and practices used by occupational health to promote healthy behaviours. These research studies highlighted the barriers and facilitators to successful health interventions and contributed towards the design, development and implementation of an activity promotion intervention. Additional research was also conducted to develop information materials based on the stage of change model. The stage approach was simplified and intervention materials were classified based on whether employees were thinking about making a change or not thinking about making a change to their activity levels. In order to test the materials, a twelve month intervention was implemented in ten work sites across the UK that were allocated to one of three groups. Two groups received information materials and one group received no information during the intervention period (control group). The difference between the two groups who received information was that one group received standard activity promotion information (standard group) and the second group received tailored information based on their stage of change construct (staged group). Participants in the staged intervention group demonstrated significant decreases in body mass index, fat percentage, waist circumference, blood pressure and resting heart rate following the twelve month intervention. In contrast, reductions were identified for the standard intervention group for waist circumference and diastolic blood pressure. Finally, there were no long-term significant improvements identified for the control group. However, group comparisons revealed there were no significant differences between the intervention conditions. The intervention also recorded self-reported psychological outcomes, which demonstrated variations throughout the intervention period for all groups. The potential reasons for these inconsistent outcomes are discussed. A process evaluation following the intervention demonstrated employees valued the health screenings and identified issues relating to knowledge, behaviour change and health implications that were important outputs of the intervention. Based on these findings, the research concludes there is scope to make physical activity interventions in the workplace more effective by applying the stage of change approach. Using the process of simplifying the stages and focusing on whether employees want to change their behaviours or not allows occupational health to deliver information that could be more meaningful and have a significant impact on behaviour change. By understanding employees readiness to change their activity behaviours and targeting information based on their beliefs, attitudes and intentions to change may produce significant improvements in health outcome measures compared to standard information. The results also suggest there is potential for this type of tailored intervention to be extended to other occupational health issues.

This programme of research was focused on developing a better understanding of pro-environmental behaviours and pro-environmental behaviour change, with consideration of the powerful effect of habits in thinking, affect and behaviour. Habit networks are discussed with reference to HOT topics (Habits, Opportunities and Thoughts), and explored empirically within the context of the FIT Framework (Fletcher & Stead, 2000). This programme of research started with a literature review on established models of pro-environmental behaviour. It found a large degree of similarity in the approaches used to conceptualise pro-environmental behaviour, and suggested the need to explore pro-environmental activity from different perspectives. The FIT Framework was then presented as an alternative approach. FIT variables measure the strength of an individual's cognitive characteristics and their degree of behavioural flexibility using the FIT Profiler (Fletcher, 1999). The empirical studies presented in this programme of research suggest that levels of personal FITness are related to engagement with pro-environmental activity and the extent to which lifestyles are environmentally sustainable. Study 1 (N = 325) explored the relationships between FITness and measures of pro-environmental activity, and Study 2 (N = 134) sought to confirm these relationships in a different sample. Both studies found positive relationships between levels of personal FITness and pro-environmental activity. Based on these results, it was suggested that FIT offers a useful alternative framework to study pro-environmental activity. Studies 3 (N = 75) and 4 (N = 100) considered the performance of pro-environmental behaviours in different sites of practice, as follow-up to the differences that emerged in Studies 1 and 2. They also explored the perceived influence of intrinsic and extrinsic variables on energy saving in home and work settings. The results suggested that the pro-environmental behaviours that are performed at home are often not transferred to the workplace and this might be because extrinsic factors in an organisational setting constrain action. Higher levels of personal FITness helped to align behaviours with intrinsic beliefs; individuals with higher levels of FITness behaved as they felt they ought to, whereas individuals with lower levels of FITness behaved as they were told to. It was suggested, therefore, that higher levels of FITness might support behavioural consistency across contexts. Study 5 (N = 95) explored the extent to which pro-environmental behaviours are characterised by habit and how the strength of habit changes according to level of personal FITness. The results suggested that people act pro-environmentally within distinct behaviour categories and this has little or no bearing on their propensity to behave in an environmentally friendly way in other areas. Habits can have a positive influence on the performance of pro-environmental behaviour but a negative influence on behaviour change. The empirical results suggest that a higher level of FITness might help people to engage more readily with pro-environmental behaviours that are performed less frequently. It was, therefore, suggested that developing levels of personal FITness might help individuals to extend their behavioural repertoire and be sufficiently flexible to include more pro-environmental behaviours, including those that are, at present, characterised less by habit. A second literature review on interventions for pro-environmental behaviour change found that many techniques have been developed from the perspective of bounded rationality and have assumed that the provision of information will initiate behaviour change. These approaches are often ineffective because of the resistive effects of habit. In light of this and the findings of the empirical studies, a FIT-based behaviour change intervention, named here as Do Something Greener, was developed as an alternative approach to address directly the problem of habit. Overall, this programme of research suggests that the influence of habits, opportunities, and thoughts should be considered in the study of pro-environmental activity. Further research exploring the effectiveness of Do Something Greener for pro-environmental behaviour change is necessary and planned as the next step in this programme of research. By exploring pro-environmental behaviour from a different perspective, it is hoped that this programme of research has also challenged some of the habitual tendencies that researchers are themselves starting to develop in relation to the study of pro-environmental behaviour, and that it has added a degree of eclecticism and pragmatism to psychological approaches to pro-environmental behaviour change. It is hoped that this will help to set a more practically oriented agenda for future research.

This thesis is based in the research area of Design for Sustainable Behaviour (DfSB), a field which seeks to reduce the social and environmental impact of products in the use phase of their life cycle. There has been significant theoretical development in this area in recent years, leading to a proliferation of intervention strategies and design methodologies. However, there has been a recognised lack of a reliable means of selecting which intervention strategy to use in a given situation, and a lack of real world intervention case studies generating measurable medium-to-long term reductions in energy consumption. Addressing these gaps was a central focus of this research. This thesis documents four distinct research phases; an extensive literature review, an in-depth user study of existing energy consuming behaviours and motivations, the development and trialling of design interventions, and the evaluation of the generated theories as a tool for designers. Literature on domestic energy consumption, human behaviour, and approaches to changing behaviour was reviewed to establish the current level of thinking and to identify opportunities for further research. This guided the undertaking of the user study with a number of families in the East Midlands of the UK, which illuminated the relevant motivational goals, and highly routinized nature, displayed in many energy consuming behaviours. Over the course of this phase of the research journey a new conceptual model of behaviour in context was developed, and refined to create the Behavioural Intervention Selection Axis (BISA). These theoretical developments were then applied to the generation of DfSB intervention concepts, one of which was selected and developed to a functional prototype stage. These prototypes were trialled in situ in family homes for an extended period, and achieved a significant change in behaviour and related energy consumption. Further evaluation of the BISA as a tool to guide designers was performed through a series of workshops with design students, which ascertained its usefulness in this respect. Both the intervention development and trialling and the design workshops showed the conceptual model and BISA to be successful in providing designers with a reliable and useful means of selecting appropriate intervention strategies to change behaviour. In addition the intervention trial provided a wealth of qualitative insight into the way in which DfSB can effect behaviour, and the range of new motivational goals it can engender.

Regular physical activity (PA) participation and fruit and vegetable (FV) consumption confers numerous positive health outcomes for cancer survivors, including prevention of cancer recurrence, second primary cancers, and other non-communicable chronic diseases. Rural-living young adult cancer survivors (YAs) possess unique barriers and concerns that influence their ability to participate in traditional face-to-face behaviour change interventions. Few researchers have explored alternative means for delivering behaviour change interventions grounded in theory utilizing a mixed-methods approach to assess processes of change and behavioural outcomes. To fill this gap and provide recommendations for future interventions and services focused on positive health behaviours in this population, the objective of the research presented in this thesis was to explore the feasibility and acceptability of a 12-week theory-based telehealth behaviour change intervention aiming to improve PA and FV consumption using a single-arm, mixed methods pilot trial. Over a 7-month period, 14 YAs self-referred. Of these 14, 5 were eligible and consented to participate with 3 completing the study. Retention to the study was 73% and adherence to the health coaching program ranged from 66.67-100% with a 40% attrition rate. Inquiry into the acceptability of the intervention offered insight into participants experiences, which was summarized within five themes: (1) the more time the better, (2) the human factor, (3) supporting access, (4) influencing the basic psychological needs, and (5) finding motivation. Collectively, the findings suggest the methods used require minor modifications before being deemed feasible despite the general acceptability of the intervention. Importantly, they highlight the necessity of more expansive recruitment strategies and a need to explore participants’ underlying intentions for participating in behaviour change interventions. Further, recommendations are made based on the findings to improve this style of intervention, including testing stepped down models of support because it may help some YAs maintain behaviour change post-intervention.