See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/230800104

Design for Behaviour Change

Chapter · January 2010

CITATIONS READS

5 6,469

3 authors:

Dan Lockton David Harrison

Eindhoven University of Technology Brunel University London
129 PUBLICATIONS   1,815 CITATIONS    183 PUBLICATIONS   4,375 CITATIONS   

SEE PROFILE SEE PROFILE

Neville A Stanton
University of Southampton
986 PUBLICATIONS   26,006 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Sleep Ecologies View project

Qualitative Interfaces View project

All content following this page was uploaded by Dan Lockton on 13 March 2014.

The user has requested enhancement of the downloaded file.

In: Advances in Psychology Research 67/69, Nova Science Publishers
Editor: A.M. Columbus

Chapter 3

DESIGN FOR BEHAVIOUR CHANGE

Dan Lockton a, David Harrison a and Neville A. Stanton b
a
Cleaner Electronics Research Group, Brunel Design, Brunel University;
Uxbridge, Middlesex, UB8 3PH, United Kingdom; Daniel.Lockton@brunel.ac.uk
b
School of Civil Engineering & the Environment, University of Southampton;
Southampton, Hampshire, SO17 1BJ, United Kingdom

ABSTRACT

The design of products, systems and environments can be used to influence user behaviour.
This idea has significant potential for social benefit, particularly where human behaviour
and product use decisions directly impact on the environment. Lessons learned and design
techniques employed to influence behaviour in one field can often find additional
application in others, but designers and other stakeholders working on ‘behaviour change’
problems are currently lacking in structured guidance, examples and empirical evidence
about the suitability of different techniques for particular kinds of behavioural influence.

Based on analysing examples of design intended to influence user behaviour from different
fields, the Design with Intent Method provides designers with a design tool suggesting
applicable design techniques, with examples, for a range of target behaviour types. This
chapter introduces the method and demonstrates it through application to an everyday
problem where intelligent changes in user behaviour, achieved through modification of a
system, can have a worthwhile impact on energy use and resource consumption, also saving
users money in the process. A method for classifying behaviour change in this field, the
Behaviour Change Barometer, is also introduced.

Keywords: design, behaviour, interaction, sustainability

 INTRODUCTION

Most designers of the products, systems and environments around us would not
immediately consider that they are in the business of ‘behaviour modification’: the world
of Skinner’s dancing pigeons seems a long way from the design studio. But if, rather than
using the term modify, we substitute influence, it becomes clearer that designers,
engineers and architects—those who bring systems into being—are, whether consciously
or otherwise, involved in influencing behaviour. On a macroscopic scale, this extends to
the way that whole societies live and evolve: new technologies bring massive
socioeconomic change; new forms of communication break down barriers to education
and political engagement; new fashions challenge existing social orders.

Nevertheless, at the much smaller scale of everyday interaction, design also influences
behaviour. As Stanton and Baber (1998) note, “[i]n designing products, designers are also
designing user activity, which does not occur independently of the product... [C]onsumer
behaviour is shaped by products as much as products are shaped by consumer
behaviour.” While intentionally influencing how consumers behave is not often explicitly
part of a design brief—except, of course, in advertising—the opportunity certainly exists,
and it is possible to identify often isolated examples from a variety of fields where the
design of products, systems, services and environments (hereinafter: systems) has been
used with the intent of influencing users’ behaviour.

Design with Intent

Defining ‘Design with Intent’ (DwI) as ‘design intended to influence or result in certain
user behaviour’, the authors (Lockton, Harrison & Stanton, (2008a, 2008b) have
reviewed and classified such examples according to various criteria, supported by a blog
(Lockton, 2005b) receiving suggestions from readers around the world. The
classifications of these examples have supported the development of a design tool for
designers involved in influencing user behaviour—the emerging field of Design for
Behaviour Change (Bhamra, Lilley, & Tang, 2008; Elias, Dekoninck, & Culley, 2007;
Lilley, Lofthouse, & Bhamra, 2005, 2006; Lockton et al., 2008b; Rodriguez & Boks,
2005; Wever, van Kuijk, & Boks, 2008).

This chapter reviews some approaches from different fields, a method for classifying
behaviour change, and introduces the tool—the Design with Intent method—
demonstrated in use through application to an everyday design problem with a specific
focus on reducing environmental impact by influencing user behaviour. Three
presuppositions underpin the research: (i) If certain design techniques have effects on
user behaviour unintentionally, they could also be applied intentionally; (ii) Differences
in design approach between environments, products (hardware/software) and services
largely are due to disciplinary boundaries rather than innate incompatibility—all are
designed systems, and many techniques, or analogues of them, recur across the board; and
(iii) It is therefore possible to abstract certain techniques from examples in one field, and
apply them in others.
 PERSPECTIVES ON DESIGN WITH INTENT

The concept of a target behaviour—aside from its role in behaviour modification—fits

well with an ‘engineering’ approach to influencing the way that people use systems. We
have an intended outcome, a particular behaviour, which we want to achieve by using the
design of something. The ‘success’ of such a design could be measured by the extent to
which the target behaviour is achieved. This is a very simplistic approach, but in terms of
desired outcomes, the target behaviour is what we ‘want’ to achieve (even if it’s qualified
by “in certain circumstances” or “for certain classes of users”).

It has, however, been argued (Buchanan, 1985; Redstrom, 2006) that all design is
intended to result in certain user behaviours. To some extent this is true: the presence of a
chair leads to our sitting down; the creation of a pen is difficult to extricate from the
expectation that someone will write with it. But thinking this narrowly tends to ignore
users’ own impacts: emergent behaviours, appropriation (Salovaara, 2008) or prior
experience (Chamorro-Koc, Popovic, & Emmison, 2008) mean that designers’ intended
use (or usability) is not always translated into user behaviour. The task-artefact cycle
(Carroll, Kellog, & Rosson, 1991) suggests that new artefacts will coevolve with
behaviours (Walker, Stanton, Jenkins, & Salmon, 2009), in turn offering new possibilities,
and so on. Equally, as Kanis (1998) shows, users operate the same products in many
ways and still achieve the desired results. In some fields, such as security (Schneier, 2003)
or health and safety, user adherence to intended behaviour is more critical than in others.

Examples of ‘Design with Intent’ in design, human-computer interaction (HCI) and

architecture are often based on environmental and ecological psychology rather than
arising from a behaviourist background. Barker’s behaviour settings (Barker, 1968;
Sommer, 1969) and Brunswik’s ecological cue validities (Brunswik, 1956) have found
some application in architecture and planning, but Gibson’s affordances (Gibson, 1986),
modified (Norman, 1988) and developed by both Norman (Norman, 1999, 2008) and
others (Gaver, 1991, 1992; McGrenere & Ho, 2000) pervade mainstream design thinking.

In parallel, the development of errorproofing, from Shingo’s poka-yoke manufacturing

quality control (Shingo, 1986) to the design of healthcare processes (Design Council et al.,
2003; Grout, 2007) has promulgated often affordance-based techniques within
organisational policy contexts where, as far as possible, one really doesn’t want errors to
occur at all (Shingo’s ‘zero quality control’). Learning through operant conditioning, in a
trial-and-error exploration of an interface with corresponding reinforcement, might be
instructional and motivational for creative graphic design software, but a bad idea for a
dialysis machine or the control room of a nuclear power station.

The errorproofing approach treats a user’s interaction with a system as a set of defined
target behaviour routes which the designer wants the user to follow, with deviations from
those routes being treated as ‘errors’. Design can help avoid the errors, either by making
it easier for users to work without making errors, or by making the errors impossible in
the first place (‘defensive design’). This is a key part of interaction design, usability and
human factors practice, much of its influence in the design profession coming from
Norman (1988). Diverse practitioners are thus aware of the strategic use of affordances as
‘interventions’ influencing user behaviour, from forcing functions reducing vehicle
misfuelling (Adams & David, 2007) to councils preventing the homeless sleeping on
benches (Lockton, 2005c).

It is worth noting a key difference between an errorproofing approach and some other
views of influencing user behaviour, such as Persuasive Technology (Fogg, 1999, 2003):
persuasion implies attitude change leading to the target behaviour, while errorproofing
doesn’t care whether or not the user’s attitude changes, as long as the target behaviour is
met. Persuasive Technology applies elements of rhetoric (Kjaer Christensen & Hasle,
2007), behavioural conditioning and the central persuasion route of Petty and Cacioppo
(1983) in the context of interactive products, social networking websites and mobile
computing. This is one of the main bodies of current work pertinent to the DwI concept,
but the criterion of persuasion, implying attitude change as a precursor to behaviour
change, necessarily excludes situations where attitude change is not required for changes
in behaviour. Beatty (2008) proposes a classification for persuasive technologies which
includes forcing functions and affordances.

With an errorproofing approach, attitude change might be an effect of the design

measures, but it doesn’t have to be. If we find that we cannot start the microwave oven
until the door is closed (due to the interlock), the target behaviour (the user closes the
door before starting the oven) is achieved regardless of whether our attitude to safety
changes. It might do, though: the act of realising that the door needs to be closed, and
why, may well cause safety to be on our minds consciously. Then again, it could do the
opposite, in the sense that we are no longer forced to consider the impact of our
behaviour on our own safety, as the product takes care of it for us. The distinction
between whether the behaviour change is mindful or not is something the authors have
attempted to capture with the ‘Behaviour Change Barometer’ described later in this
chapter.

Making it easier for users to avoid errors—whether through warnings, choice of defaults,
confirmation dialogues and so on—is slightly ’softer’ than actually forcing the user to
conform, and does perhaps offer the chance to relay some information about the
reasoning behind the measure. But the philosophy behind all of these is, inevitably “we
know what’s best”: the ‘paternalism’ component of libertarian paternalism (Sunstein &
Thaler, 2003). The fact that all of us can probably think of everyday examples where we
constantly have to change a setting from its default, or a confirmation dialogue slows us
down—process friction (Hyde, 2007)—suggests that simple errorproofing cannot entirely
stand in for an intelligent process of understanding the user’s motivations and
comprehensions of the system.

The extent to which system structures influence individuals’ behaviour is also recognised
within science and technology studies (STS) by Latour (1992) and Akrich (1992); work
by Jelsma & Knot (2002) among others applies these ideas to ‘script’ or ‘steer’ behaviour.
There is also growing appreciation of the concept in legal disciplines—particularly
cyberlaw, examining internet architecture and digital rights management (DRM) and how
they govern behaviour (Gillespie, 2007; Kesan & Shah, 2005; Lessig, 1999; Zittrain,
2008). The author (Lockton, 2005a), adopting Lessig’s term ‘architectures of control’,
linked these to affordance-based techniques in design.

Applicability to Environmentally Sensitive Design

With products which consume resources during use, users’ behaviour and operational
decisions contribute significantly to overall environmental impacts (Elias et al., 2007).
Whilst engineering advances permit greater technical efficiency, point-of-use inefficiency
driven by user behaviour—especially with resource-intensive appliances such as washing
machines, dryers, refrigerators and electric kettles—persists and can have a financial
impact on users as well as the environmental impact. Wood and Newborough (2003) and
McCalley and Midden (2002) cite studies in the UK, US and the Netherlands giving 26-
36% as the proportion of home energy usage due to user behaviour decisions, and the UK
Government’s Stern Review (Stern, 2007) identifies behaviour change as a priority.

Winter and Koger (2004) and Cone and Hayes (1984), among others, have recognised the
importance of psychological factors driving many behaviour-related environmental
problems, and proposed that understanding this better will play a significant role in
changing that behaviour: “using psychology to build a sustainable world” (Winter &
Koger, 2004). The ‘heuristics and biases’ approach in behavioural economics (Kahneman,
Slovic, & Tversky, 1982; Kahneman & Tversky, 2000; Gilovich, Griffin, & Kahneman,
2002) has recently taken a high-profile turn (Ariely, 2008; Cialdini, 2003; Thaler &
Sunstein, 2008) towards influencing behaviour, and some interventions, e.g (Cialdini,
2003; de Kort, McCalley, & Midden, 2008) specifically focus on use-related
environmental problems.

Using social marketing techniques to educate consumers about efficient product use is
often the focus of government strategy (Department for Environment, Food and Rural
Affairs, 2008), but there is a more limited literature on using design itself to influence
user behaviour. Some design researchers working on design for behaviour change (see
previous section) have begun to develop the field of ‘design interventions’ applicable by
designers as responses to user behaviour ‘problems’, particularly environmental, but also
‘pro-social’ behaviour generally. The Design with Intent method has been developed for
application in this domain, initially at least: helping people use systems more efficiently.
In most cases this can also save them money and time. A well-planned intervention will
provide an alignment of benefits—social for the community and financial for users.

Nevertheless, it is a challenge for designers to find the right forms in which to apply the
variety of techniques to ‘behaviour problems’, for example influencing more sustainable
product use. As Blevis (2007) puts it, “It is easier to state the kinds of behaviours we
would like to achieve from the perspective of sustainability than it is to account for how
such behaviours may be adequately motivated.” Outside academia, design teams do not
always find it easy to apply abstract taxonomies, or have time for incorporating this stage
into a design process, so a quick method for translating theorists’ valuable work into
practical design suggestions for tackling particular briefs would be useful. This is the aim
of the Design with Intent method. In parallel to the actual design method though, it is
important to understand what kind of behaviour change we are aiming to achieve in each
case.

A ‘BEHAVIOUR CHANGE BAROMETER’: CLASSIFYING BEHAVIOUR CHANGE

If we think about how systems are used, it is clear that changes to the overall ‘use phase’
can result from the systems themselves changing, users changing their behaviour, or a
combination of both. Elias et al. (2007) have captured these possibilities with a 2 × 2
matrix, in which ‘new products’ and ‘old products’ are combined with ‘new user
behaviour’ and ‘old user behaviour’, giving four possible scenarios. Along these lines, it
is possible to consider technology change (via design) and attitude change (via education)
as two routes to achieve overall behaviour change. Especially in the sustainable design
field, the emphasis is often on one strategy or the other, even though the routes are by no
means mutually exclusive.

Lilley et al. (2005) describe three “solutions to limit socially and environmentally
undesirable behaviours”: Educational intervention—which corresponds closely to
attitude change; Technological intervention—corresponding to technology change; and
Product-led intervention—closely aligned with Elias et al’s Design for New User
Behaviour.

Further consideration of the possibilities in this area, and how to represent them, led the
authors to the development of a ‘Behaviour Change Barometer’ (Figure 1). This diagram
attempts to illustrate somewhat more nuanced ‘cases’ of behaviour change, and which
factors are present or absent in each case. It is applicable to many kinds of behaviour
change with systems. The barometer metaphor is stretched slightly, but it seemed
appropriate given that the diagram is mapping change. The same information is presented
in Table 1: in essence, there are six variables involved, with the possibility space divided
into quadrants graphically.
4
2

3c
1c

3a
1a

3b
1b
Case

Overall behaviour

•
•
•
•
•
—
—
—
change results

User understanding/

•
•
•
•

—
—
—
—
mindfulness

Figure 1. The ‘Behaviour Change Barometer’

New user behaviour

•
•
•
•
Table 1. The barometer diagram in tabular form

—
—
—
—
change in practice

New product behaviour

•
•
•
•
—
—
—
—
in practice

Perceived affordance

•
•
•

—
—
—
—
—
change

Actual functional

•
•
•
•

—
—
—
—
change
The focus of the authors’ research is on the intersection of technology change and attitude
change (Quadrant 3): the design of systems which, through new system behaviour,
change user behaviour. Quadrant 3 will be discussed last here—before that, it is useful to
run through the other quadrants briefly.

Quadrant 1: Status quo

In the first quadrant, no overall behaviour change results. It makes sense to describe case
1b first—this is the absolute ‘no change’ case, where there is no change in the actual
functions of the systems (they might be new systems, but they don’t do anything different
to the old systems), people use them in the same way they did before, and users have no
understanding or mindfulness of the issues around behaviour change. Case 1a describes
situations where the systems’ functions have been changed, but users make no use of this,
and have no understanding or mindfulness of the issues involved (e.g. a washing machine
offers a new ‘eco’ mode alongside the other settings, but a user doesn’t use it). Therefore
no overall behaviour change results, despite product improvement. In 1c, users have an
understanding of the issues, and may be mindful of their behaviour and its impacts, but
nevertheless don’t change what they do, and continue to use systems in the same way as
before—e.g. someone who knows that leaving a television on standby wastes electricity,
but doesn’t act on this understanding. Again, no overall behaviour change results, despite
improved user understanding.

This quadrant encompasses much current behaviour with energy-using consumer

products—improved education and improved technology have raised awareness of
environmental issues, and allowed systems to be operated more efficiently, but if users
don’t act accordingly, there will be no overall change in behaviour.

Quadrant 2: New user behaviour with existing systems

Educating users about the implications of their behaviour is generally done with the
intention that users will follow through and actually change the way they use systems (if
they don’t change, this is 1c as described above). If this is successful—e.g. a campaign to
persuade people to keep their car tyres inflated correctly to save fuel—then new user
behaviour occurs with existing systems, and no design or engineering changes are needed
to the systems. Overall, there is a change in behaviour.

The scope of this quadrant corresponds closely with much current government policy of
using social marketing, public education campaigns and so on—employing persuasion
and rhetoric to drive attitude change as a foundation for behaviour change. There are
many ways that this quadrant could be subdivided into behavioural cases, but from the
point of view of the current study, this will not be explored further here.
Quadrant 4: Existing user behaviour with new system behaviour

Where new systems themselves behave differently in use, yet allow users to maintain
their existing behaviours, overall behaviour change results without users necessarily
needing to understand the issues involved. No persuasion occurs. For example, compact
fluorescent lightbulbs, from the user’s point of view, do not require any different user
behaviour to tungsten filament bulbs, but in operation they always result in new system
behaviour. A refrigerator door which automatically closes itself if left ajar does not, again,
require the user to do anything different, but the system itself behaves differently to
accommodate existing user behaviour.

This quadrant would include the major proportion of ‘eco-products’ available, most of
which are designed to allow the user to change routines and behaviours as little as
possible; there are many possible ways the category can be subdivided further according
to various other factors.

Quadrant 3: New user behaviour with new system behaviour

In the cases described by this quadrant, both system behaviour and user behaviour change,
resulting in an overall behaviour change. The behaviour change can be driven entirely by
functional changes to the system, or by mindful user understanding, or by both, but the
products are designed to lead to this. This is Design with Intent.

These are systems that enable, motivate or constrain—influence—the way that users can
interact with them. A common factor is that there is a perceived affordance change with
the system: it somehow indicates that a change in behaviour is needed (compared with
quadrant 4 where there is no such indication). This quadrant is where the authors’
research is focused.

In case 3a, the perceived affordance change does not reflect actual functional change to
the system, yet it influences users to change their behaviour. For example, a washing
machine which gives users an ‘estimated cost’ for each mode still embodies all the same
functions as one which doesn’t—the user can choose to ignore the recommendation, but
is influenced to choose the most economical mode, and thus a change in product
behaviour is likely to result from the change in user behaviour. This is where much
Persuasive Technology research (e.g. Oinas-Kukkonen, Hasle, Harjumaa, Segerstahl, &
Ohrstrom, 2008) seems to fit.

3c is the case where a user need not think about the issues involved, but will still behave
differently due to functional changes to the system—e.g. a washing machine which
automatically determines the most efficient settings for a particular load, and silently
carries them out, doesn’t require the user to understand what’s going on, but does end up
changing the user’s behaviour (removing inefficient decisions) and thus the system
behaviour changes too. These systems have the potential to be complex, especially where
automation is required, but need not be. Something as simple as removing an option from
a menu changes the user’s behaviour (prevents him or her choosing it) but doesn’t require
the user to think about it.

Finally, returning to the centre of the quadrant, 3b describes cases where user
understanding, alongside functional changes to the system and perceived affordance
change, lead to user and system behaviour change in practice: these are the real core of
what this study is about and where the authors hope they will be able to make advances in
understanding useful to designers and anyone else working in the field of influencing user
behaviour. These are interesting systems, potentially involving lots of factors and effects
but not necessarily complex in themselves.

THE DESIGN WITH INTENT METHOD

The Design with Intent concept (DwI) refers to design intended to influence or result in
certain user behaviour across a range of disciplines from architecture to software.
Techniques used in one context, suitably generalised, can be applied in others, and the
DwI Method aims to assist this process. It takes designers from a brief involving
behaviour change, to a range of applicable design techniques and examples (‘patterns’—
cf. Alexander et al., 1977; Tidwell, 2005) which can inspire concepts addressing the
problem—effectively a ‘structured brainstorming’ method.

The general structure of the method is shown in Figure 2: there are two ‘modes’,
inspiration and prescription, depending on how the designer or design team prefers to
make use of it.

Figure 2. Structure of the DwI Method

The starting point is a design brief for the system involving influencing user behaviour. In
the inspiration mode (Lockton, Harrison, & Stanton, 2009), the designer simply takes
inspiration from a set of ‘headline’ design patterns which are applicable to a wide range
of target behaviours, grouped into six different ‘lenses’ (Table 2) representing particular
‘worldviews’ on using design to influence behaviour. These allow designers to think
outside the immediate frame of reference suggested by the brief (or client). The patterns,
illustrated with examples from different fields, serve as a creative trigger. The use of
illustrated examples in an ‘idea space’ is intended to allow designers to understand the
patterns quickly, even where the terminology is unfamiliar.

Table 2. Six ‘lenses’ for looking at influencing user behaviour

Architectural lens The Architectural Lens draws on techniques used to influence user behaviour in
architecture, urban planning and related disciplines such as traffic management
and crime prevention through environmental design (see also the Security lens).
While the techniques have been developed in the built environment, many of the
ideas can also be applied in interaction and product design, even in software or
services; they are effectively about using the structure of systems to influence
behaviour.
Errorproofing lens The Errorproofing Lens represents a worldview treating deviations from the target
behaviour as ‘errors’ which design can help avoid, either by making it easier for
users to work without making errors, or by making errors impossible in the first
place. This view on influencing behaviour is often found in health & safety-
related design, medical device design and manufacturing engineering.
Persuasive lens The Persuasive Lens represents the emerging field of persuasive technology,
where computers, mobile phones and other systems with interfaces are used to
persuade users: changing attitudes and so changing behaviour through contextual
information, advice and guidance.
The major applications so far have been in influencing behaviour for social
benefit, e.g. persuading people to give up bad habits, adopt healthier lifestyles or
reduce their energy use.
Visual lens The Visual Lens combines ideas from product semantics, semiotics, ecological
psychology and Gestalt psychology about how users perceive patterns and
meanings as they interact with the systems around them. These techniques are
often applied without necessarily considering how they can influence user
behaviour.
Cognitive lens The Cognitive Lens draws on research in behavioural economics and cognitive
psychology looking at how people make decisions, and how this is affected by
‘heuristics’ and ‘biases’. If designers understand how users make interaction
decisions, that knowledge can be used to influence interaction behaviour.
Equally, where users often make poor decisions, design can help counter this,
although this risks the accusation of design becoming a tool of the ‘nanny state’
which ‘knows what’s best’.
Security lens The Security Lens represents a ‘security’ worldview, i.e. that undesired user
behaviour is something to deter and/or prevent though ‘countermeasures’
designed into products, systems and environments, both physically and online,
with examples such as digital rights management.
From a designer’s point of view, this can be an ‘unfriendly’ and, in some
circumstances unethical view to take, effectively treating users as ‘guilty until
proven innocent’.

In the prescription mode, the designer formulates the brief in terms of one of a range of
target behaviours (Table 3), seven of which describe interactions between a user and the
system, and four of which relate to influencing interaction between users, mediated by the
system. For each target behaviour, a subset of the most applicable design patterns from
each lens is presented to the designer, again illustrated with examples.

While still serving as a creative inspiration, this mode effectively ‘prescribes’ a set of
patterns which are especially applicable or have been applied to similar problems by
other designers, in other contexts. Each pattern is described and the advantages and
disadvantages discussed with notes on its implementation, user reactions and
effectiveness. From the prescribed patterns, a range of design concepts can be generated,
all of which have at least some ‘precedent’ in terms of the underlying patterns’
application in behaviour change contexts. The form of the prescription mode is loosely
modelled on that of TRIZ (Altshuller, 1996), leading the designer from a specific brief to
a more general brief (the target behaviour) and general solutions (the suggested patterns)
then serving as the inspiration for specific solutions to the original brief.

Table 3. Target behaviours with examples

User-system interaction
Influencing interactions between a user and a system Examples
S1 User follows process or path, performing Customer places order via website without missing
actions in a specified sequence out any steps
S2 User follows process or path optimized for User fills/boils kettle with right amount of water
run-time criteria
S3 Decision among alternatives: user’s choice Diners choose healthier meal in office canteen
is guided
S4 Access, use or occupation based on user Only users who know PIN can access bank account
characteristics via ATM
S5 Access, use or occupation based on user If driver speeding, next traffic lights turn red, else
behaviour green
S6 No access, use or occupation, in a specific Park bench fitted with central armrest to prevent
manner, by any user anyone lying down
S7 User provided with functionality only when Office lighting cannot be switched on if ambient
environmental criteria satisfied daylight adequate
User-user interaction
Influencing interaction between users, mediated by system Examples
U1 Separate flows and occupation: users have Traffic follows one-way system into/out of car park
no influence on each other
U2 Interaction occurs between users or groups Staff from different departments mix socially in
of users atrium
U3 No user-created blockages or congestion Wide pedestrian concourses prevent groups blocking
caused by multiple users passage for others
U4 Controlled rate of flow or passage of users Visitors to popular museum exhibit routed past it
slowly on moving walkway

Alongside the details of implementation and examples, each pattern within each lens is
classified into one or more approach: enabling, motivating or constraining behaviour
(Table 4). These are intended to help a designer appreciate the mechanisms behind the
design pattern.

Much work in Persuasive Technology has taken the approach of motivating behaviour,
with attitude change usually a precursor, but Fogg’s reduction and tunnelling (Fogg, 2003)
are arguably also about enabling particular behaviours by making them simpler (Maeda,
2006). Buckminster Fuller’s ‘trimtab’ concept—“modify[ing] the environment in such a
way as to get man moving in preferred directions” (Krausse & Lichtenstein, 2001)—also
accords with the enabling approach and provides a link to the wider field of design for
social benefit. Human factors strategies aimed at influencing behaviour in a health and
safety context often employ a constraining approach. The approach used in practice—and
hence the patterns and concepts chosen for further development—may, of course, be
dictated by the client or other stakeholders rather than being the designer’s decision.

Table 4. Three approaches to influencing behaviour by design

Enabling behaviour Enabling ‘desirable’ behaviour by making it easier for the user than the
alternatives
Motivating behaviour Motivating users to change behaviour by educating, incentivising and
changing attitudes
Constraining behaviour Constraining users to ‘desirable’ behaviour by making alternatives difficult or
impossible

The DwI Method has been developed through a series of workshop sessions with students
(e.g. Lockton, Harrison, Holley, & Stanton, 2009), applying it to problems including
influencing more efficient use of household lighting, and improving the efficiency of
printing from a user’s point of view. Development is ongoing with application to a range
of ‘sustainable behaviour’ problems; its use is best explained via an example application,
using the prescription mode.

Example: Influencing householders to close curtains at night

A significant proportion of household heat loss is via windows, both the glass and
surrounding frames. The UK’s Energy Saving Trust puts the figure at 20% (Energy
Saving Trust, n.d.), while the University of Florida estimates that “20-50% of the total
energy loss in a well-insulated structure occurs through and around the windows and
doors” (Hammer, 2003). Aside from better insulation, human behaviour has the potential
to contribute to energy saving in this context as in many others; the UK government’s
Foresight programme contends that “Human behaviour determines energy use as much as
building design” (Department for Innovation, Universities and Skills, 2008).

Specifically addressing heat loss via windows, one of the behaviour changes frequently
recommended is for householders to make sure curtains in every room are closed at night.
The US Department of Energy goes so far as to make this a featured ‘Energy Savings
Tip’ on the homepage of its public-facing website (Department of Energy, n.d.). There is
scope for providing householders with financial savings without initial extra expenditure,
unlike installing better insulation or glazing1; the fact that not all householders routinely
close their curtains at night suggests that a design intervention may be helpful.

1
The obvious technological solutions such as new glazing, or automatic systems, while ‘solving’ the
problem, are a different form of intervention, not actually about influencing user behaviour. These would
Considering the curtain problem reveals at least three possible sources of ‘error’ in terms
of deviation from the ideal behaviour: users understanding conservation but forgetting to
close curtains (oversight); users not understanding conservation or the link with curtains,
and thus taking no action; or users finding it difficult to close curtains and therefore not
carrying out the process. Once learned, the process of closing curtains at night becomes a
rule-based behaviour (Vicente & Rasmussen, 1992); the oversight error is a lapse
(Norman, 1983) or an error of omission (Matthews, Davies, Westerman, & Stammers,
2000) and the lack-of-understanding error a mistake (Norman, 1983) or perhaps a
sequential error of commission (Matthews et al., 2000). The third source of error—
inconvenience—may also be important: in certain rooms of a house it may be difficult to
reach curtains to close (or open) them, for example due to furniture in front of them, or
for people with impaired mobility. Since leaving curtains open all the time allows the
room to be naturally lit during the daytime, and, if the room is not used overnight, does
not appear to give any disadvantages (other than the heat loss), this may become the
default state. This is often the case in offices.

Regardless of the error type, the ideal outcome would be for neither problem to occur at
all, as part of a process of user interaction. A design brief might describe this simply as
“We want users to close the curtains at night to conserve heat”.

Applying the method (Fig. 2) in prescription mode, the brief is expressed as a target
behaviour (Table 3). The process of closing curtains is a user-system interaction, and of
the target behaviours listed, S1—“User follows process or path, performing actions in a
specified sequence”—is the best match for the brief. Following through what is
prescribed for the S1 target behaviour by each of the six lenses results in 19 different
patterns in total, each of which, individually or in combination, potentially suggests a
number of possible concept solutions (Table 5) to a designer working on the problem.

Table 5. The design patterns applicable to the S1 target behaviour—“User follows process or path,
performing actions in a specified sequence”—and some concept solutions based on them.

Patterns applicable to S1 Concept for new / redesigned interface or product inspired

target behaviour
Self-monitoring Curtains / windows / heating system which can inform user about their
state. Use potential benefit compared with past behaviour etc. Could use
windows / curtains as interface, e.g. projecting information / graphics
Reduction Reducing hassle / effort required by users to close curtains—e.g. a
weighted system or combined mechanism closing multiple curtains
Simulation & feedforward Suggest / simulate closing curtains at exactly right moment—when users
Kairos about to go to bed, or when they enter room and switch lights on, or if
Conditional warnings significant heat outflow detected
Operant conditioning Rewarding user for closing curtains by providing praise, ‘delight’ reward,
or explicit display of money saved
Respondent conditioning Helping user develop habit of closing curtains by association with

come under Quadrant 4 of the ‘Behaviour Change Barometer’: designed to allow the user to change
routines and behaviours as little as possible.
 another event, e.g. going to bed; embedding ‘trigger’ in environment
Commitment & consistency Interface which gets users to commit to a goal of a particular energy use
improvement
Self-monitoring Interface which points out how well user doing (e.g. “Your insulation is
only 65% effective because curtains are still open in three rooms.”)
Scarcity Interface which points out energy/financial waste of not closing curtains
Self-monitoring
Scarcity Demonstrate to users ‘precious warmth’ and how they’ll feel warmer if
Self-monitoring they close curtains
Affective engagement Emotional involvement, e.g. causing users to worry or feel guilty about
energy/financial waste or lack of privacy from open curtains
Surveillance Making it clear that other people can see into your house at night if you
don’t close curtains
Interlock If dark outside, room lighting can’t be switched off (before going to bed)
Lock-in & lock-out until curtains closed
Interlock Room lighting can’t be switched on in first place if dark outside, unless
Lock-in & lock-out curtains closed
Interlock Room heating won’t come on at night unless curtains closed
Lock-in & lock-out
Conditional warnings Warning lights / alarm / display somewhere in house if curtains not
closed
Movement & oscillation Curtain could move indicating that it should be closed
Segmentation & spacing De-segmentation of elements, so fewer curtains to close, or linking
closing mechanisms (single pull cord) so less work for user
Positioning & layout Positioning other items used as part of nightly routine, e.g. light switches,
Proximity & similarity next to curtains
Implied sequences
Prominence & visibility Prominent pull cords / other closing mechanisms
Positioning & layout Room / furniture layouts designed to make it easier to reach curtains

The patterns suggested involve a mixture of enabling, motivating and constraining

approaches, and this is reflected in the concept solutions. Some aim to use design to make
it easier for people to close curtains (enabling), to remind them to do so, or to teach them
that it is worth doing (motivating); others try to force users to adopt a routine
(constraining). Alongside financial savings from reduction in energy use, privacy can be
emphasised as a factor in motivating users to close curtains.

The decision on which concepts to develop further would of course be subject to many
more considerations: the reality of most design processes is that situational constraints
(financial, political, legal, development time, domain expertise, organizational factors)
will significantly shape the outcome. The DwI Method is expected to open up
constructive innovation rather than immediately converge on particular solutions.

CONCLUSION

This chapter has introduced the idea that the design of the systems around us can have a
role in influencing human behaviour, and discussed a range of perspectives from different
disciplines. A method for classifying behaviour change, the ‘behaviour change
barometer’ has been explored, and the ‘Design with Intent Method’, a design tool to help
designers working on behaviour change problems, has been introduced and applied
briefly to a problem involving influencing more sustainable user behaviour, to generate a
range of possible concept solutions.

Influencing user behaviour through design has significant potential for social benefit,
particularly where human behaviour and product use decisions directly impact on the
environment. Lessons learned and design techniques employed to influence behaviour in
one field can often find additional application in others. Intelligent changes in user
behaviour, achieved through modification of a system, can have a worthwhile impact on
energy use and resource consumption, in many cases also saving users money in the
process.

Design considerations should, therefore, be part of any behaviour change strategy where
design could play a part. This is a young, emerging field, and its potential for improving
human well-being has yet to be demonstrated on a large scale, but the opportunities exist
to match design solutions to a wide range of the behaviour problems facing society today
and in our shared future.

REFERENCES

Adams, P., & David, G. (2007). Light vehicle fuelling errors in the UK: The nature of the problem, its
consequences and prevention. Applied Ergonomics, 38, 499--511.
Akrich, M. (1992). The De-Scription of Technical Objects. In W. Bijker & J. Law (Eds.), Shaping
Technology / Building Society (pp. 205--224). Cambridge: MIT Press.
Alexander, C., Ishikawa, S., Silverstein, M., Jacobson, M., Fiksdahl-King, I., & Angel, S. (1977). A Pattern
Language: Towns, Buildings, Construction. New York: Oxford University Press.
Altshuller, G. (1996). And Suddenly the Inventor Appeared. Worcester: Technical Innovation Center.
Ariely, D. (2008). Predictably Irrational. London: HarperCollins.
Barker, R. G. (1968). Ecological Psychology. Stanford: Stanford University Press.
Beatty, P. C. (2008). Designing Persuasive Technologies for Human Factors Engineering: An Alternative
Classification to the Triad of Captology. In H. Oinas-Kukkonen, P. F. V. Hasle, M. Harjumaa, &
K. Segerstahl (Eds.), Persuasive 2008: Third International Conference on Persuasive Technology:
Poster Proceedings (pp. 98--101). Oulu: University of Oulu Press.
Bhamra, T., Lilley, D., & Tang, T. (2008). Sustainable use: Changing Consumer Behaviour Through
Product Design. In Changing the Change, Turin, Italy, July 2008: Proceedings. Turin.
Blevis, E. (2007). Sustainable Interaction Design: Invention & Disposal, Renewal & Reuse. In CHI 2007
Proceedings - Design Theory.
Brunswik, E. (1956). Perception and the representative design of psychological experiments. Berkeley:
University of California Press.
Buchanan, R. (1985). Declaration by Design: Rhetoric, Argument, and Demonstration in Design Practice.
Design Issues, 2
Carroll, J. M., Kellog, W. A., & Rosson, M. B. (1991). The Task-Artifact Cycle. In J. M. Carroll (Ed.),
Designing Interaction: Psychology at the Human-Computer Interface. London: Cambridge
University Press.
Chamorro-Koc, M., Popovic, V., & Emmison, M. (2008). Human experience and product usability:
Principles to assist the design of user-product interactions. Applied Ergonomics, 39, 1--9.
Cialdini, R. (2003). Crafting Normative Messages to Protect the Environment. Current Directions in
Psychological Science, 12
Cone, J. D., & Hayes, S. C. (1984). Environmental Problems / Behavioral Solutions. Cambridge:
Cambridge University Press.
Department for Environment, Food and Rural Affairs. (2008, January). A framework for pro-environmental
behaviours. Department for Environment, Food and Rural Affairs.
Department for Innovation, Universities and Skills. (2008, November 26). Energy re-think needed to make
homes and businesses greener (Press Release). Retrieved December 2, 2008, from
http://nds.coi.gov.uk/environment/fullDetail.asp?ReleaseID=385484&NewsAreaID=2.
Department of Energy. (n.d.). Department of Energy - Homepage. Retrieved November 30, 2008, from
http://www.energy.gov/.
Design Council, Department of Heath, Buckle, P., Clarkson, P., Coleman, R., Lane, R. M., et al. (2003).
Design for Patient Safety: A System-Wide Design-Led Approach to Tackling Patient Safety in the
NHS. London: Design Council / Department of Health. Retrieved December 2, 2008, from
http://www-edc.eng.cam.ac.uk/medical/downloads/report.pdf.
Elias, E. W., Dekoninck, E. A., & Culley, S. J. (2007). The Potential for Domestic Energy Savings through
Assessing User Behaviour and Changes in Design. In Proceedings of EcoDesign 2007. Tokyo.
Energy Saving Trust. (n.d.). Walls, windows, and doors: Energy saving statistics and facts. Retrieved
December 2, 2008, from http://www.energysavingtrust.org.uk/corporate/Corporate-and-
media/Media-centre/Energy-saving-statistics-and-facts/Walls-windows-and-doors.
Fogg, B. J. (1999). Persuasive Technologies. Communications of the ACM, 42
Fogg, B. J. (2003). Persuasive Technology: Using Computers to Change What We think and Do. San
Francisco: Morgan Kaufmann.
Gaver, W. W. (1991). Technology Affordances. In Proceedings of the SIGCHI conference on Human
factors in computing (p. 79
Gaver, W. W. (1992). The Affordances of Media Spaces for Collaboration. In Proceedings of the 1992
ACM conference on Computer-supported cooperative work (pp. 17-24). Toronto.
Gibson, J. (1986). The Ecological Approach to Visual Perception. Hillsdale: Lawrence Erlbaum Associates.
Gillespie, T. (2007). Wired Shut: Copyright and the Shape of Digital Culture. Cambridge: MIT Press.
Gilovich, T., Griffin, D., & Kahneman, D. (Eds.). (2002). Heuristics and Biases: The Psychology of
Intuitive Judgment. Cambridge: Cambridge University Press.
Grout, J. (2007). Mistake-Proofing the Design of Health Care Processes. Rockville: Agency for Heathcare
Research and Quality.
Hammer, M. (2003). Window Management for Energy Conservation. Institute of Food and Agricultural
Sciences, University of Florida. Retrieved December 2, 2008, from
http://edis.ifas.ufl.edu/pdffiles/FY/FY54800.pdf.
Hyde, B. (2007, May 10). Friction. Ascription is an Anathema to any Enthusiasm. Retrieved March 5, 2009,
from http://enthusiasm.cozy.org/archives/2007/05/friction.
Jelsma, J., & Knot, M. (2002). Designing environmentally efficient services: a 'script' approach. Journal of
Sustainable Product Design, 2(3-
Kahneman, D., Slovic, P., & Tversky, A. (Eds.). (1982). Judgment under Uncertainty: Heuristics and
Biases. Cambridge: Cambridge University Press.
Kahneman, D., & Tversky, A. (Eds.). (2000). Choices, Values, and Frames. Cambridge: Cambridge
University Press.
Kanis, H. (1998). Usage centred research for everyday
product design. Applied Ergonomics, 29(1), 75--82.
Kesan, J. P., & Shah, R. C. (2005). Shaping Code. Harvard Journal of Law and Technology, 18(2),

Kjaer Christensen, A., & Hasle, P. F. V. (2007). Classical Rhetoric and a Limit to Persuasion. In Y. de Kort,

de Kort, Y. A., McCalley, L. T., & Midden, C. J. H. (2008). Persuasive Trash Cans: Activation of Littering
Norms by Design. Environment and Behavior, In Press.
Krausse, J., & Lichtenstein, C. (2001). R.Buckminster Fuller: Your Private Sky: Discourse. Baden: Lars
Muller Publishers.
Latour, B. (1992). Where Are The Missing Masses? The Sociology of a Few Mundane Artifacts. In W.
Bijker & J. Law (Eds.), Shaping Technology / Building Society (pp. 225--258). Cambridge: MIT
Press.
Lessig, L. (1999). Code and Other Laws of Cyberspace. New York: Basic Books.
Lilley, D., Lofthouse, V., & Bhamra, T. (2005). Towards Instinctive Sustainable Product Use. In
Proceedings of 2nd International Conference in Sustainability, Creating the Culture.
Lilley, D., Lofthouse, V., & Bhamra, T. (2006). Towards sustainable use: An exploration of designing for
behavioural change. In L. Feijs, S. Kyffin, & B. Young (Eds.), Proceedings of DeSForM 2006:
Design and semantics of form and movement
Lockton, D. (2005a). Architectures of Control in Consumer Product Design. MPhil Dissertation, University
of Cambridge.
Lockton, D. (2005b). What is Design with Intent? Design with Intent / Architectures of Control. Retrieved
December 2, 2008, from http://architectures.danlockton.co.uk/what-is-design-with-intent/.
Lockton, D. (2005c). 'Benches' Archive. Design with Intent / Architectures of Control. Retrieved December
2, 2008, from http://architectures.danlockton.co.uk/category/benches/.
Lockton, D., Harrison, D., Holley, T., & Stanton, N. A. (2009). Influencing Interaction: Development of the
Design with Intent Method. In Persuasive Technology: Fourth International Conference,
Persuasive 2009, Claremont, California, April 27-29, 2009, Proceedings (p. to appear). Claremont,
California: ACM Press.
Lockton, D., Harrison, D., & Stanton, N. (2008a). Design with Intent: Persuasive Technology in a Wider
Context. In H. Oinas-Kukkonen, P. Hasle, M. Harjumaa, K. Segerstahl, & P. Ohrstrom (Eds.), (p.
78). Berlin: Springer.
Lockton, D., Harrison, D., & Stanton, N. (2008b). Making the user more efficient: Design for sustainable
behaviour. International Journal of Sustainable Engineering, 1(1), 3--8.
Lockton, D., Harrison, D., & Stanton, N. A. (2009). Design for Behaviour Change: The Design with Intent
Toolkit v.0.9. Uxbridge: Brunel University Press.
Maeda, J. (2006). The Laws of Simplicity. Cambridge: MIT Press.
Matthews, G., Davies, D. R., Westerman, S. J., & Stammers, R. D. (2000). Human Error. In Human
Performance: Cognition, stress and individual differences. New York: Psychology Press.
McCalley, L. T., & Midden, C. J. H. (2002). Energy conservation through
product-integrated feedback: The roles of
goal-setting and social orientation. Journal of Economic Psychology, 23, 589--603.
McGrenere, J., & Ho, W. (2000). Affordances: Clarifying and Evolving a Concept. In Proceedings of
Graphics Interface 2000. Montreal.
Norman, D. (1988). The Psychology of Everyday things. New York: Basic Books.
Norman, D. (2008). Signifiers, Not Affordances. Interactions, 15(6), 18--19.
Norman, D. A. (1983). Design rules based on analyses of human error. Communications of the ACM, 26(4),
254--258.
Norman, D. A. (1999). Affordance, Conventions, and Design. Interactions
Oinas-Kukkonen, H., Hasle, P., Harjumaa, M., Segerstahl, K., & Ohrstrom, P. (Eds.). (2008). Persuasive
Technology: Third International Conference, Persuasive 2008, Oulu, Finland, June 4-6, 2008,
Proceedings. Berlin: Springer.
Petty, R. E., & Cacioppo, J. T. (1983). Central and Peripheral Routes to Persuasion: Application to
Advertising. In L. Percy & A. Woodside (Eds.), Advertising and Consumer Psychology. Lexington:
Lexington Books.
Redstrom, J. (2006). Persuasive Design: Fringes and Foundations. In W. IJsselsteijn, Y. de Kort, C. J. H.
Midden, B. Eggen, & E. van den Hoven (Eds.), Proceedings of First International Conference on
Persuasive Technology for Human Well-Being, PERSUASIVE 2006
Springer.
Rodriguez, E., & Boks, C. (2005). How design of products affects user behaviour and vice versa: the
environmental implications. In Proceedings of Fourth International Symposium on
Environmentally Conscious Design and Inverse Manufacturing, 2005. Eco Design 2005. . Tokyo.
Salovaara, A. (2008). Inventing New Uses for Tools: A Cognitive Foundation for Studies on Appropriation.
Human Technology, 4(2), 209--228.
Schneier, B. (2003). Beyond Fear. New York: Copernicus Books.
Shingo, S. (1986). Zero Quality Control: Source Inspection and the Poka-Yoke System. Portland:
Productivity Press.
Sommer, R. (1969). Personal Space: The Behavioral Basis of Design. Englewood Cliffs: Prentice-Hall.
Stanton, N. A., & Baber, C. (1998). Designing for consumers: editorial. Applied Ergonomics, 29(1), 1--3.
Stern, N. (2007). Stern Review: The Economics of Climate Change. Cambridge: Cambridge University
Press.
 Sunstein, C. R., & Thaler, R. H. (2003). Libertarian Paternalism is Not an Oxymoron. University of
Chicago Law Review, 70
Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness.
New Haven: Yale University Press.
Tidwell, J. (2005). Designing Interfaces. Sebastopol: O'Reilly.
Vicente, K., & Rasmussen, J. (1992). Ecological interface design: theoretical foundations. IEEE
Transactions on Systems, Man and Cybernetics, 22(4), 589--606.
Walker, G. H., Stanton, N. A., Jenkins, D. P., & Salmon, P. M. (2009). From telephones to iPhones:
Applying systems thinking to networked,
interoperable products. Applied Ergonomics, 40, 206--215.
Wever, R., van Kuijk, J., & Boks, C. (2008). User-centred Design for Sustainable Behaviour. International
Journal of Sustainable Engineering, 1
Winter, D. D. N., & Koger, S. M. (2004). The Psychology of Environmental Problems (2nd ed.). Mahwah:
Lawrence Erlbaum Associates.
Wood, G., & Newborough, M. (2003). Dynamic energy-consumption indicators for domestic appliances:
environment, behaviour and design. Energy and Buildings
Zittrain, J. (2008). The Future of the Internet - And How to Stop It. London: Allen Lane.

View publication stats