A variety of social and digital media technologies are being used to influence a change in an individual's or group's behaviour. A major challenge is in understanding what leads to or prevents different forms of influence from having an effect, what those effects are, how long they take to come about and for how long they last. This research is concerned with the problem domain of climate change and with using social and digital media technologies to influence users to change their energy consumption behaviour. The objective is to understand how user information may be utilised within the development of persuasive technologies and behaviour change support systems. This paper contributes fundamental and applied research on how user values, lifestyle aspects and energy consumption behaviours may be modelled to support systems in delivering relevant and personalised information and knowledge that can influence behaviour change.
There is growing interest in understanding how social and digital media can be
used to influence an individual's or a group's behaviour in areas including health,
defence and security, climate change, and more. Research in this area seeks to
understand fundamental and applied aspects of what leads to or prevents different
forms of influence from having an effect, what those effects are, how long they
take to come about and for how long they last. This requires an understanding of
both the explanatory and predictive aspects of different forms of influence. In
addition, research into how digital technologies might play a role in influencing
behavioural change is of concern to researchers in cyber influence and persuasive
technologies [
This paper contributes fundamental and applied research on influencing behavioural change in the area of carbon reduction. There are a myriad of reasons why serious action needs to be taken to reduce our carbon footprints. A plethora of highly commendable courses of remedial action are being taken including developing alternative and/or more efficient technologies for creating, storing and using alternative forms of energy. A major concern of others and our research is to change people's understanding and behaviour towards energy usage and to bring about reductions in carbon rich energy consumption. Our means to do that is through people and society themselves, and our medium for mediating that change is personal and societal digital solutions. Our approach is to understand people's values, to recognise those values as being important influences on behaviour and to create, using the medium of digital technology, data, information and knowledge resources that are both personal and communitybased to influence people to change their behaviour towards lower carbon energy usage.
People use information that is created, accessed and stored on their laptops, tablets and smart phones to help them make both everyday and important decisions. People are enabled to create, and share information and knowledge. This information and knowledge creating, searching, and sharing, results in people and broader communities adopting new ways of acting, doing and behaving. It also supports the establishment of new communities, which in turn create and share new information and knowledge.
Persuasive technologies [
A primary aim of our research is to understand how people's values can influence long-lasting behaviour changes. From this understanding we can then investigate attributes of software systems to influence people to reduce their high carbon, energy consumption behaviours. This paper outlines how user values, lifestyle aspects and energy consumption behaviours may be modelled within software systems and utilised to deliver relevant and personalised information and knowledge that can influence behaviour change. 2
Software systems that influence energy consumption behaviour vary in their
design, and their requirements are often based on popular theories of behaviour
change, such as the behavioural model of rational choice [
Therefore we need to identify, understand and make good use of the complex information about people's values, how they affect their everyday lives, and how they might have implications on their energy consumption behaviour. In addition, while there may be common values and common aspects to people's lives we need to recognise their individual and personal forms, their differences, and their relationship to consequential behaviour. Hence we need to personalise and tailor any influences to the individual. 3
Designers of software systems that aim to influence behaviour change need to
take into account the broad spectrum of ways to frame behaviour change
interventions, in order to make them personalised and relevant to individuals and
groups of people. Based on existing work into the content and structure of values
[
Our underpinning philosophy is that saving energy is not always the governing or
guiding principle around which everyday lives are organised. Everyday lives are
88
organised primarily due to values, around contexts and ways of living. Software
systems that aim at influencing behaviour change therefore need to take into
account the complex trade-offs that are made to meet the demands and challenges
of everyday life while maintaining the values they hold. Of course the value of
nature and the environment is a value that many individuals hold [
People have many values. They are used to select and justify activities, and to
evaluate artefacts and events (including other people and themselves) [
Activities, artefacts and events are important aspects of everyday life. They are
related to the way in which individuals spend their time and are instrumental in
supporting their values. An activity may be defined as a set of actions that have a
goal [
In addition to activities, there are other types of important information. These are events and artefacts. For instance, a family member's birthday party may well require them to have a birthday cake to mark the occasion. This would give rise to very different cooking behaviour than if an individual was cooking an evening meal alone. This suggests that information about the activity alone may not be enough information to explain why behaviour is performed in a certain way. Using the example of a birthday party, it is possible to see that everyday life is connected and/or constrained by values influencing particular events (e.g. birthday) and particular artefacts (e.g. cake).
Due to the constraints of values on activities, events and artefacts, energy consumption behaviour is a consequence of the way in which individuals and groups of people make decisions and trade-offs. The decisions and trade-offs are between their important values, determining their lifestyles, motivating their activities, and contextualising events and artefacts and determining their energy related behaviours.
3.2
Value Lifestyle Aspect
Energy Consumption Behaviour that has direct implications on an individuals Behaviour carbon footprint A Value represents something that an individual or group sees as an important part of their lives. Lifestyle Aspects represent the socio-cultural context in which people live their lives. They provide patterns that may change during different periods in people's lives, they maybe self-chosen, circumstance driven -- or a mixture of both. Activities represent what people do. They can be related to work, home, entertainment, caring for others, etc. Artefacts are largely physical in some form, but include images and sounds such as in the case of music. Informational artefacts are also relevant; such as news, facts, or ideas. Events can be natural events such as sunsets and/or created events such as birthdays. Energy Consumption Behaviours are behaviours that are directly energy consumption related, they can include both using energy such as providing heating for warmth or cooking and saving energy such as turning the thermostat down or creating heat from renewable energy sources.
Activities, Artefacts and Events are all important contextual features of Energy Consumption Behaviours. Energy Consumption Behaviours are consequential to ways in which Lifestyle Aspects are performed. Lifestyle Aspects are in turn instrumental in supporting Values. These associations form an important influence structure that can subsequently be modelled within a system to influence a change in energy consumption behaviour by tailoring implemented strategies of influence. 90 3.3
Table 1 shows that the pivotal parts of the structure are Values, Lifestyle Aspects,
Activities, Events, Artefacts, and Energy Consumption Behaviour. The Energy
Consumption Behaviour objects are the part of the model that is specific to the
problem domain of energy-use. The remainder of the model is general and related
to everyday life and decision-making. In short, the model connects what people
do, why they do it, and the implications this has on energy use. In order to
visually represent this information and their associations, we exemplify their
construction in Figure 1 based on universal values, activities and energy
consumption behaviours in the literature [
An object model was chosen as the information instances are of central importance to the system design. It formalises the relationship between information that drives decision-making in real life at the same time as representing the associations between object instances implemented within a system. In other words the decisions that individuals make within their lifestyles are instrumental to their values; these are represented as associations between Lifestyle Aspect objects and Value objects. For ease of reading, the information type is shown for each of the four hierarchical layers.
Values
Health &
Wellbeing Lifestyle Aspects
Healthy Eating Energy Consumption Behaviour Activity or Event or Artefact
Baking Activity: Cooking
Hobbies Travelling / Holiday
Travelling
Behaviour Event: Birthday Party
Artefact: Cake KEY:
Object
Association
Information Type
Family Gym Class … … … …
The connection between the Lifestyle Aspect of healthy eating and the Value of health and wellbeing shows that the Lifestyle Aspect of healthy eating is instrumental to supporting the Value of health and wellbeing. The Lifestyle Aspect of healthy eating may have different ways in which it is carried out which affect how well the Value is supported. Energy Consumption Behaviour of baking is consequential to the ways of carrying out the Lifestyle Aspect of healthy eating. The important contextual information related to the Energy Consumption Behaviour of baking is the Activity of cooking, the Event of a birthday party and the Artefact of the cake.
Constructing the influence structure allows for designers of behaviour change interventions to reason logically about how to frame interventions for different situations. For instance, using the example given previously, we can see that the Energy Consumption Behaviour related to baking is consequential to the Lifestyle Aspect of healthy eating. Healthy eating is instrumental to two Values: health and wellbeing, and family. Using this influence structure, behaviour change interventions may be framed in the context of a valued aspect of a person's life that may be either directly or only indirectly related to energy. To relate this to the example, interventions may be framed by supporting valued interactions with family and/or promoting health and wellbeing through energy efficient ways of baking. The outcome of the intervention is that the person(s) is influenced to change behaviour in favour of behaviour that consumes less energy while directly supporting their highly valued activities. 4
This paper argues how the construction of a user model, based on values, lifestyle aspects and energy use behaviour may be utilised within a software system to inform strategies that influence a change in energy usage behaviour. These information types were chosen as they are the key drivers of everyday decisionmaking that lead to energy consumption behaviour. As such the content and interrelationship of these information types may inform intervention strategies that are relevant and personalised.
The model explained in this paper illustrates the main conceptual types that are important. A natural progression of this work is towards an implementation-level design. The additional implementation-level detail that is necessary may add lower level implementation specific detail but should not change the conceptual model. It is important to take into account the tensions and trade-offs between values. This is because the tensions and trade-offs are important to select and understand the consequences of a particular strategy of influence.
In future work we aim to empirically evaluate the influence structure through the design an online social media system that aims at influencing a change in energy consumption behaviour. Iterative system development will allow for the model to be specified at a lower level of granularity.
Acknowledgement. Eviz project. The work reported in this paper is funded by the Engineering and Physical Sciences Research Council (EPSRC) under the Transforming Energy Demand in Buildings through Digital Innovation (TEDDI) (grant reference EP/K002465/1).
1. 2. 92
American Occupational Therapy Association: Occupational therapy practice framework: Domain and process. American Journal of Occupational Therapy 56, 609 – 639 (2002) Blake, J.: Overcoming the value-action gap in environmental policy: Tensions between national policy and local experience. Local environment 4(3), 257–278 (1999) Brynjarsdottir, H., H˚akansson, M., Pierce, J., Baumer, E., DiSalvo, C., Sengers, P.: Sustainably unpersuaded: how persuasion narrows our vision of sustainability. In: Proc. of the 2012 ACM annual Conf. on Human Factors in Computing Systems. pp. 947–956. CHI ’12, ACM, New York, NY, USA (2012)