Tell me your successful story with tobacco! Indirect effects of narrative voice and behavioral audience-character similarity on the intention to quit smoking

Abstract

Smoking is linked to multiple health problems, but it is also the biggest preventable cause of premature death. Therefore, an important goal in health communication is to improve the effectiveness of campaigns targeted to smokers. The present study focuses on smoking prevention using narrative messages in which a former smoker described the process of quitting smoking and the improvements experienced since then. The role of two narrative devices was analyzed in an online experiment (N = 680) with a 2 (narrative voice: first vs. third-person message) x 2 (audience-character behavioral similarity: low vs. high) between-subjects factorial design. Results showed that the first-person narrative with a high-similar protagonist condition induced the highest levels of identification. Moreover, mediational analyses showed that identification with the protagonist, reactance and counterarguing were significant mediators. The present work opens an innovative line of research on the construction of narrative messages for smoking prevention.

Full text

Narrative devices for smoking prevention
1
Full paper
Tell me your Successful Story with Tobacco! Indirect Effects of Narrative
Voice and Behavioral Audience-Character Similarity on the Intention to
Quit Smoking
Juan-José IGARTUA
University of Salamanca, Spain
Isabel RODRÍGUEZ-DE-DIOS
University of Amsterdam, The Netherlands
Laura RODRÍGUEZ-CONTRERAS
University of Salamanca, Spain
María MARCOS-RAMOS
University of Salamanca, Spain
Beatriz GONZÁLEZ-DE-GARAY
University of Salamanca, Spain
Francisco-Javier FRUTOS
University of Salamanca, Spain
Paper presented to the Health Communication Division of the International Communication
Association 70th Annual Conference (“New Developments in Tobacco Risk Communication”,
standard paper session). Gold Coast (Australia), May 19–27, 2020.
Abstract
Smoking is linked to multiple health problems, but it is also the biggest preventable
cause of premature death. Therefore, an important goal in health communication is to improve
the effectiveness of campaigns targeted to smokers. The present study focuses on smoking
prevention using narrative messages in which a former smoker described the process of quitting
smoking and the improvements experienced since then. The role of two narrative devices was
analyzed in an online experiment (N = 680) with a 2 (narrative voice: first vs. third-person
message) x 2 (audience-character behavioral similarity: low vs. high) between-subjects
factorial design. Results showed that the first-person narrative with a high-similar protagonist
condition induced the highest levels of identification. Moreover, mediational analyses showed
that identification with the protagonist, reactance and counterarguing were significant
mediators. The present work opens an innovative line of research on the construction of
narrative messages for smoking prevention.
Keywords: health communication, smoking prevention, narrative persuasion, behavioral
character-audience similarity, narrative voice.

Narrative devices for smoking prevention
2
Tobacco use has been linked to
multiple health problems, such as
respiratory and cardiovascular diseases, and
different types of cancer (American Cancer
Society, 2018; World Health Organization,
2017). That said, smoking is also the
leading preventable cause of premature
death. Consequently, improving the
effectiveness of smoking cessation
campaigns is an important goal in public
health management and health
communication. The present study focuses
on smoking prevention using narrative
messages, i.e. personal stories of former
smokers that serve as a model for changing
attitudes and behaviors of current smokers
(Dunlop, Wakefield, & Kashima, 2010;
Kim, 2019; Niederdeppe, Heley, & Barry,
2015).
Narrative Messages for Smoking
Prevention
To limit the scope of this study,
firstly, we must define what a narrative is.
There are several definitions of the term, but
they all have in common the idea of a
message in which there is (at least) one
character who experiences or faces (at least)
one event or issue in a specific
spatiotemporal framework (Green, 2006;
Kreuter, Green, Cappella, Slater, Wise,
Storey, Clark, O’Keefe, DeErwin, Holmes,
Hinyard, Houston, & Woolley, 2007;
McDonald, 2014). The character is a human
being who acts motivated by certain
intentions or reasons and who aims to reach
a goal. The event or issue that the character
is facing is a transition between two states
that are temporary and causally related.
However, most narratives are composed by
multiple events that are causally related to
each other. Having said that, although
events can be presented in non-
chronological order, the underlying
structure will be a cause-effect relationship
(or action and reaction). This relationship
connects the narrative events and the
characters into a structure that takes form of
a story or narrative. These elements
(characters, events, space and time) convert
narratives into specific and concrete
messages, that is, “stories of particular
cases” in which events involving one or
several people in a specific area happen.
Focusing on specific particular cases
contrasts with, for instance, scientific
explanations, which give more general and
abstract information based on multiple
cases. Thus, non-narrative messages give
more general and abstract data that is
presented as information (“every cigarette
you smoke increases your chance of getting
lung cancer”) or in statistical form
(providing quantified data about the
prevalence of a phenomenon, “8 out 10
people with lung cancer will die over the
next three years”), compared to the
experimental style of the narrative (“the
lung cancer chemotherapy affects all your
body, it is a living death”).
In the context of our project, we will
define narrative messages for smoking
prevention as short personal stories told by
a former smoker who shares details about
their experience with tobacco. A smoking
prevention narrative refers to a former
smoker’s personal story which aims to serve
as a model for changing attitudes, beliefs
and behaviors of current smokers (e.g., the
Tips from Former Smokers campaign,
developed in the United States by the
Centers for Disease Control and Prevention,
CDC). Therefore, these narratives refer to
two different states, life as a smoker and life
after quitting smoking, that are connected
with each other in time and causal
sequence. Furthermore, the narrative can
refer to aspects such as the reasons that led
the character to stop smoking (intentions
and goals of the character), the previous
degree of tobacco dependence (how often
they smoke), the number of times that they
tried to stop smoking, the actions taken to
quit smoking, the benefits of quitting
smoking, and the disappearance of smoking
negative effects when quitting smoking.
The goal of a smoking prevention narrative
is to induce audience or target audience’s
desire to quit smoking, to reinforce their

Narrative devices for smoking prevention
3
self-efficacy expectations and to convince
them of the effectiveness of the promoted
preventive response (quit smoking).
Narrative messages are increasingly
used in health communication to achieve
public health objectives, such as diseases
detection and prevention (Frank, Murphy,
Chatterjee, Moran, & Baezconde-
Garbanati, 2015; Green, 2006; Jensen,
Yale, Krakow, John, & King, 2017;
Thompson & Kreuter, 2014). It has been
found that health narrative messages can
promote beliefs and consistent attitudes
changes, as well as healthy behaviors (De
Graaf, Sanders, & Hoeken, 2016). In the
case of smoking prevention, in recent years
there has been a growing body of research
that shows the effectiveness of narratives
(De Graaf, van den Putte, Nguyen, Zebregs,
Lammers, & Neijens, 2017; Dunlop et al.,
2010; Kim, 2019; Kim, Bigman, Leader,
Lerman, & Cappella, 2012; Kim & Lee,
2017; Kim, Shi, & Cappella, 2016;
Williams, Green, Kohler, Allison, &
Houston, 2011).
Meta-analysis studies show that
narrative interventions have significant
effects on the dependent variables
considered (on beliefs, r = .17; attitudes, r =
.19; intentions, r = .17; and behaviors, r =
.23; Braddock & Dillard, 2016). However,
there is also a significant variation in these
narrative intervention effects (Shen, Sheer,
& Li, 2015; Zebregs, Van den Putte,
Neijens, & De Graaf, 2015). These findings
suggest that, even if narrative messages can
be a promising health communication tool,
not all narratives are effective. Thus, it is
crucial to find out which narrative
ingredients are more effective from a
persuasive point of view. That is precisely
one of the main goals of the present study.
Explanatory Processes of Narrative
Persuasion
A second goal of the present study is
to understand and explain the processes or
mechanisms responsible for the persuasive
impact of narrative messages for smoking
prevention. The main theoretical models of
narrative persuasion are the Transportation-
Imagery Model by Green and Brock (2002),
the Extended Elaboration Likelihood
Model (E-ELM) by Slater and Rouner
(2002), and the Entertainment Overcoming
Resistance Model, (EORM) by Moyer-
Gusé (2008). In this context, the main
processes studied are identification with the
protagonist, narrative engagement or
transportation, counterarguing and
reactance.
Identification is an imaginative
process that implies a gradual loss of self-
awareness and the assumption of the
emotional and cognitive point of view of the
narrative’s protagonist
(Cohen, 2001; Igartua, 2010). Narrative
transportation is a psychological process
that implies an immersive or engaged state
in the story or narrative (Busselle &
Bilandzic, 2009; Green & Brock, 2000).
Counterarguing is a process of making
critical assessments during the exposure to
a message (and that implies thinking
negatively about the persuasive or
preventive proposal). It has been formally
defined as the “generation of thoughts (or
cognitive answers) that explicitly refute a
message’s intended persuasive theme”
(Niederdeppe, Kim, Lundell, Fazili, &
Frazier, 2012, p. 758). Reactance is a
second process linked to the resistance
against a persuasive attempt. It is activated
when the individual feels their freedom of
choice is being threatened (Rains, 2013).
Hence, a distinction is made
between two types of mediating processes:
a) those which are directly connected with
the impact of the message’s characteristics
(i.e., identification with the characters and
narrative transportation); and, b) those
which explain why people involved with
narratives and characters are persuaded, as
a result of the counterarguing and reactance
processes. At the beginning of the causal
chain, identification and narrative
transportation are the most relevant
processes (primary mediators), since they
initiate the process that is going to facilitate

Narrative devices for smoking prevention
4
the persuasive impact through the reduction
of counterarguments and reactance. Thus,
counterarguing and reactance are
considered secondary mediators (Banerjee
& Greene, 2012; Shen, Seung, Andersen, &
McNeal, 2017).
The model EORM by Moyer-Gusé
(2008) indicates that identification and
narrative transportation reduce
counterarguing and reactance. This is
because people who are involved with the
narrative let themselves be carried away by
the story (and experience fun and
entertainment). This weakens any stance or
critical attitude towards the message (since
they are processes incompatible with fun or
entertainment). Nonetheless, empirical
evidence on the issue is inconclusive. For
example, Moyer-Gusé and Nabi (2010)
found that transportation led to greater
counterarguing. On the contrary,
identification reduced such reactance
against the message.
Notwithstanding, since narrative
transportation and identification with the
characters are relevant processes for
changing attitudes through the consumption
of narratives, nowadays research focuses on
finding out how to increase such processes.
That is, finding which variables related to
the characteristics of the characters or to the
mode of narrative presentation have an
effect on such processes and, indirectly, on
attitudinal variables (De Graaf et al., 2016).
The present study focuses precisely on the
study of two factors that may increase
identification and narrative transportation:
audience-character similarity and narrative
voice or perspective from which the story is
told.
Similarity to the Protagonist
Similarity to the protagonist occurs
when the individual who is exposed to a
narrative message shares certain trait with
the protagonist. Similarity can be based on
objective traits (demographic aspects, such
as gender and age), but also on
psychological or subjective features (such
as personality, beliefs, opinions, values and
biographical experiences). It is assumed
that “if there is a considerable social
distance between the audience and the
characters (...) persuasion is less likely to
occur” (Walter, Murphy, & Gillig, 2018, p.
32). Nevertheless, empirical evidence on
the effect of similarity brings inconsistent
results. Tukachinsky’s review (2014) shows
that the manipulation of similarity (in
objective traits) increases narrative
transportation and the perception of
similarity, but not identification.
It is worth considering, however,
that similarity is a complex construct with
several dimensions that has not always been
manipulated in the same way, which could
explain the inconsistent results. Thus,
Tukachinsky’s review (2014) only
considered studies in which similarity had
been manipulated based on demographic
traits, such as gender, age or ethnic origin
(Chen, Bell, & Taylor, 2016, 2017; Cohen,
Weimann-Saks, & Mazor-Tregerman,
2018; Kim, 2019). In the present study we
introduce a new dimension that is relevant
for smoking prevention: behavioral
similarity considering the degree of tobacco
dependence. To manipulate similarity in
this way, information about the degree of
tobacco dependence of both the protagonist
of the narrative and participants were taken
into account. Literature on tobacco
dependence shows that success factors for
quitting smoking include, among others, the
severity of the dependence (number of
cigarettes per day, score on the Fagerström
test) (Moreno & García, 2000).
Having said that, it is also possible
that similarity will only influence
identification (and transportation) under
certain conditions of reception (Kaufman &
Libby, 2012). Consequently, it is crucial to
continue delving into this research area and
to explore in which specific conditions the
effect of similarity can be increased. In the
present study, we propose narrative voice as
one these conditions.

Narrative devices for smoking prevention
5
Narrative Voice
Narrative voice or point of view
refers to the narrator’s perspective and from
which the story is told: first, second or third
person (Chen et al., 2017; Christy, 2018;
Nan, Futerfas, & Ma, 2017). First-person
narratives focus on the protagonist’s
feelings and thoughts. They do so by
assuming the first-person voice, which
leads to the use of first-person pronouns,
such as “I”, “my”, and “me”. Moreover, the
narrator is involved in the story and
expresses their view or their direct
experience on an issue (e.g., “I felt bad for
smoking”, “I had always thought that I
would never be able to quit smoking”).
Third-person narratives have a narrator who
is not a character in the story and tells the
character’s experience from an external
perspective or from the viewer’s
perspective. This way, characters are
referred by their names or as “he” or “she”,
i.e. third person pronouns, (e.g., “John felt
bad for smoking”, “John had always
thought that he would never be able to quit
smoking”). Finally, second-person
narratives cast the reader as the protagonist
(e.g., “You felt bad for smoking”; “you had
always thought that you would never be
able to quit smoking”). The second person
perspective is useful for developing
materials, such as guides, instruction
manuals and self-help books, but it is rarely
used in narrative health campaigns (Christy,
2018).
The use of first person facilitates
audience members taking the character’s
perspective and this is a critical dimension
in the identification with the characters.
Moreover, neuropsychological research
shows that first (versus third) person
narratives are processed differently at a
neural level (Van Krieken, Hoeken, &
Sanders, 2017).
Chen et al. (2016) state that
narrative persuasion research has given
little attention to the effect of narrative
voice, despite it being a relevant formal
feature for the development of narrative
messages for smoking prevention (e.g., Tips
from Former Smokers campaign). Thus, de
Graaf et al.’s review (2016) analyzed 153
experimental studies on health-related
narrative persuasion and it only identified
four studies that had manipulated this
feature (and only one of them focused on
smoking prevention). This review stated
that first-person narratives have a greater
potential to have persuasive effects.
However, evidence was not consistent, and
the number of studies considered was low.
But what has been observed in previous
studies is that first-person messages (versus
third-person narratives) are perceived as
more personal and easier to understand,
increase perspective taking and the
identification with the character, and are
more effective for inducing risk perception
(Chen et al., 2017; De Graaf, Hoeken,
Sanders, & Beentjes, 2012; Kaufman &
Libby, 2012; Nan, Dahlstrom, Richards, &
Rangarajan, 2015).
Nevertheless, studies that combined
narrative voice and another second
independent variable in the same
experiment have not yielded as clear results.
For instance, the study by Nan et al. (2017)
showed that the first-person message was
only superior over the third-person one
when they used a text-based narrative.
There were no differences when using the
audio-based narrative. Likewise, Kaufman
and Libby (2012) observed that reading a
first-person narrative depicting a character
that is part of the audience’s in-group
significantly increased identification.
Overview and Hypotheses
Our research aims to delve into the
‘joint effect’ of behavioral similarity and
narrative voice. Thus, its original
contribution or innovation to the study of
narrative persuasion for smoking
prevention is the concept of optimal
reception condition, which implies the use
of a first-person narrative led by a character
with high behavioral similarity to the
audience. Due to the scarcity of studies that

Narrative devices for smoking prevention
6
have used such variables in smoking
prevention research, and considering the
existence of contradictory results on the
“main effects” of similarity and narrative
voice, it is relevant to study how to combine
both factors to induce greater levels of
identification and narrative transportation,
and to test the indirect effect (through such
processes and trough counterarguing and
reactance) on the persuasive impact of
smoking prevention narratives.
It can be expected that a smoking
prevention message using a first-person
voice and a character with high behavioral
similarity to the audience (considering the
degree of tobacco dependence) will
increase identification with the protagonist
and narrative transportation. It is therefore
more likely that the message will not be
perceived as a threat (it will not elicit
reactance and counterarguing), and as such
it will have an indirect effect on the
intention to quit smoking, on self-efficacy
expectations, and on the perceived
effectiveness of the preventive response.
Therefore, it is hypothesized (H1) that the
joint effect of first-person narrative and
audience-character behavioral similarity
will induce the highest levels of
identification with the protagonist (H1a)
and narrative transportation (H1b).
Secondly, it is hypothesized (H2) that the
optimal reception condition will have an
indirect effect on the intention to quit
smoking, on self-efficacy expectations, and
on the perceived efficacy of the preventive
response, which will be (serially) mediated
by identification with the character (H2a)
and narrative transportation (H2b) (primary
mediators), and counteraguing and
reactance (secondary mediators).
Figure 1. Hypothesized serial multiple mediator model
Method
With the aim of testing our
hypotheses, an online experiment using
Qualtrics was conducted among a
representative sample of 680 smoker adults,
aged 18-55 years, of both genders, residents
in Spain. To estimate the sample size
required, we conducted a previous analysis
using G*Power (Faul, Erdfelder, Lang, &
Buchner, 2007). All the participants were
exposed to a narrative message whose
protagonist was an adult smoker (a 35-year-
old man) who told his experience in the
process of quitting smoking, emphasizing
the positive consequences of doing it.
Narrative voice (first-person vs. third-
person) and the degree of tobacco
dependence of the narrative protagonist
(low vs. high) were manipulated in the
narrative. All the materials related to the
experiment (datasets, measures, narratives
and pilot study results) are available in
Open Science Framework
(https://osf.io/y3c8f/).
Sample
A sample of 680 smoker adults,
aged 18-55 years (M = 35.68, SD = 10.85),
selected from Qualtrics (opt-in online panel
provider) panels in Spain, and based on a
quota of gender (50% of each gender) and
age (20% aged 18-22 years, 30% aged 23-
35 years and 50% aged 36-55 years),
participated in the study. The fieldwork
took place between 2 and 14 October 2018.
To select participants, three screener
Optimal reception
condition
(a first-person narrative
whose protagonist is
similar to the audience)
Identification with the
protagonist
Narrative
transportation
Counterarguing
Reactance
Quit smoking intention
Self-efficacy expectations
Perceived effectiveness
of the preventive
response

Narrative devices for smoking prevention
7
questions were included at the beginning of
the online questionnaire: “in relation to
tobacco, you would define yourself as”
(never smoker, occasional smoker, regular
smoker), “have you smoked 100 or more
cigarettes in your lifetime?” (no, yes), “in
the last week…” (I have not smoked, I have
smoked less than 5 cigarettes a day, I have
smoked 5 or more cigarettes a day). We
only considered participants who were
regular smokers, who have smoked 100 or
more cigarettes in their lifetime, and who
have smoked 5 or more cigarettes a day.
Design and Procedure
In the study, a 2 x 2 between-
subjects factorial design was used. Two
independent variables were manipulated:
narrative voice (first-person vs. third-
person message) and the degree of tobacco
dependence of the narrative protagonist
(low vs. high). The fieldwork was carried
out by Qualtrics, which allowed us to use a
completely randomized design regarding
the distribution of participants into
treatment groups.
Since Qualtrics enables to
implement several quality controls, the
questionnaire was designed so that it should
be answered from beginning to end in a
single session. Moreover, only those
participants who had spent between 6 and
45 minutes answering the questionnaire (M
= 11.16 minutes, SD = 4.71), had spent
between 60 and 420 seconds reading the
narrative (M = 110.67, SD = 50.54), and had
answered correctly to the attention-check
question placed at the end of the
questionnaire, were considered as valid
cases.
Narratives and Experimental
Manipulations
A narrative whose protagonist was a
35-year-old man (Miguel) who had quit
smoking a year ago (former smoker) was
written using stories from former smokers
on forums and websites and narratives from
previous studies (Dunlop et at., 2010; Kim
et al., 2012; Kim et al., 2016; Williams et
al., 2011) as a basis. In his story, the former
smoke talked about issues such as the age at
which he started smoking, why he decided
to stop smoking, and how he managed to
quit smoking. Likewise, he described the
process of giving up tobacco and the
improvements he has noticed since that
time, but without including abstract
information or statistical data. At the end of
the narrative, the protagonist pointed out
that he had not smoked for more than a year
and described the benefits that he felt.
To manipulate narrative voice, we
modified the features that establish the
grammatical person in written narratives,
such as the choice of personal pronouns. In
the first-person narrative, we used first-
person pronouns such as “I”, “me”, and
“my” (e.g., “I quit smoking a year ago”),
whereas in the third-person narrative, we
used third-person pronouns such as “he”
and “him” (e.g., “He quit smoking a year
ago”). Moreover, in the first-person
narrative, the protagonist introduced
himself by name at the beginning: “my
name is Miguel, I am 35 years old and I
started smoking at 15”. In the third-person
narrative, the narrator of the story
introduced the protagonist by his name:
“Miguel is 35 years old and he started
smoking at 15”. This experimental
procedure to manipulate narrative voice has
been successfully employed in a great
number of studies (Banerjee & Greene,
2012; Chen et al., 2017; Chen, McGlone, &
Bell, 2015; Nan et al., 2015; Nan et al.,
2017).
To manipulate behavioral similarity,
we considered the degree of tobacco
dependence of both the protagonist and the
participants. The narrative whose
protagonist was a former smoker with low
tobacco dependence, included information
about the level of tobacco consumption
before quitting smoking (“I was not hooked
on tobacco, I only smoked 9 or 10 cigarettes
a day”), about the money he saved from
quitting smoking (“I started saving, more

Narrative devices for smoking prevention
8
than 80 euros a month, by not buying
tobacco), or vital change (“I no longer have
to worry about having tobacco, I no longer
need smoking to relax”). The narrative
whose protagonist was a former smoker
with high tobacco dependence, emphasized
a higher level of tobacco consumption
before quitting smoking (“I was completely
hooked on tobacco, I smoked more than a
pack a day”), higher savings from quitting
smoking (“I started saving, more than 150
euros a month, by not buying tobacco”), or
a more profound vital change (“I am no
longer a tobacco slave who digs through
household garbage looking for cigarette
butts, I no longer wake up at night to
smoke”).
In order to create an audience-
character behavioral similarity index (low,
high), in the pre-test measure we asked the
participants about their degree of tobacco
dependence using the Fagerström test
(Heatherton, Kozlowski, Frecker, &
Fagerström, 1991). The Fagerström test is a
6-item scale used to measure nicotine
dependence. Combining the answers to the
six items, the degree of nicotine
dependence (values between 0 and 10) was
determined. The characteristics of each
participant were matched with the
characteristics of the protagonist. Hence,
similarity was determined taking behavioral
criteria as a reference. High similarity was
determined when the narrative’s
protagonist had a high level of tobacco
dependence and the participant scored 5 or
more on the Fagerström test (median value
in the study), or when the narrative’s
protagonist had a low level of tobacco
dependence and the participant scored less
than 5 on the Fagerström test. Likewise, low
similarity was determined when the
narrative’s protagonist had a high level of
tobacco dependence and the participant
scored less than 5 on the Fagerström test, or
when the narrative’s protagonist had a low
level of tobacco dependence and the
participant scored 5 or more on the
Fagerström test.
A pilot study was conducted among
120 participants (57.5% females, aged 18-
32 years, M = 20.73, SD = 2.69).
Participants were randomly assigned to one
of the four versions of the narrative. After
reading the narrative, they answered a
questionnaire with questions about clarity
and understanding (e.g., “The message is
clear and comprehensible”), perception of
credibility, interest, and perceived realism,
on 7-point Likert scale (from 1 = strongly
disagree, to 7 = strongly agree). Results
showed that narratives were perceived as
clear and easy to understand (M = 6.13, SD
= 2.69), credible (M = 5.43, SD = 1.33),
interesting (M = 5.30, SD = 1.28) and
realistic (M = 5.92, SD = 0.97). In all the
variables considered the means were
significantly higher than the theoretical
mean (4), as tested with a one sample t-test
(p < .001).
Measures
The questionnaire included pre-test
measures and post-test measures. Pre-test
measures included the screener questions
and the Fagerström test (mentioned above),
and also basic sociodemographic
information (gender and age). After the
reading of the narrative whose protagonist
was a former smoker, post-test measures
were immediately presented. They included
scales (following this order) to evaluate
perceived similarity with the protagonist,
identification with the protagonist, narrative
transportation, counterarguing, reactance,
quit smoking intention, self-efficacy
expectations, and perceived effectiveness of
the preventive response.
Dependent variables
Quit smoking intention. A scale
composed of 3 items (e.g., “I am definitely
going to quit smoking in the future”, from
1 = strongly disagree, to 7 = strongly agree)
was developed following the work of
Dunlop et al. (2010), Kim et al. (2012), Kim
et al. (2016), Thrasher, Arillo-Santillán,

Narrative devices for smoking prevention
9
Villalobos, Pérez-Hernández, Hammond,
Carter, Sebrié, Sansores, and Regalado-
Piñeda (2012), Wehbe, Basil, and Basil
(2017), and Williams et al. (2011). An index
of quit smoking intention was obtained by
averaging the three items (α = .82).
Self-efficacy expectations related
to quitting smoking. Self-efficacy is
defined as the confidence a person has in
their ability to have and maintain a certain
behavior in a given situation (in this case,
refrain from smoking when such habit has
been left) (Spek, Lemmens, Chatrou,
Kempen, Pouwer, & Pop, 2013). To
measure self-efficacy expectations, we used
a scale composed of 6 items (e.g., “I think I
am capable of quitting smoking whenever I
want”, from 1 = strongly disagree, to 7 =
strongly agree) following the work of Chen
et al. (2015), McQueen, Waters, Kaphingst,
Caburnay, Sanders Thompson, Boyum, and
Kreuter (2016), Spek et al., (2013) and
Williams et al. (2011). An index of self-
efficacy expectations was calculated by
averaging the six items (α = .88).
Perceived effectiveness of the
preventive response (quit smoking). It
was assessed with a scale composed of 5
items (e.g., “although you have smoked for
many years, it is possible to recover your
health if you quit smoking in time”, from 1
= strongly disagree, to 7 = strongly agree)
based on the study by Chen et al. (2015). An
index of perceived effectiveness of the
preventive response was created by
averaging the five items (α = .78).
Mediator variables
Identification with the
protagonist. It was assessed with a scale
composed of 11 items whose reliability and
structural validity has been tested in a
previous study (Igartua & Barrios, 2012).
The instrument was designed to measure
retrospectively the identification with the
protagonist (e. g., “I felt as if I was Miguel”,
from 1 = not at all, to 5 = very much). An
index of identification with the protagonist
was obtained by averaging the eleven items
(α = .93).
Narrative transportation. It was
measured with the Transportation Scale–
Short Form developed by Appel, Gnambs,
Richter, and Green (2015), and composed
of 5 items (e.g., “I felt very involved or
engaged during the reading of the story”,
from 1 = strongly disagree, to 7 = strongly
agree). An index of narrative transportation
was calculated by averaging the five items
(α = .89).
Counteraguing. A scale composed
of 3 items (e.g., “while I was reading the
message, I thought that the information
provided by Miguel was inaccurate,
misleading or exaggerated”, from 1 =
strongly disagree, to 7 = strongly agree) was
developed following the scale by Moyer-
Gusé and Nabi (2010), and Igartua and
Vega (2016). An index of counterarguing
was obtained by averaging the three items
(α = .72).
Reactance. It was assessed using
the perceived threat to freedom scale
developed by Shen (2015). This scale
consists of 4 items (e.g., “the message
threatened my freedom to choose”, from 1
= strongly disagree, to 7 = strongly agree).
An index of reactance was calculated by
averaging the four items (α = .85).
Experimental manipulation check
Perceived similarity with the
protagonist of the story. With the aim of
testing the experimental manipulation,
immediately after reading the narrative,
participants answered the following
questions: “to what extent do you think you
have things in common with Miguel?”, “to
what extent do you think you are similar to
Miguel considering his degree of tobacco
dependence before he quit smoking?” (from
1 = not at all, to 5 = very much). Both items
were combined to form an index of
perceived similarity with the protagonist (r
= .53, p = .001; M = 3.41, SD = 0.84).

Narrative devices for smoking prevention
10
Results
Preliminary Analysis
The random allocation to the four
experimental conditions was successful.
There were no statistically significant
differences between the conditions in
sociodemographic terms (gender (
c
2 (3, N =
680) = 1.01, p = .798; age (F (3, 676) = 0.07,
p = .973) or in the degree of tobacco
dependence (F (3, 676) = 0.52, p = .663).
The manipulation of behavioral
similarity was also effective. Using
moderation analysis (model 1) with
PROCESS for SPSS (Hayes, 2018), we
found that there was a statistically
significant interaction effect between the
degree of tobacco dependence of the
narrative protagonist and the participants’
degree of tobacco dependence (Fagerström
test) on the index of perceived similarity (B
= 0.19, SE = 0.02, p = .001). Analysis of
conditional effects revealed that
participants with low tobacco dependence
(scored 2 or less on the Fagerström test)
reported higher perceived similarity when
reading the narrative whose protagonist had
low tobacco dependence before quitting
smoking than when reading the one whose
protagonist had high tobacco dependence
(θX
à
Y |(W = 2.0) = -0.58, SE = 0.09, p = .001).
Likewise, participants with high tobacco
dependence (scored 7 or more on the
Fagerström test) reported higher perceived
similarity when reading the narrative whose
protagonist had high tobacco dependence
than reading the one whose protagonist had
low dependence (θX
à
Y |(W = 7.0) = 0.38, SE =
0.08, p = .001). On the contrary, the effect
of protagonist’s degree of tobacco
dependence on perceived similarity was not
statistically significant among participants
with medium tobacco dependence (θX
à
Y |(W
= 5.0) = 0.01, SE = 0.06, p = .993).
Figure 2. Manipulation check: conditional effect of the degree of tobacco dependence of the
narrative protagonist on perceived similarity in participants with low, medium and high-tobacco
dependence (Fagerström test)
Note. - Conditional effects analysis with PROCESS macro for SPSS (Model 1). Covariates: narrative voice,
gender and age.

Narrative devices for smoking prevention
11
Finally, correlations between
mediators and dependent variables were
analyzed. This analysis revealed that
mediation processes showed convergent
correlations with the proposed hypotheses
(e.g., association between identification and
narrative transportation with reactance and
counterarguing). Moreover, we also
identified statistically significant
relationships between mediation processes
and the dependent variables. These results
justify the proposed mediation model that
will be tested later in this paper.
Table 1. Descriptive analysis and correlations between mediating and dependent variables
1
2
3
4
5
6
7
1 Identification
-
2 Narrative transportation
.85 ***
-
3 Counterarguing
-.31 ***
-.30 ***
-
4 Reactance
-.26 ***
-.29 ***
.44 ***
-
5 Quit smoking intention
.51 ***
.49 ***
-.18 ***
-.23 ***
-
6 Self-efficacy expectations
.10 **
.14 ***
-.04
-.14 ***
.33 ***
-
7 Perceived effectiveness of
the preventive response
.41 ***
.43 ***
-.27 ***
-.28 ***
.39 ***
.30 ***
-
Mean
3.64
5.28
2.76
2.62
4.71
4.53
5.81
Standard deviation
0.79
1.18
1.24
1.39
1.36
1.31
0.91
Note. - N = 680. In all variables, a higher score reflects a greater intensity of the process considered, from
1 “low” to 7 “high” (except in the identification scale with a theoretical range from 1 “low” to 5 “high”).
** p < .01, *** p < .001
Joint Effect of Audience-Character
Behavioral Similarity and Narrative
Voice on Identification and Narrative
Transportation (H1)
It had been hypothesized that
participants would experience the highest
levels of identification with the protagonist
(H1a) and narrative transportation (H1b) in
the optimal reception condition compared
to the reference condition. To test this
hypothesis, two analysis of variance
(ANOVA) with the experimental condition
as an independent variable and
identification and narrative transportation
as dependent variables were performed,
carrying out two planned contrast analyses
(post hoc tests) (see Walter et al., 2018, for
a similar analytical approach). The first
planned contrast (contrast coefficients: -1,
0, 0, 1) compared condition 4 or optimal
reception condition (i.e., reading a first-
person narrative with a high-similarity
character, coefficient 1) with reference
condition 1 (that is, reading a third-person
narrative with a low-similarity character,
coefficient -1). The second planned contrast
(coefficients: -1, -1, -1, 3) compared the
optimal reception condition with the
average of the three experimental
conditions.
Regarding identification, marginally
non-significant statistical differences were
observed as a function of the experimental
condition (F(3, 676) = 2.23, p = .083, hp2 =
0.010). However, the two planned contrasts
were statistically significant (contrast 1:
t(676) = 2.55, p = .011, r = .09; contrast 2:
t(676) = 2.28, p = .023, r = .08). These
results showed that the optimal reception
condition induced the highest levels of
identification with the protagonist, which
supports H1a (see Figure 3a).

Narrative devices for smoking prevention
12
Figure 3. Effect of the experimental condition on identification with the protagonist (H1a)
and narrative transportation (H1b)
(a) Effect of experimental condition on identification
(b) Effect of experimental condition on narrative transportation
With respect to narrative
transportation, no statistically significant
differences were observed depending on the
experimental condition (F(3, 676) = 0.96, p
= .410, hp2 = 0.004). In addition, the two
planned contrasts were also not statistically
significant (contrast 1: t(676) = 1.66, p =
.097, r = .06; contrast 2: t(676) = 1.18, p =
.236, r =. 04), so H1b was not confirmed.
However, the descriptive results showed
that the optimal reception condition also
induced the highest levels of narrative
transportation (see Figure 3b).
Mediation Analysis (H2)
The second hypothesis predicted an
indirect effect of the optimal reception
condition on the intention to quit smoking,
on self-efficacy expectations, and on the
perceived efficacy of the preventive
response, which would be mediated by
identification (H2a) and narrative
transportation (H2b) (primary mediators)

Narrative devices for smoking prevention
13
and counterarguing and reactance
(secondary mediators). To examine whether
the optimal reception condition induced a
significant indirect effect on the dependent
variables, the PROCESS macro for SPSS
was used (Model 6, serial multiple mediator
model, 10,000 bootstrapping samples to
generate 95% confidence intervals by the
percentile method; Hayes, 2018). The
independent variable (experimental
condition) was coded as a multicategorical
variable, resulting in three dummy variables
(X1, X2, X3) and establishing the control
condition (third-person narrative with a
low-similarity protagonist) as the category
of reference (for a similar analytical
approach, see: Bolkan, Goodboy & Myers,
2017; Walter et al., 2018). This procedure
required the estimation of the indirect effect
of the optimal reception condition (X3),
while the other two dummy variables (X1,
X2) acted as covariates (Hayes & Preacher,
2014; gender and age were also included as
covariates). Thus, regression coefficients
that estimate the indirect effect, calculate
the difference between being exposed to a
first-person narrative with a protagonist
with high behavioral similarity (1) versus
being exposed to a third-person narrative
with a protagonist with low behavioral
similarity (0). Results from the 12
mediation analyses are presented in Figure
4 and Table 2.
Figure 4. Mediation analysis (H2)
H2a (primary mediator: identification)

Narrative devices for smoking prevention
14
H2b (primary mediator: narrative transportation)
Nota. - Optimal Reception Condition (X3): first person narration with a protagonist with high behavioral
similarity with the participant (depending on the degree of tobacco dependence). The figure shows the non-
standardized regression coefficients (B). N = 680. + p < .10, ** p < .01, *** p < .001. Covariates: X1, X2,
gender and age.
It was observed that the optimal
reception condition increased the
identification with the protagonist of the
narrative, which in turn reduced reactance
and counterarguing. In addition, a lower
level of reactance was associated with a
greater intention to quit smoking, with
higher self-efficacy expectations and with a
greater perceived effectiveness of the
preventive response. However, even though
identification reduced counterarguing, this
process was only associated with one of the
dependent variables considered:
counterarguing during reading the narrative
was negative associated with the perceived
efficacy of the preventive response.
Therefore, confirming the H2a, we
observed three statistically significant
indirect effects of the optimal reception
condition on the three dependent variables
that were explained by the increase in the
identification and the reduction of the
reactance. In addition, we also observed a
statistically significant indirect effect of the
optimal reception condition on the
perceived effectiveness of the preventive
response, which was due to an increase of
identification and a reduction of
counterarguing (see Table 2).
The results were less consistent
regarding the role of narrative
transportation, mainly because the optimal
reception condition had a marginally non-
significant effect (p = .097) on this process.
However, it was observed that narrative
transportation reduced both counterarguing
and reactance. In turn, a lower level of
reactance increased the intention to quit
smoking, self-efficacy expectations, and the
perceived effectiveness of the preventive

Narrative devices for smoking prevention
15
response. However, a lower counterarguing
was only associated with a greater
perceived effectiveness of the preventive
response. Thus, the results do not support
H2b.
Table 2. Indirect effects of the optimal reception condition (ORC) on quit smoking intention,
self-efficacy expectations and perceived effectiveness of the preventive response (H2).
Mediation analysis with PROCESS macro for SPSS (Model 6).
Indirect effects
Effect
Boot SE
Boot 95% IC
ORC à Identification à Reactance à Quit smoking
intention
.0109
.0065
[.0014, .0261]
ORC à Identification à Reactance à Self-efficacy
expectations
.0115
.0068
[.0013, .0280]
ORC à Identification à Reactance à Perceived
effectiveness of the preventive response
.0121
.0057
[.0026, .0249]
ORC à Identification à Counterarguing à Quit
smoking intention
.0037
.0048
[-.0042, .0153]
ORC à Identification à Counterarguing à Self-
efficacy expectations
.0026
.0053
[-.0072, .0144]
ORC à Identification à Counterarguing à
Perceived effectiveness of the preventive response
.0129
.0062
[.0028, .0269]
ORC à Transportation à Reactance à Quit
smoking intention
.0073
.0058
[-.0011, .0211]
ORC à Transportation à Reactance à Self-
efficacy expectations
.0074
.0059
[-.0013, .0216]
ORC à Transportation à Reactance à Perceived
effectiveness of the preventive response
.0079
.0054
[-.0013, .0202]
ORC à Transportation à Counterarguing à Quit
smoking intention
.0034
.0038
[-.0019, .0127]
ORC à Transportation à Counterarguing à Self-
efficacy expectations
.0009
.0035
[-.0056, .0092]
ORC à Transportation à Counterarguing à
Perceived effectiveness of the preventive response
.0080
.0055
[-.0013, .0205]
Note. - The independent variable (X3 with PROCESS) was coded with the values 0 = being exposed to a third-
person narrative with a low behavioral similarity protagonist (reference category or control group) and 1 = being
exposed to a narration written in the first person with a high behavioral similarity protagonist (ORC). The table
shows the indirect effects. An indirect effect is considered to be statistically significant if the established
confidence interval (95% CI) does not include the value 0. If the value 0 is included in the confidence interval,
that means that the indirect effects is equal to 0, that is, there is no an association between the variables involved
(Hayes, 2018). Statistically significant indirect effects are marked in bold. Covariates: X1, X2, gender and age.
Discussion and conclusions
The present study provides evidence
on how to improve smoking prevention
interventions targeting adults. It has been
shown that variables related to the
construction of prevention messages (in
particular, the joint effect of narrative voice
and audience-character behavioral
similarity) have a significant effect on quit
smoking intention, on self-efficacy
expectations and on the perceived
effectiveness of the preventive response,
since they initiate empathic involvement
processes, which in turn reduce the
resistance towards the persuasive message.
Furthermore, the present study
presents two innovations compared to
previous studies: the concepts of behavioral
similarity and optimal reception condition.
In the past, similarity had been manipulated
only in demographic traits, such as gender,
nationality or ethnic origin, and results were
inconsistent (Cohen et al., 2018; Kaufman
& Libby, 2012; Kim, 2019; Tukachinsky,
2014). In the present study we have

Narrative devices for smoking prevention
16
demonstrated that is possible to effectively
manipulate behavioral similarity by
highlighting that the narrative’s protagonist
shares a “common story” with the audience
(in this case, about tobacco dependence),
since it has been proven that this
manipulation increases perceived
similarity. The second original contribution
of this study (within the area of narrative
persuasion for smoking prevention) is the
concept of optimal reception condition
(first-person narrative with a protagonist
with high behavioral similarity to the
audience), since previous studies had
analyzed the effect of such variables
(similarity and narrative voice) separately
(e.g., De Graaf et al., 2016).
Therefore, and in line with
hypothesis 1, it was found that the optimal
reception condition induced greater levels
of identification with the protagonist (H1a),
but it did not significantly increase narrative
transportation (H1b). Regarding the
mediation analyses (H2), the optimal
reception condition had three significant
indirect effects on the dependent variables,
which were due to the increase in
identification and the reduction in
reactance. Moreover, the optimal reception
condition had also a statistically significant
indirect effect on the perceived
effectiveness of the preventive response,
which is explained by the increase in
identification and the reduction of
counterarguing.
These results are consistent with
previous studies showing that first-person
narratives, by being perceived as more
personal (since they promote closeness
between the reader and the narrative’s
protagonist) and easier to understand,
increased identification with the protagonist
(Chen et al., 2016, 2017; Christy, 2018; De
Graaf et al., 2012; Kaufman & Libby, 2012;
Nan et al., 2015). However, the optimal
reception condition had a marginally non-
significant effect on narrative
transportation. This result is not far from
previous studies, such as the one by
Banjerjee and Greene (2012), which found
no impact of first or third-person narratives
on narrative transportation. Likewise, the
meta-analysis by Tukachinsky (2014)
concluded that narrative voice influences
identification, but not narrative
transportation. On the other hand, our study
showed that both narrative transportation
and identification reduced reactance, and
these findings are consistent with the E-
ELM (Slater & Rouner, 2002) and EORM
(Moyer-Gusé, 2008) theoretical models,
which are widely applied in narrative health
communication. In fact, our study has noted
that both narrative transportation and
identification lead to lower levels of
counterarguing. Therefore, according to
previous research, and as it was proposed in
our second hypothesis (H2a), identification
with the protagonist was a mediator
variable, but narrative transportation did not
act as a significant mediator (Cohen, Tal-
Or, & Mazor-Treerman, 2015; De Graaf et
al., 2012; Green & Brock, 2000; Hoeken &
Fikkers, 2014; Igartua, 2017; Igartua &
Barrios, 2012; Walter et al., 2018).
One of the limitations of this study
is not having properly controlled (in the
narrative design) the effect of demographic
similarity, particularly the effect of
protagonist’s gender and age. Although the
meta-analysis review by Tukachinsky
(2014) concluded that objective similarity
(based on sociodemographic traits, such as
gender, age or ethnic origin) has no
significant effect on identification (but it
has on narrative transportation and
perceived similarity), this systematic
review was based on 48 experimental
studies on narrative persuasion in general.
Therefore, future studies should verify
whether this result could be generalized to
smoking prevention.
The results obtained allow us to
think about different applications in the
field of tobacco prevention and treatment.
Firstly, the primary means of prevention of
smoking risks are the health warnings on
cigarette packets, which can be avoided and
caused reactance (Rodríguez-Contreras, &
Igartua, 2018). Therefore, we suggest that

Narrative devices for smoking prevention
17
narrative smoking prevention campaigns
using the first-person voice and designed to
stimulate audience-character behavioral
similarity should be implemented, since the
present study has shown that these variables
had a significant (indirect) effect on the quit
smoking intention, self-efficacy
expectations, and on the perceived
effectiveness of the preventive response.
This is possible because these variables
initiate mechanisms of affective connection
with the message (in this case, identification
with the characters) that reduce reactance.
We consider that a narrative
intervention such as the one suggested
could be aimed at tobacco treatment, but it
could also become a primary prevention
tool, since any attempt to reduce tobacco
consumption indirectly aims to prevent
health problems or diseases related to
smoking. In this regard, as with the Tips
from Former Smokers campaign
(developed in the United Stated), a former
smoker’s experience (i.e., someone who has
overcome their tobacco addiction) would be
used both for preventing smoking (that is,
preventing people from taking up smoking)
and for helping current smokers to quit, and
thus to prevent smoking damage.
Secondly, one strategy that is
becoming increasingly present in the field
of health and, specifically, in the smoking
prevention field, is the development of
mobile applications to quit smoking
(Lacoviello, Steinerman, Klein, Silver,
Berger, Luo, & Schork, 2017; Ubhi,
Michie, Kotz, Wong, & West, 2015). In this
regard, it would be interesting to develop
mobile applications for smoking prevention
in which the user could customize the avatar
in relation to their behavioral similarity.
Thus, we could analyze whether the
possibility of being able to choose a
character with similar characteristics makes
the application more effective. This type of
applications could be particularly relevant
for smoking prevention among young
people and adolescents, since they
intensively use smartphones.
In conclusion, the present study
yields relevant insights for health-related
narrative persuasion research by showing
how certain narrative devices (such as the
combined used of a first-person message
and the inclusion of elements to increase
behavioral similarity with the protagonist)
can be useful for improving smoking
prevention and treatment.
References
American Cancer Society. (2018). Health
risks of smoking tobacco. Retrieved from
https://www.cancer.org/cancer/cancer-
causes/tobacco-and-cancer/health-risks-
of-smoking-tobacco.html#references
Appel, M., Gnambs T., Richter T., & Green
M. C. (2015). The transportation scale–
short form (TS–SF). Media Psychology,
18(2), 243-266. doi:
10.1080/15213269.2014.987400
Banerjee, S. C., & Greene, K. (2012). Role
of transportation in the persuasion
process: cognitive and affective
responses to antridrug narratives.
Journal of Health Communication,
17(5), 564-581. doi:
10.1080/10810730.2011.635779
Bolkan, S., Goodboy, A. K., & Myers, S. A.
(2017). Conditional processes of
effective instructor communication and
increases in students’ cognitive learning.
Communication Education, 66(2), 129–
147. doi:
10.1080/03634523.2016.1241889
Braddock, K., & Dillard J. D. (2016). Meta-
analytic evidence for the persuasive
effect of narratives on beliefs, attitudes,
intentions, and behaviors.
Communication Monographs, 83(4),
446–467. doi:
10.1080/03637751.2015.1128555
Busselle, R., & Bilandzic, H. (2009).
Measuring narrative engagement. Media
Psychology, 12(4), 321–347. doi:
10.1080/15213260903287259
Chen, M., Bell, R. A., & Taylor, L. D.
(2016). Narrator point of view and
persuasion in health narratives: the role

Narrative devices for smoking prevention
18
of protagonist–reader similarity,
identification, and self-referencing.
Journal of Health Communication,
21(8), 908–918. doi:
10.1080/10810730.2016.1177147
Chen, M., Bell, R. A., & Taylor L. D.
(2017). Persuasive effects of point of
view, protagonist competence, and
similarity in a health narrative about type
2 diabetes. Journal of Health
Communication, 22(8), 702–712. doi:
10.1080/10810730.2017.1341568
Chen, M., McGlone, M. S., & Bell, R. A.
(2015). Persuasive effects of linguistic
agency assignments and point of view in
narrative health messages about colon
cancer. Journal of Health
Communication, 20(8), 977–988. doi:
10.1080/10810730.2015.1018625
Christy, K. R. (2018). I, you, or he:
examining the impact of point of view on
narrative persuasion. Media Psychology,
21(4), 700–718. doi:
10.1080/15213269.2017.1400443
Cohen, J. (2001). Defining identification: a
theoretical look at the identification of
audiences with media characters. Mass
Communication & Society, 4(3), 245–
264. doi:
10.1207/S15327825MCS0403_01
Cohen, J., Tal-Or, N., & Mazor-Tregerman,
M. (2015). The tempering effect of
transportation: exploring the effects of
transportation and identification during
exposure to controversial two-sided
narratives. Journal of Communication,
65(2), 237–258. doi:
10.1111/jcom.12144
Cohen, J., Weimann-Saks, D., & Mazor-
Tregerman, M. (2018). Does character
similarity increase identification and
persuasion? Media Psychology, 21(3),
506–528. doi:
10.1080/15213269.2017.1302344
Christy, K. R. (2018). I, you, or he:
examining the impact of point of view on
narrative persuasion. Media Psychology,
21(4), 700–718. doi:
10.1080/15213269.2017.1400443
De Graaf, A., Hoeken, H., Sanders, J., &
Beentjes, J. W. J. (2012). Identification
as a mechanism of narrative persuasion.
Communication Research, 39(6), 802–
823. doi: 10.1177/0093650211408594
De Graaf, A., Sanders, J., & Hoeken, H.
(2016). Characteristics of narrative
interventions and health effects: a review
of the content, form, and context of
narratives in health-related narrative
persuasion research. Review of
Communication Research, 4, 88–131.
doi: 10.12840/issn.2255-
4165.2016.04.01.011
De Graaf, A., van den Putte, B., Nguyen, M.
H., Zebregs, S., Lammers, J., & Neijens,
P. (2017). The effectiveness of narrative
versus informational smoking education
on smoking beliefs, attitudes and
intentions of low-educated adolescents,
Psychology & Health, 32(7), 810–825.
doi: 10.1080/08870446.2017.1307371
Dunlop, S. M., Wakefield, M., & Kashima,
Y. (2010). Pathways to persuasion:
Cognitive and experiential responses to
health-promoting mass media messages.
Communication Research, 37(1), 133–
164. doi: 10.1177/0093650209351912
Faul, F., Erdfelder, E., Lang, A. G., &
Buchner, A. (2007). G*Power 3: a
flexible statistical power analysis
program for the social, behavioural, and
biomedical sciences. Behavioural
Research Methods, 39, 175–191.
doi:10.3758/BF03193146
Frank, L. B., Murphy, S. T., Chatterjee, J.
S., Moran, M. B., & Baezconde-
Garbanati, L. (2015). Telling stories,
saving lives: creating narrative health
messages. Health Communication,
30(2), 154–163. doi:
10.1080/10410236.2014.974126
Green, M. C. (2006). Narratives and cancer
communication. Journal of
Communication, 56(1), S163–S183. doi:
10.1111/j.1460-2466.2006.00288.x
Green, M. C., & Brock, T. C. (2000). The
role of transportation in the
persuasiveness of public narratives.
Journal of Personality and Social

Narrative devices for smoking prevention
19
Psychology, 79(5), 701–721. doi:
10.1037/0022-3514.79.5.701
Green, M. C., & Brock, T. C. (2002). In the
mind’s eye: transportation-imagery
model of narrative persuasion. In M. C.
Green, J. J. Strange, & T. C. Brock
(Eds.), Narrative impact. Social and
cognitive foundations (pp. 315–341).
Mahwah, NJ: Lawrence Erlbaum
Associates.
Hayes, A. F. (2018). Introduction to
mediation, moderation, and conditional
process analysis. A regression-based
approach. New York, NY: Guilford
Press.
Hayes, A. F., & Preacher, K. J. (2014).
Statistical mediation analysis with a
multicategorical independent variable.
British Journal of Mathematical and
Statistical Psychology, 67(3), 451–470.
doi: 10.1111/bmsp.12028
Heatherton, T. F., Kozlowski, L. T.,
Frecker, R. C., & Fagerström, K. O.
(1991). The Fagerström test for nicotine
dependence. A revision of the
Fagerström tolerance questionnaire.
British Journal of Addictions, 86, 1119–
1127. doi: 10.1111/j.1360-
0443.1991.tb01879.x
Hoeken, H., & Fikkers, K. M. (2014). Issue-
relevant thinking and identification as
mechanisms of narrative persuasion.
Poetics, 44, 84–99. doi:
10.1016/j.poetic.2014.05.001
Igartua, J. J. (2010). Identification with
characters and narrative persuasion
through fictional feature films.
Communications, 35(4), 347–373. doi:
10.1515/comm.2010.019
Igartua, J. J., & Barrios, I. (2012). Changing
real-world beliefs with controversial
movies: processes and mechanisms of
narrative persuasion. Journal of
Communication, 62(3), 514–531. doi:
10.1111/j.1460-2466.2012.01640.x
Igartua, J. J., & Vega, J. (2016).
Identification with characters,
elaboration, and counterarguing in
entertainment-education interventions
through audiovisual fiction. Journal of
Health Communication, 21(3), 293–300.
doi: 10.1080/10810730.2015.1064494
Jensen, J. D., Yale, R. N., Krakow, M.,
John, K. K., & King, A. J. (2017).
Theorizing foreshadowed death
narratives: examining the impact of
character death on narrative processing
and skin self-exam intentions. Journal of
Health Communication, 22(1), 84–93.
doi: 10.1080/10810730.2016.1252816
Kaufman, G. F., & Libby, L. K. (2012).
Changing beliefs and behavior through
experience-taking. Journal of
Personality and Social Psychology,
103(1), 1–19. doi: 10.1037/a0027525
Kim, M. (2019). When similarity strikes
back: conditional persuasive effects of
character-audience similarity in anti-
smoking campaign. Human
Communication Research, 45(1), 52–77.
doi: 10.1093/hcr/hqy013
Kim, H. S., Bigman, C. A., Leader, A. E.,
Lerman, C., & Cappella, J. N. (2012).
Narrative health communication and
behavior change: the influence of
exemplars in the news on intention to
quit smoking. Journal of
Communication, 62(3), 473–492. doi:
10.1111/j.1460-2466.2012.01644.x
Kim, H. K., & Lee, T. K. (2017).
Conditional effects of gain-loss framed
narratives among current smokers at
different stages of change. Journal of
Health Communication, 22(12), 990–
998, doi:
10.1080/10810730.2017.1396629
Kim, M., Shi, R., & Cappella J. N. (2016).
Effect of character–audience similarity
on the perceived effectiveness of
antismoking PSAs via engagement.
Health Communication, 31(10), 1193–
1204. doi:
10.1080/10410236.2015.104821
Kreuter, M. W., Green, M. C., Cappella, J.
N., Slater, M. D., Wise, M. E., Storey,
D., Clark, E. M., O’Keefe, D. J., Erwin,
D. O., Holmes, K., Hinyard, L. J.,
Houston, T., & Woolley, S. (2007).
Narrative communication in cancer
prevention and control: a framework to

Narrative devices for smoking prevention
20
guide research and application. Annals of
Behavioral Medicine, 33(3), 221–235.
doi: 10.1007/BF02879904
Lacoviello, B. M., Steinerman, J. R., Klein,
D. B., Silver, T. L., Berger, A. G., Luo,
S. X., & Schork, N. J. (2017). Clickotine,
a personalized smartphone app for
smoking cessation: initial evaluation.
JMIR mHealth and uHealth, 5(4), e56.
doi: 10.2196/mhealth.7226
McDonald, D. G. (2014). Narrative
research in communication: key
principles and issues. Review of
Communication Research, 2(1), 115–
132. doi: 10.12840/issn.2255-
4165.2014.02.01.005
McQueen, A., Waters, E. A., Kaphingst, K.
A., Caburnay, C. A., Sanders Thompson,
V. L., Boyum, S., & Kreuter, M. W.
(2016). Examining interpretations of
graphic cigarette warning labels among
us youth and adults. Journal of Health
Communication, 21(8), 855–867. doi:
10.1080/10810730.2016.1177142
Moreno, J. J., & García, F. J. (2000).
Factores asociados al éxito en programas
para dejar de fumar. Adicciones, 12(3),
329–341. doi: 10.20882/adicciones.644
Moyer-Gusé, E. (2008). Toward a theory of
entertainment persuasion: explaining the
persuasive effects of entertainment-
education messages. Communication
Theory, 18(3), 407–425. doi:
10.1111/j.1468-2885.2008.00328.x
Moyer-Gusé, E., & Nabi, R. L. (2010).
Explaining the effects of narrative in an
entertainment television program:
overcoming resistance to
persuasion. Human Communication
Research, 36(1), 26–52. doi:
10.1111/j.1468-2958.2009.01367.x
Nan, X., Dahlstrom, M. F., Richards, A., &
Rangarajan, S. (2015). Influence of
evidence type and narrative type on HPV
risk perception and intention to obtain
the HPV vaccine. Health
Communication, 30(3), 301–308. doi:
10.1080/104102336.2014.888629
Nan, X., Futerfas, M., & Ma, Z. (2017).
Role of narrative perspective and
modality in the persuasiveness of public
service advertisements promoting HPV
vaccination. Health
Communication, 32(3), 320–328. doi:
10.1080/10410236.2016.1138379
Niederdeppe, J., Heley, K., & Barry, C. L.
(2015). Inoculation and narrative
strategies in competitive framing of three
health policy issues. Journal of
Communication, 65(5), 838–862.
doi:10.1111/jcom.12162
Niederdeppe, J., Kim, H. K. Lundell, H.,
Fazili, F., & Frazier, B. (2012). Beyond
counterarguing: simple elaboration,
complex integration, and
counterelaboration in response to
variations in narrative focus and
sidedness. Journal of
Communication, 62(5), 758–777. doi:
10.1111/j.1460-2466.2012.01671.x
O’Keefe, D. J. (2003). Message properties,
mediating states and manipulation
checks: claims, evidence, and data
analysis in experimental persuasive
message effects research.
Communication Theory, 13, 251–274.
doi: 10.1111/j.1468-
2885.2003.tb00292.x
Rains, S. A. (2013). The nature of
psychological reactance revisited: A
meta-analytic review. Human
Communication Research, 39(1), 47–73.
doi:10.1111/j.1468-2958.2012.01443.x
Rodríguez-Contreras, L., & Igartua, J. J.
(2018). Apelaciones al miedo en
advertencias sanitarias para la
prevención del tabaquismo: ¿una imagen
vale más que mil palabras? Revista
Española de Comunicación en Salud,
9(2), 138–151. doi:
10.20318/recs.2018.4492
Shen, F., Sheer, V. C., & Li, R. (2015).
Impact of narratives on persuasion in
health communication: A meta-analysis.
Journal of Advertising, 44(2), 105–113.
doi: 10.1080/00913367.2015.1018467
Shen, L. (2015). Targeting smokers with
empathy appeal antismoking public
service announcements: a field
experiment. Journal of Health

Narrative devices for smoking prevention
21
Communication, 20(5), 573–580. doi:
10.1080/10810730.2015.1012236th
Shen, L., Seung, S., Andersen K. K., &
McNeal, D. (2017). The psychological
mechanisms of persuasive impact from
narrative communication. Studies in
Communication Sciences, 17(2), 165–
181. doi: 10.24434/j.scoms.2017.02.003
Slater, M. D., & Rouner, D. (2002).
Entertainment-education and elaboration
likelihood: understanding the processing
of narrative persuasion. Communication
Theory, 12(2), 173–191. doi:
10.1111/j.1468-2885.2002.tb00265.x
Spek, V., Lemmens, F., Chatrou, M.,
Kempen, S., Pouwer, F., & Pop, V.
(2013). Development of a smoking
abstinence self-efficacy
questionnaire. International Journal of
Behavioral Medicine, 3(20), 444–449.
doi: 10.1007/s12529-012-9229-2
Thoemmes, F. (2015). Reversing arrows in
mediation models does not distinguish
plausible models. Basic and Applied
Social Psychology, 37(4), 226–234. doi:
10.1080/01973533.2015.1049351
Thompson, T., & Kreuter, M. W. (2014).
Using written narratives in public health
practice: a creative writing perspective.
Preventing Chronic Disease, 11, e94.
doi: 10.5888/pcd11.130402
Thrasher, J. F., Arillo-Santillán, E.,
Villalobos, V., Pérez-Hernández, R.,
Hammond, D., Carter, J., Sebrie, E.,
Sansores, R., & Regalado-Piñeda, J.
(2012). Can pictorial warning labels on
cigarette packages address smoking-
related health disparities? Field
experiments in Mexico to assess pictorial
warning label content. Cancer Causes &
Control, 23(1), 69–80. doi:
10.1007/s10552-012-9899-8
Tukachinsky, R. (2014). Experimental
manipulation of psychological
involvement with media.
Communication Methods and Measures,
8(1), 1–33. doi:
10.1080/19312458.2013.873777
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), e17.
doi: 10.2196/jmir.3479
Van Krieken, K., Hoeken, H., & Sanders, J.
(2017). Evoking and measuring
identification with narrative characters: a
linguistic cues framework. Frontiers in
Psychology, 8, e1190. doi:
10.3389/fpsyg.2017.01190
Walter, N., Murphy, S. T., & Gillig, T. K.
(2018). To walk a mile in someone else’
s shoes: how narratives can change
causal attribution through story
exploration and character customization.
Human Communication Research,
44(1), 31–57. doi: 10.1111/hcre.12112
Wehbe, M. S., Basil, M., & Basil, D.
(2017). Reactance and coping responses
to tobacco counter-advertisements.
Journal of Health Communication,
22(7), 576–583. doi:
10.1080/10810730.2017.1329853
Williams, J. H., Green, M. C., Kohler, C.,
Allison, J. J., & Houston, T. K. (2011).
Stories to communicate risks about
tobacco: development of a brief scale to
measure transportation into a video
story: the ACCE project. Health
Education Journal, 70(2), 184–191. doi:
10.1177/0017896910373171
World Health Organization. (2017).
Tobacco. Recuperado en
http://www.who.int/mediacentre/factshe
ets/fs339/en/.
Zebregs, S., Van den Putte, B., Neijens, P.,
& De Graaf, A. (2015). The differential
impact of statistical and narrative
evidence on beliefs, attitude, and
intention: a meta-analysis. Health
Communication, 30(3), 282–289. doi:
10.1080/10410236.2013.842528

Narrative devices for smoking prevention
22
Appendix – Narratives (Spanish versions)
First-person narrative and a protagonist with low tobacco dependence (N1)
Mi nombre es Miguel, tengo 35 años y empecé a fumar a los 15. A los pocos meses ya
fumaba a diario, aunque nunca fui de esas personas que se encienden un cigarro recién
levantadas, no estaba tan enganchado, sólo fumaba 9 o 10 cigarrillos al día. Durante casi veinte
años fumé prácticamente a diario, y me daba igual que me dijeran que el tabaco mata. Yo era
muy joven para pensar en eso. Hace un año sufrí un amago de infarto. Esto me hizo entender
que ya no era tan joven y que mi cuerpo ya no respondía como antes. Me asusté. El médico me
dijo que lo primero que tenía que hacer era dejar de fumar para que mi cuerpo se recuperara y
que con el tiempo sería como si nunca hubiera fumado.
Las primeras semanas fueron muy difíciles. Me dolía muchísimo la cabeza, tenía
ansiedad y me costaba mucho concentrarme en el trabajo (no podía parar de pensar en
encenderme un cigarro, en darle una última calada). Además, como estaba todo el día ansioso
e inquieto, me costaba dormir. Creía que no iba a ser capaz, sólo el miedo a sufrir un infarto me
hacía tirar para adelante. A los tres meses empecé a ver algunos cambios en mi cuerpo: le sacaba
sabor a la comida, mi olfato también mejoró e incluso notaba que respiraba mejor. Pensé que
hacer deporte me vendría bien y así fue… me sentía más relajado, ya no tenía problemas para
dormir, mi tensión arterial bajó y desapareció por completo la tos que siempre tenía al
levantarme. Además, empecé a ahorrar, más de 80 euros al mes, al dejar de comprar tabaco.
Ya hace un año que dejé de fumar. Ahora me siento mucho mejor, más libre: ya no
tengo que estar pendiente de tener tabaco, ya no tengo la necesidad de fumarme un cigarro para
“relajarme”, o de esperar hasta el descanso del trabajo para fumarme un cigarro.
Sé que esto sólo es el principio y que para que los pulmones estén completamente
limpios tienen que pasar diez años, pero no me voy a rendir, no quiero volver a fumar un cigarro.
Gracias a dejar de fumar mi salud ha mejorado y ha disminuido el riesgo de volver a sufrir un
infarto. Todo esto hace que no quiera volver a fumar, ya no me atrae, incluso cuando veo a otras
personas fumar no me llama. Esto me da aún más fuerzas para no volver a fumar. Mi estilo de
vida ha cambiado y el tabaco se ha convertido en algo incompatible con mi día a día.

Narrative devices for smoking prevention
23
First-person narrative and a protagonist with high tobacco dependence (N2)
Mi nombre es Miguel, tengo 35 años y empecé a fumar a los 15. A los pocos meses ya
no podía empezar el día sin fumarme un par de cigarrillos, incluso antes de desayunar. Estaba
muy enganchado y fumaba más de una cajetilla de tabaco al día. Durante casi veinte años fumé
a diario, incluso cuando estaba enfermo, y me daba igual que me dijeran que el tabaco mata.
Yo era muy joven para pensar en eso. Hace un año sufrí un amago de infarto. Esto me hizo
entender que ya no era tan joven y que mi cuerpo ya no respondía como antes. Me asusté. El
médico me dijo que lo primero que tenía que hacer era dejar de fumar para que mi cuerpo se
recuperara y que con el tiempo sería como si nunca hubiera fumado.
Las primeras semanas fueron muy difíciles. Me dolía muchísimo la cabeza, tenía
ansiedad y me costaba mucho concentrarme en el trabajo (no podía parar de pensar en
encenderme un cigarro, en darle una última calada). Además, como estaba todo el día ansioso
e inquieto, me costaba mucho dormir. Creía que no iba a ser capaz, sólo el miedo a sufrir un
infarto me hacía tirar para adelante. A los tres meses empecé a ver algunos cambios en mi
cuerpo: le sacaba sabor a la comida, mi olfato también mejoró e incluso notaba que respiraba
mejor. Pensé que hacer deporte me vendría bien y así fue… me sentía más relajado, ya no tenía
problemas para dormir, mi tensión arterial bajó y desapareció por completo la tos que siempre
tenía al levantarme. Además, empecé a ahorrar, más de 150 euros al mes, al dejar de comprar
tabaco.
Ya hace un año que dejé de fumar. Ahora me siento mucho mejor, más libre: ya no soy
un esclavo del tabaco que hurga en la basura de casa en busca de colillas, ya no me despierto
por las noches para fumar y tampoco me levanto pensando en ese primer cigarro.
Sé que esto sólo es el principio y que para que los pulmones estén completamente
limpios tienen que pasar diez años, pero no me voy a rendir, no quiero volver a caer en el vicio.
Gracias a dejar de fumar mi salud ha mejorado y ha disminuido el riesgo de volver a sufrir un
infarto. Todo esto hace que no quiera volver a fumar, ya no me atrae, incluso cuando veo a otras
personas fumar no me llama. Esto me da aún más fuerzas para no volver a fumar. Mi estilo de
vida ha cambiado y el tabaco se ha convertido en algo incompatible con mi día a día.

Narrative devices for smoking prevention
24
Third-person narrative and a protagonist with low tobacco dependence (N3)
Miguel tiene 35 años y empezó a fumar a los 15. A los pocos meses ya fumaba a diario,
aunque nunca fue de esas personas que se encienden un cigarro recién levantadas, no estaba tan
enganchado, sólo fumaba 9 o 10 cigarrillos al día. Durante casi veinte años fumó prácticamente
a diario, y le daba igual que le dijeran que el tabaco mata. Miguel era muy joven para pensar en
eso. Hace un año sufrió un amago de infarto. Esto le hizo entender que ya no era tan joven y
que su cuerpo ya no respondía como antes. Se asustó. El médico le dijo que lo primero que
tenía que hacer era dejar de fumar para que su cuerpo se recuperara y que con el tiempo sería
como si nunca hubiera fumado.
Las primeras semanas fueron muy difíciles para Miguel. Le dolía muchísimo la cabeza,
tenía ansiedad y le costaba mucho concentrarse en el trabajo (no podía parar de pensar en
encenderse un cigarro, en darle una última calada). Además, como estaba todo el día ansioso e
inquieto, le costaba dormir. Creía que no iba a ser capaz, sólo el miedo a sufrir un infarto le
hacía tirar para adelante. A los tres meses empezó a ver algunos cambios en su cuerpo: le sacaba
sabor a la comida, su olfato también mejoró e incluso notaba que respiraba mejor. Pensó que
hacer deporte le vendría bien y así fue… se sentía más relajado, ya no tenía problemas para
dormir, su tensión arterial bajó y desapareció por completo la tos que siempre tenía al
levantarse. Además, empezó a ahorrar, más de 80 euros al mes, al dejar de comprar tabaco.
Ya hace un año que Miguel dejó de fumar. Ahora se siente mucho mejor, más libre: ya
no tiene que estar pendiente de tener tabaco, ya no tiene la necesidad de fumarse un cigarro
para “relajarse”, o de esperar hasta el descanso del trabajo para fumarse un cigarro.
Miguel sabe que esto sólo es el principio y que para que sus pulmones estén
completamente limpios tienen que pasar diez años, pero no se va a rendir, no quiere volver a
fumar un cigarro. Gracias a dejar de fumar su salud ha mejorado y ha disminuido el riesgo de
volver a sufrir un infarto. Todo esto hace que no quiera volver a fumar, ya no le atrae, incluso
cuando ve a otras personas fumar no le llama. Esto le da aún más fuerzas para no volver a
fumar. Su estilo de vida ha cambiado y el tabaco se ha convertido en algo incompatible con su
día a día.

Narrative devices for smoking prevention
25
Third-person narrative and a protagonist with high tobacco dependence (N4)
Miguel tiene 35 años y empezó a fumar a los 15. A los pocos meses ya no podía empezar
el día sin fumarse un par de cigarrillos, incluso antes de desayunar. Estaba muy enganchado y
fumaba más de una cajetilla de tabaco al día. Durante casi veinte años fumó a diario, incluso
cuando estaba enfermo, y le daba igual que le dijeran que el tabaco mata. Miguel era muy joven
para pensar en eso. Hace un año sufrió un amago de infarto. Esto le hizo entender que ya no era
tan joven y que su cuerpo ya no respondía como antes. Se asustó. El médico le dijo que lo
primero que tenía que hacer era dejar de fumar para que su cuerpo se recuperara y que con el
tiempo sería como si nunca hubiera fumado.
Las primeras semanas fueron muy difíciles para Miguel. Le dolía muchísimo la cabeza,
tenía ansiedad y le costaba mucho concentrarse en el trabajo (no podía parar de pensar en
encenderse un cigarro, en darle una última calada). Además, como estaba todo el día ansioso e
inquieto, le costaba mucho dormir. Creía que no iba a ser capaz, sólo el miedo a sufrir un infarto
le hacía tirar para adelante. A los tres meses empezó a ver algunos cambios en su cuerpo: le
sacaba sabor a la comida, su olfato también mejoró e incluso notaba que respiraba mejor. Pensó
que hacer deporte le vendría bien y así fue… se sentía más relajado, ya no tenía problemas para
dormir, su tensión arterial bajó y desapareció por completo la tos que siempre tenía al
levantarse. Además, empezó a ahorrar, más de 150 euros al mes, al dejar de comprar tabaco.
Ya hace un año que Miguel dejó de fumar. Ahora se siente mucho mejor, más libre: ya
no es un esclavo del tabaco que hurga en la basura de casa en busca de colillas, ya no se despierta
por las noches para fumar y tampoco se levanta pensando en ese primer cigarro.
Miguel sabe que esto sólo es el principio y que para que sus pulmones estén
completamente limpios tienen que pasar diez años, pero no se va a rendir, no quiere volver a
caer en el vicio. Gracias a dejar de fumar su salud ha mejorado y ha disminuido el riesgo de
volver a sufrir un infarto. Todo esto hace que no quiera volver a fumar, ya no le atrae, incluso
cuando ve a otras personas fumar no le llama. Esto le da aún más fuerzas para no volver a
fumar. Su estilo de vida ha cambiado y el tabaco se ha convertido en algo incompatible con su
día a día.