A touching app voice thinking about ethics of persuasive technology through an analysis of mobile smoking-cessation apps

Abstract

We study smoking-cessation apps in order to formulate a framework for ethical evaluation, analyzing apps as ‘medium’, ‘market’, and ‘genre’. We center on the value of user autonomy through truthfulness and self-understanding. Smoking-cessation apps usually communicate in an anonymous ‘app voice’, with little presence of professional or other identified voices. Because of the fast-and-frugal communication, truthfulness is problematic. Messages in the ‘quantification’ modules may be read as deceitfully accurate. The app voice frames smoking as a useless, damaging habit indicative of weakness of will, in a ‘cold-turkey’ frame of individual mind-over-body heroism. Thus apps contribute to a stigmatization of smokers and culpabilization of relapses. The potential to support user autonomy through diverse meaningful voices and personalized communication remains yet unused.

Full text

1
A touching app voice
Thinking about ethics of persuasive technology through an
analysis of mobile smoking-cessation apps
Cosima Rughiniş
Răzvan Rughiniş
Ştefania Matei
University of Bucharest
cosima.rughinis@sas.unibuc.ro
University Politehnica of Bucharest
razvan.rughinis@cs.pub.ro
University of Bucharest
stefania.matei@sas.unibuc.ro
The article in published in: Ethics and Information Technology, 2015, 17:4, pp. 295-309
The final version is available at http://link.springer.com/article/10.1007%2Fs10676-016-9385-1
Abstract
We study smoking-cessation apps in order to formulate a framework for ethical evaluation,
analyzing apps as ‘medium’, ‘market’, and ‘genre’. We center on the value of user autonomy
through truthfulness and self-understanding. Smoking-cessation apps usually communicate
in an anonymous ‘app voice’, with little presence of professional or other identified voices.
Because of the fast-and-frugal communication, truthfulness is problematic. Messages in the
‘quantification’ modules may be read as deceitfully accurate. The app voice frames smoking
as a useless, damaging habit indicative of weakness of will, in a ‘cold-turkey’ frame of
individual mind-over-body heroism. Thus apps contribute to a stigmatization of smokers and
culpabilization of relapses. The potential to support user autonomy through diverse
meaningful voices and personalized communication remains yet unused.
Keywords
Persuasive technology, ethics, mobile apps, smoking-cessation, app voice, autonomy
1 Introduction: technology and ethical inquiry
In this paper we aim to highlight ethical issues in order to guide design and use of smoking
cessation apps, starting from a review of current Android solutions.
Smoking cessation apps are an instance of persuasive technology (B. Fogg, 1998)
developed in the growing field of mobile health solutions, aiming to support users in their
decision of quitting. Digital persuasive technology refers to “digital products designed to
change what we think and do” (B. J. Fogg, 2009, p. 1). Persuasive technology may be
designed in pursuit of specific beliefs or actions, or may aim to support users’ decision
making at a broader level, by enhancing awareness, motivation, self-efficacy, memory etc.
Unlike other monitoring and tracking apps (Lupton, 2013), especially those part of the
Quantified Self movement (Nafus & Sherman, 2014), smoking cessation programs put less
emphasis on generating data on user behaviors, and more on issuing eloquent voices to
guide users in this difficult stage of life. These voices sometimes speak in numbers, through
what we have called the ‘quantification modules’ – but these numbers are largely

2
projections, rather than measurements. They may also speak in pieces or advice, as well as
through visual means. Mobile smoking cessation apps literally talk to their users, and thus
raise the question on how are we to evaluate this conversation from an ethical perspective.
We use the concept of voice following its career in the social sciences after the original
formulation of M. Bakhtin (1984). A voice refers to “a speaking subject's perspective,
conceptual horizon, intention, and world view” (Wertsch, 1993, p. 51) – for any given
utterance, in any medium - aural or written. Subjects often speak with multiple voices,
referring to and addressing others’ speeches. Their voices are shaped by what Bakhtin
termed social languages: “social dialects, characteristic group behavior, professional jargons,
generic languages, languages of generations and age groups, tendentious languages,
languages of the authorities of various circles and of passing fashions, languages that serve
the specific sociopolitical purposes of the day" (M. M. Bakhtin, 1981, p. 262, apud Wertsch,
1993, p. 58). Individual subjects speak through social languages, appropriating their
vocabularies and styles into distinctive voices. Research in medical humanities has relied on
the concept of voice to identify multiple ways of interpreting and speaking about one’s
condition, grounded in different experiences, concerns and vocabularies – see for example
Puustinen’s analysis of shifting physician’s voices (Puustinen, 2000), Engestrom classification
of voices in medical encounters (Engestrom, 1995), or Lewis’ discussion of the multiplicity of
voices interpreting depression in Chekhov (Lewis, 2006). With the rise of persuasive
software, artificial voices become increasingly common – ventriloquizing medical and other
expert languages to fuel users’ decision making processes with discursive resources. We pay
attention to apps’ voices in order to identify ethical issues relevant for design.
Researchers in engineering ethics have gradually enlarged the scope of ethical inquiry
beyond safety concerns, for example in the Value Sensitive Design (VDS) movement
(Friedman, Kahn, & Borning, 2008; Friedman, 1996). VDS rests on the theoretical insight that
technology is not a morally neutral instrument: ‘values emerge from the tools that we build
and how we choose to use them’ (Friedman, 1996). Lessig’s controversial assertion that
‘code is law’ (Lessig, 2000) has furthered discussions on the moral valence of technology.
While the morality of tools is often implicit in designers’ projects, the Persuasive
Technology (PT) movement (B. Fogg, 1998) or Design with Intent (DwI) (Lockton, Harrinson,
& Stanton, 2008) have brought technologically-enabled (or disabled) moral action to the
forefront of design.
1.1 Framing reality through technological lenses
Technologies shape the choice situations in which people find themselves, re-creating
them as moral subjects. Designers are taking on the job of a choice architect (Thaler,
Sunstein, & Balz, 2010) – influencing ways in which users define their situations and act upon
them, aiming to routinize some actions and take them out of decisional spotlight, and
proposing forms of feedback to keep actions in line with desired goals. This is
At the same time, technological framing reverberates through multiple aspects of a
person’s life, beyond stated objectives. Verbeek relies on Don Ihde’s post-phenomenological
philosophy of technology to argue for the mediating role of technology in the reciprocal
constitution of the acting subjects and their world (Verbeek, 2008), thus writing about the
‘technological mediation of morality’. Verbeek examines how technology makes possible
interpretations of the world (hermeneutic mediation) and ways of action (pragmatic

3
mediation) (Verbeek, 2006: 3) that influence our engagement with the world at large. To
illustrate:
“The FoodPhone, to stick to the example, may persuade its users to develop more healthy eating
habits.(…) Beside having the desired effect of stimulating a more healthy eating pattern, the
FoodPhone can e.g. make eating something stressful; it can stimulate humans to interpret their
health exclusively in terms of their eating pattern while neglecting the importance of other factors
like having enough exercise; and taking pictures of all food consumed will definitely organize
social relations at the table.” (Verbeek, 2008: 4)
These observations bring into question the moral acceptability and risks of framing
actions through the lenses of such technologies. Observing oneself and acting through
technologies that privilege a focus on the value of health, and measurement of health-
defining indicators may lead to medicalization (Conrad, 1992) or healthism (Lupton, 2013).
When these technologies create a sharp, moralized distinction between the healthy and the
ill, one may also witness stigmatization of the ‘unhealthy’ selves or others (Courtwright,
2013; Guttman & Salmon, 2004) .
1.2 Truthfulness of technological voices
Technologies frame situations of choice through multiple forms of communication –
including signals, reminders, notes to self, but also messages initiated within the
technological product or conversations with others enabled by the product. This raises the
issue of truthfulness in communication.
Some values, such as safety and privacy, are shared by many types of technologies, be
they explicitly persuasive or not. Enhancing user autonomy through truthfulness and honesty
in communicated messages are distinctively central values in PT discussions – deriving from
an ambivalent understanding of persuasion, between ‘manipulation’ and ‘convincing’.
As Spahn observes, truthfulness refers both to messages that are formulated to the user,
and to the feedback that the application gives about the user (Spahn, 2012: 639-640). Still,
some bias may considered contextually acceptable. Users judge truthfulness in the context
of a specific genre that may allow various forms of reasonable re-elaborations of what would
count as ‘factual truth’. For example, accuracy of self-presentation in the context of online
dating profiles may be better judged as the truthfulness of a promise (Ellison, Hancock, &
Toma, 2011), than that of a medical or judicial record. Apps may also offer clues to guide a
specific interpretation of truthfulness, through a gamified appearance, inviting a playful
relativity, or through appeals to scientific objectivity, leading to expectations of rigor.
Still, assessing autonomy through transparency of intent and truthfulness leaves open the
issue of unforeseen consequences (Atkinson, 2006), or the ‘not reasonably predictable’
(Berdichevsky & Neuenschwander, 1999) outcomes.
Various authors discuss the morality of PT also by inquiring whether the intentions of the
persuader are transparent, and whether they are shared by users (Atkinson, 2006;
Berdichevsky & Neuenschwander, 1999) – in other words, whether PT is an instance of self-
persuasion or other-persuasion:
“With regard to autonomy, it is furthermore important to distinguish whether the user persuades
himself with the PT, or whether the PT is used to persuade him by someone else. Cases of self-
persuasion are most likely cases, in which the user already shares the value in question and uses
the PT only to overcome a ‘weakness of the will’” (Spahn, 2012: 645).

4
2 Mobile apps for health persuasion
Smartphones have become an increasingly common device for health persuasion (Klasnja &
Pratt, 2012; Mosa, Yoo, & Sheets, 2012). Mobile phones have been already used in smoking-
cessation programs, in a wide range of SMS-based interventions (Meier, Tackett, &
Wagener, 2013). Smoking-cessation applications (apps) for smartphones are now available in
ever larger varieties on software markets for various operating systems (Abroms et al., 2011;
Abroms et al., 2013).
Apps have also become an instrument for large scale public health interventions: the
European Commission has initiated the ExSmokers campaign, developed by Saatchi &
Saatchi, that uses two mobile apps - the ExSmokers iCoach and the FCB iCoach - besides
visual messages, public events, and social media communication. The campaign received two
Euro Effie awards, as well as an EACA Care award (2013) and the European Lung Foundation
Award (2013).
Randomized studies offer some indication of effectiveness for SMS-based interventions in
smoking cessation (Meier et al., 2013). As regards smoking cessation apps, a content-analysis
review of features for 47 iPhone apps (L. Abroms et al., 2011) and a follow up analysis for 47
iPhone and 51 Android apps (L. C. Abroms et al., 2013) concluded that, as a rule, they do not
adhere to established clinical guidelines for smoking cessation; for example, they did not
recommend approved medication or counseling. Similar results obtain in a content analysis
of content analysis of 225 Android smoking cessation apps (Hoeppner et al., 2015): apps
cover on average only 2-3 of the 5 recommended clinical practice steps and provide few
opportunities for tailoring. A study looking at smoking cessation apps in light of motivation
theory concluded that, as a rule, apps are limited in complexity and are mostly focused on
extrinsic motivators (Choi, Noh, & Park, 2014). There is very little information, to date,
concerning effectiveness for smoking-cessation apps. On study of a specific app concluded
that it might be helpful for at least some of its users, as almost 19% of them recorded
abstinence periods of 28 days or longer (Ubhi, Michie, Kotz, Wong, & West, 2015). It is
beyond our scope to assess effectiveness for individual apps or specific collections, but we
notice that this absence of information is also an ethical concern.
Smoking-cessation apps purport to guide users through a difficult personal
transformation, which has both short- and long-term implications for self and others.
Attempting to renounce smoking involves a transformation in one’s lifestyle, and, whether
successful or unsuccessful, it impacts the individual’s health and emotional wellbeing.
Moreover, attempts to quit influence relationships with significant others, who may
welcome or disapprove of the attempt: ex-smokers lose some opportunities of cigarette-
focused sociability, and gradually acquire an interest in smoking cessation or abstinence for
those around them. Therefore, given the increasing acceptance of mobile apps for health
communication among the general public, and their use in public health campaigns, it is
important to highlight specific ethical issues.
2.1 Typical app modules
We analyze 27 free smoking-cessation apps for Android available in the Google Play market,
ranging from the most popular (more than 4.000 user reviews at the time of app selection)
to the more recent or less popular. As of October 15, 2013, when we started our analysis, a
total of 6 apps had more than 1000 user reviews, and another 7 had more than 300 user
reviews (Figure 1). In what follows, we refer to specific apps by means of their rank in the

5
top of user reviews, as presented in Figure 1. We do not aim to quantify the relative
frequency of features, but to present a repertoire of ethical considerations that can inform
users’ choice, engineering education, design work, public health interventions, and
counselors’ recommendations. Starting from ethical analyses of ubiquitous healthcare
(Brown & Adams, 2007) and public health communications (Guttman & Salmon, 2004), we
evaluate the relevance of typical ethical concerns and we identify additional, specific moral
risks.
Figure 1. Smoking-cessation applications included in analysis
Source: Authors’ analysis of data on Google Play, 15 October 2013.
Apps include one or several typical modules for supporting smokers in giving up or
diminishing smoking. Most applications are dedicated to persons who have just given up
smoking, and confront the initial, difficult cigarette-free hours and days. Some apps also
address people who aim to control their smoking and decrease it gradually, or people who
contemplate giving up smoking but they have not made any decision yet.
The most common module for supporting recent ex-smokers consists in a ‘calculator’ –
that is, a device that estimates various parameters of users’ health state and risks, as well as
savings through not smoking, and displays them as a function of time. Users are required to
input initial information such as: the number of cigarettes smoked per day, duration of
smoking, number of cigarettes per pack, price per pack. This data is used to customize app
messages concerning the benefits of cessation: immediate improvements in one’s physical
condition, decreased risks, saved money, and saved time through not smoking. Figure 2
illustrates three different calculator displays for ‘Quit Smoking – Quit Now!’ (app 3):
statistics, progress bars for health improvements, and ‘achievements’.

6
Figure 2. Quit Smoking – Quit Now! Calculator module: statistics, health improvements, achievements
(a) Quit Smoking – Quit Now!
(b) Quit Smoking – Quit Now!
(c) Quit Smoking – Quit Now!
Source: Screenshots published on Google Play.
Photo courtesy of Fewlaps [http://fewlaps.com]
Calculator modules require users to report their ‘relapses’ (cigarettes smoked after the
quit date), with several types of reactions. Most calculators reset all progress indicators,
considering that a ‘relapse’ cigarette turns back the user to the status of ‘smoker’, with its
associated risks (Figure 3a, app 1). Less frequently, apps count the number of ‘relapse
cigarettes’ as a distinct indicator, without any modification in the gradual improvement of
health indicators associated with the status of ex-smoker. Apps can also attempt to alleviate
the urge to smoke through temporization, asking users to push a button whenever they
want to smoke, then asking whether they can wait, and advising abstinence (Figure 3b, app
1, and c, app 3).
Figure 3. Reactions to reported cigarettes
(a) Quit Smoking - Azati
(b) Quit Smoking - Azati
(c) Quit Smoking - Azati
Source: Screenshots from authors’ phones.
Photo courtesy of Azati [www.azati.com]

7
A second typical app device consists in ‘coach’ modules dedicated to the management of
cravings and relapses. Apps may offer textual tips for such difficult moments. A typical
collection includes circa 30 pieces of written advice, expressed in the app anonymous voice,
of one to several sentences in length, that may be presented on users’ request (upon
pushing a ‘panic’ button, for example), as random messages, or on a dedicated static page.
Some apps also include internal forums, in which users share experiences and advice. Apps
also encourage users to tell about their efforts on social media – for example, by publishing
their medals on Facebook.
The most popular apps focus on the calculator module; some do not offer any tips for
situations of cravings, while others have a rather limited collection of tips. Other apps focus
on advice with little display of statistics, including the European Council solutions (apps 9
and 14).
2.2 Apps as situations of choice
Apps shape situations of choice for designers and users, thus mediating moral action and
moral subjectivity. We propose to study these situations at four levels of generality: (1) the
app medium, (2) the market, (3) the genre, and (4) the individual piece of software. For each
type we discuss specific ethical issues, while examining how apps are currently working, and
also how they could work, given the potentialities of the medium.
Smartphone apps represent, first and foremost, a medium of communication, enabling
the formulation and dissemination of multiple messages – relying on different sensorial
modalities, various degrees of gamification, more or less interactivity and personalization,
individual or community relationships etc. Individual apps can vary greatly; still, it is possible
to examine the ethical specificities of the medium itself.
Secondly, users do not typically encounter a single app, but a multitude of apps available
at the same time and in similar access conditions. Therefore, it is important to take into
account the variability of the solutions on offer, and the interactional features of the
platform that mediates users’ contact with apps – that is, the app market. In this paper we
examine free Android apps available on the Google Play platform.
Thirdly, app components can be classified in genres according to their communication and
interaction patterns. In this study we focus on two smoking-cessation genres: the
‘calculator’, which presents users with quantitative indicators of health and finance, and the
‘coach’, which offers advice for difficult moments of craving a cigarette. Any individual app
may combine several genres; most smoking-cessation apps that we have analyzed combine
calculators, coaches, and some elements of visual persuasion through anti-smoking posters.
There are other genres as well in smoking-cessation apps, based on hypnosis, subliminal
messages, smoking simulation, presentation of scary images, audio and video coaching. They
appear in dedicated products rather than integrated with calculators and coaches, and we
did not include them in this analysis because they rely on different theoretical assumptions
and communication strategies; also, they are significantly less popular (as indicated by the
number of user reviews).
Last but not least, the most specific objects of ethical evaluation are individual apps as
pieces of software which interact directly with users, shaping their daily situations of choice.

8
2.3 App voices
Apps introduce in users’ life new voices, communicating through various social languages
(M. M. Bakhtin, 1981, apud Wertsch, 1993, p. 58). Calculator modules communicate in a
voice that relies heavily on quantification and on scientific legitimation from medical
research, employing a vocabulary of health risks and benefits. This voice is also present in
textual advice in the coach module – combined with much more colloquial and moralizing
injunctions (see examples in section 3.3.5). Still, as previous research has pointed out (L. C.
Abroms et al., 2013; L. Abroms et al., 2011), this quasi-medical voice stops short of including
clinical guidelines for smoking cessation – it appears rather as a distinctive ‘app voice’ that
ventriloquizes numbers from research as a resource to deliver quantitative encouragement.
Apps may also bring forward peers’ voices, typically in pseudonymous form, in the
internal app forums. Other than that, the apps communicate in their own anonymous voice
– it is the app speaking to the user, throughout its operation.
Other voices that could tell about their experiences or share advice are notably missing –
such as known peers, writers, doctors, humorists, psychologists, or any other person
communicating from within a personal standpoint. The EC app FCB iCoach is an exception,
including FCB players’ and members’ voices as personally assumed advice.
App voices are also confined to textual messages, usually ‘fast and frugal’ – with little
reliance on actual human voices or other aural means of persuasion, such as music
(‘hypnosis’ is the app genre which relies on images and sound). Visual communication
employs mostly poster-type images - either inspirational or scary.
3 Ethical issues for smoking-cessation apps
We propose a frame of reference for ethical assessment of apps inquiring into all four
situations of choice: medium, market, genre and piece of software.

9
Table 1 presents an overview.
Besides moral issues such as privacy, truthfulness, autonomy and equity, we have also
included effectiveness (impact) as an ethical concern at the market, genre and individual
level, given that smoking-cessation apps operate under the explicit promise to support users
in a difficult transformation with important consequences for their and significant others’
wellbeing. Whether or not apps actually deliver this support is of clear ethical import. We
can define effectiveness most clearly for individual apps: at this level, it could be evaluated
through various research designs, experimental or experiential. At the level of app genre, it
would be possible to estimate the aggregated or typical impact, through designs such as
meta-analyses of experimental evaluations, or other styles of review.

10
Table 1. A framework for ethical assessment of smoking-cessation apps
Apps as…
Ethical concerns
1. Medium
 Privacy:
o Use of personal information
o Security vulnerabilities
 Truthfulness: communication based on ‘fast and frugal’
messages in ‘app voice’
 Equity and social gaps
2. Market
 Effectiveness (reliance on medical knowledge and expertise,
medical authorization)
 Free choice: costs
 Informed choice:
o Visible credentialing or authorization information
o Transparency of commercial interests
o Quality of labeling information
o Quality of feed-back from users
 Equity and social gaps
3. Communication
genre
 Encouraging autonomy:
o Users’ control and app interactivity
o Self-persuasion through personalization
o Qualities of information: diversity and reach of hosted
voices, truthfulness, depth of understanding, moral
values
o Risk of stigmatization
 Effectiveness
o Adherence to medical guidelines
o Theory-informed design
o Evidence from evaluation of aggregated or typical
impact (review, meta-analysis)
4. Piece of software
 All concerns listed above
 Effectiveness
o Adherence to medical guidelines
o Theory-informed design
o Evidence from case-study evaluation of impact
(experimental, experiential)
At the market level, the relevant considerations for effectiveness is whether any process
of medical authorization is in place, whether scientific knowledge and expertise (medical,
psychological, sociological) are present in the diversity of solutions, and the extent to which
informed choice concerning expected effectiveness is possible (that is, whether evidence
about app usefulness to past and present users, and its reliance on scientific knowledge and
professional expertise is easily available for would-be users).
In what follows we illustrate discussions of ethical issues at medium, market and genre
levels, for smoking cessation apps. We do not analyze in detail individual apps, given that
higher-level issues are shared with individual app level issues, and the in-depth case study of
specific apps is beyond our scope of inquiry.

11
3.1 Medium-specific ethical issues
The most common concerns regarding the medium of mobile apps refer to privacy and the
use of personal information. In order to install and use an app, users consent to various
levels of access to their information and smartphone services. This opens to a more or less
extent the possibility of misuse of personal information, by companies or persons who
manage app users’ data or by mobile communication carriers, as well as the possibility of
hackers’ acquiring unwanted access through security breaches (Meier et al., 2013).
Mobile apps are a medium that favors ubiquitous access to short, ‘fast and frugal’
messages – usually formulated in an anonymous ‘app voice’, or in pseudonymous peer
voices on forums (see section 2.3). This raises potential concerns for truthfulness versus
deceitfulness, and reasonable versus manipulative approximation: there is often no time and
space for qualifications of certainty and the details needed in medical communication.
3.2 Market-specific ethical issues
As observable in its name, the Google Play app market does not claim any medical authority
whatsoever. While it hosts many apps purporting to address health issues, it also hosts apps
for entertainment, lifestyle, office work, social networking etc. People search and find
smoking-cessation apps as elements of this heterogeneous collection, and it is likely that
they are not entrusting them with the legitimacy typical for professional medical settings. It
is also likely that health-related apps are used for a variety of tasks, including entertainment,
self-presentation on social networks, or tech experimentation. The social situation in which
apps become available is relevant for contextualizing the weight of ethical considerations in
relation to other issues (of technical, aesthetical, social, or entertainment value).
3.2.1 Effectiveness: medical presence and authorization
There is, to date, no process of authorization for health-related apps according to proven
impact, adherence to medical procedures or other criteria. Conditions for their public
release are similar with those for entertainment, office, or social networking apps. A
question rises, then, whether medical expertise is represented in the variety of apps on the
market. Are users able to choose a medically-informed intervention, if they want one?
Our analysis corroborates conclusions in previous research (L. C. Abroms et al., 2013; L.
Abroms et al., 2011): there is a general low compliance with clinical guidelines for smoking
cessation. Similar conclusions apply to apps for panic disorders (Van Singer, Chatton, &
Khazaal, 2015), alcohol-control (Cohn, Hunter-Reel, Hagman, & Mitchell, 2011), diabetes
self-management (Breland, Yeh, & Yu, 2013), weight loss and fitness (Cowan et al., 2013;
Pagoto, Schneider, Jojic, DeBiasse, & Mann, 2013). The European Council apps are the only
ones in our collection that recommend use of approved medication, appeal to counseling,
and offer to connect the user with a Helpline. ‘Cold turkey’ is the dominant approach of
other apps, either explicitly (‘Nicotine replacement therapy might not be your best option…
Cold turkey is the most logical option!’- app 12) or implicitly - see the qualitative analysis of
app messages in Rughiniș, Matei, & Rughiniș, 2014 and Matei, Rughiniș, & Rughiniș, 2014.
Also, we found little psychological insight in most instances of textual advice – with the
notable exception of the European Council apps.

12
3.2.2 Free and informed choice
An important ethical feature of markets is freedom of choice. This depends on costs,
information, and access, among others.
As a rule, most apps offer a free version, and costs per individual apps are relatively low
(in the range of USD 1-7). A second dimension of free choice consists in access to
information about available options.
Before installing an app from Google Play market, users have access to several types of
information:
a) Name of the producer, date of publication, and producers’ description that usually
lists: the main features, a series of several screenshots illustrating functionalities
and aesthetics, and, occasionally, a video presentation;
b) Price;
c) Users’ comments and ratings; users are, as a rule, anonymous or writing under
pseudonyms; therefore, credibility is at stake;
d) Quantitative indicators of popularity: user reviews (illustrated in Figure 1); ratings
on Google Plus; a range for the number of installs (eg: 100,000 – 500,000 for app
8).
As mentioned before, smoking-cessation apps have not been comprehensively evaluated
as regards their effectiveness. Still, apps do occasionally present themselves with reference
to medical criteria of performance or credibility. For example, app 8 mentions that it is
‘Featured at Healthline.com in their list of top quit smoking apps of 2013!’. While this is
accurate, Healthline.com also includes a note specifying that ‘Healthline Networks does not
endorse or warrant for fitness of purpose any of these applications. These apps have not
been evaluated for medical accuracy by Healthline Networks and unless otherwise indicated,
haven’t been approved by the Federal Drug Administration (FDA)’. App 3 mentions that
‘Also, QuitNow! will provide you with W.H.O.-based (UN's World Health Organization)
indicators on your health improvement process (…)’. A question rises, then, to the
appropriateness of reference to medical authority in app claims of effectiveness.
Another issue regards the transparency of commercial interests. Many of the free apps
include advertisements, a practice that is expected by users. Still, app 11 offers ‘E-cig
coupons’ as part of smoking cessation strategy, without marking this as a commercial
interest. The offer includes the following encouragement: ‘Having an electronic cigarette can
help you during those tough times when you really want to smoke. (…) Just take a few pulls
from the electronic cigarette to hold yourself off until the next scheduled smoke time!’ Users
are thus exposed to unmarked advertisement.
A content analysis of Android smoking cessation apps (Hoeppner et al., 2015) concludes
that user reviews and the number of downloads correlate positively with the degree of
customization and the number of clinically recommended steps covered within the app;
thus, it follows that users can rely with some predictive success on common market
indicators in their selection of individual apps. This is important especially given the results
of a study on users of a smoking cessation app, stating that about three quarters of those
who have used other health related apps did not check the credibility of the app publishers
before downloading it (BinDhim, McGeechan, & Trevena, 2014). Thus market metrics such as
the number of user-awarded stars and the number of downloads remain important criteria
of choice.

13
3.2.3 Social gaps
By selectively targeting sections of the population, public health interventions risk increasing
social gaps (Brown & Adams, 2007; Guttman & Salmon, 2004). This can become a vicious
circle dynamic if the intervention relies on stigmatization and imputation of individual
responsibility, while at the same time failing to reach social categories that are already
lacking in financial, human and social capital: in such a situation, the intervention may leave
them even more disempowered in confronting their health vulnerabilities.
There is a case to be made that the current market for calculator and coach apps targets
mostly smokers with higher human capital. For example, a study of the users of one specific
smoking-cessation app concludes that “compared with smokers trying to quit in England,
they had higher consumption, and were younger, more likely to be female, and had a non-
manual rather than manual occupation” (Ubhi et al., 2015). This is, on the one hand, a
property of the medium: while mobile phones are virtually universally accessible,
smartphones have not yet reached the same degree of penetration. Moreover, calculator
modules do require a certain degree of statistical literacy and medical vocabulary, while
coach modules rely heavily on written text communication; in particular, the European
Commission’s solution ExSmokers iCoach delivers relatively lengthy pieces of written
information (covering one or more full screens). This may account for its relatively low
popularity, despite being part of a massive intervention program (ranking 9 in Figure 1).
While the market is dominantly English speaking, the most popular apps offer versions in
other languages (such as apps 1, 2, 3 in Figure 1); also, the ExSmokers iCoach (app 9) is
available in a variety of languages. Therefore, language is becoming less and less a barrier, at
least in theory. In practice, users still need to search for apps on Google play in English,
install apps in English, and then select other language from the Settings menu. Apps that
offer multiple languages cannot be actually found on the market by searching with keywords
in French (‘fumer’), Romanian (‘fumat’), German (‘rauchen’), etc. This is a clear access barrier
for people with less formal education, in non-English speaking countries.
3.3 Genre-specific ethical issues for ‘calculator’ and ‘coach’ apps
One of the key moral considerations in persuasive technology and in health interventions
refers to personal autonomy: to what extent the intervention is enabling the person to make
free and informed decisions, to gain understanding of and control over her life, to improve
her wellbeing according to her own moral worldview? Typical moral risks consist in: a)
paternalism, when individual preferences and decisions are ignored or overridden in the
name of wellbeing; b) dependence, when individuals are kept in a subordinate position by
not receiving the knowledge, information, decision-making power, or other resources
required to make their own choices; c) problematic agency – when individuals experience a
change in their personalities through intervention, and it becomes unclear for them who is
the source of agency in their own actions (Brown & Adams, 2007); d) culpability – when
individuals are held responsible for success or failure of actions which are only partly under
their control, without being offered the required support; culpability may lead to e)
stigmatization, a negative moral portrayal of persons that decreases their possibility to
engage in social interaction and to legitimately sustain their points of view (Guttman &
Salmon, 2004).

14
3.3.1 Dimensions of autonomy
We can examine apps in relation to several dimensions of autonomy (Figure 4). The most
concrete sign refers to the actual involvement of individuals in the app-intervention: what
degree of control do they have? The second layer refers to personalization - the extent to
which users can communicate their circumstances and preferences in order to customize
treatment, allowing them to direct their own behavior change as a form of self-persuasion
(Spahn, 2012). The third layer points to the degree that apps enhance the information base
on which individuals ground their decisions. The fourth and the fifth layers refer to enhanced
self-understanding and self-direction; this distinction is analytical rather than psychological.
Self-understanding refers to the messages that users receive concerning their agency, the
forces that shape their actions, the imputation of responsibility for various outcomes. At the
highest level, moral deliberation refers to the moral values highlighted by the app, and
instantiated in the actions and lifestyle that it recommends.
Figure 4. Layers of app-supported autonomy
Source: Authors’ analysis.
3.3.2 Involvement and personalization
Apps have a high potential for interactivity and user control of interaction. In most cases,
users choose when they want to see app-related information and advice.
Interactivity is a resource for personalization: apps can rely on specific information
provided by individual users in order to tailor communication to their situation. Calculator
modules employ data about users’ smoking patters, although, as we discuss in section 3.3.3
dedesubt, their level of individualization as regards health is make-believe rather than
substantive. Apps may also position users on an evolution trajectory, passing through
various stages depending on their smoking behavior and willingness to quit; European
Council apps customize advice to users’ self-declared stage. As regards the content of tips
and advice, as a rule, they are not customized to users’ preferences. Although it would be
possible to enable people to rate advice and adjust content to their preferences, such
features have not yet been implemented. Customization would also require a much larger

15
collection of advice and motivational messages, in order to enable meaningful
personalization on a variety of dimensions (such as positive versus fear-based motivators,
humorous vs. inspirational, text vs. audio or video, short vs. long, focused on the body, mind
or socio-material surroundings, etc).
Personalization is rarely pursued, even in forms of addressing the users: apps do not
employ their name, and often formulate tips as general rather than specific advice. A more
recent review also concludes that tailoring is used “sparingly” - on average for only about
one of the five As (“ask”, “advise”, “assess”, “assist” and “arrange follow-up”) recommended
by clinical practice guidelines (Hoeppner et al., 2015).
While users can customize to some extent their app-based smoking-cessation program at
the market level, when selecting a genre and a particular solution, there is little space for
personalization and increasing self-persuasion in interactions with individual apps.
3.3.3 Information and truthfulness
Apps offer a wide array of information concerning smoking risks and their evolution after
cessation, as well as information concerning nicotine addiction, withdrawal symptoms, and
strategies for managing cravings. While smoking-cessation app users have considerable
opportunities to broaden their knowledge basis as regards nicotine dependence,
truthfulness remains problematic.
Calculator modules require users to input summary information on their smoking
patterns and offer, in exchange, quasi-personalized estimates of health improvement and
risk reduction. These indicators are offered with high numerical precision: for example, users
are informed about the exact number of days, hours, minutes and seconds in which their risk
of heart disease will be reduced by half (Figure 2). Given that apps have no access to
individual level information about health status, this is at best an estimate at a highly
aggregate level, and the decimals and seconds on display are of aesthetic rather than
medical relevance. Users are offered personalized health information of high numerical
precision that bear little actual relation to their current individual situation. This raises the
moral issue of truthfulness and its reverse, deceitful persuasion (Guttman & Salmon, 2004).
On the one hand, it may be argued that users understand that app communication relies on
a rhetorical convention of personalization, while its content is essentially impersonal. Still,
such an assumption cannot hold for all possible users. Thus, even if only a small proportion
of users start to believe that those risks actually describe their individual health condition,
persuasion turns into deceit. Several specific recommendations can be formulated here, for
designers interested in enhancing app truthfulness. Firstly, apps should inform users about
the nature and sources of the information that is presented as being ‘about them’. Secondly,
apps should not translate aggregated estimates of average individual chances and risks into
a numerically precise formulation – rather, they should consistently communicate that
numbers include a degree of approximation. Thirdly, users should be reminded that app
messages refer not to “you” but, rather, to “people like you” – where this likeness is
determined on the basis of input information. Users may also be invited to create avatars,
for example, in order to create a slight distance between their real, flesh-and-blood persons
and the targets of app communication; they would thus become guides and observers of
their half-fictive-half-real avatars influenced by aggregate risks of smoking and various
probabilities of withdrawal symptoms.

16
Apps also engage in other types of persuasive manipulation. For example, app 1 offers to
assess users’ level of nicotine dependence through a questionnaire. After answering
questions in all possible combinations, it becomes clear that the app only offers two
diagnostics: ‘low’ dependence, and ‘moderate’ dependence. There is no inbuilt possibility to
be diagnosed with ‘high’ dependence. The ‘low’ diagnosis brings the following clarification:
‘Mild nicotine dependence. It is mostly psychological. You can handle it. Stay firm and do not
give up’. The ‘moderate’ diagnosis explains: ‘Moderate nicotine dependence. You should
make an effort to overcome both psychological and nicotine dependence’. The absence of a
‘high’ level may be deemed encouraging, but at the same time it is deceitful for those
persons that are strongly addicted; the ‘moderate’ message puts the burden of resistance on
the individual (‘you should make an effort…’), ignoring the availability of counseling and of
pharmacotherapy, including nicotine replacement, that can alleviate withdrawal symptoms.
This strategy of persuasion may thus lead to the recrimination of users for an eventual
failure to quit, and to a misunderstanding of their condition – in effect undermining their
scope of action.
It is also noteworthy that apps do not use established psychological scales which are
available online to describe nicotine dependence (Etter, Le Houezec, & Perneger, 2003;
Wellman et al., 2011), motives for smoking (Smith et al., 2010) or withdrawal symptoms
(Bolt et al., 2009; Welsch et al., 1999). Apps could make better use of available scientific
tools for identifying and measuring smoking dependence and withdrawal experiences. This
brings us back to the abovementioned absence of psychological voices, be it of quantitative
or experiential persuasion.
Last but not least, our discussion of truthfulness risks to omit the fact that some users
may actually prefer to be scared rather than accurately informed – that is, they may prefer
to be a little pushed into believing whatever it takes so that they can quit. People often
engage in self-deception, and they may also deliberately choose a scaring app over a
scientifically sound app. Designers are left to arbitrate the conflicting requirements of
truthfulness, playfulness, and persuasive (self-)deception, in a rhetorical situation that is
under-specified. There is no explicit promise for scientific soundness on Google Play – other
than avoiding malware and fraud, and apps’ occasional, explicit or implicit claims of
evidence-based advice. The clearest recommendation that one may derive from this
rhetorical situation is that, whatever balance designers choose for their app between
conflicting communication styles (factual, playful, threatening or another), they should
communicate it effectively to users.
3.3.4 Self-understanding: ‘mind over body’ and culpability
Smoking cessation engages people in a process of self-understanding, as they are confronted
with issues of physical dependence on nicotine and of psychological dependence on smoking
as a coping strategy and a social habit. The ethical issue is, then, what ‘theory of self’ do
smoking apps promote?
The dominant model of most calculator and coach apps relies on a ‘cold turkey’ strategy,
ignoring pharmacotherapy and medical counseling and advocating individual self-control as
the central resource for quitting . The implicit model of human action is that of ‘mind over
body’, glorifying individual control over bodily reactions, with little if any external support.
This view is often shared by app users and it is consolidated through app forum talk (Figure 5
c).

17
Peer support is the only social resource commonly enrolled, since users are encouraged to
take part in dedicated forums and to share their challenges and victories on Facebook,
Twitter etc. This stress on the importance of community support counterbalances to some
extent the individualistic bent of the app voice.
A consequence of the individualist and mentalist representation of action is that relapses
are often framed as a personal failure, as a reset of the smoking-cessation process that
should be imputed to the individual. There is variability in the treatment of relapses, with
some apps allowing for occasional cigarettes as part of cessation (‘Have you tried so many
times already? Your chance of succeeding the next time is greater. You’re more aware of the
pitfalls’, app 9), and micromanaging cravings by encouraging users to report relapses and to
keep on going. The relevant ethical issue concerns the definition of failure and the
imputation of responsibility: how can individuals be supported by encouraging self-efficacy
and avoiding moral recrimination for instances of relapse?
3.3.5 Self-understanding and stigmatization
A widely discussed issue in relation to anti-smoking public health campaign refers to
whether the negative moral, cognitive, and aesthetic portrayal of smoking results in a
stigmatization of smokers – and, if yes, whether this stigmatization is morally acceptable.
From a contractualist perspective, acceptability of stigmatization as an incentive for behavior
change revolves on the extent to which it leads to cross-situational social isolation of the
targeted persons, preventing their participation in decisions that shape their lives
(Courtwright, 2013). It becomes an empirical task for ethical assessment to determine
whether the redefinition of smoking in a specific social setting has led to a trans-situational
stigmatization of smokers, diminishing their social standing, or has constrained strictly their
smoking behaviors, without affecting their voice in various relationships and communities.
For example, in authors’ social surroundings smoking does not seem to function as a stigma,
at least in our interpretation – but this may well be different in other parts of the social
world.
If smoking turns into a ‘morally defective’ behavior in a social group, this may also affect
ex-smokers, in at least two ways:
a) If they view smoking as morally flawed, they may end up devaluing their own past
self;
b) Ex-smokers may gradually devalue and / or avoid interaction with some of their
significant others (colleagues, friends, family) who are still smoking.
Apps rely occasionally on the resource of stigmatizing discourses to uphold ex-smokers
through the cessation period. There is considerable variability: some apps do not include
such negative messages, or very rarely. Some examples are:
a) Defining smokers as wrongdoers: ‘Quitting smoking means: You will no longer
hurt yourself and others’ (app 4); ‘Young women who are pregnant and who
smoke put their fetuses at increased risk for decreased birth weight, premature
birth, and perinatal mortality’ (app 12);
b) The aesthetization of smoking as disgusting: ‘You’ve taken the first steps towards
busting this disgusting habit’ (app 11); ‘Smoking is a disgusting and stinking habit’
(app 4); the olfactory sense plays a powerful role in the process of Othering
smokers (Gavriluță, 2002);

18
c) A focus on bodily disfigurement: ‘Are you worried about your sex appeal? Studies
have shown a clear link between smoking and impotence and reduced sexual
pleasure. Fancy a cigarette?’ (app 9); ‘Your teeth, your breath and your skin
thanks you. Smoking leads to tooth loss, gives you bad breath and a sallow
complexion’ (app 22);
d) An evaluation of smoking as stupid (‘When you haven’t smoked for a month or
more you will realize how stupid it was to spend all that money on an addiction
that was literally killing you! Never again!’, app 22), and of smokers as foolish
(‘Smokers are comparable to alcoholics and heroin addicts. You fool yourself, and
tell yourself it’s not that bad an addiction to have. Treat yourself like an alcoholic
and never touch another cigarette again!’, app 12).
As part of their persuasive approach, most apps rely on framing smoking as a useless
behavior, driven by biological addiction, which incurs massive losses with no gain, and is to
be understood as a sign of lack of will, irrationality, or disease. This understanding stands in
contrast to the multiple psychological and social uses that smoking can have – as attested in
social research as well as biographical accounts. Smoking is used as a powerful resource for
sociability, self-presentation and identity work (Bottorff, Oliffe, Kalaw, Carey, & Mroz, 2006;
Desantis, 2003; Fry, Grogan, Gough, & Conner, 2008; Macnaughton, Carro-Ripalda, &
Russell, 2012). The value of smoking may be even higher for socially marginal groups (Lawn,
Pols, & Barber, 2002). While smoking cessation may not be the best stage of life to extol
cigarettes as tools for friendship, creativity, and self-control, the ethical issue remains as to
the limitations in self-understanding, understanding of other smokers, and appreciation of
difficulties in giving up smoking, induced by a simplistic discourse on the nature of
‘psychological dependence’ (that remains largely undefined in the app body of knowledge).
The two European Council apps differ from the other couch-type modules insofar the
body of advice reflects a broader understanding of the resorts of action. They include more
tips on emotional work, guiding users to a more attentive observation of their thoughts and
feelings, encouraging introspection and even introducing some concepts: ‘When you crave a
cigarette, you're more susceptible to 'rationalisation': justifying a bad habit. Don't cling to it,
seek a temporary distraction!’, ‘It's normal to feel panic from time to time. Are you afraid
you'll lose part of your identity? Rest assured, you won't!’; ‘Stay calm! Is that anger you're
feeling? Or is it fear? Don't walk away from your feelings. Observe them. Then they're easier
to let go’; ‘A little resistance is normal. Put that feeling under a microscope. Why do you
smoke? Is this a rational or more of an emotional choice?’; ‘You might sometimes glorify the
past and miss smoking. This is nothing more than the 'rose-tinted spectacles’ phenomenon!’
(app 9). These examples are an illustration of the potential of apps to promote a sharper
self-understanding - a potential that, for now, remains largely unrealized.
3.3.6 Moral deliberation and medicalization
At the higher level of moral deliberation, the question is: what values are promoted,
explicitly or implicitly, by smoking-cessation app-based interventions? One of the relevant
risks derives from medicalization (Conrad, 1992): promoting health and lack of disease as the
dominant vocabulary of wellbeing, rendering other sources of value in life less visible.
Calculator modules are heavily dependent on health and also on financial indicators,
adding consumerism to medicalization as an implicit moral orientation; smokers are
continuously encouraged to consider the state of their body, in the present and in the

19
future, and the state of their savings. Coach modules also focus on health and finance, but
add encouragements concerning beauty and interpersonal relationships (‘Without smoking
you will stay attractive much longer’, app 4; ‘When a craving comes, call a friend and take a
few minutes to connect with him. Your spirits will be lifted and chances are you’ll perk them
too’, app 22). The dominant virtue of smoking-cessation apps is strength of will, to which tips
and pieces of advice refer as a resource to overcome cravings.
App 2 (‘Get Rich or Die Smoking’) is an interesting attempt to overcome a money-focused
view of cessation benefits, encouraging users to convert their savings in a list of products
(Figure 5 a). While the application title and the product list encourage a consumerist view of
wellbeing, the app forum highlights users’ plans for their savings that often include travel,
spending time with their significant others, or helping their family and friends, and other
diverse projects – thus enlarging the view of what smoking cessation can contribute to the
value of life (Figure 5 b).
Figure 5. Communication devices: product lists and forums in smoking-cessation apps
(a) Get Rich or Die Smoking
(b) Get Rich or Die Smoking
Source: Screenshots published on Google Play.
Photo courtesy of Tobias Gruber [grubsen@gmail.com]
As a consequence of their medicalizing perspective, most apps are also focused on the
individual smoker, with little representation of the adverse effects that second-hand smoke
has on others – humans but also non-humans, such as pet cats or dogs. Smoking is framed,
implicitly, as an individual action that primarily affects the individual – and quitting, likewise,
is framed as an individual’s decision about her or his own life. There is significant scope for
improving apps representation of the effects of smoking and smoking cessation on other
beings other than the individual smoker (C. Rughiniș & Rughiniș, 2014).
4 Conclusions
We examine how smoking-cessation apps mediate moral action for designers and users by
shaping situations of choice as medium, market, and genre. We highlight specific moral

20
issues, taking into account both the current state of app-based smoking-cessation
persuasion, and the potential of the app medium.
In dialogue with previous assessments of ethics in persuasive technology and in public
health interventions, we focus on the core value of autonomy for people who interpret the
world and act with smoking-cessation apps. Apps communicate mostly in an anonymous
‘app voice’, complemented with pseudonymous peer voices in forums. Messages are fast
and frugal as a property of the medium, thus raising issues of truthfulness. Moreover,
smoking cessation apps open an ambivalent situation, in which game elements combine
with scientific rhetoric, complicating users’ expectations of reasonable verity. Outside
numerical messages, there is little presence of professional voices in the app environment,
be it medical, psychological, sociological, literary or otherwise. There is also a virtual absence
of personal voices of people speaking from experience under a personal name, either
through text or audio or otherwise.
The app voice in the couch modules has relatively little to contribute to smokers’
self-understanding qua smokers, outside health-related trivia. The app voice relies more
frequently than not on messages that stigmatize smoking and smokers, thus limiting the
potential for understanding self and others as smokers. The dominant moral frame relies on
the ‘cold-turkey’ lay medical theory, framing the ex-smoker as an individual hero who
overcomes addiction through strength of will. App incentives often promote medicalization
and commercialization of personal wellbeing.
The app medium offers substantial potential for assembling knowledge and advice about
smoking and smoking cessation, for expressing diverse voices, including peer experiences
and professional insights, and for encouraging users to personalize app messages shaping
them into self-persuasion. Still, to date much of this potential remains unused.
5 References
Abroms, L. C., Lee Westmaas, J., Bontemps-Jones, J., Ramani, R., & Mellerson, J. (2013). A
content analysis of popular smartphone apps for smoking cessation. American Journal
of Preventive Medicine, 45(6), 732–6. http://doi.org/10.1016/j.amepre.2013.07.008
Abroms, L., Padmanabhan, N., Thaweethai, L., & Phillips, T. (2011). iPhone Apps for Smoking
Cessation: A Content Analysis. American Journal of Preventive Medicine, 40(3), 279–
285.
Atkinson, B. M. C. (2006). Captology: A Critical Review. In W. Ijsselsteijn, Y. de Kort, & E. van
den Hoven (Eds.), Persuasive Technology. Lecture Notes in Computer Science Vol. 3962
(pp. 171–182). Springer.
Bakhtin, M. (1984). Problems of Dostoevsky’s Poetics. University of Minnesota Press.
Bakhtin, M. M. (1981). The dialogical imagination: Four essays. (M. Holquist, Ed.). Austin:
University of Texas Press.
Berdichevsky, D., & Neuenschwander, E. (1999). Toward an ethics of persuasive technology.
Communications of the ACM, 42(5), 51–58. http://doi.org/10.1145/301353.301410
BinDhim, N. F., McGeechan, K., & Trevena, L. (2014). Who Uses Smoking Cessation Apps? A
Feasibility Study Across Three Countries via Smartphones. JMIR mHealth and uHealth,
2(1).
Bolt, D. M., Piper, M. E., McCarthy, D. E., Japuntich, S. J., Fiore, M. C., Smith, S. S., & Baker, T.
B. (2009). The Wisconsin Predicting Patients’ Relapse questionnaire. Nicotine & Tobacco
Research, 11(5), 481–492.

21
Bottorff, J. L., Oliffe, J., Kalaw, C., Carey, J., & Mroz, L. (2006). Men’s constructions of
smoking in the context of women's tobacco reduction during pregnancy and
postpartum. Social Science & Medicine, 62(12), 3096–108.
Breland, J. Y., Yeh, V. M., & Yu, J. (2013). Adherence to evidence-based guidelines among
diabetes self-management apps. Translational Behavioral Medicine, 3(3), 277–86.
Brown, I., & Adams, A. A. (2007). The ethical challenges of ubiquitous healthcare.
International Review of Information Ethics, 8, 53–60.
Choi, J., Noh, G.-Y., & Park, D.-J. (2014). Smoking Cessation Apps for Smartphones: Content
Analysis With the Self-Determination Theory. Journal of Medical Internet Research,
16(2), e44. http://doi.org/10.2196/jmir.3061
Cohn, A., Hunter-Reel, D., Hagman, B., & Mitchell, J. (2011). Promoting Behavior Change
from Alcohol Use through Mobile Technology: The Future of Ecological Momentary
Assessment. Alcoholism: Clinical and Experimental Research, 35(12), 2209–2215.
Conrad, P. (1992). Medicalization and Social Control. Annual Review of Sociology, 18, 209–
232.
Courtwright, A. (2013). Stigmatization and public health ethics. Bioethics, 27(2), 74–80.
Cowan, L. T., Van Wagenen, S. A., Brown, B. A., Hedin, R. J., Seino-Stephan, Y., Hall, P. C., &
West, J. H. (2013). Apps of steel: are exercise apps providing consumers with realistic
expectations?: a content analysis of exercise apps for presence of behavior change
theory. Health Education & Behavior : The Official Publication of the Society for Public
Health Education, 40(2), 133–9.
Desantis, A. D. (2003). A Couple of White Guys Sitting Around Talking: The Collective
Rationalization of Cigar Smoking. Journal of Contemporary Ethnography , 32(4), 432–
466.
Ellison, N. B., Hancock, J. T., & Toma, C. L. (2011). Profile as promise: A framework for
conceptualizing veracity in online dating self-presentations. New Media & Society,
14(1), 45–62. http://doi.org/10.1177/1461444811410395
Engestrom, R. (1995). Voice as Communicative Action. Mind, Culture, and Activity, 2(3), 192–
214.
Etter, J.-F., Le Houezec, J., & Perneger, T. (2003). A Self-Administered Questionnaire to
Measure Dependence on Cigarettes: The Cigarette Dependence Scale.
Neuropsychopharmacology, 28, 359–370.
Fogg, B. (1998). Persuasive computers. In Proceedings of the SIGCHI conference on Human
factors in computing systems - CHI ’98 (pp. 225–232). New York, New York, USA: ACM
Press. http://doi.org/10.1145/274644.274677
Fogg, B. J. (2009). Creating Persuasive Technologies: An Eight-Step Design Process. In
Persuasive ’09 (pp. 1–6). Claremont, California.
Friedman, B. (1996). Value-sensitive design. Interactions, 3(6), 16–23.
http://doi.org/10.1145/242485.242493
Friedman, B., Kahn, P. H., & Borning, A. (2008). Value Sensitive Design and Information
Systems. In K. E. Himma & H. T. Tavani (Eds.), The Handbook of Information and
Computer Ethics (pp. 69–101). Hoboken, NJ: John Wiley & Sons.
Fry, G., Grogan, S., Gough, B., & Conner, M. (2008). Smoking in the lived world: how young
people make sense of the social role cigarettes play in their lives. The British Journal of
Social Psychology, 47(Pt 4), 763–80.
Gavriluță, N. (2002). On Tolerance and Acceptance of the Other. Journal for the Study of
Religions and Ideologies, (3), 22–27.

22
Guttman, N., & Salmon, C. T. (2004). Guilt, fear, stigma and knowledge gaps: ethical issues in
public health communication interventions. Bioethics, 18(6), 531–52.
Hoeppner, B. B., Hoeppner, S. S., Seaboyer, L., Schick, M. R., Wu, G. W. Y., Bergman, B. G., &
Kelly, J. F. (2015). How Smart are Smartphone Apps for Smoking Cessation? A Content
Analysis. Nicotine & Tobacco Research.
Klasnja, P., & Pratt, W. (2012). Healthcare in the pocket: Mapping the space of mobile-phone
health interventions. Journal of Biomedical Informatics, 45(1), 184–198.
Lawn, S. J., Pols, R. G., & Barber, J. G. (2002). Smoking and quitting: a qualitative study with
community-living psychiatric clients. Social Science & Medicine, 54(1), 93–104.
Lessig, L. (2000). Code is Law. On Liberty in Cyberspace.
Lewis, B. (2006). Listening to Chekhov: Narrative Approaches to Depression. Literature and
Medicine, 25(1), 46–71.
Lockton, D., Harrinson, D., & Stanton, N. (2008). Design with Intent: Persuasive Technology in
a Wider Context.
Lupton, D. (2013). Quantifying the body: monitoring and measuring health in the age of
mHealth technologies. Critical Public Health, 23(4), 393–403.
Macnaughton, J., Carro-Ripalda, S., & Russell, A. (2012). “Risking enchantment”: how are we
to view the smoking person? Critical Public Health, 22(4), 455–469.
Matei, Ș., Rughiniș, C., & Rughiniș, R. (2014). Refreshing Quantification and other Ploys to
Give Up the Habit A Repertoire of Relations, Identities, and Rhetorical Devices in
Smoking Cessation Applications. In The 16th International Conference on Human-
Computer Interaction HCII 2014. Applications and Services Lecture Notes in Computer
Science Volume 8512 (pp. 265–276). Crete, Greece: Springer Berlin Heidelberg.
Meier, E., Tackett, A. P., & Wagener, T. L. (2013). Effectiveness of Electronic Aids for Smoking
Cessation. Current Cardiovascular Risk Reports.
Mosa, A. S. M., Yoo, I., & Sheets, L. (2012). A systematic review of healthcare applications for
smartphones. BMC Medical Informatics and Decision Making, 12(1), 67.
Nafus, D., & Sherman, J. (2014). This One Does Not Go Up to 11: The Quantified Self
Movement as an Alternative Big Data Practice. International Journal of Communication,
8, 1784–1794.
Pagoto, S., Schneider, K., Jojic, M., DeBiasse, M., & Mann, D. (2013). Evidence-Based
Strategies in Weight-Loss Mobile Apps. American Journal of Preventive Medicine, 45(5),
576–582.
Puustinen, R. (2000). Voices to be heard—the many positions of a physician in Anton
Chekhov’s short story, A Case History. Journal of Medical Ethics: Medical Humanities,
26, 37–42.
Rughiniș, C., & Rughiniș, R. (2014). Influence of daily smoking frequency on passive smoking
behaviors and beliefs: Implications for self tracking practices and mobile applications.
Revista de Cercetare și Intervenție Socială, 44, 116–131.
Rughiniș, R., Matei, Ș., & Rughiniș, C. (2014). “Smoking Does Not Make You Happy”.
Unlearning Smoking Habits through Mobile Applications on Android OS. In The 6th
International Conference on Computer Supported Education CSEDU 2014. Barcelona:
INSTICC.
Smith, S. S., Piper, M. E., Bolt, D. M., Fiore, M. C., Wetter, D. W., Cincciripini, P. M., & Baker,
T. B. (2010). Development of the Brief Wisconsin Inventory of Smoking Dependence
Motives. Nicotine & Tobacco Research, 12(5), 489–499.
Spahn, A. (2012). And lead us (not) into persuasion…? Persuasive technology and the ethics

23
of communication. Science and Engineering Ethics, 18(4), 633–50.
http://doi.org/10.1007/s11948-011-9278-y
Thaler, R. H., Sunstein, C. R., & Balz, J. P. (2010). Choice Architecture. SSRN Electronic
Journal.
Ubhi, H. K., Michie, S., Kotz, D., Wong, W. C., & West, R. (2015). A Mobile App to Aid Smoking
Cessation: Preliminary Evaluation of SmokeFree28. Journal of Medical Internet
Research, 17(1).
Van Singer, M., Chatton, A., & Khazaal, Y. (2015). Quality of Smartphone Apps Related to
Panic Disorder. Frontiers in Psychiatry, 6, 1–7. http://doi.org/10.3389/fpsyt.2015.00096
Verbeek, P.-P. (2006). Persuasive Technology and Moral Responsibility. Toward an ethical
framework for persuasive technologies.
Verbeek, P.-P. (2008). Obstetric Ultrasound and the Technological Mediation of Morality: A
Postphenomenological Analysis. Human Studies, 31(1), 11–26.
Wellman, R. J., Di Franza, J. R., Morgenstern, M., Hanewinkel, R., Isensee, B., & Sabiston, C.
M. (2011). Psychometric Properties of the Autonomy over Tobacco Scale in German.
European Addiction Research, 18, 76–82.
Welsch, S. K., Smith, S. S., Wetter, D. W., Jorenby, D. E., Fiore, M. C., & Baker, T. B. (1999).
Development and Validation of the Wisconsin Smoking Withdrawal Scale. Experimental
and Clinical Psychopharmacology, 7(4), 354–361.
Wertsch, J. V. (1993). Voices of the Mind. A Sociocultural Approach to Mediated Action.
Cambridge, MA: Harvard University Press.