Digital Behaviour Change Intervention: Personalizing Persuasive Games for Wellbeing
Lead Research Organisation:
University of Aberdeen
Department Name: Computing Science
Abstract
Digital behaviour change interventions (also known as Persuasive Technology) encourage people to change their attitudes and behaviours. My research project aims to investigate the design of a gamified persuasive intervention as a proactive response for preventing mental health problems and improving subjective wellbeing in a student population. I am particularly interested in 1) encouraging students to engage in positive activities that induce happiness and 2) adapting persuasive strategies to different personalities, active stressors, attitudes and past behaviours.
Students are continually exposed to a variety of stressors during their academic career, which can induce high levels of psychological distress, depression, anxiety and loneliness. Moreover, students who experience mental health problems might feel nervous about receiving care or seeking support from their institution and local practitioners. This can lead to numerous health complications and have a major influence on student confidence, academic performance and subjective wellbeing. It is, therefore, critical that students receive appropriate support, tailored to their needs, throughout their academic journey. Persuasive Technology can motivate, shape and reinforce beneficial behaviours, helping individuals to reduce the negative impact of risk factors or stressors and to enhance the effect of protective factors. Thus, digital behaviour change interventions can facilitate the integration of preventive models that would allow early detection of symptoms and enable a larger number of individuals to access support remotely. Using persuasive technology to promote wellbeing would reduce time costs and financial expenses for mental healthcare providers and support services, as well as encourage initiative, empower individuals, promote self-care and improve self-management skills.
Inspired by research in positive psychology, my PhD project involves the design of algorithms for a gamified persuasive intervention that encourages engagement in positive activities. The main focus of my work is tailoring persuasive strategies for motivating student engagement in meaningful, achievable and enjoyable activities which increase happiness and support students in managing stressors more effectively. My project also involves performing user studies (both qualitative and quantitative) to inform and validate those algorithms, as well as controlled experiments to evaluate the perceived and actual effectiveness of the gamified intervention. While digital behaviour change interventions can be delivered using various approaches, I chose to focus on persuasive games due to their strong motivational pull. Persuasive games are very interactive and require active engagement from participants, which can increase the emotional quality of the intervention and contribute as an incentive to keep users engaged. As a result, an increasing number of persuasive games have been developed in recent years as novel solutions for motivating healthier behaviours, such as encouraging physical activity and balanced nutrition. Despite the growing interest in persuasive games, there remains a need for further research into their application in the wellbeing domain and the different aspects of design and personalisation of games which promote happiness. My PhD builds on work in the areas of Persuasive Technology (including Gamification), Intelligent User Interfaces, User Modelling and Personalisation. The findings of this work will have implications for the personalisation and design of persuasive technology for health and wellness.
Students are continually exposed to a variety of stressors during their academic career, which can induce high levels of psychological distress, depression, anxiety and loneliness. Moreover, students who experience mental health problems might feel nervous about receiving care or seeking support from their institution and local practitioners. This can lead to numerous health complications and have a major influence on student confidence, academic performance and subjective wellbeing. It is, therefore, critical that students receive appropriate support, tailored to their needs, throughout their academic journey. Persuasive Technology can motivate, shape and reinforce beneficial behaviours, helping individuals to reduce the negative impact of risk factors or stressors and to enhance the effect of protective factors. Thus, digital behaviour change interventions can facilitate the integration of preventive models that would allow early detection of symptoms and enable a larger number of individuals to access support remotely. Using persuasive technology to promote wellbeing would reduce time costs and financial expenses for mental healthcare providers and support services, as well as encourage initiative, empower individuals, promote self-care and improve self-management skills.
Inspired by research in positive psychology, my PhD project involves the design of algorithms for a gamified persuasive intervention that encourages engagement in positive activities. The main focus of my work is tailoring persuasive strategies for motivating student engagement in meaningful, achievable and enjoyable activities which increase happiness and support students in managing stressors more effectively. My project also involves performing user studies (both qualitative and quantitative) to inform and validate those algorithms, as well as controlled experiments to evaluate the perceived and actual effectiveness of the gamified intervention. While digital behaviour change interventions can be delivered using various approaches, I chose to focus on persuasive games due to their strong motivational pull. Persuasive games are very interactive and require active engagement from participants, which can increase the emotional quality of the intervention and contribute as an incentive to keep users engaged. As a result, an increasing number of persuasive games have been developed in recent years as novel solutions for motivating healthier behaviours, such as encouraging physical activity and balanced nutrition. Despite the growing interest in persuasive games, there remains a need for further research into their application in the wellbeing domain and the different aspects of design and personalisation of games which promote happiness. My PhD builds on work in the areas of Persuasive Technology (including Gamification), Intelligent User Interfaces, User Modelling and Personalisation. The findings of this work will have implications for the personalisation and design of persuasive technology for health and wellness.
Organisations
People |
ORCID iD |
| Judith Masthoff (Primary Supervisor) | |
| Ana Ciocarlan (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509814/1 | 01/10/2016 | 30/09/2021 | |||
| 1957285 | Studentship | EP/N509814/1 | 01/10/2016 | 30/09/2019 | Ana Ciocarlan |
| Description | This research project investigates the design of an adaptive persuasive intervention that encourages people to participate in positive activities and perform more acts of kindness, in order to improve their subjective wellbeing and prevent mental health problems. The intervention builds on our initial findings and related work in personalising Persuasive Technology. The aims of my PhD are outlined in the following research questions: 1. How can personalised persuasive interventions encourage people to perform more acts of kindness and improve wellbeing? (a) What are the different types of kindness activities which promote subjective wellbeing? (b) What behaviour determinants influence people's kindness behaviour? (c) What persuasive strategies are the most effective for encouraging kindness behaviour? (d) How can we map behaviour deteraminants and persuasive strategies to intervention design elements? (e) What do we need to know about users to personalise design elements, persuasive strategies and kindness activities effectively? (f) What is an effective algorithm for adapting design elements, persuasive strategies and kindness activities to users' characteristics? 2. How effective is the persuasive intervention in motivating and sustaining behaviour change? (a) How can the effectiveness of the persuasive intervention be evaluated? (b) What is the perceived persuasiveness of the strategies in motivating behaviour change? (c) What is the actual persuasiveness of the strategies in motivating behaviour change? (d) What is the impact of the intervention on behavioral intention? (e) What is the impact of the intervention on actual behaviour? (f) What is the impact of the intervention on subjective wellbeing? My research work is focused on experimental investigation, developing an persuasive system for wellbeing and gathering of knowledge. The key activities followed in my research project so far are outlined below: • Reviewing literature on behaviour change, persuasive interventions, user modelling, recommendations and personalisation. • Conducting qualitative studies to investigate how to adapt activity complexity to personality, level of wellbeing and behavioural intention. • Conducting crowdsourcing studies to collect kindness activities which can be recommended in the intervention. • Validating kindness activities in different categories: positivity, generosity, gratitude, friendliness and self-kindness. • Conducting a small-scale pilot study to investigate the impact of different persuasive messages to encourage acts of kindness. The findings from the pilot study were used to inform the design of a larger study to investigate persuasive interventions for subjective wellbeing. • Designing a persuasive intervention system which encourages people to perform acts of kindness. • Conducting a study using the system to investigate the impact of different personality types, different persuasive messages and different activity categories on how people engage with the intervention The studies conducted so far considered how different personality traits of the Five Factor Model (Conscientiousness, Extraversion, Agreeableness, Openness and Emotional Stability) impact how people engage with a behaviour change intervention that encourages acts of kindness. I investigated which persuasive messages are more effective for different personality traits and I have conducted some initial qualitative research on how to adapt activity complexity to personality, stressors and attitude. The messages were created by using Cialdini's principles of persuasion (Social Proof, Liking, Authority, Scarcity, Reciprocity, Unity and Commitment). I developed an online prototype and in a study with 45 participants, I also investigated the impact on user wellbeing, behaviour and behavioural intention. Participants were assigned 5 daily kindness activities for one week and received persuasive messages. A total of 1193 activities were completed, and participants showed significant increases in wellbeing and behavioural intention. This suggests that the intervention has a positive effect and encourages people to perform more acts of kindness. My results suggest that people who are high in conscientiousness tend to engage in more kindness activities overall, so they may require less persuasive messages. However, a negative correlation was found for individuals who are high in extraversion. The results show that they complete less kindness activities, even though in the pre-questionnaires they indicate their intention to do more acts of kindness is very high. Individuals with high levels of agreeableness tend to complete more activities that involve interactions and being friendly towards others. As a result, the intervention could adapt to assign more acivities from the friendliness category and persuade them to complete activities from other categories as well. The results also showed that people high in the conscientiousness trait are more persuaded by messages which use the authority principle, followed by messages using the scarcity principle. For people with high agreeableness, the most effective messages were the ones using the commitment and scarcity principles. However, people high in extraversion seem to be negatively influenced by persuasive messages which use the social proof or reciprocity principle. For these studies, kindness activities were developed manually. However, a larger variety of activities is required for our future experiments, so we conducted a crowdsourcing experiment to create a large collection of kindness activities that can be recommended to users. Subsequently, we ran a validation study to classify the activities in the five different categories of kindness activities: positivity, friendliness, generosity, gratitude, self-kindness. The data which was gathered from questionnaires will be used to develop user models and learn about interests and experiences. Participants express interests and values directly through interactions, which also contributes to the development of user models. The novel design of this intervention also allows me to conduct valid self-assessment, in support of objective behavioural outcomes, to measure the effectiveness of different strategies for different personalities. This new approach enables me to investigate and compare perceived persuasion and actual persuasion, identifying which strategies best support di?erent people in behaviour change. To further investigate actual persuasion, I conducted a study to observe the impact of Cialdini's seven persuasive principles on people's direct choices, actions and behaviour. Furthermore, I investigated if people's susceptibility to different persuasive messages is consistent over time. The findings suggest that certain persuasive principles have a greater influence on a person's actions than others, with Reciprocity and Liking being the most effective, and Authority and Unity being the least effective. Our results differ from work investigating perceived persuasiveness, suggesting that what people perceive to be more persuasive is not necessarily what will persuade them to perform an action. Moreover, the study showed that people's susceptibility to different principles is dependent on their personality traits, and it remains constant with time. This could be explained by the fact that people's personality does not change and, therefore the level of influence different persuasive principles has on them remains constant. Based on our results, a behaviour change intervention could adapt the persuasive messages and activity categories to user personality and interests. I have planned a series of future experiments to inform the development of a personalisation algorithm which can be incorporated in the system. Future studies will explore how behavioural intention changes over time, and the effect on actual user behaviour after taking part in the intervention. I will perform evaluations for the effectiveness of the intervention and further refine my proposed model based on the obtained findings. |
| Exploitation Route | The findings of this research could support future work in personalising persuasive strategies and designing digital behaviour change interventions for wellbeing. Further investigation is required to identify differences between perceived and actual persuasiveness and to explore how behavioural intention changes over time. We could also explore actual persuasiveness in a different domain, such as persuading people to engage in other health related or sustainable behaviours. An additional study could also investigate if susceptibility to different persuasive messages remains constant after a longer period of time. It is important to identify the most effective way of eliciting user preferences and personality types. Future studies could explore whether other attributes such as an individual's mood states or need for cognition can impact susceptibility to persuasive principles. Moreover, a new study could investigate whether the messages produced by the participants during the intervention can elicit positive emotions in a reader, thus contributing to the overall improvement in levels of subjective wellbeing. |
| Sectors | Digital/Communication/Information Technologies (including Software),Healthcare |
| Description | UMAP Student Travel Grant |
| Amount | $1,000 (USD) |
| Organisation | Association for Computing Machinery Special Interest Group on Computer–Human Interaction (ACM SIGCHI) |
| Sector | Charity/Non Profit |
| Country | United States |
| Start | 07/2018 |
| End | 07/2018 |
| Description | UMAP Student Travel Grant |
| Amount | $500 (USD) |
| Organisation | Association for Computing Machinery Special Interest Group on Computer–Human Interaction (ACM SIGCHI) |
| Sector | Charity/Non Profit |
| Country | United States |
| Start | 07/2017 |
| End | 07/2017 |
| Title | Kindness is Contagious: Digital Behaviour Change Intervention |
| Description | The digital behaviour change intervention encourages people to engage in more positive activities in order to improve their subjective wellbeing and prevent symptoms of commmon mental health problems, including stress, anxiety or depression. A personalisation algorithm is currently under development to be added to the intervention, in order to apply the best behaviour change strategies for people of different personality types. |
| Type | Preventative Intervention - Behavioural risk modification |
| Current Stage Of Development | Refinement. Non-clinical |
| Year Development Stage Completed | 2019 |
| Development Status | Under active development/distribution |
| Impact | Mental health problems are a growing public concern around the world and healthcare services are struggling to meet the overwhelming rising demand for care provision. Common mental health conditions such as stress, anxiety and depression can significantly influence an individual's health, confidence, and performance. Moreover they lead to large direct costs for the society and the economy, as they are are one of the leading causes of sick absence in the UK. This research project investigates the design of an adaptive persuasive intervention that encourages people to participate in positive activities and perform more acts of kindness. Recent research work has shown that intentional engagement in positive activities, such as practicing kindness, showing generosity or expressing gratitude, increases happiness and improves people's wellbeing. Our digital intervention uses persuasion to motivate engagement in meaningful, achievable and enjoyable activities, which improve subjective wellbeing and support individuals in managing stressors effectively. Persuasive Technology can motivate, shape, and reinforce beneficial behaviours, whilst personalising interventions facilitates the integration of preventive and proactive responses to promote initiative, empower people and improve self-management skills. Adaptation is particularly important in the mental health domain, as users have subjective opinions and their needs and emotions are continuously evolving. Furthermore, what motivates change for some individuals, could demotivate others. The findings of this research will have implications for the personalisation and design of persuasive interventions for wellbeing, providing new insights and contributing to the areas of Persuasive Technology, Artificial Intelligence and Personalisation. The intervention will allow a higher number of individuals to access healthcare support remotely and facilitate detection of symptoms, reducing time costs and ?nancial expenses for healthcare providers and users. The intervention will help employees prevent work-related stress and mental health problems, bringing numerous benefits to the industry and society. The research is focused on delivering an innovative intelligent technology, which helps people and adresses key societal and economical concerns worldwide. A total of 45 unique participants took part in a pilot experiment. The study was conducted over a period of 7 days, with participants being assigned 5 activities each day. All daily activities involved performing simple acts of kindness and were centered around 5 key categories that inspire positivity and promote wellbeing: being positive, being generous, expressing gratitude, being friendly and self-kindness. A different persuasive message was shown to participants every day to encourage the completion of a higher number of activities. Therefore, activity category and message type were within-subject variables. A pre-questionnaire and post-questionnaire were also completed by the participants, as well as daily questionnaires in which participants rated the activities and described their experiences in more detail. Participants' pre-intention and post-intention to perform random acts of kindness pertaining to the key categories were also evaluated. A total of 1193 activities were completed during the experiment. Participants showed an increase from the baseline level of wellbeing reported in the beginning as well as an increase in behavioural post-intention from the baseline pre-intention. This suggests that the persuasive intervention can encourage people to perform more acts of kindness and have a positive effect on people's levels of happiness and wellbeing. |
| Title | Kindness is Contagious (Digital Behaviour Change Intervention) |
| Description | This adaptive digital behaviour change intervention uses persuasive principles to encourage people with different personality types to perform more positive activities and acts of kindness in order to improve their subjective wellbeing. |
| Type Of Technology | Webtool/Application |
| Year Produced | 2018 |
| Impact | In a study using the application 45 participants were assigned 5 daily activities for 7 days, and asked to complete as many as possible by the end of each day. Participants received different persuasive notifications every day to encourage them to complete a higher number of activities. We investigated how participation levels are influenced by different personality types, different persuasive message types and different categories of activities. Furthermore, we analysed the influence of the intervention on participant behaviour and the effect on behavioural intention, by comparing pre-intention and post-intention to perform different kinds of positive activities. At the end of the study participants showed an increase from the baseline level of wellbeing reported in the beginning. This suggests that a digital intervention to encourage acts of kindness can have a positive effect on people's levels of happiness and wellbeing. We found a significant average difference, as participants reported wellbeing scores which were on average 0.522 higher than those reported in the beginning. Overall, behavioural post-intention also showed an increase from the baseline pre-intention, suggesting that a persuasive intervention can encourage people to perform more acts of kindness. Post-intention scores were .906 points higher than pre-intention scores. A demonstration at the UMAP 2018 Conference in Singapore showed the system in action, encouraging conference delegates to perform acts of kindness. Ethical consent for the installation was obtained from the Physical Sciences and Engineering ethics board at the University of Aberdeen. |