ArticlePDF Available

“It Happened to Me and It’s Serious”: Conditional Indirect Effects of Infection Severity Narrated in Testimonial Tweets on COVID-19 Prevention

MDPI
International Journal of Environmental Research and Public Health (IJERPH)
Authors:

Abstract and Figures

The health crisis caused by COVID-19 resulted in societal breakdowns around the world. Our research is based on determining which features of testimonial messages are most relevant in increasing persuasive impact. An online experiment with a 2 (severity infection narrative: low vs. high) × 2 (infection target: narrative’s protagonist vs. protagonist’s father) between-subject factorial design was carried out. Young people between 18 and 28 years (N = 278) were randomly assigned to one of the four experimental conditions, where they were asked to read a narrative message in the form of a Twitter thread describing a COVID-19 infection (with mild or severe symptoms) that affected either the protagonist of the message (a 23-year-old young person) or their father. After reading the narrative message, the mediating and dependent variables were evaluated. A message describing a severe COVID-19 infection affecting their protagonist to increase the perception of personal risk increased the persuasive impact through an increase in cognitive elaboration and a reduction in reactance. Our study highlights that creating persuasive messages based on social media targeted at young people that describe a careless behavior resulting in a severe COVID-19 infection can be an appropriate strategy for designing prevention campaigns.
Content may be subject to copyright.
Citation: Igartua, J.-J.; Rodríguez-
Contreras, L.; Guerrero-Martín, Í.;
Honorato-Vicente, A. “It Happened
to Me and It’s Serious”: Conditional
Indirect Effects of Infection Severity
Narrated in Testimonial Tweets on
COVID-19 Prevention. Int. J. Environ.
Res. Public Health 2023,20, 6254.
https://doi.org/10.3390/
ijerph20136254
Academic Editor: Xiaoquan Zhao
Received: 28 April 2023
Revised: 16 June 2023
Accepted: 28 June 2023
Published: 29 June 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
International Journal of
Environmental Research
and Public Health
Article
“It Happened to Me and It’s Serious”: Conditional Indirect
Effects of Infection Severity Narrated in Testimonial Tweets on
COVID-19 Prevention
Juan-JoséIgartua * , Laura Rodríguez-Contreras , Íñigo Guerrero-Martín and Andrea Honorato-Vicente
Department of Sociology and Communication, Faculty of Social Sciences, Campus Unamuno (Edificio FES),
University of Salamanca, 37007 Salamanca, Spain; laurarodriguezcontreras@usal.es (L.R.-C.);
i.guerrero@usal.es (Í.G.-M.); andreahonorato@usal.es (A.H.-V.)
*Correspondence: jigartua@usal.es; Tel.: +0034-670-801-074; Fax: +0034-923-294-573
Abstract:
The health crisis caused by COVID-19 resulted in societal breakdowns around the world.
Our research is based on determining which features of testimonial messages are most relevant in
increasing persuasive impact. An online experiment with a 2 (severity infection narrative: low vs.
high)
×
2 (infection target: narrative’s protagonist vs. protagonist’s father) between-subject factorial
design was carried out. Young people between 18 and 28 years (N = 278) were randomly assigned
to one of the four experimental conditions, where they were asked to read a narrative message in
the form of a Twitter thread describing a COVID-19 infection (with mild or severe symptoms) that
affected either the protagonist of the message (a 23-year-old young person) or their father. After
reading the narrative message, the mediating and dependent variables were evaluated. A message
describing a severe COVID-19 infection affecting their protagonist to increase the perception of
personal risk increased the persuasive impact through an increase in cognitive elaboration and a
reduction in reactance. Our study highlights that creating persuasive messages based on social media
targeted at young people that describe a careless behavior resulting in a severe COVID-19 infection
can be an appropriate strategy for designing prevention campaigns.
Keywords:
narrative persuasion; COVID-19; testimonial messages; Twitter; health communication;
cognitive processes
1. Introduction
Testimonial messages have become a popular strategy for designing prevention cam-
paigns in the field of health communication [
1
4
]. A testimonial message describes a
personal story in which the protagonist comments on their situation or experience in a
health-related context. For example, the CDC in the United States, in its “Tips From Former
Smokers” campaign (https://www.cdc.gov/tobacco/campaign/tips/index.html, accessed
on 12 April 2021) developed a series of testimonial messages (written and in video format)
where former smokers with different profiles or characteristics described how they started
smoking and the consequences this had on their own health, highlighting the problems they
personally suffered from (e.g., cancer). In this case, and in other similar cases (as well as in
connection with other risky behaviors), the consequences of unhealthy behaviors (such as
smoking) on the appearance of health problems (for example, developing cancer) in the
person who is the protagonist of the message are described. Additionally, a loss-framing
approach is adopted, seeking to elicit negative emotions in the message audience (such
as fear or worry) and induce a greater perception of personal risk (“if I smoke, I may get
sick”), activate the severity perception (“smoking has serious consequences for health”)
and stimulate self-protective behaviors (for example, “I will try to quit smoking in the next
three months”).
Int. J. Environ. Res. Public Health 2023,20, 6254. https://doi.org/10.3390/ijerph20136254 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2023,20, 6254 2 of 22
However, our work focuses on a different context: COVID-19 prevention. The health
crisis caused by COVID-19 resulted in societal breakdowns around the world, infecting to
date more than 700 million people and causing more than 6 million deaths worldwide [
5
].
Due to its rapid transmission, infection rate and mortality in those who contracted it, the
World Health Organization (WHO) declared a pandemic situation on 11 March 2020 [
6
].
Without a vaccine available, the virus could only be curbed through changes in daily behav-
iors (quarantine) and social coordination (social distancing, the use of masks, disinfection of
buildings and mobility and transportation restrictions) to reduce the number of COVID-19
infections and avoid overloading healthcare systems [79].
Some of the earliest research during the pandemic showed that there were large
differences in people’s ability to adopt measures to reduce virus transmission [
10
]. In this
sense, it can be stated that COVID-19 has posed a communication challenge on how to
approach prevention. It is well known that COVID-19 infection occurs when an infected
person comes into contact with an uninfected person [
11
]. For example, a nurse working
in a hospital (lacking appropriate equipment) could become infected simply by being in
contact with an infected person whom they have to attend to. It follows that many of the
COVID-19 prevention measures were based on social distancing, asking the population
to avoid risky situations such as attending celebrations, social gatherings or large events
(such as going to the theater or cinema). Moreover, contagion could occur, even if people
protected themselves by reducing social contact, as long as there was a person in their
social network who did not follow this pattern. Thus, a young person living with their
parents could act as a vector of contagion within their family if they did not follow the
recommendation to avoid social contact. For example, if this young person went to a party
with friends and became infected, they could later infect their parents.
The challenge posed by COVID-19 at the level of medical treatment was, therefore,
joined by the challenge faced by health communication professionals in designing pre-
vention campaigns. Specifically, one of the most critical elements was to determine how
health campaigns should be designed to raise awareness and encourage people to avoid
risky social contacts or to refrain from attending celebrations, parties or similar events in
enclosed spaces. The work presented here contributes to health communication research
and focuses on analyzing the most effective strategies for preventing COVID-19, a viral
infection that is easily transmitted through social contact. Given that COVID-19 spreads
when an infected person exhales droplets and very small respiratory particles containing
the virus [
11
], it is vitally important to avoid contact with infected people, knowing that
many people may have the virus and not show symptoms. Therefore, it is recommended
to avoid crowded places and maintain a social distance.
We present the results of an online experiment in which testimonial messages for
the prevention of COVID-19 were created by manipulating two independent variables.
The testimonial message, written in the first person and designed as a Twitter “thread”,
featured a young man who indicated having celebrated their birthday by organizing a
party with their friends without adopting any preventive measures. As a consequence of
such a party, a COVID-19 infection affecting the young person themselves or a relative
with whom they lived, their father (target of the infection), was mentioned. On the other
hand, the narrative described the symptoms experienced by the infected person as mild
or severe (severity of infection). In this context, the current study draws on research on
narrative persuasion, focusing specifically on the role of identification with the protagonist,
narrative transportation, reactance and cognitive elaboration. The aim is to shed light
on the psychological mechanisms underlying the impact of message characteristics on
preventive measures, including the perceived personal risk of contracting COVID-19,
perceived severity of the disease and intentions to adopt preventive behaviors.
Int. J. Environ. Res. Public Health 2023,20, 6254 3 of 22
2. Narrative Persuasion with Testimonial Messages: Advancing the Study of
Psychological Mechanisms
Empirical evidence has demonstrated the effectiveness of narrative messages for health
communication [
1
,
12
14
]). Narrative messages represent a communication format consist-
ing of a temporal sequence of causally linked events involving one or more characters [
15
].
Meta-analysis reviews have found that narrative messages produce significant effects on
attitudes, beliefs, behavioral intention and behaviors, although significant variation in
these effects has also been observed [
16
,
17
]. This means that it is necessary to determine
which features of narrative messages are most relevant for eliciting a series of psychological
processes that ultimately lead to an impact on measures related to risk perception, severity
perception and the adoption of preventive behaviors.
The main theoretical models of narrative persuasion [
18
20
] have established that
narrative messages induce persuasive effects on individuals through a set of mechanisms,
such as identification with the protagonist, narrative transportation and reactance [
18
20
].
Identification with the protagonist is defined as a multidimensional construct linked
to emotional empathy, cognitive empathy, the feeling of merging with the character and
adopting their goals [
21
,
22
]. It constitutes a psychological phenomenon whereby audience
members mentally adopt the perspective of the protagonist of the narrative [
23
]. Identi-
fication allows individuals to overcome the natural tendency to limit their thoughts and
feelings to a single perspective [
21
]. Therefore, research in this field has focused on deter-
mining the factors that increase identification with the protagonist of a message, as this
can impact the persuasive effectiveness of messages [
24
]. In this sense, it has been found
that characters with positive attributes generate greater identification than characters with
negative attributes [25,26].
In the context of our research, the testimonial message manipulated who was infected
with COVID-19 (either the young protagonist narrating the story or their father). However,
both versions of the story mentioned that the infection had resulted from the protagonist’s
imprudent behavior. Engaging in imprudent behavior that could result in a family mem-
ber’s illness can be regarded as a negative attribute of the story’s protagonist. Therefore, it
could be expected that identification with the protagonist of the message would be lower
when the person infected by COVID-19 was the protagonist’s father, especially when the
symptoms caused by the disease were severe. This led to the first hypothesis of the study:
H1.
When the narrative describes that the person infected with COVID-19 is the protagonist’s
father, there will be lower identification with the protagonist, especially when the symptoms caused
by the illness are severe.
Narrative transportation is a psychological process that involves a state of immersion
in the story and integrates three components: the focusing of attention on the story, af-
fective induction and the formation of highly vivid mental images of the characters and
situations described in the message. It is also one of the main mechanisms that explain how
stories persuade people [
27
]. Narrative transportation can be influenced by the emotional
tone of the message [
24
]. Indeed, narrative messages are especially powerful when they
evoke strong emotions, as narrative transportation is affected by the emotional tone of the
message [
28
]. In the context of the present study, the severity of symptoms associated with
COVID-19 described in the testimonial message was manipulated. It could be expected that
the version describing severe symptoms (such as high fever and difficulty breathing) com-
pared to the one that described mild symptoms (such as “some cough” or mild headache)
would trigger greater emotional engagement that would manifest itself in experiencing
greater narrative transportation. This led to the second hypothesis of the study:
H2.
The narrative message describing a COVID-19 infection with severe symptoms will induce a
higher degree of narrative transportation compared to the one describing mild symptoms.
Reactance is a process associated with resistance to persuasion attempts that is acti-
vated when the individual perceives that their freedom of choice is being threatened [
29
,
30
].
Int. J. Environ. Res. Public Health 2023,20, 6254 4 of 22
Moyer-Gusé[
19
] indicates that narrative messages elicit low levels of reactance because
people become immersed in the story. Thus, reducing reactance weakens any critical stance
or attitude towards the message, leading to an effective persuasive impact. However, in a
recent study on the acquisition of preventive behaviors during the pandemic, it was ob-
served that a greater perception of threat to freedom was linked to higher reactance, which,
in turn, was associated with lower levels of adherence to preventive behavior in the face of
COVID-19 [
31
]. In this context, it has been suggested that guilt is a mechanism that holds
great persuasive power [
32
]. Guilt is also linked to transgression and the negative effects
that irresponsible or inappropriate behavior can have on others. However, it is important to
explore the conditions under which the impact of guilt-based appeals leads to persuasion
and their relationship to reactance. In this regard, Bessaravoba et al. [
33
] conducted a
study with high school students (aged 16–18 years) and observed that persuasive messages
designed to induce guilt resulted in a decrease in their effectiveness by increasing reactance.
In fact, the authors of this study concluded that the use of guilt-based appeals in media
campaigns aimed at adolescents could be counterproductive.
Taking as reference the role of guilt and reactance and their effect on the younger
population, we proposed that reactance may be an important process in the prevention
of COVID-19 through testimonial messages. We assumed that a narrative that attributed
to the protagonist the responsibility for the COVID-19 infection of a family member (the
protagonist’s father) would induce greater reactance than a narrative that alluded that the
target of the contagion was the person who committed the irresponsible behavior, and
this effect would be greater when the narrative mentioned that the infection symptoms
were severe.
H3.
A narrative describing that the person infected with COVID-19 is the protagonist’s father will
produce higher reactance, especially when the infection is described as severe.
According to the extended parallel processing model (EPPM), which has previously
been applied to explain the effects of fear appeals, the degree to which a person feels
threatened by a health problem determines their motivation to act preventively. Thus,
the greater the perceived threat, the greater that motivation [
34
36
]. In the context of our
research, we assumed that the narrative message describing a severe COVID-19 infection
would induce a greater perception of severity than the message describing a mild infection,
especially when the contagion target was the protagonist of the story. In a certain sense,
a narrative message that alluded to a severe infection and that the person suffering from
the disease was the protagonist of the story could be considered a fear appeal and, for this
reason, we assumed that this could increase the perceived severity of the infection narrated
in the testimonial. In this context, the fourth hypothesis was established:
H4.
The perceived severity of the infection narrated in the message will be higher when a severe
(versus mild) infection is described and the target of the contagion is the protagonist of the story
(versus their father).
A recent study found that the components of the EPPM (perceived efficacy and
perceived threat) positively influenced the intention to stay home during the COVID-19
pandemic [
37
]. Likewise, that study also found that the relationship between the perceived
efficacy of the preventive response and the intention to stay home was moderated by
perceived threat. Thus, people who assessed the virus as more severe and perceived
preventive behaviors as effective in avoiding virus spread were more likely to adopt
preventive behaviors against COVID-19. Similar results have been obtained in other studies
that highlight the relevance of the perceived severity of the infection in the prevention of
COVID-19 [
38
,
39
]. These findings suggest that messages that increase the perceived severity
of the infection narrated in the message can motivate people to engage in preventive
behaviors for themselves, thus, reducing the spread of the virus (see also [40]).
Therefore, we assumed that for the message to exert a significant effect on prevention
measures, it had to elicit the perception that the COVID-19 infection narrated in the message
Int. J. Environ. Res. Public Health 2023,20, 6254 5 of 22
was serious and that the person suffering from it was the young protagonist. This process, in
turn, would activate different routes to explain the persuasive impact: through identification
with the protagonist, through narrative transportation and through reactance. However, in
this work, we considered a fourth mechanism: cognitive elaboration [
41
43
]. Cognitive
elaboration is defined as a process of reflection on the content of the message and constitutes
a measure of the intensity of such reflection during the reception process [
44
]. Cognitive
elaboration is the mechanism that has been least explored to date in narrative persuasion
research in health communication (see [
45
]), despite the fact that health messages are
designed to stimulate active cognitive processing in individuals. In this sense, we ventured
that the experience narrated in the message could serve as an inspiration [
46
] and stimulate
deep cognitive processing in people so that they questioned their previous opinions and
adjusted their attitudes towards the pandemic and prevention behaviors.
At this point, we held the assumption that in order for a narrative message to have
a persuasive impact, it had to trigger a sequence of psychological processes that acted as
mediator mechanisms. In this context, the proposed relevant mediators were the perception
of the severity of the infection described in the message (primary mediator), identifica-
tion with the protagonist, narrative transportation, reactance and cognitive elaboration
(secondary mediators). In this given context, it was proposed that when the testimonial
narrative described a severe COVID-19 infection (as opposed to a mild one), in which the
person telling the story was directly involved (the young protagonist versus their father),
the perceived severity of the infection would increase. This heightened perception, in turn,
would be associated with a greater identification with the protagonist, increased narrative
transportation, lower reactance and greater cognitive elaboration. In turn, these processes
would be associated with a greater perceived personal risk of contracting COVID-19, a
greater perceived severity of the disease and a greater intention to engage in preventive
behavior. Therefore, our fifth hypothesis proposed a moderated serial–parallel mediation
model (see Figure 1):
Int. J. Environ. Res. Public Health 2023, 20, 6254 6 of 23
Figure 1. Hypothesized moderated serial–parallel mediation model.
3. Method
3.1. Participants
A total of 435 individuals participated in the present study. However, the sample was
reduced to 278 persons who met the following criteria: aged between 18 and 28, residing
in Spain and not having had a past or current COVID-19 infection. In addition, the nal
selection of the sample also took into account a series of quality controls that were
described below (see Design and Procedure). Convenience sampling was used by
distributing the link to the experimental questionnaire designed with Qualtrics through
major media social platforms (Facebook, Twier, Instagram, WhatsApp) and via email.
The link to the experimental questionnaire remained active from 19 May to 1 June 2021.
Participants who were included in the nal sample had a mean age of 23.14 years (SD =
2.90), with 70.5% being women, 27.7% being men and 1.8% indicating another response
(see Table 1).
Table 1. Characteristics of the study participants (N = 278).
Variables Mean (SD) or Percentage Range
Age
M
= 23.14
SD = 2.90
18–28
Sex Male: 77 (27.7%)
Female: 196 (70.5%)
Other response: 5 (1.8%)
Perceived knowledge about
COVID-19
M
= 3.25
SD = 0.67
1 (low)5 (high)
Level of personal concern about
the COVID-19 pandemic
situation
M
= 6.85
SD = 1.75
0 (low)10 (high)
Level of concern for a possible
COVID-19 infection
M
= 3.53
SD = 0.94
1 (low)5 (high)
Do any close relatives have or
have had the COVID-19?
No: 147 (52.9%)
Yes: 131 (47.1%)
3.2. Design and Procedure
An online experiment with a 2 (severity infection narrative: low vs. high) × 2
(infection target: narrative’s protagonist vs. protagonists father) between-subject factorial
design was carried out. Participants were randomly assigned to one of the four
experimental conditions, where they were asked to read a narrative message in the form
Figure 1. Hypothesized moderated serial–parallel mediation model.
H5.
It is hypothesized that there will be significant indirect effects on the severity of the infection
narrative on several outcomes related to COVID-19 prevention, namely, perceived personal risk
of COVID-19 infection (H5a), perceived severity of the disease (H5b) and intention to engage
in preventive behavior (H5c). These effects are expected to be mediated by several psychological
processes, namely, the perception of severity of the infection narrated in the story, identification
with the protagonist, narrative transportation, reactance and cognitive elaboration. However, such
indirect effects will only manifest when the infected individual is the protagonist of the story.
Int. J. Environ. Res. Public Health 2023,20, 6254 6 of 22
3. Method
3.1. Participants
A total of 435 individuals participated in the present study. However, the sample was
reduced to 278 persons who met the following criteria: aged between 18 and 28, residing
in Spain and not having had a past or current COVID-19 infection. In addition, the final
selection of the sample also took into account a series of quality controls that were described
below (see Section 3.2). Convenience sampling was used by distributing the link to the
experimental questionnaire designed with Qualtrics through major media social platforms
(Facebook, Twitter, Instagram, WhatsApp) and via email. The link to the experimental
questionnaire remained active from 19 May to 1 June 2021. Participants who were included
in the final sample had a mean age of 23.14 years (SD = 2.90), with 70.5% being women,
27.7% being men and 1.8% indicating another response (see Table 1).
Table 1. Characteristics of the study participants (N = 278).
Variables Mean (SD) or Percentage Range
Age M = 23.14
SD = 2.90
18–28
Sex Male: 77 (27.7%)
Female: 196 (70.5%)
Other response: 5 (1.8%)
Perceived knowledge about COVID-19 M = 3.25
SD = 0.67
1 (low)–5 (high)
Level of personal concern about the
COVID-19 pandemic situation
M = 6.85
SD = 1.75
0 (low)–10 (high)
Level of concern for a possible
COVID-19 infection
M = 3.53
SD = 0.94
1 (low)–5 (high)
Do any close relatives have or have had
the COVID-19?
No: 147 (52.9%)
Yes: 131 (47.1%)
3.2. Design and Procedure
An online experiment with a 2 (severity infection narrative: low vs. high)
×
2 (infection
target: narrative’s protagonist vs. protagonist’s father) between-subject factorial design
was carried out. Participants were randomly assigned to one of the four experimental
conditions, where they were asked to read a narrative message in the form of a Twitter
thread describing a COVID-19 infection (with mild or severe symptoms) that affected either
the protagonist of the message (a 23-year-old young person) or their father.
The questionnaire used was divided into three main blocks: pretest measures, the
narrative messages in the form of a Twitter thread, and post-test measures. The first block
included an introduction and informed consent. In addition, demographic information
was requested (gender, age, country of residence) and included measures on the degree
of perceived knowledge about COVID-19 (how they would rate their level of knowledge
about coronavirus), the level of personal concern about the COVID-19 pandemic situation
(to what extent they felt concerned about the current situation of the COVID-19 pandemic),
the level of concern for a possible COVID-19 infection (how often they worried about the
possibility of contracting coronavirus) and the question “Do any close relatives have or
have had the COVID-19?”.
Immediately after completing the pretest measures, participants were randomly as-
signed to the experimental conditions, which involved reading a testimonial message. Since
two aspects of the message were manipulated (severity infection narrative and infection
target), four different versions of the same message were constructed. All versions showed
a similar beginning and ending, and only the elements related to the severity of the disease
symptoms and the infection target varied. The developed messages (see Independent
Int. J. Environ. Res. Public Health 2023,20, 6254 7 of 22
Variables and Stimulus Materials) were between 213 and 219 words in total, and the esti-
mated reading time was calculated to be 90 s (with a text readability tool, https://legible.es,
accessed on 19 April 2021).
After reading the testimonial message, participants completed post-test measures
to assess the mediating variables (perceived severity infection, identification with the
protagonist, narrative transportation, cognitive elaboration, reactance) and dependent
variables (perceived personal risk of COVID-19 infection, perceived severity of COVID-19
and protective behavioral intent against COVID-19). In addition, participants were asked a
number of questions to evaluate their retention of key details from the narrative, such as
the name and age of the protagonist and who in the story had contracted COVID-19.
Since Qualtrics allowed for a series of quality controls to be implemented, the ques-
tionnaire was designed in such a way that it could only be completed in a single session.
The average duration to complete the entire process (pretest, message reading and post-test)
was approximately 10 min (M = 9.93 min, SD = 4.55, Mdn = 8.67), and only participants
who had completed the questionnaire between 6 and 45 min were considered. In addition,
only the results from participants who took between 30 and 180 s to read the message
(
M = 53.74 s
, SD = 20.61, Mdn = 48.85) were counted as valid cases. Moreover, only partic-
ipants who remembered the central details of the testimonial message were considered
valid cases.
All materials related to the online experiment (testimonial messages, measures, datasets
and syntax files) are available via the Open Science Framework (OSF): https://osf.io/cz3dt/
(accessed on 27 April 2023).
3.3. Independent Variables and Stimulus Materials
The two experimental manipulations were applied to a first-person testimonial nar-
rative message. The message was constructed on the Twitter platform, specifically as a
Twitter thread consisting of 7 tweets, each no longer than 280 characters. The credibility
and realism of the message were maximized by simulating the interface of the Twitter
platform and preserving the structure of the social network (see Supplementary Materials
in the OSF).
The testimonial was written in an informal and youthful tone to grant the message
greater realism and credibility, with particular attention paid to using gender-neutral
language that did not provide clues about the protagonist’s gender. In this sense, prior
to the experiment, a pilot study was carried out (N = 28) to select a unisex name for the
protagonist of the narrative message. A unisex name is one that is valid for both male and
female individuals and does not impose a gender identity. The protagonist’s gender was
intended to be neutral in order to avoid any gender-based biases in the reception of the
narrative. Participants were asked to select the most suitable unisex name from a list of
15 names, taking into account the Spanish context. Based on the results of the pilot study,
Alex (35.7%) was selected as the name of the protagonist in the Twitter testimonial narrative.
The testimonial message elaborated for this study had a clear and causal structure, in
such a way that a series of events were presented that were connected in space and time
(see Table 2). It was narrated in the first person, since previous research had concluded that
the use of the first person was more effective than the second- or third-person narrative
voice [
42
,
47
]. The message’s protagonist was a 23-year-old individual named Alex Sanchez
(@sanchez98_alex). The message described a COVID-19 infection due to risky behavior
and the resulting consequences.
Int. J. Environ. Res. Public Health 2023,20, 6254 8 of 22
Table 2. Testimonial messages used as experimental stimuli.
Severity Infection Narrative: Low
-Infection Target: Narrative’s Protagonist (215 Words)
Severity Infection Narrative: High
-Infection Target: Narrative’s Protagonist (224 Words)
Today I wanted to talk to you about the situation I’m going
through at the moment. A thread.
Like most of my friends, I had always thought that coronavirus
was not something I had to worry about. I thought it was like a
flu, nothing more. That’s why my social life had changed little
in the last few months.
I turned 23 last week and of course, I threw a party to celebrate.
We weren’t too many, but we didn’t pay much attention to
safety. Being with your friends makes you forget about COVID,
masks, distance . . .
Two days after the party I started to feel sick. I had the test, a
PCR, and they told me I had COVID. The news broke me.
I’ve been in bed for four days, with fever, headache, congestion,
and some coughing. Let’s see how the disease evolves in the
next few days.
The doctor has said that between the fifth and eighth day of the
infection is when everything can change and determine if the
evolution is going to get worse, so I don’t know what will
happen and it scares me.
Now I think this could have been avoided. I see this COVID
thing is more serious than I thought. Take care, protect
yourselves, we are not immortal.
Today I wanted to talk to you about the situation I’m going
through at the moment. A thread.
Like most of my friends, I had always thought that coronavirus
was not something I had to worry about. I thought it was like a
flu, nothing more. That’s why my social life had changed little
in the last few months.
I turned 23 last week and of course, I threw a party to celebrate.
We weren’t too many, but we didn’t pay much attention to
safety. Being with your friends makes you forget about COVID,
masks, distance . . .
Two days after the party I started to feel sick. I had the test, a
PCR, and they told me I had COVID. The news broke me.
I have been in the hospital for four days, with a very high fever
and difficulty breathing, they even put me on oxygen. Let’s see
how the disease evolves in the next few days.
The doctor has said that between the fifth and eighth day of the
infection is when everything can change and determine if the
evolution is going to get worse, so I don’t know what will
happen and it scares me.
Now I think this could have been avoided. I see this COVID
thing is more serious than I thought. Take care, protect
yourselves, we are not immortal.
Severity Infection Narrative: Low
-Infection Target: Protagonist’s Father (218 Words)
Severity Infection Narrative: High
-Infection Target: Protagonist’s Father (226 Words)
Today I wanted to talk to you about the situation I’m going
through at the moment. A thread.
Like most of my friends, I had always thought that coronavirus
was not something I had to worry about. I thought it was like a
flu, nothing more. That’s why my social life had changed little
in the last few months.
I turned 23 last week and of course, I threw a party to celebrate.
We weren’t too many, but we didn’t pay much attention to
safety. Being with your friends makes you forget about COVID,
masks, distance . . .
Two days after the party my father started to feel sick. He had
the test, a PCR, and they told him he had COVID. The news
broke me.
My father has been in bed for four days, with fever, headache,
congestion, and some coughing. Let’s see how the disease
evolves in the next few days.
The doctor has said that between the fifth and eighth day of the
infection is when everything can change and determine if the
evolution is going to get worse, so I don’t know what will
happen and it scares me.
Now I think this could have been avoided. I see this COVID
thing is more serious than I thought. Take care, protect
yourselves, we are not immortal.
Today I wanted to talk to you about the situation I’m going
through at the moment. A thread.
Like most of my friends, I had always thought that coronavirus
was not something I had to worry about. I thought it was like a
flu, nothing more. That’s why my social life had changed little
in the last few months.
I turned 23 last week and of course, I threw a party to celebrate.
We weren’t too many, but we didn’t pay much attention to
safety. Being with your friends makes you forget about COVID,
masks, distance . . .
Two days after the party my father started to feel sick. He had
the test, a PCR, and they told him he had COVID. The news
broke me.
My father has been in the hospital for four days, with a very
high fever and difficulty breathing, they even put him on
oxygen. Let’s see how the disease evolves in the next few days.
The doctor has said that between the fifth and eighth day of the
infection is when everything can change and determine if the
evolution is going to get worse, so I don’t know what will
happen and it scares me.
Now I think this could have been avoided. I see this COVID
thing is more serious than I thought. Take care, protect
yourselves, we are not immortal.
The protagonist of the testimonial began by stating that, like most of his friends, they
thought that coronavirus was nothing to worry about, that it was like a flu, and that was
why their social life had changed little in the last few months. They then recounted that they
had celebrated a party with friends to celebrate their birthday and had not paid attention
to COVID-19 prevention measures (mask use and social distancing). The use of the “party”
factor as a trigger for the infection and transmission of the virus was inspired by the short
film “Compromiso en 60 Segundos”, directed by Willy Suárez, winner of the first prize
Int. J. Environ. Res. Public Health 2023,20, 6254 9 of 22
in the Cultura Inquieta microfilm contest and which went viral on social media in Spain
at the end of 2020 (https://youtu.be/YolGcDOkL7E, accessed on 8 February 2021). In a
third Twitter thread, the protagonist narrated that several days after the party, they (or their
father) began to feel unwell, and a PCR test confirmed that it was a COVID-19 infection.
They then described the symptoms (mild versus severe) that they (or their father) had
four days after the COVID-19 infection was confirmed. The mild symptoms mentioned
were having a fever, headache, congestion and “some cough”, and it was emphasized that
they had been in bed for four days. In contrast, the version referring to severe symptoms
mentioned having a very high fever and difficulty breathing; additionally, it was stated
that they had been in the hospital for four days and that medical treatment with oxygen
had been necessary. The message also indicated that the doctor had said that the infection
could vary between the fifth and eighth day, so anything could be expected to happen,
thus, creating a scenario of uncertainty. Faced with this situation, the protagonist expressed
feeling fear and concern, stating that they thought that all of this could have been avoided,
now valuing that COVID-19 was more serious than they had thought, and ending with
a prevention message directed at young people: “Take care, protect yourselves, we are
not immortal”.
3.4. Measures
Perceived infection target (manipulation check). To contrast the effectiveness of the experi-
mental manipulation related to the infection target (who was the person suffering from the
COVID-19 infection), two items were included: “the person affected by COVID-19 who
narrated the message was a young person” and “the person affected by COVID-19 who
narrated the message had become infected due to imprudence” (from 1 = strongly disagree
to 7 = strongly agree).
Perceived severity infection. An ad hoc scale was created consisting of two items: “the
person infected with COVID-19 experienced severe symptoms” and “it is very unlikely
that the disease described in the message will eventually endanger the life of the person
infected with COVID-19” (from 1 = strongly disagree to 7 = strongly agree). Both items
showed a negative correlation (r(276) =
0.18, p<.001) and were combined (after recoding
the second item) to form an index of perceived severity infection (M = 4.90, SD = 1.28). This
index was used as an outcome variable to test H4 and as a mediating variable to test H5
(see [48]).
Identification with the protagonist. Identification was assessed using an 11-item scale [
49
]
that measured the degree of identification with a specific character (e.g., “I felt as if I were
Alex”; from 1 = not at all to 5 = very much). The 11 items were averaged into a reliable
scale (α= 0.88, M = 2.87, SD = 0.75).
Narrative transportation. It was assessed by means of the transportation scale—short
form [
50
]—which consisted of 5 items (e.g., “I was mentally involved in the narrative while
reading it”; from 1 = strongly disagree to 7 = strongly agree). The 5 items were averaged
into a reliable scale (α= 0.74, M = 4.18, SD = 1.17).
Cognitive elaboration. An adapted version of the cognitive elaboration scale developed
by Igartua and Rodríguez-Contreras [
42
] was used, consisting of four items (e.g., “while
reading the narrative, I intensely reflected on the topic of the coronavirus”; from 1 = strongly
disagree to 7 = strongly agree). The 4 items were averaged into a reliable scale (
α
= 0.84,
M = 4.66, SD = 1.28).
Reactance. This was assessed with the perceived threat to freedom scale created by
Shen (2015) [
51
], comprising 4 items (e.g., “the message was trying to pressure me”; from
1 = strongly disagree to 7 = strongly agree). A reactance index was constructed from
calculating the average across the four items (α= 0.84, M = 2.06, SD = 1.29).
Perceived personal risk of COVID-19 infection. Taking as a reference the study conducted
by Jahangiry et al. [
52
], participants were asked to indicate their degree of agreement or
disagreement with the statement “my risk of contracting coronavirus is very high” (from
1 = strongly disagree to 7 = strongly agree; M = 3.34, SD = 1.53).
Int. J. Environ. Res. Public Health 2023,20, 6254 10 of 22
Perceived severity of COVID-19. A scale composed of 5 items (e.g., “I think coronavirus
is more severe than influenza”, “COVID-19 can cause serious health problems”; from
1 = strongly disagree to 7 = strongly agree) was developed based on two previous
studies [
52
,
53
]. A perceived severity of COVID-19 index was constructed from calculating
the average across the five items (α= 0.71, M = 6.04, SD = 0.82).
Protective behavioral intent against COVID-19. To assess the participants’ protective
behaviors, the following question was included (based on the study by
Rosero-Bolaños et al. [
54
]): “on a scale of 0 (not at all likely) to 10 (very likely), what
is the probability that you will not attend parties and/or meetings in the next 4 weeks, for
fear of infecting others with coronavirus (COVID-19)” (M = 6.46, SD = 2.99).
Recall of the details of the narrative. To assess the retention of essential information from
the testimonial message, participants were asked the following questions at the end of the
questionnaire: a) “What was the name of the person who was the protagonist of the story
you read?” (1 = I do not remember; 2 = Dani; 3 = Alex (correct); 4 = Rosa); b) “How old
was the protagonist of the message?” (1 = I do not remember; 2 = 18 years old; 3 = 23 years
old (correct): 4 = 28 years old); c) “who was the person who, according to the message,
had health problems after being infected with coronavirus?” (1 = I do not remember;
2 = the person who narrated the message; 3 = a friend of the person who narrated the
message; 4 = the father of the person who narrated the message; 5 = the mother of the
person who narrated the message).
3.5. Data Analysis
Data analyses were conducted using IBM SPSS 28 statistical software. Descriptive
analyses were calculated to examine the sample demographics and basic statistical in-
formation of the measures (means and standard deviations). The internal consistency
reliability was calculated for all self-report scales composed of 3 or more items using Cron-
bach’s alpha coefficient. A one-way analysis of variance (ANOVA), Student’s t-test and
chi-squared test were used to check for successful randomization and to test the efficacy
of experimental manipulations. Bivariate correlations between the mediating and depen-
dent variables were analyzed with Pearson’s product moment coefficient of correlation. A
factorial ANOVA was performed to determine the impact of the severity of the infection
on narrative transportation (H2), including the infection target as a second independent
variable. To test hypotheses 1, 3, 4 and 5, the PROCESS macro (version 4.3) for SPSS was
used. This macro made it possible to test different moderation, mediational and condi-
tional process models [
55
]. To carry out the analyses with PROCESS, the independent
variables (severity infection and infection target) were coded
0.5 (for “low-severity in-
fection narrative” and for “infection target: protagonist’s father”, respectively) and 0.5
(for “high-severity infection narrative” and for “infection target: narrative’s protagonist”,
respectively). This coding approach is called the main effect parameterization because of
the use of
0.5 and 0.5 for coding the values of the focal predictor and the values of the
moderating variable that produces regression coefficient estimates that correspond to the
main effects of each independent variable from a 2
×
2 ANOVA [
56
]. To test H1, H3 and
H4, the PROCESS macro was used, applying model 1 (simple moderation). To test H5
(a moderated serial–parallel mediation model), the PROCESS macro was used, applying
a customized conditional process model (10,000 bootstrapping samples to generate 95%
confidence intervals using the percentile method). According to the bootstrapping method,
an indirect effect is statistically significant if the confidence interval established (CI at 95%)
does not include the value 0. If the value 0 is included in the CI, the indirect effect is equal
to 0, that is, there is no association between the variables considered.
4. Results
4.1. Preliminary Analysis
The randomization was successful: the conditions did not differ significantly on
gender (
χ2
(6, N = 278) = 3.37, p= 0.761), age (F(3, 274) = 1.65, p= 0.287), duration (in
Int. J. Environ. Res. Public Health 2023,20, 6254 11 of 22
minutes) of participation in the experiment (F(3, 274) = 0.29, p= 0.839), reading time of the
testimonial message (F(3, 274) = 0.38, p= 0.764), perceived knowledge about COVID-19
(F(3, 274) = 1.65, p= 0.178), level of personal concern about the COVID-19 pandemic
situation (F(3, 274) = 0.09, p= 0.961) and level of concern for a possible COVID-19 infection
(F(3, 274) = 0.27, p= 0.844). There were also no statistically significant differences between
experimental conditions in the percentage of participants with close family members
who currently had, or have had in the past, COVID-19 infection. (
χ2
(3, N = 278) = 1.13,
p= 0.769).
The experimental manipulation of the severity of the infection was effective. Par-
ticipants exposed to the high-severity message (M = 5.44, SD = 1.36) showed a higher
degree of agreement with the statement “the person infected with COVID-19 experienced
severe symptoms” than those exposed to the low-severity message (M = 4.34, SD = 1.63;
t(255.89) =
6.03, p< 0.001; Levene’s test F = 11.96, p< 0.001). In addition, persons exposed
to the low-severity narrative showed a higher degree of agreement (M = 3.37, SD = 1.78)
with the statement “it is very unlikely that the disease will end up endangering the life of
the COVID-19 infected person referred to in the message” compared to persons exposed to
the high-severity narrative (M = 2.86, SD = 1.65; t(276) = 2.49, p= 0.007).
The experimental manipulation of the infection target was also effective. Participants
exposed to the message referring to the protagonist’s infection showed a higher degree of
agreement with the statement “the person affected by COVID-19 who narrates the message
was a young person” (M = 6.77, SD = 0.68) than those exposed to the message referring
to the protagonist’s father’s infection (M = 5.03, SD = 2.17; t(176.07) =
9.20, p< 0.001;
Levene’s test F = 220.18, p< 0.001). Moreover, those exposed to the message alluding to
the protagonist’s infection showed a higher degree of agreement with the statement “the
person affected by COVID-19 who narrates the message had been infected by an imprudent
action” (M = 6.36, SD = 1.10) than those exposed to the message alluding to the protagonist’s
father’s infection (M = 5.71, SD = 1.69; t(251.64) =
3.81, p< 0.001; Levene’s test F = 17.77,
p< 0.001).
Correlations between the mediating variables and the dependent variables were
also analyzed (see Table 3). This analysis confirmed that the mediating processes showed
convergent correlations with each other. For example, the perceived severity of the infection
was associated with greater cognitive processing. In addition, the mediating processes were
also significantly associated with the dependent variables (e.g., a positive and significant
correlation was observed between cognitive elaboration and perceived personal risk).
Table 3. Descriptive analysis and correlations between mediating and dependent variables.
1 2 3 4 5 6 7 8
1 Perceived severity infection - - - - - - - -
2 Identification 0.10 * - - - - - - -
3 Narrative transportation 0.17 ** 0.73 *** - - - - - -
4 Cognitive elaboration 0.23 *** 0.46 *** 0.45 *** - - - - -
5 Reactance 0.15 ** 0.02 0.04 0.10 * - - - -
6 Perceived personal risk of
COVID-19 infection 0.04 0.24 *** 0.22 *** 0.25 *** 0.01 - - -
7 Perceived severity
of COVID-19 0.17 *** 0.16 ** 0.15 ** 0.27 *** 0.26 *** 0.16 ** - -
8 Protective behavioral intent
against COVID-19 0.02 0.09 + 0.13 * 0.06 0.17 *** 0.02 0.14 ** -
Mean 4.90 2.87 4.18 4.66 2.06 3.34 6.04 6.45
Standard deviation 1.28 0.75 1.17 1.28 1.19 1.53 0.82 2.99
Note. N = 278. For all the variables, a higher score indicates a greater intensity of the considered process, from
1 = low to 7 = high (except for the identification scale, which has a theoretical range from 1 = low to 5 = high).
+p< 0.10; * p< 0.05; ** p< 0.01; *** p< 0.001.
Int. J. Environ. Res. Public Health 2023,20, 6254 12 of 22
4.2. H1: The Impact of the Infection Target on Identification with the Protagonist Is Moderated by
the Severity of the COVID-19 Infection
H1 posited that the narrative alluding to a COVID-19 infection of the protagonist’s
father would induce less identification with the protagonist than the narrative alluding
to an infection on the protagonist himself, specifically when the infection was described
as severe. The PROCESS macro (Model 1) was used to test this hypothesis of modera-
tion, which yielded a non-statistically significant interaction effect (B = 0.11, SE = 0.18,
p= 0.533). In addition, neither statistically significant main effects of the infection target
(B =
0.11, SE = 0.09, p= 0.216) nor of the severity infection described in the message
(B = 0.13, SE = 0.18, p= 0.620) were observed on identification with the protagonist. There-
fore, H1 did not receive empirical support.
4.3. H2: Narrative Describing an Infection with Severe Symptoms Induces Greater Narrative
Transportation Than the One That Describe Mild Symptoms
H2 proposed that a message mentioning a severe COVID-19 infection would result
in greater narrative transportation compared to a message referring to a mild infection.
Although no main effect was predicted for the infection target (nor an interaction effect
between the two independent variables), an ANOVA was conducted regardless. The results
showed that the severity of the COVID-19 infection did not significantly influence the
degree of narrative transportation (F(1, 274) = 0.63, p= 0.427). The message describing
severe symptoms (M = 4.23, SE = 0.09) induced a similar level of narrative transporta-
tion as the message describing mild symptoms (M = 4.12, SE = 0.10). Thus, H2 did not
receive empirical support. There was no statistically significant main effect observed
for the infection target (F(1, 274) = 3.34, p= 0.068) nor an interaction effect between the
two independent variables (F(1, 274) = 1.11, p= 0.291).
4.4. H3: The Impact of the Infection Target on Reactance Is Moderated by the Severity of the
COVID-19 Infection
According to H3, the narrative mentioning that the person infected with COVID-19
was the protagonist’s father would induce greater reactance compared to the narrative
mentioning that the infected person was the young person, especially if the infection was
described as severe. The PROCESS macro (model one) was used to test this hypothesis
of moderation, which yielded a non-statistically significant interaction effect (B =
0.32,
SE = 0.31, p= 0.292). In addition, neither statistically significant main effects of the infection
target (B =
0.00, SE = 0.15, p= 0.994) nor of the severity of the infection described in the
message (B = 0.16, SE = 0.15, p= 0.292) were observed on reactance. Therefore, H3 was not
empirically supported.
4.5. H4: The Effect of the Severity of the COVID-19 Infection on Perceived Severity Is Moderated
by the Infection Target
H4 predicted that the narrative message describing a severe infection would lead to
a greater perceived severity, especially when the infection target was the protagonist of
the story. Therefore, we expected to find an interaction effect between the severity of the
infection described in the message and the infection target on the perception of the severity
of the infection narrated in the story. The PROCESS macro (Model 1) was used to test
this hypothesis of moderation, which yielded a statistically significant interaction effect
(B = 0.90, SE = 0.28, p= 0.001). The conditional effect analysis (see Figure 2) revealed that
the severity of the infection symptoms described in the message significantly increased
the perceived severity only when the infection target was the narrative’s protagonist
(
θXY | (infection target = “narrative’s protagonist)
= 1.25, SE = 0.20, p< 0.001), but not when
it was the protagonist’s father (
θXY | (infection target = protagonist’s father)
= 0.34, SE = 0.19,
p= 0.073). In conclusion, the severity of the infection increased the perceived severity of
the infection narrated in the message only when the infection target was the protagonist of
the testimonial. These results provided empirical support for H4.
Int. J. Environ. Res. Public Health 2023,20, 6254 13 of 22
Int. J. Environ. Res. Public Health 2023, 20, 6254 14 of 24
protagonist (θ
X
Y
|
(infection
target
=
“narrative’s
protagonist)
= 1.25, SE = 0.20, p < 0.001), but not when
it was the protagonist’s father (θ
X
Y
|
(infection
target
=
protagonist’s
father)
= 0.34, SE = 0.19, p = 0.073).
In conclusion, the severity of the infection increased the perceived severity of the infection
narrated in the message only when the infection target was the protagonist of the
testimonial. These results provided empirical support for H4.
Figure 2. Moderating eect of infection target on the relationship between the severity of the
symptoms described in the Twier narrative and the perceived severity of the infection.
4.6. H5: Testing a Moderated SerialParallel Mediation Model
H5 proposed a moderated serialparallel mediation model. This model predicted
indirect (specic and conditional) eects of the severity of the infection described in the
message on the perceived personal risk of COVID-19 infection (H5a), on the perceived
Figure 2.
Moderating effect of infection target on the relationship between the severity of the
symptoms described in the Twitter narrative and the perceived severity of the infection.
4.6. H5: Testing a Moderated Serial–Parallel Mediation Model
H5 proposed a moderated serial–parallel mediation model. This model predicted
indirect (specific and conditional) effects of the severity of the infection described in the
message on the perceived personal risk of COVID-19 infection (H5a), on the perceived
severity of COVID-19 (H5b) and on the intention of preventive behavior (H5c) mediated
by the perception of the severity of the infection narrated in the story (primary mediator),
identification with the protagonist, narrative transportation, reactance and cognitive elabo-
ration (secondary mediators). However, it was predicted that these indirect effects would
only manifest themselves when the infected person was the protagonist of the message. To
test this model (which included five mediating variables, one independent variable and
one moderating variable), the PROCESS macro (customized model) was used.
The results of the analyses provided partial support for H5, as cognitive elaboration
and reactance constituted relevant mediating mechanisms for explaining the impact of
the severity of the COVID-19 infection narrated in the message on the outcome variables.
However, neither identification with the protagonist nor narrative transportation constitute
significant mediating mechanisms. The results were presented individually for each de-
pendent variable in the form of tables (where the specific conditional indirect effects were
reported) and figures (where the unstandardized regression coefficients quantifying the
relationship between the different variables were reported).
Regarding the first dependent variable considered (perceived personal risk of COVID-
19 infection), it was observed that when the testimonial narrative referred to a severe
infection in which the person narrating the story was involved (the young protagonist
who had engaged in risky behavior), the perceived severity infection increased (B = 0.90,
SE = 0.28, p= 0.001). In turn, the perceived severity infection narrated in the message was
associated with greater narrative transportation (B = 0.16, SE = 0.05, p= 0.003), greater
cognitive elaboration (B = 0.23, SE = 0.05, p< 0.001) and lower reactance (B =
0.15,
SE = 0.05, p= 0.010). However, only cognitive elaboration showed a significant effect on
perceived personal risk of contracting COVID-19 (B = 0.19, SE = 0.08, p= 0.014). The indirect
effect of the severity of the infection narrated in the message on perceived personal risk,
through the perceived severity of the infection and cognitive elaboration (serial mediation),
was statistically significant only when the infection target was the young protagonist
(effect = 0.0575, SE = 0.0298, 95% CI: 0.0071, 0.1234) (see Table 4and Figure 3a).
Int. J. Environ. Res. Public Health 2023,20, 6254 14 of 22
Table 4.
Results of the conditional specific indirect effects of the severity of the symptoms described
in the Twitter narrative on perceived personal risk of COVID-19 infection (H5).
Conditional Specific Indirect Effects Effect Boot SE Boot 95% CI
Severity infection narrative Perceived severity infection
Identification Perceived personal risk
- Infection target: Protagonist’s father 0.0056 0.0065 [0.0039, 0.0218]
- Infection target: Narrative’s protagonist 0.0201 0.0204 [0.0098, 0.0701]
IMM = 0.0145 (95% CI: 0.0065, 0578)
Severity infection narrative Perceived severity infection
Narrative transportation Perceived personal risk
- Infection target: Protagonist’s father 0.0042 0.0090 [0.0088, 0.0281]
- Infection target: Narrative’s protagonist 0.0152 0.0269 [0.0320, 0.0782]
IMM = 0.0109 (95% CI: 0.0246, 0.0581)
Severity infection narrative Perceived severity infection
Cognitive elaboration Perceived personal risk
- Infection target: Protagonist’s father 0.0160 0.0125 [0.0026, 0.0466]
-Infection target: Narrative’s protagonist 0.0575 0.0298 [0.0071, 0.1234]
IMM = 0.0415 (95% CI: 0.0041, 0.0993)
Severity infection narrative Perceived severity infection
Reactance Perceived personal risk
- Infection target: Protagonist’s father 0.0004 0.0045 [0.0084, 0.0106]
- Infection target: Narrative’s protagonist 0.0014 0.0134 [0.0268, 0.0284]
IMM = 0.0010 (95% CI: 0.0200, 0.0203)
Note. Significant specific conditional indirect effects in bold. IMM = index of moderated mediation (difference
between conditional indirect effects).
Int. J. Environ. Res. Public Health 2023, 20, 6254 15 of 23
message. To test this model (which included ve mediating variables, one independent
variable and one moderating variable), the PROCESS macro (customized model) was
used.
The results of the analyses provided partial support for H5, as cognitive elaboration
and reactance constituted relevant mediating mechanisms for explaining the impact of the
severity of the COVID-19 infection narrated in the message on the outcome variables.
However, neither identication with the protagonist nor narrative transportation
constitute signicant mediating mechanisms. The results were presented individually for
each dependent variable in the form of tables (where the specic conditional indirect
eects were reported) and gures (where the unstandardized regression coecients
quantifying the relationship between the dierent variables were reported).
Regarding the rst dependent variable considered (perceived personal risk of
COVID-19 infection), it was observed that when the testimonial narrative referred to a
severe infection in which the person narrating the story was involved (the young
protagonist who had engaged in risky behavior), the perceived severity infection
increased (B = 0.90, SE = 0.28, p = 0.001). In turn, the perceived severity infection narrated
in the message was associated with greater narrative transportation (B = 0.16, SE = 0.05, p
= 0.003), greater cognitive elaboration (B = 0.23, SE = 0.05, p < 0.001) and lower reactance
(B = 0.15, SE = 0.05, p = 0.010). However, only cognitive elaboration showed a signicant
eect on perceived personal risk of contracting COVID-19 (B = 0.19, SE = 0.08, p = 0.014).
The indirect eect of the severity of the infection narrated in the message on perceived
personal risk, through the perceived severity of the infection and cognitive elaboration
(serial mediation), was statistically signicant only when the infection target was the
young protagonist (eect = 0.0575, SE = 0.0298, 95% CI: 0.0071, 0.1234) (see Table 4 and
Figure 3a).
(a) Dependent variable: perceived personal risk of COVID-19 infection
(b) Dependent variable: perceived severity of COVID-19
Figure 3. Cont.
Int. J. Environ. Res. Public Health 2023,20, 6254 15 of 22
Int. J. Environ. Res. Public Health 2023, 20, 6254 15 of 23
message. To test this model (which included ve mediating variables, one independent
variable and one moderating variable), the PROCESS macro (customized model) was
used.
The results of the analyses provided partial support for H5, as cognitive elaboration
and reactance constituted relevant mediating mechanisms for explaining the impact of the
severity of the COVID-19 infection narrated in the message on the outcome variables.
However, neither identication with the protagonist nor narrative transportation
constitute signicant mediating mechanisms. The results were presented individually for
each dependent variable in the form of tables (where the specic conditional indirect
eects were reported) and gures (where the unstandardized regression coecients
quantifying the relationship between the dierent variables were reported).
Regarding the rst dependent variable considered (perceived personal risk of
COVID-19 infection), it was observed that when the testimonial narrative referred to a
severe infection in which the person narrating the story was involved (the young
protagonist who had engaged in risky behavior), the perceived severity infection
increased (B = 0.90, SE = 0.28, p = 0.001). In turn, the perceived severity infection narrated
in the message was associated with greater narrative transportation (B = 0.16, SE = 0.05, p
= 0.003), greater cognitive elaboration (B = 0.23, SE = 0.05, p < 0.001) and lower reactance
(B = 0.15, SE = 0.05, p = 0.010). However, only cognitive elaboration showed a signicant
eect on perceived personal risk of contracting COVID-19 (B = 0.19, SE = 0.08, p = 0.014).
The indirect eect of the severity of the infection narrated in the message on perceived
personal risk, through the perceived severity of the infection and cognitive elaboration
(serial mediation), was statistically signicant only when the infection target was the
young protagonist (eect = 0.0575, SE = 0.0298, 95% CI: 0.0071, 0.1234) (see Table 4 and
Figure 3a).
(a) Dependent variable: perceived personal risk of COVID-19 infection
(b) Dependent variable: perceived severity of COVID-19
Int. J. Environ. Res. Public Health 2023, 20, 6254 16 of 23
(c) Dependent variable: protective behavioral intent against COVID-19
Figure 3. Results of the moderated serial–parallel mediation model (H5). The gures show the non-
standardized regression coecients, B. The dashed line represents non-signicant coecients. + p a
0.10, * p < 0.05, ** p < 0.01, *** p < 0.001. (a) Dependent variable: perceived personal risk of COVID-
19 infection. (b) Dependent variable: perceived severity of COVID-19. (c) Dependent variable:
protective behavioral intent against COVID-19.
Table 4. Results of the conditional specic indirect eects of the severity of the symptoms described
in the Twier narrative on perceived personal risk of COVID-19 infection (H5).
Conditional Specific Indirect Effects Effect Boot SE Boot 95% CI
Severity infection narrative Perceived severity infection
Identification Perceived personal risk
- Infection target: Protagonists father 0.0056 0.0065 [0.0039, 0.0218]
- Infection target: Narrative’s protagonist 0.0201 0.0204 [0.0098, 0.0701]
IMM = 0.0145 (95% CI: 0.0065, 0578)
Severity infection narrative Perceived severity infection
Narrative transportation Perceived personal risk
- Infection target: Protagonists father 0.0042 0.0090 [0.0088, 0.0281]
- Infection target: Narrative’s protagonist 0.0152 0.0269 [0.0320, 0.0782]
IMM = 0.0109 (95% CI: 0.0246, 0.0581)
Severity infection narrative Perceived severity infection
Cognitive elaboration Perceived personal risk
- Infection target: Protagonists father 0.0160 0.0125 [0.0026, 0.0466]
- Infection target: Narrative’s protagonist 0.0575 0.0298 [0.0071, 0.1234]
IMM = 0.0415 (95% CI: 0.0041, 0.0993)
Severity infection narrative Perceived severity infection
Reactance Perceived personal risk
- Infection target: Protagonists father 0.0004 0.0045 [0.0084, 0.0106]
- Infection target: Narrative’s protagonist 0.0014 0.0134 [0.0268, 0.0284]
IMM = 0.0010 (95% CI: 0.0200, 0.0203)
Note. Signicant specic conditional indirect eects in bold. IMM = index of moderated mediation
(dierence between conditional indirect eects).
In relation to the second dependent variable, it was observed that both cognitive
elaboration (B = 0.13, SE = 0.04, p = 0.001) and reactance (B = 0.15, SE = 0.03, p < 0.001)
showed signicant eects on the perceived severity of COVID-19. Additionally, two
signicant conditional specic indirect eects were obtained through the mediation of
cognitive elaboration (eect = 0.0398, SE = 0.0186, 95% CI: 0.0102, 0.0820) and reactance
Figure 3.
Results of the moderated serial–parallel mediation model (H5). The figures show the
non-standardized regression coefficients, B. The dashed line represents non-significant coefficients.
+pa 0.10, * p< 0.05, ** p< 0.01, *** p< 0.001. (
a
) Dependent variable: perceived personal risk
of COVID-19 infection. (
b
) Dependent variable: perceived severity of COVID-19. (
c
) Dependent
variable: protective behavioral intent against COVID-19.
In relation to the second dependent variable, it was observed that both cognitive
elaboration (B = 0.13, SE = 0.04, p= 0.001) and reactance (B =
0.15, SE = 0.03, p< 0.001)
showed significant effects on the perceived severity of COVID-19. Additionally, two
significant conditional specific indirect effects were obtained through the mediation of
cognitive elaboration (effect = 0.0398, SE = 0.0186, 95% CI: 0.0102, 0.0820) and reactance
(effect = 0.0296, SE = 0.0156, 95% CI: 0.0054, 0.0652), in both cases when the infection target
was the young protagonist (see Table 5and Figure 3b).
Finally, concerning the third dependent variable, it was also observed that both
cognitive elaboration (B = 0.32, SE = 0.15, p= 0.041) and reactance (B =
0.35, SE = 0.13,
p= 0.009) showed a significant effect on protective behavioral intent against COVID-19. In
addition, two specific conditional indirect effects were observed through the mediation of
cognitive elaboration (effect = 0.0946, SE = 0.0608, 95% CI: 0.0006, 0.2365) and reactance
(effect = 0.0690, SE = 0.0384, 95% CI: 0.0079, 0.1557), in both cases when the infection target
was the young protagonist (see Table 6and Figure 3c).
Int. J. Environ. Res. Public Health 2023,20, 6254 16 of 22
Table 5.
Results of the conditional specific indirect effects of the severity of the symptoms described
in the Twitter narrative on perceived severity of COVID-19 (H5).
Conditional Specific Indirect Effects Effect Boot SE Boot 95% CI
Severity infection narrative Perceived severity infection
Identification Perceived severity of COVID-19
- Infection target: Protagonist’s father 0.0010 0.0029 [0.0030, 0.0088]
- Infection target: Narrative’s protagonist 0.0037 0.0089 [0.0094, 0.0271]
IMM = 0.0027 (95% CI: 0.0069, 0.0208)
Severity infection narrative Perceived severity infection
Narrative transportation Perceived severity of COVID-19
- Infection target: Protagonist’s father 0.0015 0.0042 [0.0059, 0.0116]
- Infection target: Narrative’s protagonist 0.0054 0.0129 [0.0199, 0.0330]
IMM = 0.0039 (95% CI: 0.0154, 0.0248)
Severity infection narrative Perceived severity infection
Cognitive elaboration Perceived severity of COVID-19
- Infection target: Protagonist’s father 0.0111 0.0086 [0.0013, 0.0320]
-Infection target: Narrative’s protagonist 0.0398 0.0186 [0.0102, 0.0820]
IMM = 0.0287 (95% CI: 0.0057, 0.0662)
Severity infection narrative Perceived severity infection
Reactance Perceived severity of COVID-19
- Infection target: Protagonist’s father 0.0083 0.0074 [0.0010, 0.0273]
-Infection target: Narrative’s protagonist 0.0296 0.0156 [0.0054, 0.0652]
IMM = 0.0213 (95% CI: 0.0032, 0.0495)
Note. Significant specific conditional indirect effects in bold. IMM = index of moderated mediation (difference
between conditional indirect effects).
Table 6.
Results of the conditional specific indirect effects of the severity of the symptoms described
in the Twitter narrative on protective behavioral intent against COVID-19 (H5).
Conditional Specific Indirect Effects Effect Boot SE Boot 95% CI
Severity infection narrative
Perceived severity infection
Identification
Protective behavioral intent against COVID-19
- Infection target: Protagonist’s father 0.0026 0.0106 [0.0294, 0.0171]
- Infection target: Narrative’s protagonist 0.0094 0.0328 [0.0834, 0.0562]
IMM = 0.0068 (95% CI: 0.0620, 0.0420)
Severity infection narrative Perceived severity infection Narrative
transportation Protective behavioral intent against COVID-19
- Infection target: Protagonist’s father 0.0241 0.0234 [0.0845, 0.0052]
- Infection target: Narrative’s protagonist 0.0864 0.0641 [0.2429, 0.0022]
IMM = 0.0623 (95% CI: 0.1977, 0.0015)
Int. J. Environ. Res. Public Health 2023,20, 6254 17 of 22
Table 6. Cont.
Conditional Specific Indirect Effects Effect Boot SE Boot 95% CI
Severity infection narrative Perceived severity infection Cognitive
elaboration Protective behavioral intent against COVID-19
- Infection target: Protagonist’s father 0.0264 0.0242 [0.0050, 0.0872]
-Infection target: Narrative’s protagonist 0.0946 0.0608 [0.0006, 0.2365]
IMM = 0.0682 (95% CI: 0.0002, 0.1865)
Severity infection narrative Perceived severity infection Reactance
Protective behavioral intent against COVID-19
- Infection target: Protagonist’s father 0.0192 0.0176 [0.0023, 0644]
-Infection target: Narrative’s protagonist 0.0690 0.0384 [0.0079, 0.1557]
IMM = 0.0497 (95% CI: 0.0048, 0.1189)
Note: significant specific conditional indirect effects in bold. IMM = index of moderated mediation (difference
between conditional indirect effects).
5. Discussion
The pandemic caused by COVID-19 has posed a global challenge that has required
the coordinated action of local, state and international institutions. The main challenge
during the pandemic was to reduce the number of infections and cases of COVID-19 to
avoid overwhelming healthcare systems [
57
]. The second major challenge was biomedical
in nature: it was necessary to develop an effective vaccine to curb the pandemic and
protect the entire population [
58
]. Finally, the third challenge encompassed all individuals
and institutions that aimed to prevent COVID-19 by increasing the awareness of modes
of transmission and preventive strategies [
59
]. This last challenge is closely linked to
communication activities, the dissemination of scientifically validated information and the
design of persuasive strategies to convince the population to adopt infection prevention
practices (such as social distancing, mask wearing, building disinfection and mobility and
transportation restrictions) [79].
This paper focused precisely on the communicative challenge posed by the pandemic
to stimulate prevention behaviors. Building on research on narrative persuasion and, par-
tially, on the impact of fear appeals (see [
60
]), an experiment was designed using testimonial
messages in the form of a Twitter “thread” as a stimulus. Social media played a decisive
role during the pandemic. Social networks such as Twitter constituted a primary platform
for disseminating health information from institutions [
61
]. However, they were also used
to spread fake news, conspiracy theories denying the existence of the virus and informa-
tion opposing the use of masks or social distancing measures [
62
64
]. Nonetheless, we
acknowledged that Twitter could also serve as an optimal platform for sharing testimonial
messages for raising awareness around prevention measures during the pandemic. We
consider that testimonial messages disseminated through social networks such as Twitter
could constitute narrative vaccines [
49
,
65
] or narrative pills [
66
] that spread personal stories
that facilitated a persuasive impact, especially among the younger population. In our case,
the testimonial message was led by a young person who narrated a risky behavior (hold-
ing a party without taking COVID-19 prevention measures) that resulted in a COVID-19
infection that affected either the young person or a family member with whom they lived
(contagion target). Furthermore, the narration described the symptoms experienced by the
infected person as either mild or severe (infection severity). We opted for this narrative per-
suasion approach because awareness messages directed at young people led by authority
figures (such as healthcare professionals or institutional actors) can provoke rejection from
this group [67].
Int. J. Environ. Res. Public Health 2023,20, 6254 18 of 22
The experiment, carried out with a convenience sample (N = 278) of young people aged
18 to 28 years, allowed us to advance our knowledge on the psychological mechanisms that
explained the impact of two message features (the severity of the infection described in the
testimonial and the infection target) on preventive measures, such as the perceived personal
risk of contracting COVID-19, the perceived severity of COVID-19 and the intention to
engage in preventive behavior. Five hypotheses were established, although only two of
them received empirical support.
The first hypothesis predicted that when the narrative described the person infected
with COVID-19 as the protagonist’s father, there would be less identification with the
protagonist of the message, especially when the symptoms caused by the disease were
severe. This hypothesis was based on research on the (positive or negative) characteristics
of narrative message protagonists and their effects on identification [
24
]. In this context,
engaging in an imprudent behavior (organizing a party and not wearing a mask) that could
potentially result in a family member becoming ill was considered a negative trait of the
protagonist of the story. However, our hypothesis did not receive empirical support, since
no statistically significant interaction effect was observed between the two manipulated
independent variables on identification. Moreover, identification was not affected by who
was the target of the contagion or by the severity of the symptoms. However, it should
be noted that the level of identification with the protagonist of the message was low
(M = 2.87, SD = 0.75), placing it below the theoretical midpoint of the scale (value of three;
t(277) =
2.68, p= 0.004). This may mean that, overall, participants felt that a young
person who had acted irresponsibly during the pandemic was not a positive role model to
identify with.
The second hypothesis predicted an effect of the severity of the symptoms narrated
in the message on narrative transportation. However, this hypothesis did not receive
empirical support either. Nonetheless, in this case, it was observed that, overall, the
testimonial message induced a level of narrative transportation (M = 4.18, SD = 1.17)
above the theoretical midpoint of the scale (value of four; t(277) = 2.58, p= 0.005). The
third hypothesis also did not receive empirical support, as the narrative mentioning that
the person infected with COVID-19 was the father and that the infection was severe did
not induce the highest level of reactance. Again, it was observed that, overall, the level
of reactance experienced by the participants was low (M = 2.06, SD = 1.29), below the
theoretical midpoint of the scale (value of four; t(277) =
24.98, p< 0.001). This result was
convergent with Moyer-Gusé’s [
19
] entertainment overcoming resistance model (EORM),
which proposed that narrative messages have the capacity to reduce reactance, given that
they are not perceived as aiming to persuade.
The fourth hypothesis did receive empirical support. It was hypothesized that a
narrative message describing a serious infection affecting the protagonist of the story could
be considered a fear appeal. Therefore, it was considered that such a combination of
elements in the message would elicit a heightened perception of the severity of the infection
described in the testimonial. The results were consistent with this prediction, which was in
line with the postulates of the EPPM [
36
] on the effects of fear on the activation of protective
motivation against a particular disease.
Finally, a moderated serial–parallel mediation model was proposed. It was considered
that for a narrative message to have a persuasive impact on measures related to COVID-19
prevention, it had to stimulate a series of psychological processes acting as mediating
mechanisms. The analysis of the specific conditional indirect effects showed that both
cognitive elaboration and reactance were relevant mechanisms to explain the impact of the
characteristics of the message on the dependent variables considered (perceived personal
risk of contracting COVID-19, perceived severity of the disease, and intention to engage
in preventive behavior). Therefore, it was noted that for a message describing a severe
COVID-19 infection affecting their protagonist to increase the perception of personal risk,
the perception that COVID-19 is a serious disease and that preventive action is necessary
(not attending parties or gatherings for fear of contracting coronavirus), a dual process had
Int. J. Environ. Res. Public Health 2023,20, 6254 19 of 22
to be activated. Thus, two routes were established that would explain the persuasive impact,
first through activating the perception that the infection described in the message was
severe and the subsequent increase in cognitive processing (e.g., “reading the message has
made me think deeply about measures to prevent the transmission of the coronavirus”) and
second through the activation of the perception that the infection narrated in the message
was severe and the subsequent reduction in reactance (e.g., “the message was trying to
manipulate me”). More importantly, the results of this study showed that identification
and narrative transportation did not constitute relevant mediating mechanisms.
The present study had two major limitations. Firstly, it did not include any measure of
the emotional impact provoked by the reading of the testimonial messages (fear and guilt,
in particular). It would have been interesting to contrast the effect of the infection target
on the experience of guilt and, subsequently, to analyze the relationship between guilt
and reactance. The second limitation of this work was that the proposed mediators were
measured rather than experimentally manipulated, which prevented drawing conclusions
with complete certainty regarding the proposed causal sequence between the different
psychological mechanisms. However, although temporal precedence is an important
element for establishing a causal inference, it is also necessary to propose a theoretical
argument about the relationship between the mediating mechanisms, a condition that our
work fulfilled by relying on theoretical models of narrative persuasion (EORM) [
19
] and
on the EPPM model [
36
]. Nevertheless, future research should use other methodological
approaches to address such causal inference problems [68].
6. Conclusions
As a general conclusion, our study highlighted that creating persuasive messages
based on social media (Twitter) targeted at young people that describe a careless behavior
resulting in a severe COVID-19 infection could be an appropriate strategy for designing
prevention campaigns (see also [
69
,
70
] on the role of Twitter in the dissemination of medical
information and misinformation during the COVID-19 pandemic). Such a message could
inspire the audience (by stimulating deep reflection and reducing reactance), leading to
the adoption of self-protective and socially responsible prevention measures [
46
]. The dual
mediation model outlined highlighted the need to stimulate the perception of severity,
which was congruent with the EPPM model [
36
,
60
]. Such a process would act as a catalyst
for activating cognitive elaboration and reducing reactance. This model could serve as a
blueprint for developing social media campaigns aimed at addressing other health or social
issues that young people face today, such as HIV–AIDS prevention, sexually transmitted
diseases, alcohol and tobacco consumption and obesity, among others.
Supplementary Materials:
The following supporting information can be downloaded at: https:
//osf.io/cz3dt/ (accessed on 27 April 2023).
Author Contributions:
Conceptualization, J.-J.I. and A.H.-V.; methodology, J.-J.I., A.H.-V. and Í.G.-M.;
formal analysis, J.-J.I. and A.H.-V.; investigation, J.-J.I., L.R.-C., Í.G.-M. and A.H.-V.; writing—original
draft preparation, J.-J.I., L.R.-C., Í.G.-M. and A.H.-V.; writing—review and editing, J.-J.I., L.R.-C.,
Í.G.-M. and A.H.-V.; supervision, J.-J.I. All authors have read and agreed to the published version of
the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement:
The study was carried out following the rules of the Dec-
laration of Helsinki of 1975 (https://www.wma.net/what-we-do/medical-ethics/declaration-of-
helsinki/, 15 February 2021), revised in 2013. Since this work does not constitute an intervention
study, was not carried out with the support of a funded research project and was based on a project
linked to the completion of a master’s thesis; the research project was not submitted for approval by
the Ethics Committee of the University of Salamanca.
Informed Consent Statement:
Informed consent was obtained from all participants involved in the
study: (a) they were informed of the objectives of the study; (b) prior to participation, all individuals
Int. J. Environ. Res. Public Health 2023,20, 6254 20 of 22
provided their informed consent for inclusion in the study; (c) there were not risks associated with
participating in the study; (d) anonymity was assured.
Data Availability Statement:
Datasets and syntax files are available via the Open Science Framework
(OSF): https://osf.io/cz3dt/ (accessed on 27 April 2023).
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Chen, M.; Bell, R.A.; Taylor, L.D. Narrator point of view and persuasion in health narratives: The role of protagonist–reader
similarity, identification, and self-referencing. J. Health Commun. 2016,21, 908–918. [CrossRef]
2.
De Wit, J.B.F.; Das, E.; Vet, R. What works best: Objective statistics or a personal testimonial? An assessment of the persuasive
effects of different types of message evidence on risk perception. Health Psychol. 2008,27, 110–115. [CrossRef]
3.
Kim, M. When similarity strikes back: Conditional persuasive effects of character-audience similarity in anti-smoking campaign.
Hum. Commun. Res. 2019,45, 52–77. [CrossRef]
4.
Watts, J.; Slater, M. Eudaimonic testimonial vs. didactic presentation impact on willingness to engage in conversations about
end-of-life care: The moderating role of modeling. J. Health Commun. 2021,26, 137–146. [CrossRef]
5.
Statista. Número Acumulado de Casos de Coronavirus en el Mundo Desde el 22 de Enero de 2020 Hasta el 16 de Marzo de. 2023.
Available online: https://es.statista.com/estadisticas/1104227/numero-acumulado-de-casos-de-coronavirus-covid-19-en-el-
mundo-enero-marzo/ (accessed on 20 March 2023).
6.
Word Health Organization. WHO Director-General’s Opening Remarks at the Media Briefing on COVID-19—11 March 2020.
Available online: https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-
media-briefing-on-covid-19{-}{-}-11-march-2020 (accessed on 16 June 2023).
7. Anderson, R.M.; Heesterbeek, H.; Klinkenberg, D.; Hollingsworth, T.D. How will country-based mitigation measures influence
the course of the COVID-19 epidemic? Lancet 2020,395, 931–934. [CrossRef] [PubMed]
8. Coccia, M. Pandemic prevention: Lessons from COVID-19. Encyclopedia 2021,1, 433–444. [CrossRef]
9.
Fong, M.W.; Gao, H.; Wong, J.Y.; Xiao, J.; Shiu, E.Y.C.; Ryu, S.; Cowling, B.J. Nonpharmaceutical measures for pandemic influenza
in nonhealthcare settings—Social distancing measures. Emerg. Infect. Dis. 2020,26, 976–984. [CrossRef] [PubMed]
10.
Lunn, P.; Belton, C.; Lavin, C.; McGowan, F.; Timmons, S.; Robertson, D. Using behavioural science to help fight the coronavirus:
A rapid, narrative review. J. Behav. Public Adm. 2020,3, 1–22. [CrossRef]
11.
Rahman, H.S.; Aziz, M.S.; Hussein, R.H.; Othman, H.J.; Omer, S.H.S.; Khalid, E.S.; Abdulrahman, N.A.; Amin, K.; Abdullah, R.
The transmission modes and sources of COVID-19: A systematic review. Int. J. Surg. Open 2020,26, 125–136. [CrossRef]
12.
De Graaf, A.; Sanders, J.; Hoeken, H. Characteristics of narrative interventions and health effects: A review of the content, form,
and context of narratives in health-related narrative persuasion research. Rev. Commun. Res. 2016,4, 88–131. [CrossRef]
13.
Murphy, S.T.; Frank, L.B.; Chatterjee, J.S.; Baezconde-Garbanati, L. Narrative versus nonnarrative: The role of identification,
transportation, and emotion in reducing health disparities. J. Commun. 2013,63, 116–137. [CrossRef]
14.
Shen, F.; Sheer, V.; Li, R. Impact of narratives on persuasion in health communication: A meta-analysis. J. Advert.
2015
,
44, 105–113. [CrossRef]
15.
Dahlstrom, M.F.; Ho, S.S. Ethical considerations of using narrative to communicate science. Sci. Commun.
2012
,34, 592–617.
[CrossRef]
16.
Braddock, K.; Dillard, J.P. Meta-analytic evidence for the persuasive effect of narratives on beliefs, attitudes, intentions, and
behaviors. Commun. Monogr. 2016,83, 446–467. [CrossRef]
17.
Zebregs, S.; Van den Putte, B.; Neijens, P.; de Graaf, A. The differential impact of statistical and narrative evidence on beliefs,
attitude, and intention: A meta-analysis. Health Commun. 2015,30, 282–289. [CrossRef] [PubMed]
18.
Green, M.C.; Brock, T.C. In the mind’s eye: Transportation-imagery model of narrative persuasion. In Narrative Impact. Social
and Cognitive Foundations; Green, M.C., Strange, J.J., Brock, T.C., Eds.; Lawrence Erlbaum Associates: Mahwah, NJ, USA, 2002;
pp. 315–341.
19.
Moyer-Gusé, E. Toward a theory of entertainment persuasion: Explaining the persuasive effects of entertainment-education
messages. Commun. Theory 2008,18, 407–425. [CrossRef]
20.
Slater, M.D.; Rouner, D. Entertainment-education and elaboration likelihood: Understanding the processing of narrative
persuasion. Commun. Theory 2002,12, 173–191.
21.
Cohen, J. Defining identification: A theoretical look at the identification of audiences with media characters. Mass Commun. Soc.
2001,4, 245–264. [CrossRef]
22.
Igartua, J.J.; Barrios, I. Changing real-world beliefs with controversial movies: Processes and mechanisms of narrative persuasion.
J. Commun. 2012,62, 514–531. [CrossRef]
23.
Cohen, J.; Tal-Or, N. Antecedents of identification: Character, text, and audiences. In Narrative Absorption; Hakemulder, F.,
Kuijpers, M.K., Bálint, K., Tan, E.S., Doicaru, M.M., Eds.; John Benjamins Publishing Company: Amsterdam, The Netherlands,
2017; pp. 133–153.
24.
Tal-Or, N.; Cohen, J. Unpacking engagement: Convergence and divergence in transportation and identification. Ann. Int. Commun.
Assoc. 2016,40, 33–66. [CrossRef]
Int. J. Environ. Res. Public Health 2023,20, 6254 21 of 22
25.
Hoffner, C.; Buchanan, M. Young adults’ wishful identification with television characters: The role of perceived similarity and
character attributes. Media Psychol. 2009,7, 325–351. [CrossRef]
26.
Lee, T.K.; Kim, H.K. An enjoyable story, a persuasive story: Exploring narrative enjoyment in narrative persuasion. J. Media
Psychol. Theor. Methods Appl. 2022,34, 361–372. [CrossRef]
27.
Green, M.C.; Brock, T.C. The role of transportation in the persuasiveness of public narratives. J. Personalit. Soc. Psychol.
2000
,
79, 701–721. [CrossRef] [PubMed]
28.
Green, M.C.; Fitzgerald, K. Transportation theory applied to health and risk messaging. In Oxford Research Encyclopedia of
Communication; Oxford University Press: Oxford, UK, 2017; Volume 25. [CrossRef]
29.
Rains, S.A. The nature of psychological reactance revisited: A meta-analytic review. Hum. Commun. Res.
2013
,39, 47–73.
[CrossRef]
30.
Wehbe, M.S.; Basil, M.; Basil, D. Reactance and coping responses to tobacco counter advertisements. J. Health Commun.
2017
,
22, 576–583. [CrossRef]
31.
Ball, H.; Wozniak, T.R. Why do some Americans resist COVID-19 prevention behavior? An analysis of issue importance, message
fatigue, and reactance regarding COVID-19 messaging. Health Commun. 2022,37, 1812–1819. [CrossRef] [PubMed]
32.
O’Keefe, D.J. Guilt as a mechanism of persuasion. In The Persuasion Handbook: Developments in Theory and Practice; Dillard, J.P.,
Pfau, M., Eds.; Sage: Thousand Oaks, CA, USA, 2002; pp. 329–344.
33.
Bessaravoba, E.; Turner, M.M.; Fink, E.L.; Blustein, N.B. Extending the theory of reactance to guilt appeals. Z. Für Psuchologie
2015,223, 215–224. [CrossRef]
34.
Carrera, P.; Aguilar, P.; Fernández, I.; Caballero, A. Health or wealth? The influence of perceived health and wealth threats and
style of thinking on protective behaviours and well-being during the COVID-19 pandemic in Spain. Int. J. Soc. Psychol.
2023
,
38, 66–91. [CrossRef]
35.
Popova, L. The extended parallel process model: Illuminating the gaps in research. Health Educ. Behav.
2012
,39, 455–473.
[CrossRef] [PubMed]
36.
Witte, K. Putting the fear back into fear appeals: The extended parallel process model. Commun. Monogr.
1992
,59, 329–349.
[CrossRef]
37.
Tsoy, D.; Godinic, D.; Tong, Q.; Obrenovic, B.; Khudaykulov, A.; Kurpayanidi, K. Impact of social media, Extended Parallel
Process Model (EPPM) on the intention to stay at home during the COVID-19 pandemic. Sustainability
2022
,14, 7192. [CrossRef]
38. Kowalski, R.M.; Black, K.J. Protection motivation and the COVID-19 virus. Health Commun. 2021,36, 15–22. [CrossRef]
39.
Okuhara, T.; Okada, H.; Kiuchi, T. Predictors of staying at home during the COVID-19 pandemic and social lockdown based on
Protection Motivation Theory: A cross-sectional study in Japan. Healthcare 2020,8, 475. [CrossRef]
40.
Andrews, M.E.; Mattan, B.D.; Richards, K.; Moore-Berg, S.L.; Falk, E.B. Using first-person narratives about healthcare workers
and people who are incarcerated to motivate helping behaviors during the COVID-19 pandemic. Soc. Sci. Med.
2022
,299, 114870.
[CrossRef] [PubMed]
41.
Hoeken, H.; Fikkers, K.M. Issue-relevant thinking and identification as mechanisms of narrative persuasion. Poetics
2014
,
44, 84–99. [CrossRef]
42.
Igartua, J.J.; Rodríguez-Conteras, L. Narrative voice matters! Improving smoking prevention with testimonial messages through
identification and cognitive processes. Int. J. Environ. Res. Public Health 2020,17, 7281. [CrossRef]
43.
Igartua, J.J.; Vega, J. Identification with characters, elaboration, and counterarguing in entertainment-education interventions
through audiovisual fiction. J. Health Commun. 2016,2, 293–300. [CrossRef] [PubMed]
44.
Petty, R.E.; Cacioppo, J.T. Communication and Persuasion: Central and Peripheral Routes to Attitude Change, 1st ed.; Springer:
Wiesbaden, Germany, 1986; ISBN 978-1-4612-9378-1.
45.
De Graaf, A.; Van Leeuwen, L. The role of absorption processes in narrative health communication. In Narrative Absorption;
Hakemulder, F., Kuijpers, M.K., Bálint, K., Tan, E.S., Doicaru, M.M., Eds.; John Benjamins Publishing Company: Amsterdam, The
Netherlands, 2017; pp. 271–292.
46.
Oliver, M.B.; Raney, A.A.; Bartsch, A.; Janicke-Bowles, S.; Appel, M.; Dale, K. Model of inspiring media. J. Media Psychol.
2021
,
33, 191–201. [CrossRef]
47.
Chen, M.; Bell, R.A. A meta-analysis of the impact of point of view on narrative processing and persuasion in health messaging.
Pychol. Health 2022,37, 545–562. [CrossRef]
48.
O’Keefe, D.J. Message properties, mediating states, and manipulation checks: Claims, evidence, and data analysis in experimental
persuasive message effects research. Commun. Theory 2003,13, 251–274. [CrossRef]
49.
Igartua, J.J.; Guerrero-Martín, I. Personal migrant stories as persuasive devices: Effects of audience– character similarity and
narrative voice. J. Soc. Political Psychol. 2022,10, 21–34. [CrossRef]
50.
Appel, M.; Gnambs, T.; Richter, T.; Green, M.C. The Transportation Scale–Short Form (TS–SF). Media Psychol.
2015
,18, 243–266.
[CrossRef]
51.
Shen, L. Targeting smokers with empathy appeal antismoking public service announcements: A field experiment. J. Health
Commun. 2015,20, 573–580. [CrossRef] [PubMed]
52.
Jahangiry, L.; Bakhtari, F.; Sohrabi, Z.; Reinhani, P.; Samei, S.; Ponnet, K.; Montazeri, A. Risk perception related to COVID-19
among the Iranian general population: An application of the extended parallel process model. BMC Public Health
2020
,20, 1571.
[CrossRef]
Int. J. Environ. Res. Public Health 2023,20, 6254 22 of 22
53.
Ning, L.; Niu, J.; Bi, X.; Yang, C.; Liu, Z.; Wu, Q.; Ning, N.; Liang, L.; Liu, A.; Hao, Y.; et al. The impacts of knowledge, risk
perception, emotion and information on citizens’ protective behaviors during the outbreak of COVID-19: A cross-sectional study
in China. BMC Public Health 2020,20, 1751. [CrossRef]
54.
Rosero-Bolaños, A.D.; Carvajal, J.L.; Bolaños-Benavides, E.F. perception of risk in relation to COVID-19 in Colombian schooled
adolescents. Rev. Boletín Redipe 2021,10, 201–2017.
55.
Hayes, A.F. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach, 3rd ed; The Guilford
Press: New York, NY, USA, 2022; ISBN 9781462549030.
56.
Igartua, J.J.; Hayes, A.F. Mediation, moderation, and conditional process analysis: Concepts, computations, and some common
confusions. Span. J. Psychol. 2021,24, e49. [CrossRef]
57.
Arshed, N.; Meo, M.S.; Farooq, F. Empirical assessment of government policies and flattening of the COVID19 curve. J. Public
Affairs. 2020,20, e2333. [CrossRef]
58.
Li, M.; Wang, H.; Tian, L.; Pang, Z.; Yang, Q.; Huang, T.; Fan, J.; Song, L.; Tong, Y.; Fan, H. COVID-19 vaccine development:
Milestones, lessons and prospects. Signal Transduct. Target. Ther. 2022,7, 146. [CrossRef] [PubMed]
59.
Plohl, N.; Musil, B. Modeling compliance with COVID-19 prevention guidelines: The critical role of trust in science. Psychol.
Health Med. 2021,26, 1772988. [CrossRef]
60.
Chen, L.; Tang, H. Examining the persuasion process of narrative fear appeals on health misinformation correction. Inf. Commun.
Soc. 2022,in press. [CrossRef]
61.
Pulido-Polo, M.; Hernández-Santaolalla, V.; Lozano-González, A.A. Institutional use of Twitter to combat the infodemic caused
by the COVID-19 health crisis. Prof. Inf. 2021,30, e300119. [CrossRef]
62.
Ahmed, W.; Vidal-Alaball, V.; Downing, J.; López Seguí, F. COVID-19 and the 5G conspiracy theory: Social network analysis of
Twitter data. J. Med. Internet Res. 2020,22, e19458. [CrossRef]
63.
Pérez-Dasilva, J.A.; Meso-Ayerdi, K.; Mendiguren-Galdospín, T. Fake news and coronavirus: Detecting key players and trends
through analysis of Twitter conversations. Prof. Inf. 2020,29, e290308. [CrossRef]
64.
Yang, K.C.; Pierri, F.; Hui, P.M.; Axelrod, D.; Torres-Lugo, C.; Bryden, J.; Menczer, F. The COVID-19 infodemic, Twitter versus
Facebook. Big Data Soc. 2021,8, 20539517211013861. [CrossRef]
65.
Leader, A.E.; Miller-Day, M.; Rey, R.T.; Selvan, P.; Pezalla, A.E.; Hecht, M.L. The impact of HPV vaccine narratives on social
media: Testing narrative engagement theory with a diverse sample of young adults. Prev. Med. Rep.
2022
,29, 101920. [CrossRef]
66.
Igartua, J.J.; Cachón-Ramón, D. Personal narratives to improve attitudes towards stigmatized immigrants: A parallel-serial
mediation model. Group Process. Intergroup Relat. 2023,26, 96–119. [CrossRef]
67.
Susmann, M.W.; Xu, M.; Clark, J.K.; Wallace, L.E.; Bñankenship, K.L.; Philipp-Muller, A.Z.; Luttrell, A.; Wegener, D.T.;
Petty, R.E. Persuasion amidst a pandemic: Insights from the Elaboration Likelihood Model. Eur. Rev. Soc. Psychol.
2022
,
33, 323–359. [CrossRef]
68. Pirlott, A.; MacKinnon, D.P. Design approaches to experimental mediation. J. Exp. Soc. Psychol. 2016,66, 29–38. [CrossRef]
69.
Thakur, N.; Han, C.Y. An exploratory study of tweets about the SARS-CoV-2 Omicron variant: Insights from sentiment analysis,
language interpretation, source tracking, type classification, and embedded URL detection. COVID
2022
,2, 1026–1049. [CrossRef]
70.
Rosenberg, H.; Syed, S.; Rezaie, S. The Twitter pandemic: The critical role of Twitter in the dissemination of medical information
and misinformation during the COVID-19 pandemic. Can. J. Emerg. Med. 2020,22, 418–421. [CrossRef] [PubMed]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
... La propia exposición a modelos en redes sociales consolida la autoeficacia (Bandura, 2004; Rimer y Glanz, 2005), es decir, el convencimiento de poder desarrollar una determinada actividad, y que trasladada al ámbito sanitario presenta infinidad de aplicaciones en lo concerniente a prevención, sensibilización y adquisición de nuevos hábitos saludables (Alkhafagy et al., 2023;Camelo-Guarín et al., 2021;Igartua et al., 2023). ...
Article
This research examines the creation of knowledge on social networks about blood donation, the role of virtual communities and their potential to attract donors. We applied a two-step methodology, starting by interviews with blood donation management institutions in Spain (N=21) and concluding with the analysis of comments from 61 donor testimonials on TikTok (N=1,606). This social network was chosen because of the scarce institutional presence and the predominance of Generation Z, which is necessary for generational change. The results show that blood donation in Spain is seen from a positive perspective, where complaints are rare or relativised and the community encourages the behaviour imitation thanks to the feeling of pride and gratitude to donors. Knowledge is generated both from the contributions of creators and audiences and is focused on informing about the process, conditions and consequences, supporting its voluntary and altruistic nature. Overall, social, group and collaborative learning takes place, with greater success in those contents shared by health profiles but supported by all users. We conclude that TikTok plays a relevant role in the literacy and awareness of blood donation based on the virtual community collective intelligence, which will affect the donor relay thanks to the behavioural models that obtain social recognition. Esta investigación examina la creación de conocimientos en redes sociales sobre la donación de sangre, el papel de las comunidades virtuales y su capacidad para atraer donantes. Se aplica una metodología en dos pasos que parte de entrevistas a las instituciones gestoras de la hemodonación en España (N=21) y concluye con el análisis de los comentarios de 61 testimonios de donantes en TikTok (N=1.606), por ser la red social con menor presencia institucional y donde predomina la Generación Z, necesaria para el relevo generacional. Los resultados muestran que la donación de sangre en España se observa desde una óptica positiva donde las quejas escasean o se relativizan y la comunidad favorece la imitación del comportamiento gracias al sentimiento de orgullo y agradecimiento a donantes. El conocimiento se genera tanto de aportaciones de los creadores como de las audiencias y se enfoca a informar sobre el proceso, condiciones y consecuencias, defendiendo su carácter voluntario y altruista. En conjunto, se produce un aprendizaje social, grupal y colaborativo, con mayor éxito en aquellos contenidos compartidos por perfiles sanitarios, pero apoyado en todos los usuarios. Se concluye que TikTok ejerce un papel relevante en la alfabetización y en la concientización sobre la hemodonación a partir de la inteligencia colectiva de la comunidad virtual que repercutirá en el relevo de donantes gracias a los modelos de comportamiento que obtienen reconocimiento social en la red.
Article
Persuasive testimonials are common in commercial, nonprofit, and public health contexts. They pose challenges to existing theories of narrative persuasion because they are typically both narrative and overtly persuasive. Prior research has suggested testimonials may be effective with counter-attitudinal recipients by decreasing negative affective responses and increasing meaningful affect. Often, however, testimonials may address behaviors that are anxiety provoking rather than counter-attitudinal; prior research provides little theoretical or empirical guidance concerning message influence in the face of such anxiety. An experiment comparing a testimonial versus a non-narrative message advocating end-of-life conversations found that the testimonial message increased behavioral intentions via meaningful affect and self-efficacy. The testimonial did not decrease anxiety, and there was no differential impact on high versus low anxiety recipients. The authors conclude that a eudaimonic testimonial may serve as a motivator of behavior regardless of anxiety concerning the message topic, as well as a means of increasing self-efficacy.
Article
Full-text available
Health misinformation could distort people's health perceptions and related behaviors. Combating such misinformation has become the common concern of the whole society. Based on the extended parallel process model (EPPM) and narrative theories, the current study recruited 415 health experts in China to participate in a 2 (threat: high vs. low) ×2 (efficacy: high vs. low) between-subjects factorial experiment. Results revealed that narrative fear appeal messages are effective in promoting health experts to correct online health misinformation for the public. Message threat can arouse the emotion of fear through the identification with the narrative character. Moreover, message efficacy plays a moderating role between fear and health misinformation correction. Theoretical implications for EPPM development and practical message strategies for motivating health experts to correction misinformation are discussed.
Article
Full-text available
This paper presents the findings of an exploratory study on the continuously generating Big Data on Twitter related to the sharing of information, news, views, opinions, ideas, knowledge, feedback, and experiences about the COVID-19 pandemic, with a specific focus on the Omicron variant, which is the globally dominant variant of SARS-CoV-2 at this time. A total of 12,028 tweets about the Omicron variant were studied, and the specific characteristics of the tweets that were analyzed include sentiment, language, source, type, and embedded URLs. The findings of this study are manifold. First, from sentiment analysis, it was observed that 50.5% of tweets had a ‘neutral’ emotion. The other emotions—‘bad’, ‘good’, ‘terrible’, and ‘great’—were found in 15.6%, 14.0%, 12.5%, and 7.5% of the tweets, respectively. Second, the findings of language interpretation showed that 65.9% of the tweets were posted in English. It was followed by Spanish or Castillian, French, Italian, Japanese, and other languages, which were found in 10.5%, 5.1%, 3.3%, 2.5%, and <2% of the tweets, respectively. Third, the findings from source tracking showed that “Twitter for Android” was associated with 35.2% of tweets. It was followed by “Twitter Web App”, “Twitter for iPhone”, “Twitter for iPad”, “TweetDeck”, and all other sources that accounted for 29.2%, 25.8%, 3.8%, 1.6%, and <1% of the tweets, respectively. Fourth, studying the type of tweets revealed that retweets accounted for 60.8% of the tweets, it was followed by original tweets and replies that accounted for 19.8% and 19.4% of the tweets, respectively. Fifth, in terms of embedded URL analysis, the most common domain embedded in the tweets was found to be twitter.com, which was followed by biorxiv.org, nature.com, wapo.st, nzherald.co.nz, recvprofits.com, science.org, and other domains. Finally, to support research and development in this field, we have developed an open-access Twitter dataset that comprises Tweet IDs of more than 500,000 tweets about the Omicron variant, posted on Twitter since the first detected case of this variant on 24 November 2021.
Article
Full-text available
Rates of human papillomavirus (HPV) infection are highest in young adults, who can be vaccinated against HPV if they were not vaccinated as adolescents. Since young adults increasingly access health information on social media, we tested the impact of a social media campaign with narrative-based health information on intentions related to HPV vaccination. We also aimed to understand which ads resonated most with young adults and led to higher survey completion rates. We created social media posts featuring videos promoting HPV vaccination. We launched a sponsored ad campaign on Facebook to reach young women, ages 18-26, across the country. Participants were randomly assigned one of 6 videos and then completed a brief survey about video engagement and intentions to: talk with a health care professional, talk with friends or family, and vaccinate against HPV. A descriptive correlational design and a test for moderation were used to explore hypothesized relationships. Across all ads, 1332 link clicks led to 991 completed surveys that were reduced to 607 surveys (95% ages 18-26, 63% non-Caucasian; 58% sexually active). Higher video engagement was associated with stronger intentions to talk with a health care professional (r = .44, p = .01), talk with friends/family (r = .52, p = .01), and vaccinate against HPV (r = .43, p = .01). Young adults were receptive to watching narrative-based health information videos on social media. When promoting HPV vaccination, more engaging information leads to greater intentions to talk about the vaccine and get vaccinated.
Article
Full-text available
Considering how communities perceive the threat and risks of COVID-19, it is essential to examine how emotional regulation stimulated through intrinsic and extrinsic incentive mechanisms via social media can reinforce ‘Stay at home’ intentions. The conceptual framework was developed using the elements of the Extended Parallel Process Model (EPPM). A self-reported questionnaire was used to measure individuals’ intention to stay at home during the pandemic based on the perceived locus of causality as a part of self-determination theory. The empirical research was conducted on a sample of 306 USA respondents. The study results indicate that both components of the EPPM—efficacy and threat—positively affect ‘stay at home’ intentions. Moreover, a positive effect of efficacy on threats was found, as was a moderating effect of threats on the relationship between efficacy and the intention to stay at home. Meanwhile, the influence of social media exposure on threats and behavioral intentions was not significant. People are likely to stay at home as a preventive measure during COVID-19 if there is enough threats and efficacy. However, the abundance of information and opinions in social media can lead to a decreased perceived threat and might disrupt the acceptance of preventive actions.
Article
Full-text available
With the constantly mutating of SARS-CoV-2 and the emergence of Variants of Concern (VOC), the implementation of vaccination is critically important. Existing SARS-CoV-2 vaccines mainly include inactivated, live attenuated, viral vector, protein subunit, RNA, DNA, and virus-like particle (VLP) vaccines. Viral vector vaccines, protein subunit vaccines, and mRNA vaccines may induce additional cellular or humoral immune regulations, including Th cell responses and germinal center responses, and form relevant memory cells, greatly improving their efficiency. However, some viral vector or mRNA vaccines may be associated with complications like thrombocytopenia and myocarditis, raising concerns about the safety of these COVID-19 vaccines. Here, we systemically assess the safety and efficacy of COVID-19 vaccines, including the possible complications and different effects on pregnant women, the elderly, people with immune diseases and acquired immunodeficiency syndrome (AIDS), transplant recipients, and cancer patients. Based on the current analysis, governments and relevant agencies are recommended to continue to advance the vaccine immunization process. Simultaneously, special attention should be paid to the health status of the vaccinees, timely treatment of complications, vaccine development, and ensuring the lives and health of patients. In addition, available measures such as mix-and-match vaccination, developing new vaccines like nanoparticle vaccines, and optimizing immune adjuvant to improve vaccine safety and efficacy could be considered.
Article
Full-text available
The design of campaigns for the improvement of intergroup attitudes requires innovative approaches that consider both the characteristics of the messages and the psychological processes they evoke. This work addresses the study of factors that increase the persuasive effectiveness of testimonial messages aimed at improving attitudes towards stigmatized immigrants. An experiment was conducted using a representative sample of 443 participants of Spanish origin on the effect of similarity to the protagonist and the narrative voice. Two mediating mechanisms (identification with the protagonist and cognitive elaboration) were evaluated, and the indirect effect of the two independent variables was studied with respect to two dependent variables: the attitude towards immigration and the intention to collaborate with NGOs to support immigrants. Similarity to the protagonist of the narrative message increased identification only when the participants read the version written in the first person. In addition, a conditional process model was tested, revealing that identification increased cognitive elaboration, which, in turn, was associated with a more favorable attitude towards immigration and a greater intention to collaborate with immigrant support organizations. This study highlights the relevance of the characteristics of narrative messages to increase affective (identification) and cognitive (elaboration) processes that explain their persuasive impact. The results are discussed in the context of research on narrative persuasion and the design of campaigns for the prevention of racism and xenophobia.
Article
Full-text available
Applying disposition theory to narrative persuasion, this study examined how audience members’ enjoyment of a narrative promotes persuasion differently than transportation and identification. In a 2 (affective disposition: liked vs. disliked story character) × 2 (framing: gain vs. loss framed story) between-subject experiment, participants (N = 295) read a story in which a liked or disliked character has either a positive outcome (a gain frame) or a negative outcome (a loss frame) dependent on the story character’s engagement in sun protection behaviors. Consistent with disposition theory, participants enjoyed the story more when a liked character was in a gain-framed (vs. loss-framed) narrative; however, no framing effect was found for a disliked character. This interactive effect on enjoyment, in turn, mediated participants’ intentions to engage in sun protection behaviors. Affective disposition and framing independently influenced transportation and identification. Transportation mediated the effect of affective disposition on behavioral intention, but identification did not. This study demonstrates distinctive narrative conditions that prompt enjoyment, transportation, and identification in different ways and, in turn, lead to persuasive effects.
Article
Rational Overcoming the COVID-19 pandemic requires large-scale cooperation and behavior change on an unprecedented scale. Individuals can help reduce the burden of the pandemic by participating in behaviors that benefit people whose life circumstances make them especially vulnerable. Objective We tested the effect of reading narrative (i.e., story-like) as opposed to expository (i.e., factual recounting) messages on beliefs about protecting others in groups vulnerable during the pandemic through increased message transportation (i.e. immersing the reader into the story). Additionally, we examined if reading narratives, as opposed to expository messages, increased intentions to engage in prosocial behaviors that benefit these groups through increased transportation. Methods The study used a between-subjects design where participants either read narrative or expository messages about the experiences of people who were at greater exposure to SARS-CoV-2 due to social and political factors, namely people who were incarcerated or working in healthcare during the onset of the COVID-19 pandemic. Results In line with pre-registered hypotheses, participants in the narrative (vs. expository) condition reported greater transportation into the message. We also observed indirect effects of narrative (vs. expository) messages, through increased message transportation, on: (1) beliefs that by physical distancing, one can protect vulnerable people (2) beliefs that members of the target groups (i.e., healthcare workers and people who are incarcerated), were vulnerable during the pandemic, (3) intentions to engage in prosocial behaviors that help family and friends, and (4) intentions to engage in prosocial behaviors that help members of vulnerable groups. Conclusion Together these results suggest that narratives can be used to motivate prosocial action during the COVID-19 pandemic to the extent that the narratives elicit transportation.