<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Persuasive Cities: Health Behavior Change at Scale</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Agnis Stibe</string-name>
        </contrib>
        <contrib contrib-type="author">
          <string-name>MIT Media Lab agnis@mit.edu</string-name>
        </contrib>
      </contrib-group>
      <abstract>
        <p>Can you imagine a city that feels, understands, and cares about your wellbeing? Future cities will reshape human behavior in countless ways. New strategies and models of urban spaces are required for creating future cities to properly respond to human activity, environmental conditions, and market dynamics. Persuasive urban systems will play an important role in making cities more livable and resource-efficient by addressing current environmental problems and enabling healthier routines. Drawing on socio-psychological theories and integrating them with new concepts for urban design, the persuasive cities research focuses on improving wellbeing across societies. This research presents an ecosystem of future cities, describes three generic groups of people depending on their susceptibility to persuasive technology, explains the process of defining behavior change, and provides tools for social engineering of persuasive cities. Further research should continue exploring how urban design in combination with socially influencing systems could encourage healthy and sustainable behaviors at scale.</p>
      </abstract>
      <kwd-group>
        <kwd>persuasive cities</kwd>
        <kwd>social engineering</kwd>
        <kwd>socially influencing systems</kwd>
        <kwd>behavior change</kwd>
        <kwd>wellbeing</kwd>
        <kwd>health</kwd>
        <kwd>persuasive technology</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>
        As population in cities continue grow exponentially the architecture and design of
future urban places will become more dominant in impacting human behavior.
According to social cognitive theory [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ], any well-designed environment can become a
strong influencer of what people think and do. There is an endlessly dynamic
interaction between a person, a particular behavior, and an environment in which that
behavior is performed. The persuasive cities research leverages this knowledge to engineer
persuasive environments for altering human behavior on societal levels.
      </p>
      <p>
        The proposed research reflects on novel ways of how persuasive technology [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]
and socially influencing systems [
        <xref ref-type="bibr" rid="ref3 ref4">3-4</xref>
        ] enable mechanisms to perpetually support
motivation of individuals comparing to conventional methods, such as those that are
based on carrots and sticks. Instead, persuasive urban systems harness social influence
from crowd behavior to craft influential messaging aimed at shifting behavior and
attitude of an individual, who naturally is an integral part of the same crowd. Such
continuous interplay can ultimately result in an ongoing process that reshapes
communities and societies without any other incentives.
      </p>
    </sec>
    <sec id="sec-2">
      <title>Emergence of Persuasive Cities</title>
      <p>Ongoing research streams focus on sensible cities (researching sensing technologies
to read human behavior in urban spaces) and smart cities (analyzing big data to
classify groups of people based on their distinct behavioral patterns), however there is a
lack of knowledge about perspective ways to achieve persistent behavioral changes at
scale. Therefore, the proposed research extends an ecosystem of future cities (Table
1) by introducing the notion of persuasive cities that aims to advance and refine
influential strategies designed for intentionally reshaping how people think and act in
urban environments.
Each layer of future cities has its role, character, and supportive technology. Sensible
cities employ sensor networks to read crowd behaviors. In other words, these cities
feel human movements. These crowd behaviors further serve as an input for big data
analytics that smart cities apply to classify groups of people according to similar
behavioral patterns (profiles). When that is accomplished, the groups having better
routines can be exemplified to other underperforming groups through intentionally
designed socially influencing systems, which are at the core of persuasive cities.
3</p>
    </sec>
    <sec id="sec-3">
      <title>Susceptibility to Persuasive Technology</title>
      <p>People generally can fall into one of the three generic categories depending on their
susceptibility to persuasive technology (Fig. 1). Self-contained people (the red circle)
most likely are not open for changing anything in them. They are fully satisfied with
who they are and what they do on daily basis, thus many behavioral interventions
might fail in attempts to influence this group of individuals. Self-driven people (the
green circle) typically have comparatively high levels of motivation and can achieve
everything that they have envisioned. Thus, these people most likely are not looking
for additional sources of encouragement, and therefore persuasive technologies might
become unnecessary for this group.</p>
      <p>However, there is another group of people that oftentimes would like to change
their routines, but rarely succeed in doing so. That reminds of New Year’s resolutions
that in many cases end around February. Therefore, this group is entitled as January
1st (the yellow circle) and seem to be the most welcoming towards technology
supported behavioral interventions designed to help achieving target behaviors.
Although, Fig. 1 presents all three groups as equal circles, in reality the size of each
group might significantly vary depending on the context and particular behavior.</p>
      <sec id="sec-3-1">
        <title>Self-Contained</title>
      </sec>
      <sec id="sec-3-2">
        <title>January 1st</title>
      </sec>
      <sec id="sec-3-3">
        <title>Self-Driven</title>
        <p>FAIL</p>
        <p>APPRECIATE</p>
        <p>
          UNNECESSARY
To achieve an envisioned target behavior, the process and components of behavior
change have to be well understood and clearly defined. In the process of defining
behavior change, there are three main components, namely the target group, its
present behavior, and its envisioned future behavior (Table 2).
Earlier research on persuasive technology [
          <xref ref-type="bibr" rid="ref2">2</xref>
          ] describes several ways how social
dynamics can influence human behavior, which have been further refined and structured
as a framework for Socially Influencing Systems (SIS) [
          <xref ref-type="bibr" rid="ref3">3</xref>
          ], depicted in Fig. 2. The SIS
framework is a useful tool for scholars and practitioners aiming at improving future
cities by introducing persuasive urban interventions targeted to support wellbeing.
        </p>
        <sec id="sec-3-3-1">
          <title>PARTICIPATION</title>
          <p>SF
Social
facilitation
SL
Social
learning
SC</p>
          <p>Social
comparison</p>
        </sec>
        <sec id="sec-3-3-2">
          <title>ENGAGEMENT</title>
          <p>NI
Normative
influence</p>
        </sec>
        <sec id="sec-3-3-3">
          <title>INVOLVEMENT</title>
          <p>
            The framework describes seven socially influencing principles that can support
persuasive urban interventions. The principles are interlinked and have potential to exert
stronger effects depending on the context of a particular behavioral challenge.
Normative influence and social comparison seem to be more effective to achieve
involvement of the target group as the two principles focus on attitudinal changes.
Cooperation and social facilitation seem to be more effective to make individuals participate
and do the envisioned future behavior even without a formed attitude towards it.
Competition and recognition seem to be more effective in engaging the target group
to do the future behavior as the principles focus on both attitude and behavior
simultaneously. For example, the effects several socially influencing principles have
already been studied in the context of urban mobility, e.g. bicycling [
            <xref ref-type="bibr" rid="ref5">5</xref>
            ].
6
          </p>
        </sec>
      </sec>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Bandura</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>Social Foundations of Thought and Action: A Social Cognitive Theory</article-title>
          . Prentice Hall, Englewood
          <string-name>
            <surname>Cliffs</surname>
          </string-name>
          (
          <year>1986</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Fogg</surname>
            ,
            <given-names>B. J.</given-names>
          </string-name>
          : Persuasive Technology:
          <article-title>Using Computers to Change What We Think and Do</article-title>
          . San Francisco: Morgan Kaufmann (
          <year>2003</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>Stibe</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>Towards a Framework for Socially Influencing Systems: Meta-Analysis of Four PLS-SEM Based Studies</article-title>
          . In: MacTavish,
          <string-name>
            <given-names>T.</given-names>
            ,
            <surname>Basapur</surname>
          </string-name>
          , S. (eds.)
          <source>Persuasive Technology. LNCS</source>
          , vol.
          <volume>9072</volume>
          , pp.
          <fpage>171</fpage>
          -
          <lpage>182</lpage>
          . Springer, Heidelberg (
          <year>2015</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Stibe</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>Advancing Typology of Computer-Supported Influence: Moderation Effects in Socially Influencing Systems</article-title>
          . In: MacTavish,
          <string-name>
            <given-names>T.</given-names>
            ,
            <surname>Basapur</surname>
          </string-name>
          , S. (eds.)
          <source>Persuasive Technology. LNCS</source>
          , vol.
          <volume>9072</volume>
          , pp.
          <fpage>251</fpage>
          -
          <lpage>262</lpage>
          . Springer, Heidelberg (
          <year>2015</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Wunsch</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Stibe</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Millonig</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Seer</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Dai</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Schechtner</surname>
            ,
            <given-names>K</given-names>
          </string-name>
          , and
          <string-name>
            <surname>Chin</surname>
          </string-name>
          , R.C.C.
          <article-title>What Makes You Bike? Exploring Persuasive Strategies to Encourage Low-Energy Mobility</article-title>
          .
          <source>Lecture Notes in Computer Science</source>
          ,
          <volume>9072</volume>
          ,
          <string-name>
            <surname>Persuasive</surname>
            <given-names>Technology</given-names>
          </string-name>
          , pp.
          <fpage>53</fpage>
          -
          <lpage>64</lpage>
          . (
          <year>2015</year>
          )
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>