ArticlePDF Available

Message fatigue: Conceptual definition, operationalization, and correlates


Abstract and Figures

Despite growing concern over the public’s fatigue towards inundated health messages, communication research has largely neglected such ramifications of prolonged, real-life campaign exposure. This paper offers an initial conceptual and empirical treatment of message fatigue, an important, but understudied, side effect of campaigns. Specifically, it proposes conceptual and operational definitions of the construct and examines psychometric characteristics of a proposed message fatigue scale. The findings from two studies concerning safe-sex (N =412) and anti-obesity messages (N =396) demonstrated solid support for the scale’s unidimensionality. In support of construct validity, the scale exhibited significant associations with message avoidance, annoyance, information seeking, and desensitization. Moreover, in an experimental setting in Study 2, message fatigue negatively predicted attention and message elaboration, while positively predicting counterargument.
Content may be subject to copyright.
Message Fatigue: Conceptual Definition, Operationalization, and Correlates
Jiyeon So, Ph.D.
Soela Kim, M.A.
Heather Cohen, M.A.
University of Georgia
Author notes. Jiyeon So is an assistant professor in the Department of Communication
Studies at the University of Georgia. Soela Kim and Heather Cohen are doctoral students in
the Department of Communication Studies at the University of Georgia.
The authors thank Drs. Lijiang Shen, Lucy Popova, and Hyunyi Cho for their helpful
comments on the manuscript. This work was supported by the University of Georgia
Research Foundation under Grant 2964.
Correspondence to Jiyeon So, Department of Communication Studies, University of Georgia,
603 Caldwell Hall, Athens, GA, USA 30602. Phone: 706-542-4893; Fax: 706-542-3245;
Despite growing concern over the public’s fatigue towards inundated health messages,
communication research has largely neglected such ramifications of prolonged, real-life
campaign exposure. This paper offers an initial conceptual and empirical treatment of
message fatigue, an important, but understudied, side effect of campaigns. Specifically, it
proposes conceptual and operational definitions of the construct and examines psychometric
characteristics of a proposed message fatigue scale. The findings from two studies
concerning safe-sex (N =412) and anti-obesity messages (N =396) demonstrated solid support
for the scale’s unidimensionality. In support of construct validity, the scale exhibited
significant associations with message avoidance, annoyance, information seeking, and
desensitization. Moreover, in an experimental setting in Study 2, message fatigue negatively
predicted attention and message elaboration, while positively predicting counterargument.
Key words: message fatigue, health campaigns, message avoidance, information
seeking, desensitization, counterargument, scale validation
Message Fatigue: Conceptual Definition, Operationalization, and Correlates
“Even the most powerful symbols lose their impact if they are constantly repeated” – Downs
In modern life, we are bombarded with health messages –be it from news reports or
public service announcements– communicating the importance of healthy behaviors, of
which we are already well aware. Commonly referred to as message fatigue, the
phenomenon of feeling tired of receiving similar messages has become increasingly prevalent,
particularly in the health message domain due to the omnipresence of health messages that
promote similar, if not the same, health behaviors. More important, as prolonged and
repeated exposure is an essential antecedent to fatigue (Calder & Sternthal, 1980), available
research on this phenomenon suggests that it may be particularly relevant for health issues
that have received disproportionate public health attention for decades, such as obesity,
tobacco use (Shanahan & Elliot, 2009), and safe sex (Frew et al., 2013).
Contrary to the seeming prevalence of message fatigue in everyday life, however,
research on the topic is quite nascent. No research to date clearly defines the construct on
either a conceptual or operational level, let alone examines possible effects of it. There
seems to be two major reasons for the paucity of communication research on this subject.
First, research on message effects has primarily focused on the effects of one-time exposure
to a single message studied in isolation and, consequently, largely neglected the ramifications
of real-life exposure to multiple, similar messages over time (Cho & Salmon, 2007). As a
result, we have not yet generated much knowledge on the cumulative effects of exposure to
numerous iterations of similar messages, or to a “class of messages,” over an extended period
of time (Slater, 2004, p. 171). Second, many communication scholars seem to assume that
more message exposure is better or more effective than less exposure, without recognizing
the potential risks of overexposure. This assumption has led scholars to become much more
concerned about audiences’ lack of repeated exposure to health campaign messages (e.g.,
Hornik, 2002) than about overexposure to such messages. To be sure, repetition may lead to
greater awareness and recognition (Jeong, Tran, & Zhao, 2012), but a growing body of
research suggests that excessive repetition may be counterproductive (e.g., Baseman et al.,
2013). In sum, while it is vital to ensure substantive exposure conducive to positive
outcomes, it is equally imperative to recognize, understand, and guard against potential
consequences of too much exposure, especially in a media-saturated society.
This paper offers an initial conceptual and empirical treatment of message fatigue,
which is one of the important, but understudied, side effects of prolonged message exposure.
Specifically, this research sheds light on what message fatigue is, how it can be
operationalized, and how it might influence processing of related health messages. The
article starts with an overview of research on multiple-message exposure to establish a basis
for understanding the contexts conducive to message fatigue. The paper then offers
conceptual and operational definitions of message fatigue, followed by an examination of the
psychometric characteristics of the message fatigue scale developed, utilizing data obtained
from two studies concerning fatigue towards safe sex (Study 1) and obesity-related messages
(Study 2, wave 1), respectively. Moreover, to further test the scale’s construct validity, the
relationships between message fatigue and a set of key message processing variables–
attention, message elaboration, and counterargument– are examined in an experimental
setting (Study 2, wave 2).
Message Fatigue: A Consequence of Overexposure!
Is More Exposure Necessarily Better?
As the premise that more exposure may not necessarily translate into greater effects
is central to the notion of message fatigue, we review extant research on multiple-message
exposure effects that provides evidence for the premise. Although not directly concerning
“message” fatigue per se, a foundation for understanding the effects of multiple message
exposures can be traced back to Zajonc’s (1968) work on mere exposure. Consistent with
Zajonc’s hypothesis, considerable research demonstrated that repeated, unreinforced
exposure to an unfamiliar stimulus (e.g., Chinese characters) can lead to positive evaluation
of the stimulus. As empirical tests of the hypothesis accumulated, however, evidence of an
inverted-U shape relationship between exposure frequency and evaluation emerged,
indicating an existence of a threshold level of exposure frequency, after which more exposure
damaged the evaluation of the stimuli (see Bornstein, 1989 for a meta analysis). Despite the
difference in the stimuli types employed in mere exposure studies (e.g., Chinese characters)
and those in advertisement research, advertisement research largely replicated these previous
findings. Initial repetitions led to positive evaluation of the advertisements, but once the
exposure frequency reached a certain threshold point, additional exposure rendered audiences
worn out, (Calder & Sternthal, 1980) and decreased the effectiveness of the advertisements
(e.g., Belch, 1982; Cacioppo & Petty, 1979).1
More relevant to the current study, there have been notable lines of media effects and
campaign research that speak meaningfully to the potential risks of excessive exposure. For
example, the inverted-U shape relationship has also been observed between exposure to news
reports containing a crime-related stereotype and related stereotypical perceptions (Arendt,
2015), and social media messages on hazards and risk perceptions (Lu, Xie, & Liu, 2015). In
exploring possible causes of the observed inverted-U shape relationships, Koch and Zerback
(2013) found that perceived persuasive intent decreased the overall positive effects of
exposure frequency on message effects. This finding underscores the importance of
considering risks of overexposure in campaign contexts, wherein persuasive intent is
inarguably present. Although far from conclusive due to the paucity of research employing
multiple exposures (Arendt, 2015), a growing body of campaign research also suggests that
overexposure to campaign messages can lead to unfavorable outcomes, such as a decrease in
anti-tobacco attitude (Reinhard, Schindler, Raabe, Stahlberg, & Messner, 2014), perceived
threat of AIDS (Schindler, Reinhard, & Stahlberg, 2011), and recall of the health message
content (Baseman et al., 2013).
Boundary Condition for Message Fatigue!
As the aforementioned review suggests, while repetition may be beneficial in
increasing familiarity (Jeong et al., 2012), considerable research suggests that excessive
message exposure can render the message ineffective in terms of persuasion. Given this
thesis, we now turn to message fatigue, the focal concept of the current research, which we
argue is one of the crucial consequences of overexposure to a class of messages, and a
psychological state that facilitates the decline in message effectiveness or “diminishing
returns” (Silk, Atkin, & Salmon, 2011, p. 217).
Because the concept of message fatigue has been used rather loosely and
indiscriminately across disciplines, we lay out a set of boundary conditions of the concept
used in our inquiry. First, the current study focuses on fatigue toward similar, as opposed to
identical, messages. By similar messages, we mean messages that share a common,
overarching theme or goal, such as inducing a global behavior (e.g., weight management),
and therefore can be considered a “class of messages” (Slater, 2004, p. 171). Such distinction
between message types is useful because some domains of research (e.g., media effects,
campaign research) focuses on fatigue caused by excessive exposure to a class of similar
messages, while others (e.g., advertisement research) consider fatigue caused by repeated
exposure to an identical message. Given that we are interested in the real-life fatigue
experience caused by inundation of various messages addressing a common health concern,
we take a more ecologically sound approach and focus on the former.
Second, research on the latter type of fatigue, which has its foundation in mere
exposure research, typically involves fatigue occurring in a relatively short period of time.
We will refer to this type of fatigue as acute message fatigue. Acute message fatigue is akin
to message wearout studied widely in advertising research. To the contrary, the type of
message fatigue we focus on in the current study is caused by excessive accumulation of
exposure to similar messages, which typically takes place after a prolonged period of time.
We refer to this type of message fatigue as chronic message fatigue. Because of the reasons
articulated earlier in the introduction, research on chronic message fatigue has been rather
Third, the “message” component of chronic message fatigue concerns both mediated
and interpersonal messages. This is because campaign messages often stimulate other forms
of mediated (e.g., news reports) and interpersonal communication (e.g., conversation with a
family member) about a campaign issue in the long term and these various modes of
communication collectively determine the overall campaign effects (Atkin, 2001; Hornik,
2002). Moreover, available research has documented that message fatigue can be generated
by both mediated (e.g., Kinnick, Krugman, & Cameron 1996) and interpersonal sources (e.g.,
Stockman et al., 2004), suggesting that the concept is not limited to one type of
communication and should encompass various modes of communication. Thus, we elected
not to specify the source of messages as one or the other, and left “message” purposefully
broad and general in the conceptual and operational definitions. Note that, in comparison,
research on acute message fatigue has focused on mediated messages (e.g., advertisements)
only. The difference between acute and chronic message fatigue identified using the three
criteria is summarized in Figure 1.
Conceptual Definition
Although research on message fatigue is dispersed and quite scarce, the existing
research converges on a number of core, defining characteristics of the concept that represent
different, but related, facets of the overarching construct. We drew from existing research on
both acute and chronic message fatigue to identify conceptual dimensions of message fatigue
because the essence of being fatigued by messages is shared across the two subtypes and
research on the acute one can meaningfully inform the conceptualization of the chronic one.
Extant research suggests that message fatigue encompasses some combinations of the
following four dimensions: 1) perceived overexposure, or perception that one has been
exposed to a class of messages beyond desired frequency (Frew et al., 2013; Herbst et al.,
2007), 2) perceived redundancy, or perception that the messages are repetitive and
overlapping (Frew et al., 2013; Kinnick et al., 1996), 3) exhaustion, or a feeling of being
burned out (Gorn & Goldberg, 1980; Kinnick et al., 1996), and 4) tedium, or lack of
enthusiasm (Schumann & Clemons, 1989).
The four dimensions of message fatigue are often intertwined and appear together in
definitions or descriptions of message fatigue experience. Perceived overexposure and
redundancy of messages have been identified across various domains of research as key
aspects of message fatigue. For example, in the public health domain, in which much
research on the concept has focused on safe sex messages, message fatigue has been defined
as the loss of interests in HIV prevention messages or counseling due to overexposure to safe
sex messages (Herbst et al., 2007) and a phenomenon of being weary of too many prevention
messages (Ogden & Bergmann, 2010). Message repetitiveness or redundancy has also been
noted as a critical aspect of the message environment conducive to safe sex message fatigue
(Frew et al., 2013).
A media effects perspective on message fatigue offered by Kinnick et al. (1996)
sheds more light on the construct. They focused on a specific type of message fatigue caused
by inundation of mass media messages about tragic social issues (e.g., child abuse), which
they labeled as compassion fatigue. Kinnick et al. (1996) described fatigued audiences as
those “grown weary of unrelenting media coverage of human tragedy and ubiquitous fund-
raising appeals” (p. 687). They further posited that the primary antecedent to message fatigue
is message saturation caused by pervasive media messages about a given issue and “flooding
of multiple communication channels with redundant [italics added] messages” (p. 689). To
clarify, while overexposure and redundancy are identified as the two crucial environmental
factors that induce message fatigue, it is the audiences’ subjective perceptions of those
environmental factors (i.e., perceived overexposure, perceived redundancy) that should be
conceptualized as key elements of the audiences’ fatigue experience.
On the more experiential side of message fatigue, Kinnick et al. (1996) described
emotional exhaustion or burnout as an important way in which fatigue manifests itself.
Likewise, being tired or burned out by overwhelming amounts of repetitive messages often
appears in focus group interviews with fatigued individuals (e.g., Frew et al., 2013; Gorn &
Goldberg, 1980). This aspect of fatigue is also reflected in Kinnick et al.’s (1996) definition
of compassion fatigue, which was stated succinctly as “burnout toward social problems” (p.
687) portrayed in the mass media.2 Emphasizing the exhaustive state one experiences upon
being fatigued by a host of messages, marketing scholars went so far as to label the
phenomenon as message “wearout” (Calder & Sternthal, 1980, p. 173).
In addition to exhaustion, tedium reaction or boredom has been identified as another
central element of fatigue experience. For example, Cacioppo and Petty's (1979) two-stage
attitude modification model posits that overexposure to repetitive messages leads to tedium
reactions, which, they argue, is responsible for the loss of message effectiveness. Likewise,
Schumann and Clemons (1989, p. 529) referred to the phenomenon of people being tired of
repetitive messages as “tedium reaction,” emphasizing the centrality of this dimension in the
fatigue experience. Relatedly, drawing from two decades of mere exposure research,
Bornstein (1989) identified boredom generated by excessive exposure as the primary
mechanism for the decline of evaluation that occurs after the apex in the exposure-evaluation
relationship has been reached.
Given the four conceptual dimensions of message fatigue extracted from the extant
research, we define message fatigue as an aversive motivational state of being exhausted and
bored by overexposure to similar, redundant messages over an extended period of time. The
construct is defined as a motivational state because the extant research suggests that it
energizes two important message-processing tendencies: disengagement with the message
(e.g., inattention, avoidance; Calder & Sternthal, 1980) and resistance toward persuasive
messages (e.g., counterargument; Cacioppo & Petty, 1979).3 With the conceptual definition
and the four dimensions in mind, we now review the existing instruments measuring message
fatigue with an eye toward examining if the four dimensions are appropriately reflected in the
existing operationalization of the construct.
Operational Definition
Besides advertisement research, in which fatigue has been typically operationalized
in terms of its immediate effects (e.g., inattention) as opposed to the actual state and
experience of being “worn out” (Belch, 1981), the vast majority of empirical research on
chronic message fatigue has relied on either anecdotal evidence or evidence from qualitative
interviews (e.g., Chillag et al., 2002; Frew et al., 2013; Shanahan & Elliott, 2009).
Consequently, operationalization of the construct as a psychological state has remained
rudimentary. Although there has been no research proposing and validating a message
fatigue scale, there have been notable efforts to operationalize the construct. For example,
Kinnick et al. (1996) used a single-item measure asking respondents to report their “feelings
of burnout” towards various social issues (p. 692). Focusing on the message environment
aspect, Stockman et al. (2004) measured message fatigue with an item, “I have heard enough
about AIDS and don't want to hear any more about it.” Similarly, Arnold et al. (2014)
measured HIV message fatigue with two items: “I am tired of thinking about HIV” and
“There is too much focus on HIV.” Although Arnold et al.’s items do not directly address
fatigue from messages per se, they are nonetheless useful in identifying recent development
on the operationalization of the construct. Of the existing measures, a 5-item scale developed
by Frew et al. (2013) seems to offer the most nuanced understanding of the complex nature of
the construct. The scale reflects three dimensions of message fatigue identified in the
previous section: One item concerns overexposure (“At this point, I’ve heard that using
condoms prevents AIDS more than I ever needed to”), two items concern redundancy (“After
seeing them for years, safe sex brochures and pamphlets seem repetitive,” and “Messages
about safe sex are all beginning to sound the same to me”) and exhaustion (“I am weary of
hearing how important it is to use condoms,” and “People I know are burned out on safe sex
The existing instruments, however, suffer from several issues. First, some of the
items lack semantic consistency and content validity. They concern different subjects
experiencing fatigue (e.g., “People I know” versus “I”) or do not directly concern fatigue
from messages per se (“I am tired of thinking about HIV”). Second, despite the literature
pointing to tedium reaction as an integral part of fatigue, this dimension is not represented in
any of the operational definitions reviewed. Lastly, and most importantly, a systematic
validation of the scale as a unidimensional construct has not been performed. Frew et al.
(2013) conducted an exploratory factor analysis and reported the reliability coefficient of
their scale as evidence for internal consistency. However, internal consistency is an indicator
of reliability, and as such, has no bearing on the validity or unidimensionality of a scale. A
confirmatory factor analysis, as opposed to an exploratory one, must be performed to
examine the factor structure (Brown, 2015; Levine, 2005). Furthermore, nomological
network construct validation (Cronbach & Meehl, 1955), in which empirical relationships
between the construct in question and measures of other supposedly related constructs are
examined, has not been performed to date. Given the methodological issues, we developed a
message fatigue scale based on the existing instruments and conceptual definitions (see Table
1) and performed a series of validation tests. As a first step toward the nomological network
construct validation, we describe the variables below, with which message fatigue scale
should show significant associations.
Study 1
The extant research suggests a number of constructs that are expected to correlate
with message fatigue. First, in Kinnick et al.'s (1996) study, selective avoidance of related
media content was one of the strongest correlates of message fatigue. Similarly, interviews
conducted by Chillag et al. (2002) shows that gay men “don’t want to go to another condom-
on-a-banana demonstration” (p. 32), indicating that fatigued individuals desire to avoid
stimuli related to the fatigue. Relatedly, audiences who are overwhelmed by repetitive
advertisements are shown to actively seek devices that enable them to avoid those
advertisements (Roberts & Alpert, 2010). Thus, we expect that more fatigued individuals
will exhibit stronger tendency to avoid exposure to related messages.
Second, fatigued individuals who, by definition, have been exposed to similar
messages beyond desired frequency, are also likely to express annoyance toward related
messages. Since annoyance or frustration is evoked when one’s goals are obstructed by
situational factors (Roseman, 1984), we expect exposure to undesired messages to cause
annoyance. Indeed, Edwards, Li, and Lee (2002) reported that people expressed irritation
toward exposure to unwanted advertisements. Similarly, Gorn and Goldberg (1980) observed
that children exposed to repetitive advertisement expressed annoyance by making remarks
such as “Oh no, not again” or “not another one” (p. 424).
Third, another important correlate of message fatigue is information seeking. When
one is tired of overexposure to similar messages, it is plausible to expect him/her to be
unmotivated to seek further information about the message subject. Consistent with this
expectation, Kinnick and colleagues (1996) found that more fatigued individuals were less
likely to seek information about the subject. Similarly, in the safe sex message fatigue
context, fatigued individuals are equated with those who do not show interests in obtaining
safe-sex-related information (e.g., Herbst et al., 2007).
In order to test the divergent validity of the scale, we identified health-related
variables that are expected to show no significant association with message fatigue. First,
health locus of control, defined as one’s belief that he or she is in control of his or her health
(Wallston, Wallston, & DeVellis, 1978), is predicted to have no significant association with
message fatigue. Second, there is no reason or relevant research to anticipate that health
literacy, or individual’s ability to process, understand, and evaluate information needed to
make public health decisions (Freedman et al., 2009), to be significantly related to message
fatigue. In sum, we predict:
H1 (convergent validity): Message fatigue will be positively associated with a) message
avoidance, b) annoyance, and negatively associated with c) information seeking.
H2 (divergent validity): Message fatigue will not be significantly associated with a)
health locus of control and b) health literacy.
Scale development. The research team initially developed scale items based on a
review of existing research and measurements. The items were then reviewed by a content
expert in health communication for face and content validity. After the expert review, a pilot
test of the items was conducted with undergraduate students at a large public university in the
U.S. (N = 105) to identify ambiguous or unclear items. The scale used in both the pilot study
and Study 1, which employed college student samples, concerned fatigue toward safe sex
messages as safe sex messages typically target young adults.
Participants and procedure. The initial set of scale items reflecting the four
dimensions of message fatigue toward safe sex messages were administered to college
students via an online survey software Qualtrics. A total of 412 participants were recruited
through a research subject pool at a large public university in the U.S. There were more
females (67%) than males, and an average participant was 18.60 years old (SD = 3.29). The
participants received class credit for their participation. The majority described themselves
as Caucasian (70.6%), with the remainder identifying as African-American (6.6%), Hispanic
(1.7%), Asian (12.9%), multiracial (7.0%), and others (1.0%; 0.2% missing data). A little
less than half (42.72%) reported that they were sexually active.
Measures. Unless noted otherwise, all items were measured with a seven-point
Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). The questionnaire
contained items in the order they are presented in this manuscript.
Message fatigue. Initially, a total of 16 items were developed to measure the four
dimensions of message fatigue. Five items from Frew et al.’s study were included (see Table
1). Results of an initial confirmatory factor analysis indicated that three reverse items did not
load well onto respective factors (all λ .37). Thus, these three items were excluded from the
subsequent analyses, resulting in a total of 13 items (M = 4.31, SD = 1.16, α = .93; see Table
Avoidance. In order to measure avoidance toward safe sex messages, we created
four items based on Kinnick et al.’s (1996) study that examined the relationship between
compassion fatigue and avoidance of media messages on social issues. The four items
included: “If I could avoid hearing about issues related to unprotected sex, I would,” “I often
try to walk away from a situation in which issues related to unprotected sex are discussed,”
“If I see a program on TV that concerns safe sex, I would turn the channel to something
different,” and “If I see campaign posters promoting safe sex, I turn away” (M = 3.15, SD =
1.22, α = .91).
Annoyance. Annoyance toward safe sex messages was assessed with two items
including, “When I hear safe sex messages, I get annoyed” and “When I hear safe sex
messages, I get irritated” (M = 2.29, SD = 1.24, r = .88).
Information seeking. Information seeking about safe sex was measured with three
items adopted from Niederdeppe et al.'s (2007) study. The items assessed the extent to which
the participants made active efforts to obtain information about safe sex from their doctors,
on the Internet, and from magazines within a year (M = 2.43, SD = 1.57, α = .76).
Health locus of control. Six items from Wallston and colleagues (1978) health
locus of control scale were adopted to measure the construct. Sample items included “I am in
control of my healthand “When I get sick I am to blame” (M = 4.91, SD = 1.11, α = .85).
Health literacy. The Single Item Literacy Screens (SILS; Wallace, Rogers, Roskos,
Holiday, & Weiss, 2006) was used to measure health literacy. The item was "How confident
are you filling out medical forms by yourself?” measured on a five-point Likert-type scale
ranging from 1 (not at all confident) to 5 (as confident as I can be; M = 3.73, SD = .82).
The dimensionality of the message fatigue scale was examined with confirmatory
factory analysis (CFA), and with tests of external and internal consistency (Hunter & Gerbing,
1982). The scale also underwent tests of face, content, and construct validity as guided by
DeVellis (2012).
Confirmatory factor analysis. CFA using Mplus version 7.0 (Muthén & Muthén,
1998–2012) was conducted to examine the scale's dimensionality. Specifically, in order to
examine the feasibility of the scale's unidimensionality, a higher-order CFA was pursued (see
Brown, 2015). Four steps were taken following Brown (2015): 1) fit the first-order four
factor model, 2) examine the magnitude and pattern of associations among the first-order
factors, 3) fit the higher order factor model, and 4) compare the fit of the lower- and higher-
order models. The first-order four factor model showed good fit to data, χ2(231) = 486.01, p
< .001, CFI = .95, RMSEA = .05 (.05, .06), SRMR = .05. Factor loadings ranged from .54
to .96 and correlations between each pair of four factors were significant and high, ranging
from .48 to .75 (all p < .001). Since significant associations among the four factors were
observed, higher-order factor was deemed plausible (Brown, 2015). In a close examination
of the correlation between the first-order factors, a pattern of correlations indicative of two
second-order factors was observed: While the overexposure-redundancy pair (r = .75) and
exhaustion-tedium pair (r = .73) correlated highly within the pair, the tedium dimension
showed relatively lower associations with both overexposure (r = .48) and redundancy (r =.
57). As overexposure and redundancy can be conceptualized as aspects of message fatigue
concerning perceptions about one’s message environment, and exhaustion and tedium focus
on audience's responses to those perceptions, the observed two higher-order dimensions
deemed conceptually plausible as well (see Brown, 2015). Thus, examinations of these two
higher-order dimensions were included in the subsequent analyses.
Since empirical feasibility of two second-order factors was observed, a total of five
factor structures were examined for comparison: first-order single factor, first-order four
factor, second-order single factor, second-order two factor, and third-order single factor.
Since the third-order factor in the third-order single factor model had two indicators (i.e.,
audience response and message environment dimensions), it was under-identified. Thus,
following convention (e.g., Boster, Kotowski, Andrews, & Serota, 2011; Shen, Condit, &
Wright, 2009), three external variables that were expected to correlate with message fatigue
(H1a-c) were included in the CFA model to permit comparisons of the model fit across the
five structural models. In the first-order single factor model, all items loaded onto a single
factor in the first-order without any specification of dimensions. This model purported to be
used in comparison to other structural models with four dimensions specified, in order to
empirically examine the feasibility of the four dimensions. In the first-order four factor
model, all four factors were allowed to be associated with each other and all of the external
variables. In the second-order two factor model, the two second-order factors were allowed
to be associated with each other and the external variables. In both second-order and third-
order single factor model, only the highest-order factor was allowed to be associated with the
external variables. The three external variables were specified as single-indicator latent
variables in all five models, following Bollen (1989). As the models estimated covariance
between message fatigue factors and the external variables, the procedure also served as tests of
external consistency of the scale.
The results indicated that all factor models except the first-order single factor model
had good fit to the data (see Table 2). The first-order single factor model with no dimensions
specified did not fit the data well, providing empirical evidence for the existence of the four
dimensions. Thus, the following discussion will focus on the comparison of the four
structural models that contains the four dimensions (see Figure 2). Higher-order models
cannot improve model fit relative to first-order models, in which factors are freely
intercorrelated (Brown, 2015). Thus, if statistical equivalence between a first-order and a
higher-order model can be demonstrated, the higher-order model can be accepted as a
plausible account of the data (Brown, 2015). Although all of the fit indices of all four models
were comparable, the third-order single factor model was a particularly close approximation
to the first-order model in terms of fit: CFI, RMSEA, and SRMR were identical between the
two models. Moreover, the BIC difference between the two models was 16.35, in favor of
the third-order single factor model (see Raftery, 1995). Taken together, the result showed
that the third-order unidimensional model is statistically equivalent to the first-order four
factor model and, therefore, can be accepted as the final factor structure of the message
fatigue scale.
Construct validity. Construct validity was examined by testing convergent (H1)
and divergent (H2) validity. Consistent with H1, message fatigue showed positive
associations with message avoidance (r = .50; H1a) and annoyance (r = .43; H1b) and a
negative association with information seeking (r = -.20; H1c) in the CFA model (all p < .001).
A series of partial correlation analyses controlling for age and sex was conducted to further
test H1 and H2. Consistent with the results of CFA, the pattern of associations predicted in
H1 was also observed in the partial correlation analyses (see Table 3). Thus, the results
collectively support H1a, H1b, and H1c and, consequently, the scale's convergent validity. In
addition, as predicted by H2, message fatigue did not show significant associations with
health locus of control (rp = .04, p = .43; H2a) and health literacy (rp = .05, p = .34; H2b).
Thus, both H2a and H2b were supported, demonstrating the scale's divergent validity.
Overall, the findings support the construct validity of the message fatigue scale.
External consistency. Another piece of evidence supportive of unidimensionality of
a scale comes from both lower-order factors and higher-order factors producing similar
patterns of correlations with external variables (Hunter & Gerbing, 1982). Overall, the CFA
results show that the lower order and higher order factors exhibited similar patterns of
associations with the external variables both in terms of magnitude and direction (see Table
3). One exception was found in the first-order four factor CFA, in which overexposure had a
non-significant relationship with information seeking (r = -.12, p = .09). However, the
direction of association was consistent with the expectation (H1c) and the magnitude of
association was similar to that between redundancy and information seeking (r = -.13, p
= .04). Furthermore, the result of the partial correlation analysis demonstrated a significant
association between overexposure and information seeking as expected (see Table 3). Thus,
it is concluded that lower-order factors exhibited reasonable consistency in their relationships
with external variables in comparison to each other and higher-order factors..
Internal consistency. The four first-order factors showed good reliabilities ranging
from .76 to .94 (see Table 3). The two second-order factors also showed good
reliabilities, .86 for message environment dimension, which combined overexposure and
redundancy items, and .94 for audience response dimension, which combined exhaustion and
tedium items. Finally, the entire message fatigue scale showed good reliability (α = .93).
Study 2
In Study 2, the message fatigue scale was reexamined with a national adult sample
and a different message topic (anti-obesity messages). The purpose of Study 2 was twofold:
First, it sought to replicate the findings from Study 1 with a different health message topic
that has also received much public health attention, which renders message fatigue relevant.
Second, we aimed at further substantiating the construct validity of the scale by examining its
relationships to message processing variables that are expected to significantly correlate with
message fatigue. In what follows, we describe the additional external variables examined in
Study 2.
Desensitization, or the phenomenon of becoming apathetic or numb to an issue due
to a prolonged, repeated exposure to issue-related messages (Cho & Salmon, 2007), was
included as an additional external variable to be tested for construct validity. Kinnick et al.
(1996, p. 689) described this “narcotizing dysfunction” of inundated messages as one of the
characteristic manifestations of burnout. Likewise, in safe sex message fatigue research,
complacency or lack of fear towards consequences of unsafe sex, which is essentially
desensitization, appears quite frequently as a consequence of fatigue (e.g., Chilag et al., 2002).
Thus, we expect more fatigued individuals to show greater levels of desensitization.
In addition, as message fatigue is expected to relate to two types of message
processing tendencies– disengagement from and resistance towards related messages– we
sought to further corroborate the scale’s construct validity by examining the scale’s
relationships to relevant message processing variables assessed in an experiment involving an
exposure to an anti-obesity message. This test also served to broaden our understanding of
the fatigue construct and its possible effects on message processing.
First, disengagement from anti-obesity messages was operationalized as low levels of
attention and message elaboration. Attention paid to an anti-obesity message was expected to
correlate negatively with fatigue towards such messages. Much research shows that greater
frequency of repeated exposure to a message leads to a substantial decrease in attention paid
to the message (e.g., Grass & Wallace, 1969). Similarly, research on safe sex message
fatigue also identifies tendency to “tune out” as a consequence of being weary of too many
HIV prevention messages (e.g., Ogden & Bergmann, 2010). Thus, greater fatigue will
negatively predict attention paid to an anti-obesity message. Relatedly, message elaboration
is also expected to negatively correlate with message fatigue. As we observed that more
fatigued individuals desired to avoid exposure to related messages in Study 1, it is logical to
expect them to be less motivated to invest in thinking and elaborating on a related message.
Indeed, Shi and Smith (2016) found that the proportion of message-related thoughts
decreased significantly when a fear appeal message was repeated.
Second, we also examined the scale’s relationship to counterargument, a form of
resistance toward message acceptance, which is expected to positively correlate with message
fatigue. Counterargument, or “thoughts that dispute or are inconsistent with the persuasive
argument’’ (Slater & Rouner, 2002, p.180), has been studied widely as an important type of
resistance toward persuasive messages. Research examining the effects of message repetition
shows that more fatigued individuals are more likely to produce more counterarguments
toward the message. For example, Belch (1982) reported that negative cognitions including
source derogation and counterargument were more pronounced when a message was repeated
at a greater frequency. Miller (1976) also found that high levels of exposure to political
campaign messages failed to induce attitude change, arguably because (unwanted) high
exposure loads caused audiences to feel that their freedom was threatened, which induced
reactance, which is, in part, comprised of counterargument (Dillard & Shen, 2005). Similarly,
Cacioppo and Petty (1979) found that counterargument initially decreased, but then increased
as frequency of message repetition increased. The foregoing discussion leads to the
following hypotheses.
H1 (convergent validity, replication): Message fatigue will be positively associated with a)
message avoidance, b) annoyance, and negatively associated with c) information seeking.
H2 (divergent validity, replication): Message fatigue will not be significantly associated
with a) health locus of control and b) health literacy.
H3 (convergent validity, addition in an experiment): Message fatigue will be negatively
associated with a) attention and b) message elaboration, and positively associated with c)
desensitization, and d) counterargument towards a fatigue-related message.
Participants and procedure. A national adult sample was recruited through a
sampling service provided by Survey Sampling International. Data utilized in Study 2 were
collected as a part of a larger project involving an experiment on overweight and obese
subjects. Individuals who were over 18 years old and either overweight or obese were
eligible to participate. Study 2 incorporated two waves, which both utilized the online survey
software Qualtrics. In the first wave survey, the subjects completed an online questionnaire
containing items measuring fatigue towards anti-obesity messages, demographic variables,
and other external variables. Those who completed the first wave survey were invited to
participate in the second wave online experiment one day after their participation in the first
wave survey. In the second wave, the participants were exposed to an obesity-related
message that discussed reasons (frames manipulated in a 2 [health vs. life satisfaction
enhancement frame] by 2 [loss vs. gain frame] between subjects design) and behavioral
recommendations for weight management (e.g., using smaller plate in meal time). The
subjects received monetary compensation for their time upon participation in each phase of
the study. Of the 396 respondents who completed the first wave survey, 55.3% were female.
An average respondent was 44.92 years old (SD = 13.05) and the majority described
themselves as Caucasian (77.5%), with the remainder identifying as African-American
(9.3%), Hispanic (3.0%), Asian (2.0%), Native American (1.0%), multiracial (4.3%), and
others (1.8%; 1.0 % missing data). 46.9% were overweight and 53.1% were obese, with a
body mass index (BMI) averaging at 34.41 (SD = 25.74). In the second wave, a total of 312
subjects participated, resulting in an attrition rate of 21.21%.
Measures. The same items in Study 1 were adopted in Study 2 to measure
avoidance (M = 3.40, SD = 1.73, α = .92), annoyance (M = 3.24, SD = 1.83, r = .95),
information seeking about obesity (M =3.33, SD =1.79, α = .82), health locus of control (M =
5.09, SD = 1.04, α = .81), and health literacy (M = 5.68, SD = 1.22; all items measured with a
seven-point scale).
Message fatigue. In addition to the 13 items retained in Study 1, four new items
(one item for each dimension) were added to the scale. In addition, one of the three reversed
items that did not load well in Study 1 was also included in Study 2 for a reexamination.
However, the initial first-order four factor CFA of the 18 items showed that the reversed item
did not load well again in Study 2 (λ = .28). Thus, this item was excluded and the subsequent
analyses were conducted with a 17-item scale (M = 4.14, SD = 1.43, α = .96; see Table 1).
Desensitization. Four items were created based on Cho and Salmon’s (2007) and
Kinnick et al.'s (1996) conceptualization of desensitization. These items included: “Obesity-
related messages no longer arouse any emotions in me,” “I am immune (not responsive) to
obesity-related messages,” “I feel indifferent about obesity-related messages,” and “I feel
numb to obesity-related messages” (M = 3.67, SD = 1.78, α = .95).
Message elaboration. Reynolds’s (1997) seven-item scale was used to measure
message elaboration. Of the seven items, two were reversed. Sample items included: While
reading the message, “I was attempting to analyze the issues in the messageand “I was
extending a good deal of cognitive effort” (M = 4.73, SD = 1.17, α = .77).
Attention. Four items from So, Kuang, and Cho's study (2016) were utilized to
measure participants’ attention paid while reading the message. Of the four items, two were
reversed items. Sample items included: “I rushed through the message without being really
attentive to the information provided” (reverse-coded) and “I paid great attention to the
information provided” (M = 5.28, SD = 1.33, α = .82).
Counterargument. Four items from Nabi, Moyer-Gusé, and Byrne’s (2007) study
were adopted to measure participants’ counterargument to the messages. Sample items
included: “I found myself looking for flaws in the message presented,” and “I found myself
thinking of ways I disagreed with what was being presented” (M = 2.74, SD = 1.71, α = .93).
Confirmatory factor analysis. The same CFA procedure employed in Study 1 was
undertaken to reexamine the dimensionality of the scale. Consistent with Study 1, the first-
order single factor model did not show good fit to data (see Table 2). Thus, the remaining
four structural models that contain the four dimensions were considered in the subsequent
analyses (see Figure 2). The first-order four factor model showed good fit to data, χ2(436) =
782.28, p < .001, CFI = .96, RMSEA = .05 (.04, .05), SRMR = .05. Factor loadings ranged
from .60 to .97 and correlations between each pair of four factors were significant and high,
ranging from .65 to .85 (all p < .001). Again, significant associations among the four factors
were observed and higher-order factor was pursued. A pattern of covariance similar to that
found in Study 1 was observed: Exhaustion and Tedium dimensions showed stronger
association with each other (r = .85) than with the other two dimensions (r ranging from .65
to .75). Thus, the four different factor structures examined in Study 1 were reexamined in
Study 2 (see Figure 2).
In order to test the third-order single factor model, four external variables measured
in the first wave survey (H1a-c, H3c) were included in the CFA. The results indicated that all
four models had good fit to data (see Table 2). Although all of the fit indices of the four
models were comparable, the third-order single factor model again showed a very close
approximation to the first-order model in terms of fit. The third-order single factor model
had identical fit indices with the first-order four factor model in terms of CFI, RMSEA, and
SRMR. Again, BIC of the third-order single factor model was the lowest of the four models
and, specifically, lower than the first-order model by 15.05, which suggests superiority of the
third-order model (Raftery, 1995). Thus, with the evidence indicating the statistical
equivalence of the third-order single factor model to the first-order four factor model, the
former was embraced as the final structure of the scale (see Figure 3 for the final third-order
single factor model).
Construct validity. Construct validity was examined by testing convergent (H1, H3)
and divergent validity (H2). As predicted, message fatigue was positively associated with
message avoidance (r = .77; H1a), annoyance (r = .72; H1b), desensitization (r = .72; H3c)
and negatively associated with information seeking (r = -.27; H1c) in the CFA model (all p
< .001). In addition, a series of partial correlation analyses was conducted, controlling for
age and sex to test construct validity further. The pattern of associations predicted in H1 was
also observed in the partial correlation analyses (see Table 3). In addition, consistent with H3,
message fatigue had a significant negative relationship with attention to an anti-obesity
message (rp = -.16, p = .004; H3a) and message elaboration (rp = -.13, p = .03; H3b).
Message fatigue positively predicted counterargument (rp = .31, p < .001; H3d) as well.
Since message frames were manipulated in a 2 x 2 factorial design in the experimental data, a
more rigorous test of the associations that controls for the possible influence of the message
frames was in order. Thus, a series of multiple regression analyses, in which message fatigue,
age, sex, two frame conditions, and interaction term of the two frame conditions were entered
as predictors for each dependent variable, was conducted. The results remained similar to
those of the partial correlation analyses: Message fatigue was negatively associated with
attention to the message (β = -.18, p = .005) and cognitive elaboration (β = -.13, p = .04) and
positively associated with counterargument (β = .35, p < .001). Thus, H1a-c and H3a-d were
supported, demonstrating the scale's convergent validity further. Lastly, message fatigue did
not show significant associations with health locus of control (rp = - .07, p = .16; H2a) and
health literacy (rp = .02, p = .67; H2b) as expected. Thus, both H2a and H2b were supported.
Together, the findings support the construct validity of the scale.
External consistency. The lower-order and higher-order factors produced similar
patterns of covariance with the external variables with the exception of first-order factor
overexposure, which showed a significant relationship with health literacy (rp = .12, p = .02;
see Table 3). Given the overall consistency in the pattern of associations with the external
variables, the scale can be considered to be externally consistent.
Internal consistency. The four first-order factors, as well as the two second-order
factors, showed good reliabilities ranging from .82 to .97 (see Table 3). The overall message
fatigue scale also showed good reliability (α = .96), providing evidence for internal
Despite its relevance in the public’s everyday life, as well as in campaign research,
message fatigue has remained largely an elusive concept with little empirical scrutiny. This
study took an important step toward elucidating the construct and sensitizing communication
scholars to an area of research that has been overlooked. Specifically, this study proposed
conceptual and operational definitions of message fatigue and empirically examined the
psychometric characteristics of the scale proposed.
The findings from two studies employing two different samples and message topics
demonstrated solid support for the proposed message fatigue scale. The four dimensions of
message fatigue– perceived overexposure, perceived redundancy, exhaustion, and tedium–
formed a unidimensional structure on the third-order with two second-order dimensions of
message environment and audience response. The scale also exhibited strong associations
with external variables including message avoidance, annoyance, and information seeking in
both Study 1 and Study 2. In addition, in the experiment employing an anti-obesity message
in Study 2, anti-obesity message fatigue assessed in wave 1 negatively predicted attention to
and cognitive elaboration of the message assessed in wave 2. Furthermore, message fatigue
positively predicted counterargument to the message, substantiating the view that
conceptualizes message fatigue as a facilitator of motivated resistance toward persuasive
While a good deal of empirical evidence for the scale’s validity was observed, there
were two minor deviations from the expectations in the test of external consistency.
Interestingly, the two unexpected findings both concerned the perceived overexposure
dimension. While all other three lower-order dimensions exhibited significant negative
relationships with information seeking, the relationship between overexposure and
information seeking fell short of the significance level of α = .05 in Study 1. As low
reliability can lead to attenuation of associations (DeVellis, 2012), we speculate that it may
be due to the relatively low reliability of the overexposure dimension (α = .76). This
speculation seems quite plausible since we found a significant relationship between
overexposure and information seeking in Study 2, in which the alpha reliability of the
overexposure dimension improved (α = .82).4
Another unexpected finding pertains to the relationship between overexposure and
health literacy found in Study 2. While we do not have a satisfactory explanation for it, we
suspect that more health literate people might have had greater access to health-related
resources, which might have led to greater exposure to obesity-related messages. However,
since this relationship was not observed in Study 1, we refrain from making a firm conclusion.
Implications for Future Research
This research offers several avenues for future research on health message design and
planning. First, of the external variables that correlated significantly with fatigue, message
avoidance warrants serious scholarly attention because exposure is a necessary condition for
campaign effects (Hornik, 2002). A good starting point for exploring this issue might be to
identify the profile of the individuals who are tired of hearing about a given health message to
the point where they simply want to walk away from a situation in which the issue is
communicated (i.e., avoidance). Profiling highly fatigued individuals in terms of
demographic, psychological traits (e.g., boredom susceptibility; Farmer & Sundberg, 1986),
and risk status (e.g., BMI, men who have sex with men) would not only provide us with a
more nuanced understanding of the message fatigue construct, but also help us address issues
related to selective avoidance to campaign messages.
Second, this study leads us to an important question that stems from a dilemma we
are facing: Although excessive exposure may generate message fatigue, reducing the volume
of health messages is not an option. There exists a diverse group of audiences, some of
which are in grave need of influence but are difficult to reach. For these individuals, we need
to maintain certain levels of persistent messaging to inform and alert them. On the other
hand, however, the persistent messaging may simultaneously render other subgroups of
individuals who are, for example, relatively easy to reach or more susceptible to boredom, to
become fatigued. Knowledge on the optimal level of exposure frequency that can facilitate
learning and acceptance in individuals who need influence, while keeping fatigue at bay
would be indispensable. This inquiry is not new. This study merely emphasizes the
importance of it. This inquiry has been addressed to some extent by studies that documented
substantial decline of advertisement effectiveness after three to five repetitions (e.g., Belch,
1982; Gorn & Goldberg, 1980). However, when it comes to addressing fatigue toward health
messages, a slightly different approach that concerns overexposure to a class of similar, and
not necessarily identical, messages purporting to induce a common, overarching behavior is
needed. In other words, we have ample empirical evidence regarding the threshold level for
acute fatigue from advertisement research but we currently do not have comparable
knowledge regarding chronic fatigue. While we believe acute and chronic message fatigue
likely have much in common, there could be a nuanced difference that has not yet been
explored. For example, unlike acute message fatigue, the chronic one involves a much more
complicated set of contextual factors (e.g., mode of communication, prolonged exposure) that
we need to consider when identifying the optimal exposure.
While the extant research clearly indicates that exposure frequency is a critical factor
conducive to fatigue, it has not adequately addressed whether or not some properties of
messages may also make individuals more fatigued. Identification of content-specific factors
that might induce greater levels of fatigue would be instrumental in designing messages that
can minimize fatigue and related negative outcomes. For example, does the type of campaign
message– awareness, instruction, or persuasion messages (Atkin, 2001)–matter? Given the
role perceived persuasive intent plays in the effects of repeated exposure (Koch & Zerback,
2013), we speculate that persuasion messages, or messages featuring why the advocated
behavior should be adopted (Atkin, 2001), may generate greater levels of fatigue than other
types. An answer to this inquiry may also guide the decisions regarding optimum
combination of the three types of message (Atkin, 2001).
Another message-specific factor that may play a significant role in generating fatigue
is message frame. Research on advertisement repetition suggests that message variation in
parts of the content could delay fatigue (e.g., Schumann & Clemons, 1989). In the health
communication context, in which disproportionate attention is given to select health issues, a
great volume of messages inevitably concerns promotion of similar if not the same behavior
(e.g., smoking cessation). In these situations, the reasons for the behavioral adoption
communicated via different frames, as opposed to the behaviors themselves, could serve as a
point of variation. For instance, a vast majority of messages promoting health behaviors (e.g.,
reducing meat consumption) use frames related to health consequences (e.g., heart disease)
when the behavior can be framed with relatively more novel reasons (e.g., being
environmentally friendly). Theory of perspective by incongruity (Burke, 1954) suggests that
change to a perspective that views the same subject from a radically different angle can take
the subject from a context in which it has been habitually construed. Thus, the change in
perspective by reframing can refresh audiences’ perceptions about the behavior and,
consequently, circumvent cognitive habituation and resultant fatigue. From a public health
perspective, health consequences may be the actual reasons why related health behaviors are
being advocated. However, if they render the audiences more fatigued and lead to message
avoidance, non-health frames that can break the habits of cognitive association (or over-used
schema) and energize them to attend to and elaborate on the message to a greater extent could
be a strategic move for health communicators.
Lastly, in addition to elucidating on message fatigue, we hope this research will
stimulate more thoughts about unintended effects of communication in general. It is not
enough to examine if messages have succeeded or failed to generate intended outcomes, but
also important to examine if they ended up generating unintended outcomes, which are often
undesirable (Cho & Salmon, 2006). We can examine a certain effect only when we are aware
of the possibility of its existence. In this regard, this study joins a burgeoning body of
communication research that aim at alerting scholars to different types of possible unintended
effects such as reactance (Dillard & Shen, 2005), boomerang effect (Byrne & Hart, 2009),
and information overload (Jensen et al., 2014). As the discipline introduces different types of
unintended effects, researchers will benefit from a more nuanced discussion about the
commonalities, differences, and interrelationships among the unintended consequences of
communication. For example, in this study, message fatigue positively predicted
counterargument, which is conceptualized as a cognitive dimension of reactance (Dillard &
Shen, 2005). Given the relationship observed, what can we say about the relationship
between message fatigue and reactance? Does message fatigue make one more reactive to
the message? In addition, we articulated how message fatigue differs from information
overload on a conceptual level (see Footnote 3). An empirical test of this argument would be
insightful in enhancing our understanding of the two types of related but distinct unintended
effects. In fact, the discriminant validity of the message fatigue scale could be tested by
showing how message fatigue and information overload predict different variance in related
outcomes (e.g., annoyance).
To conclude, this study offered a substantial foundation for message fatigue, an
understudied construct, importance of which has been alluded to in previous research. As
demonstrated in this study, it is an aversive motivational state associated with unfavorable
outcomes. A number of important questions that were beyond the scope of the current
manuscript remain and were posed for future research. Given the detrimental nature of
message fatigue, answers to these questions would be instrumental in devising ways to
minimize or circumvent fatigue and, consequently enhance effectiveness of communication
Arendt, F. (2015). Toward a dose-response account of media priming. Communication
Research, 42, 1089-1115. doi: 10.1177/0093650213482970
Arnold, M. P., Andrasik, M., Landers, S., Karuna, S., Mimiaga, M. J., Wakefield, S., et al.
(2014). Sources of racial/ethnic differences in awareness of HIV vaccine trials.
American Journal of Public Health, 104, 112-118. doi:10.2105/AJPH.2014.301893
Atkin, C. K. (2001). Theory and principles of media health campaigns. In R. E. Rice & C. K.
Atkin (Eds.), Public communication campaigns (3rd ed., pp. 49–68). Thousand Oaks,
CA: Sage.
Baseman, J.G., Revere, D., Painter, I., Toyoji, M., Thiede, H., & Duchin, J. (2013). Public
health communications and alert fatigue. BMC Health Services Research, 13, 295.
Bawden, D., & Robinson, L. (2009). The dark side of information: Overload, anxiety, and other
paradoxes and pathologies. Journal of Information Science, 35, 180–191. doi:
Belch, G. E. (1981). An examination of comparative and noncomparative TV commercials: The
effects of claim variation and repetition on cognitive response and message acceptance.
Journal of Marketing Research, 18, 333- 349. doi: 10.2307/3150974
Belch, G. E. (1982). The effects of television commercial repetition on cognitive response and
message acceptance. Journal of Consumer Research, 9, 56–65.
Bollen, K.A. (1989). Structural equations with latent variables. New York: Wiley.
Bornstein, R. F. (1989). Exposure and affect: overview and meta-analysis of research, 1968–
1987. Psychological Bulletin, 106, 265–289.
Boster, F. J., Kotowski, M., Andrews, K., & Serota, K. (2011). Identifying influence:
Development and validation of the connectivity, persuasiveness, and maven scales.
Journal of Communication, 61, 178–196. doi:10.1111/j.1460-2466.2010.01531.x
Brown, T. A. (2015). Confirmatory factor analysis for applied research (2nd ed.). New York,
NY: Guilford Press.
Burke, K. (1954). Permanence and change: An anatomy of purpose. Los Altos, CA: Hermes.
Byrne, S., & Hart, P. S. (2009). The “boomerang” effect: A synthesis of findings and a
preliminary theoretical framework. Annals of the International Communication
Association, 33, 3-37.
Cacioppo, J. T., & Petty, R. (1979). Effects of message repetition and position on cognitive
response, recall and persuasion. Journal of Personality and Social Psychology, 37, 97-
109. doi:10.1037/0022- 3514.37.1.97
Calder, B., & Sternthal, B. (1980). Television commercial wearout: An information processing
view. Journal of Marketing Research, 17, 173–186.
Charles, E. P. (2005). The correction for attenuation due to measurement error: clarifying
concepts and creating confidence sets. Psychological Methods, 10, 206-226. doi:
Chillag, K., Bartholow, K., Cordeiro, J., Swanson, S., Patterson, J., Stebbins, S., Woodside, C.,
& Sy, F. (2002). Factors affecting the delivery of HIV/AIDS prevention programs by
community-based organizations. AIDS Education and Prevention, 14, 27-37.
Cho, H., & Salmon, C.T. (2007). Unintended effects of health communication campaigns.
Journal of Communication, 57, 293-317. doi:10.1111/j.1460-2466.2007.00344.x
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological
Bulletin, 52, 281-302.
DeVellis, R. (2012). Scale development: Theory and applications (3rd ed.). Thousand Oaks,
CA: Sage.
Dillard, J.P., & Shen, L. (2005). On the nature of reactance and its role in persuasion health
communication. Communication Monographs, 72, 144-168.
Edwards, S. M., Li, H., & Lee, J. H. (2002). Forced exposure and psychological reactance:
Antecedents and consequences of the perceived intrusiveness of pop-up Ads. Journal of
Advertising, 31, 83–96. doi:10.1080/00913367.2002.10673678
Eppler, M. J., & Mengis, J. (2004). The concept of information overload: A review of literature
from organization science, accounting, marketing, MIS, and related disciplines. The
Information Society, 20, 325–344. doi:10.1080/01972240490507974
Farmer, R., & Sundberg, N. D. (1986). Boredom proneness: The development and correlates of
a new scale. Journal of Personality Assessment, 50, 4 –17.
Freedman, D. A., Bess, K. D., Tucker, H. A., Boyd, D. L., Tuchman, A. M., & Wallston, K. A.
(2009). Public health literacy defined. American Journal of Preventive Medicine, 36,
446-451. doi: 10.1016/j.amepre.2009.02.001
Frew, P. M., Williams, V. A., Shapiro, E. T., Sanchez, T., Rosenberg, E. S., Fenimore, V. L., &
Sullivan, P. S. (2013). From (un)willingness to involveMENt: Development of a
successful study brand for recruitment of diverse MSM to a longitudinal HIV research.
International Journal of Population Research, 1–9.
Gorn, G. G., & Goldberg, M. E. (1980). Children's responses to repetitive television
commercials. Journal of Consumer Research, 6, 421-425.
Grass, R., & Wallace H., (1969). Satiation effects of TV commercials. Journal of Advertising
Research, 9, 3-8.
Herbst, J., Beeker, C., Mathew, T., McNally, W., Passin, L., Kay, N., et al. (2007). The
effectiveness of individual-, group- and community-level HIV behavioral risk reduction
interventions for adult men who have sex with men: A systematic review. American
Journal of Preventive Medicine, 32, 38-67. doi:10.1016/j.amepre.2006.12.006
Hornik, R. (2002). Public health communication: Making sense of contradictory evidence. In R.
Hornik (Ed.), Public health communication: Evidence for behavior change (3rd ed., pp.
1-19). Mahwah, MJ: Lawrence Earlbaum.
Hunter, J. E., & Gerbing, D. W. (1982). Unidimensional measurement, second order factor
analysis, and causal models. In B. M. Staw & L. L. Cummings (Eds.), Research in
organizational behavior (Vol. 4, pp. 267–320). Greenwich, CT: JAI Press.
Jensen, J. D., Carcioppolo, N., King, A. J., Scherr, C. L., Jones, C. L., & Niederdeppe, J. (2014).
The cancer information overload (CIO) scale: Establishing predictive and discriminant
validity. Patient Education and Counseling, 94, 90–96. doi:10.1016/j.pec.2013.09.016
Jeong, Y., Tran, H., & Zhao, X. (2012). How much is too much? The collective impact of
repetition and position in multi-segment sports broadcast. Journal of Advertising
Research, 52, 87–101. doi:10.2501/JAR-52-1-087-101
Kinnick, K. N., Krugman, D. M., & Cameron, G. T. (1996). Compassion fatigue:
Communication and burnout toward social problems. Journalism & Mass
Communication Quarterly, 73, 687-707. doi:10.1177/107769909607300314
Koch, T., & Zerback, T. (2013). Helpful or harmful? How frequent repetition affects perceived
statement credibility. Journal of Communication, 63, 993-1010. doi:10.1111/jcom.12063
Levine, T. R. (2005). Confirmatory factor analysis and scale validation in communication
research. Communication Research Reports, 22, 335-338.
Lu, X., Xie, X., & Liu, L. (2015). Inverted U-shaped model: How frequent repetition affects
perceived risk. Judgment and Decision Making, 10, 219–224. Retrieved from
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of
Psychology, 52, 397-422. .psych.52.1.397
Miller, R. C. (1976). Mere exposure, psychological reactance, and attitude change. Public
Opinion Quarterly, 40, 229-233. doi: 10.1086/268290
Muthén, L. K., & Muthén, B. O. (1998–2012). Mplus user’s guide (7th ed.). Los Angeles, CA:
Muthén & Muthén.
Nabi, R. L., Moyer- Gusé, E., & Byrne, S. (2007). All joking aside: A serious investigation into
the persuasive effect of funny social issue messages. Communication Monographs, 74,
29–54. doi:10.1080/03637750701196896
Niederdeppe, J., Hornik, R.C., Kelly, B.J., Frosch, D.L., Romantan, A., Stevens, R.S., Schwartz,
J.S. (2007). Examining the dimensions of cancer-related information seeking and
scanning behavior. Health Communication, 22, 153-167.
Ogden, J., & Bergmann, H. (2010). “Don’t let your loved ones get involved with a Fataki!”
Addressing intergenerational sex in Tanzania through the Fataki campaign. Case Study
Series. AIDS Support and Technical Assistance Resources, Task Order 1. Retrieved
Raftery, A. E. (1995). Bayesian model selection in social research. In P. V. Marsden (Ed.),
Sociological methodology (pp. 111-163). Cambridge: Basil Blackwell.
Reinhard, M-A., Schindler, S., Raabe, V., Stahlberg, D., & Messner, M. (2014). Less is
sometimes more: How repetition of an antismoking advertisement affects attitudes
toward smoking and source credibility. Social Influence, 9, 116-132.
Reynolds, R. A. (1997). A validation test of a message elaboration measure. Communication
Research Reports, 14, 269–278.
Roberts, C., & Alpert, F. (2010). Total Customer Engagement: Designing and aligning strategic
elements to achieve growth. Journal of Product & Brand Management, 19, 199-209.
Roseman, I. J. (1984). Cognitive determinants of emotions: A structural theory. In P. Shaver
(Ed.), Review of personality and social psychology (Vol. 5, pp. 11-36). Beverly Hills,
CA: Sage Publications.
Schumann, D., & Clemons, D. (1989). The repetition/variation hypotheses: Conceptual and
methodological issues. Advances in Consumer Research, 16, 529-534.
Schindler, S., Reinhard, M-A., & Stahlberg, D. (2011). Repetition of educational AIDS advertising
affects attitudes. Psychological Reports, 108, 693698. doi: 10.2466/01.13.PR0.108.3.693-
Shanahan, P., & Elliott, D. (2009). Evaluation of the graphic warnings on tobacco product packaging
2008: Executive summary, Canberra: Department of Health and Aging. Retrieved from
Shen, L., Condit, C. M., & Wright, L. (2009). The psychometric property and validation of a
fatalism scale. Psychology and Health, 24, 597–613. doi:10.1080/08870440801902535.
Shi, J., & Smith, S. W. (2016). The effects of fear appeal message repetition on perceived threat,
perceived efficacy, and behavioral intention in the extended parallel process model. Health
communication, 31, 275-286.
Silk, K. J., Atkin, C.K., & Salmon, C.T. (2011). Developing effective media campaigns for
health promotion. In T. L. Thompson, R. Parrott, & J. F. Nussbaum (Eds.), The
Routledge handbook of health communication (2nd ed., pp. 203–219). New York, NY:
Slater, M. D. (2004). Operational and analyzing exposure: The foundation of media effects
research. Journalism & Mass Communication Quarterly, 81, 168-183.
doi: 10.1177/107769900408100112
Slater, M. D., & Rouner, D. (2002). Entertainment-education and elaboration likelihood:
Understanding the processing of narrative persuasion. Communication Theory, 12, 173–
191. doi: 10.1111/j.1468-2885.2002.tb00265.x
So, J., Kuang, K., & Cho, H. (2016). Reexamining fear appeal models from cognitive appraisal
theory and functional emotion theory perspectives. Communication Monographs, 83,
120-144. doi: 10.1080/03637751.2015.1044257
Spearman, C. (1904). The proof and measurement of association between two things. The
American Journal of Psychology, 15, 72-101.
Stockman, J. K., Schwarcz, S. K., Butler, L. M., De Jong, B., Chen, S. Y., Delgado, V., &
McFarland, W. (2004). HIV prevention fatigue among high-risk populations in San
Francisco. Journal of Acquired Immune Deficiency Syndromes, 20, 432–434. Retrieved
Wallace, L.S., Rogers, E.S., Roskos, S.E., Holiday, D.B., & Weiss, B.D. (2006). Brief report:
Screening items to identify patients with limited health literacy skills. Journal of
General Internal Medicine, 21, 874-877. doi:10.1111/j.1525-1497.2006.00532.x
Wallston, K.A., Wallston, B.S., & DeVellis, R. (1978). Development of the multidimensional
health locus of control (MHLC) scales. Health Education & Behavior, 6, 160-170.
Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social
Psychology, 9, 1-27. doi:10.1037/h002584
1Bornstein’s (1989) meta-analysis concerned studies that employed persuasive effects
(e.g., affect or attitude towards the stimuli) as the outcome variable. Note that the observed
inverted U-shape relationship between message exposure frequency and (persuasive) outcome
may not apply to other types of outcomes such as familiarity or brand recognition (for example,
see Jeong et al., 2012 for the difference in the exposure effects on brand recognition and
advertising liking).
2Note that the term “burnout” used in the extant message fatigue literature has a rather
colloquial meaning, especially in comparison to the definition of “burnout” used in the job
burnout literature, which defines (job) burnout as “a prolonged response to chronic emotional
and interpersonal stressors” (Maslach, Schaufeli, & Leiter, 2001, p. 397).
3Message fatigue should be carefully distinguished from other types of unintended
effects of health messages that may seem similar, such as information overload. Information
overload refers to a state in which an individual is overwhelmed by the amount of relevant
information available to them (Bawden & Robinson, 2009; Jensen et al., 2014). Message
fatigue and information overload are similar in that both are caused by excessive exposure to
information and messages in the environment. However, the key difference is that message
fatigue concerns excessive exposure to similar messages over time, whereas information
overload is often generated by exposure to overwhelming amount of different, diverse, or
contradicting information (Eppler & Mengis, 2004). The latter type of information environment
hampers one’s ability to select the most useful information (Bawden & Robinson, 2009), as
opposed to making one bored and worn-out via repetition and/or overexposure.
4We tested this speculation by conducting an additional analysis correcting for
attenuation due to measurement error guided by Spearman (1904). Confidence intervals around
the corrected correlations were estimated by applying Charles’ (2005) approach. The corrected
correlation between overexposure and information seeking improved from -.12 to -.16, 95% CI
[-.28, -.04], thus providing empirical evidence that the lack of significant correlation likely have
been due to low reliability of the overexposure dimension in Study 1.
Table 1
Dimensions and Items of Message Fatigue Scale
M (SD)
Study 1
Study 2
Message Environment
(safe sex)
1. I have lost track of the amount of times I have heard that unprotected sex (obesity) is a serious problem.
5.75 (1.45)
4.96 (1.90)
2. At this point, I've heard about problems related to unprotected sex (obesity) more than I ever needed to.a
4.56 (1.69)
4.40 (1.79)
3. I have heard enough about how important it is to practice safe sex (maintain a healthy weight).
5.06 (1.60)
4.98 (1.76)
4. There are simply too many health messages about safe sex (obesity) nowadays.b
3.23 (1.60)
3.71 (1.87)
5. The importance of maintaining a healthy weight is overtaught.
2.98 (1.76)
6. Obesity-related messages rarely provide new information.
4.62 (1.81)
7. After hearing them for years, messages about safe sex (obesity) seem repetitive.a
4.86 (1.47)
4.83 (1.81)
8. Messages about safe sex (obesity) are all beginning to sound the same to me.a
4.66 (1.58)
4.88 (1.77)
9. I can predict what a message about safe sex (obesity) is going to say.
5.10 (1.48)
4.76 (1.75)
Audience Response
10. I am burned out from hearing that unprotected sex (obesity) is a serious problem.a
3.90 (1.67)
3.63 (1.96)
11. I'm sick of hearing about consequences of unprotected sex (problems associated with obesity).
3.81 (1.71)
3.61 (1.93)
12. I'm tired of hearing about the importance of maintaining a healthy weight.a
3.76 (1.67)
3.60 (1.98)
13. Obesity-related messages make me want to sigh.
3.74 (2.02)
14. Health messages about safe sex (obesity) are boring.
3.98 (1.59)
3.94 (1.85)
15. Safe sex (obesity-related) messages make me want to yawn.
3.64 (1.61)
3.69 (1.90)
16. I find messages about safe sex (obesity) to be dull and monotonous.
3.79 (1.59)
4.08 (1.88)
17. Obesity-related messages are tedious.
3.94 (1.85)
Fatigue Index
4.31 (1.16)
4.14 (1.43)
Note. a = items adopted from Frew et al. (2013). b = items adopted from Arnold et al. (2014).
Table 2
Model Fit Indices of the Four Different Structural Models of Message Fatigue Scale
Chi-square (df)
Study 1
First-order single factor
1457.90 (246)
.11 (.11, .12)
College (n = 412)
First-order four factor
486.01 (231)
.05 (.05, .06)
Second-order single factor
558.19 (242)
.06 (.05, .06)
Second-order two factor
515.30 (238)
.05 (.05, .06)
Third-order single factor
519.39 (240)
.05 (.05, .06)
Study 2
First-order single factor
1872.69 (454)
.09 (.09, .10)
Adult (n = 396)
First-order four factor
782.28 (436)
.05 (.04, .05)
Second-order single factor
846.06 (450)
.05 (.04, .05)
Second-order two factor
829.48 (445)
.05 (.04, .05)
Third-order single factor
833.33 (449)
.05 (.04, .05)
Note. External variables included in Study 1 were avoidance, annoyance, and information seeking. In Study 2, desensitization was added.
Table 3
Correlations between Factors and External Variables
Info seek
Study 1 (N = 412)
one factor
two factor
four factor
Study 2 (N = 396)
one factor
- .07
two factor
four factor
Note. Alpha reliability of each factor is on the diagonal. DS = desensitization, HLC = health locus of control, HL = health literacy. Values in the
parentheses are covariance from CFA. All other values are partial correlation coefficients with age and sex controlled. *p < .05, **p < .01, ***p
Figure 1. Two Different Types of Message Fatigue.
Exposure Period
Message Types
Communication Mode
Message Fatigue
Message Fatigue
Identical messages
Short period of time
Various modes of
Prolonged period of time
Similar messages
Figure 2. Factor Structures of the Four Factor Models Tested.
Figure 3. Standardized Parameter Estimates of the Third-order Single-factor Model for the
Message Fatigue Scale.
Note. Parameter estimates from the college student sample is on the left of the slash and those
from the adult sample is on the right of the slash. *Since the residual variance of “audience
response” latent factor was initially a small negative value and was creating a heywood case,
it was fixed to zero in the final model.
... Message fatigue is the aversive motivational state that results from excessive exposure to campaign messages or similar information over an extended period of time. When fatigued, individuals become less attentive, less responsive, and more resistant to campaign messages and related information (Kim & So, 2020;Koh et al., 2020;Reynolds-Tylus et al., 2020;So et al., 2017). Thus, understanding the bases and functioning of fatigue in persuasive health campaigns has obvious value. ...
... How do these multiple sources work together to influence fatigue? Second, as explicated, message fatigue is the consequence of repeated exposure to campaign messages over time (e.g., So et al., 2017). With few exceptions, however, fatigue research has employed only cross-sectional designs, which preclude conclusions about the dynamic behavior of fatigue (e.g., Ball & Wozniak, 2021). ...
... To the extent that it is viable to view the fatigue literature as a theory, its logic is embodied in just two propositions: (a) at some point, exposure causes fatigue and (b) fatigue produces one or more forms of resistance to the campaign. Regarding the second proposition, the literature has linked increased fatigue empirically with perceived overload as well as decreases in issue involvement and increases in perceived tedium (Gurr & Metag, 2021;So et al., 2017). Fatigued individuals become less attentive, less responsive, and more resistant to campaign messages and related information (Koh et al., 2020;So et al., 2017). ...
Message fatigue is the aversive motivational state that results from excessive exposure to campaign messages or similar information over an extended period of time. When fatigued, individuals become less attentive, less responsive, and more resistant to campaign messages and related information. Thus, understanding the bases and functioning of fatigue in persuasive health campaigns has obvious value. Despite considerable interest in this important topic, major questions remain under-studied. One such question hinges on the observation that campaigns are implemented in social systems, not laboratories. Apart from any direct effects that a campaign might produce, there is the potential for secondary exposure via individuals or other media that can yield distinct influences. How do these multiple sources work together to influence fatigue? Second, as explicated, message fatigue is the consequence of repeated exposure to campaign messages over time. With few exceptions, however, fatigue research has employed only cross-sectional designs, which preclude conclusions about the dynamic behavior of fatigue. How does fatigue change over the course of a campaign? Finally, the bases of fatigue are not entirely clear. Whereas fatigue is defined as a subjective judgment of excessive exposure, little is known about the affective processes underlying that judgment. How do emotional responses to a campaign amplify or attenuate fatigue? We examined these questions in the context of a campus COVID-19 safety behaviors campaign.
... Research in neighboring disciplines has demonstrated that individuals be come fatigued from health messages (So, Kim, & Cohen, 2017) and worn out by ad vertising (Craig, Sternthal, & Leavitt, 1976) when exposed too often. Similar mecha nisms have been found for social problems and human interest stories during media hypes; parts of the population become chronically fatigued and avoid these issues (Beyer & Figenschou, 2014;Kinnick, Krug man, & Cameron, 1996). ...
... Reactance as a negative cognitive and / or emotional reaction to exposure to different media stimuli has been found to correlate with avoidance behavior (Mar cinkowski & Došenovic´, 2020). Fatigue from health messages -which has cogni tive and emotional components (So et al., 2017) -leads to avoiding mediated health information. Finally, news overload can lead to information avoidance strategies, such as selective scanning (Lee, Holton, & Chen, 2019;Song, Jung, & Kim, 2017). ...
... overexposure Overexposure is the key factor for under standing the emergence of fatigue from news issues. It was first discussed within theoretical models of repetition effects (Berlyne, 1970;Cacioppo & Petty, 1979, 1989Stang, 1975), the concept of adver tising wearout (Cacioppo & Petty, 1980;Calder & Sternthal, 1980;Craig et al., 1976), for health messages (So et al., 2017), and persuasive statements in news articles (Koch & Zerback, 2013). After a thresh old point of repetition, the evaluation of a stimulus, such as an ad, can shift from positive to negative. ...
Full-text available
A significant amount of political communication research is grounded in the dynamics of the media’s and the public’s attention to public issues, assuming that the news media draw the public’s attention to issues, thereby fostering an informed and participating citizenry. However, there is evidence from several countries that this mechanism is disrupted for issues with high shares of news coverage during a period. Against this background, this article scrutinizes the idea that recipients become fatigued from these issues in the news. Having transferred findings on overexposure from other media stimuli to the news environment, issue fatigue is defined as a negative cognitive and affective state consisting of decreasing issue-specific information processing involvement, perceived information overload, and increasing boredom, annoyance, and anger toward an issue. Issue fatigue can lead to the avoidance of information about the issue, thus serving as a new explanatory approach to avoidance of media information at an issue level. Further consequences, causes, and the development of issue fatigue are discussed.
... Message fatigue has been studied in various contexts including safe-sex messages (So et al., 2017), anti-obesity/weight management messages (Kim & So, 2018;So et al., 2017), electronic cigarettes prevention messages (Keating & Galper, 2021), and bystander intervention to prevent sexual violence on campuses (Reynolds-Tylus et al., 2021). Recently, message fatigue has been applied to the context of COVID-19 related messages (Ball & Wozniak, 2021;Buneviciene et al., 2021). ...
... Message fatigue has been studied in various contexts including safe-sex messages (So et al., 2017), anti-obesity/weight management messages (Kim & So, 2018;So et al., 2017), electronic cigarettes prevention messages (Keating & Galper, 2021), and bystander intervention to prevent sexual violence on campuses (Reynolds-Tylus et al., 2021). Recently, message fatigue has been applied to the context of COVID-19 related messages (Ball & Wozniak, 2021;Buneviciene et al., 2021). ...
... We conducted a confirmatory factor analysis of the message fatigue construct. Specifically, we tested the factor structure proposed and validated by So et al., (2017) in its original formulation. So et al. (2017) proposed a third-order single factor structure, in which message fatigue is broken down into four subdimensions in the first order (i.e., "overexposure," "redundancy," "exhaustion," and "tedium"). ...
Based on the Risk Information Seeking and Processing Model, the present study examines whether COVID-19 message fatigue leads to greater information avoidance and heuristic processing, and consequently greater acceptance of misinformation. We conducted a survey of 821 Korean adults regarding their information seeking and processing regarding COVID-19 vaccination. Results of SEM analyses showed that COVID-19 message fatigue was (a) negatively related to information insufficiency and (b) positively related to information avoidance and heuristic processing. Information avoidance and heuristic processing were subsequently related to greater levels of misinformation acceptance. Theoretical and practical implications are discussed.
... So-called 'wear-out effects' (e.g., Chen et al. 2016) show associations between stimulus repetition and decreased purchase intention, among other variables. Recently, for example, prolonged exposure to health campaigns was found to reduce or eliminate the amount of attention paid to advertising campaigns (Kim and So 2018;So et al. 2017). ...
... Health research describes fatigue as an experience that has cognitive and emotional dimensions (Ream and Richardson 1996; for an overview, see also Patel 2010). In the context of information consumption, emotional and/or cognitive responses due to repetition have been identified, e.g., from a prolonged exposure to health messages (Kim and So 2018;So et al. 2017), from repetition of persuasive statements in news articles (Koch 2017;Koch and Zerback 2013), or as a response to certain characteristics of political news, such as negativity (Schemer 2014). Similarly, psychology of human perception and information processing has shown that excessive repetition of stimuli-in our context, a news issue-provokes negative emotions (Bornstein 1989) and a decline in cognitive engagement (Berlyne 1970). ...
Full-text available
This paper scrutinizes the phenomenon of issue fatigue and its consequences. Issue fatigue results from overexposure to a news topic that has been on the media’s agenda for an extended period of time. Fatigued recipients become annoyed, and no longer wish to be exposed to the topic. Based on the findings of an explorative qualitative study, a quantitative online survey was conducted in Germany, Mexico, and Pakistan (N = 481). Using cluster analysis, we identified an emotional and a cognitive type of issue fatigue, and investigated how these types react. Both types of fatigued recipients avoided further news about the respective issue in traditional news media (= information avoidance). Differences were observed concerning the strategies to handle fatigue (= coping strategies): recipients of the emotional type posted about their fatigue in social media; recipients of the cognitive type turned to information in sources other than the mainstream news. Taking into account country-specific differences, we concluded that, generally, issue fatigue—via information avoidance—results in an uninformed citizenry. This can be a hurdle for the functioning of an established democracy or for the success of democratic transitions. Posting about issue fatigue, which was more frequent in Mexico and Pakistan, might ‘infect’ others, and intensify problems resulting from issue fatigue. Turning to alternative sources can be either beneficial or problematic for the development of a well-informed citizenry, depending on whether alternative sources provide reliable and truthful information.
... The SDDOH and some municipal governments also implement advertising campaigns that encourage prevention measures such as using mosquito repellants. Targeting this messaging when and where WNV risk is highest may help to limit the phenomenon of message fatigue, in which repeated warnings are eventually ignored because of overexposure (So et al. 2017). ...
Background: West Nile virus (WNV), a global arbovirus, is the most prevalent mosquito-transmitted infection in the United States. Forecasts of WNV risk during the upcoming transmission season could provide the basis for targeted mosquito control and disease prevention efforts. We developed the Arbovirus Mapping and Prediction (ArboMAP) WNV forecasting system and used it in South Dakota from 2016 to 2019. This study reports a post hoc forecast validation and model comparison. Objectives: Our objective was to validate historical predictions of WNV cases with independent data that were not used for model calibration. We tested the hypothesis that predictive models based on mosquito surveillance data combined with meteorological variables were more accurate than models based on mosquito or meteorological data alone. Methods: The ArboMAP system incorporated models that predicted the weekly probability of observing one or more human WNV cases in each county. We compared alternative models with different predictors including a) a baseline model based only on historical WNV cases, b) mosquito models based on seasonal patterns of infection rates, c) environmental models based on lagged meteorological variables, including temperature and vapor pressure deficit, d) combined models with mosquito infection rates and lagged meteorological variables, and e) ensembles of two or more combined models. During the WNV season, models were calibrated using data from previous years and weekly predictions were made using data from the current year. Forecasts were compared with observed cases to calculate the area under the receiver operating characteristic curve (AUC) and other metrics of spatial and temporal prediction error. Results: Mosquito and environmental models outperformed the baseline model that included county-level averages and seasonal trends of WNV cases. Combined models were more accurate than models based only on meteorological or mosquito infection variables. The most accurate model was a simple ensemble mean of the two best combined models. Forecast accuracy increased rapidly from early June through early July and was stable thereafter, with a maximum AUC of 0.85. The model predictions captured the seasonal pattern of WNV as well as year-to-year variation in case numbers and the geographic pattern of cases. Discussion: The predictions reached maximum accuracy early enough in the WNV season to allow public health responses before the peak of human cases in August. This early warning is necessary because other indicators of WNV risk, including early reports of human cases and mosquito abundance, are poor predictors of case numbers later in the season.
... The effect of previous exposure to similar content has largely been neglected in mass communication studies [49,50]. Message fatigue can sometimes trigger an aversive motivational state [51]; it can lead to passive forms of resistance, e.g., message avoidance and inattention [52], as well as more active types of resistance, e.g., counterarguments [53]. Our results are consistent with the existing fatigue effect theory: the greater the fatigue of a message or exposure to repetitive promotional campaigns, the lower is the consumer's interest in related messages, and the lower is the cognitive effort required to understand it. ...
Full-text available
This study investigated the determinants of digital music success in South Korea. We identified information sources and factors that may influence the consumption of digital music, including song and artist factors, record label and distributor influence factors, promotional media influence factors, and electronic word of mouth (e-WOM). The analysis was conducted using music download data and ranking chart data from a major music platform. First, we found that traditional promotional media, such as the number of television or radio exposures and appearances in an audition program, significantly affected the success of music. Second, regarding social promotional media factors, promotional videos on YouTube mainly affected short-term success, whereas Twitter mentions showed an increasing influence over time. Third, an artist with a long career positively affected a song's early success, while frequent song releases had a negative impact. Our results offer marketers insights into promoting digital music in this new era.
The global outbreak of COVID-19 in 2020 has led to the dominance of COVID-19 prevention information on all media channels. Drawing on the ability–motivation model of information processing, this study examined how such an information overabundance hampered individuals’ ability and motivation to process in the era of COVID-19. With a survey conducted from 493 participants, we found that less message elaboration of COVID-19 prevention information was predicted by greater message fatigue, a state of low motivation due to information overabundance. In addition, greater message fatigue was accompanied by greater information overload, a state of low ability due to information overabundance. Moreover, certain motivation-related (i.e. health status, trait reactance and frequency of information seeking) and ability–related factors (i.e. health literacy, health status, trait anxiety and information quality) were found to be associated with message fatigue and information overload, respectively. The theoretical and practical implications are discussed.
Purpose Graphic health warnings (GHWs) on tobacco product packaging constitute one component within a multifaceted set of tobacco control measures. This study aims to understand whether consumers’ attention to GHWs will be associated with recall and quit intentions, using Australia as the case for this study. Design/methodology/approach Using the 14 GHWs currently in market as visual stimuli, non-probability intercept sampling was conducted, eye tracking and post-survey datasets were collected from a total of 419 respondents across three Australian cities. Findings Results show the front graphic image areas draw initial attention and the Quitline message area holds the longest attention duration. Attention is highly correlated with better quality of recall of health warning information, emotive responses, believability ratings among smokers and smokers’ perception of health risks and quit intentions. Associations are also noted with perceived health risk and quitting intentions. Originality/value To the best of the authors’ knowledge, this is the first study that has objectively tested the effectiveness of in-market tobacco GHWs in Australia and highlights eye tracking as a valid measurement approach that can enhance and drive new insights to evaluate consumer behaviour towards visual stimuli. This study extends new knowledge around the physiological relationships between viewing behaviours, health vulnerability perceptions and intentions to quit smoking, which has theoretical implications for the extended parallel process model which underpins this research.
The global social restrictions necessitated by the COVID-19 pandemic has resulted in a dramatic increase in the use of video conferencing for activities such as work, study, and personal relationships. Alongside its many benefits, video conferencing can also have adverse effects on users. Video conferencing fatigue is a commonly cited problem, especially for those individuals forced by COVID-19 to adopt the technology. Drawing from the technostress perspective, this paper examines the causes and consequences of VCF during a pandemic situation. A research model is developed and tested quantitively with data collected from 429 users of common video conferencing tools such as Zoom, Teams, and WebEx. The results suggest the relationship between video conferencing stressors and the outcomes of user satisfaction and continuance intentions, are mediated by video conferencing fatigue. In addition, the strengths of these relationships vary depending on whether video conferencing is mainly used for work, study, or social purposes.
Issue addressed: In Australia, cancer is the leading contributor to disease burden, with breast and bowel cancer among the most commonly diagnosed cancers. Despite the presence of community wide health promotion activities and screening programs, people living in regional and rural locations experience a number of factors which reduce breast and bowel cancer survival outcomes. This study investigates the ways that high-risk community members in a regional area of Australia interact with health messaging about breast and bowel cancer screening. Methods: A qualitative research method was used to conduct 31 in-depth one-on-one interviews with community members, leaders and essential service providers. A thematic approach was used to analyse data. Results: Findings provided insight to the ways that health is spoken about within the community, what prompts discussion of health, trustworthy sources of health information, and the significance of peer to peer communication. Conclusions: Existing community communication lines can be used to assist in delivering public health messages among high-risk and vulnerable population groups. Utilising community ambassadors is identified as a health promotion method for hard to reach populations. SO WHAT?: Conversations about health and screening amongst community members, and led by community members, play a key role in shaping engagement with cancer screening programs and represent an important site for health promotion activities. These findings have implications for future public health initiatives amongst high risk groups in regional locations.
Full-text available
We asked how repeated media reports on technological hazards influence an individual's risk perception. We looked for two contradictory effects, an increasing effect of repetition on perceived risk with the first few repetitions and a decreasing effect with later repetitions, leading to the inverted U-shaped pattern. In an experiment, we demonstrated the inverted U-shaped relationship between the repetition and perceived risk in the context of food risk. The finding broadens the range of mere-exposure effects and indicates that exposure to risk information can be a double-edged sword, which brings either an increasing or a decreasing perceived risk.
Full-text available
This study examined the effect of moderately repeated exposure (three times) to a fear appeal message on the Extended Parallel Processing Model (EPPM) variables of threat, efficacy, and behavioral intentions for the recommended behaviors in the message, as well as the proportions of systematic and message- related thoughts generated after each message exposure. The results showed that after repeated exposure to a fear appeal message about preventing melanoma, perceived threat in terms of suscept- ibility and perceived efficacy in terms of response efficacy significantly increased. The behavioral intentions of all recommended behaviors did not change after repeated exposure to the message. However, after the second exposure the proportions of both systematic and all message-related thoughts (relative to total thoughts) significantly decreased while the proportion of heuristic thoughts significantly increased, and this pattern held after the third exposure. The findings demonstrated that the predictions in the EPPM are likely to be operative after three exposures to a persuasive message.
Full-text available
Recognizing inconsistencies between the extant fear appeal theories and emotion literature, this research integrated cognitive appraisal theory and functional emotion theory into a fear appeal literature and proposed a model that describes a process through which both fear and anxiety can contribute to adaptive responses. Findings from an experiment (N = 927) supported the predictions. Fear and anxiety emerged as distinct constructs. Perceived susceptibility was a stronger predictor of anxiety than fear, while perceived severity was a stronger predictor of fear than anxiety. In addition, greater fear and anxiety led to greater response efficacy through increased motivation to obtain protection-related information and heightened attention to such information, thus mediating the threat and coping appraisal processes. The SEM model testing the predictions showed that perceived susceptibility had the strongest total effects on protection intention, followed by anxiety, perceived severity, and fear.
Communicative messages are often constructed strategically. In many cases, the creators of such messages strive to curtail specific anti-social or unhealthy attitudes and behaviors held by the target audience. However, these messages are not always successful in achieving the intended effect. Messages with a specific intent can backfire and cause an increase in the unhealthy or anti-social attitude or behavior targeted for change. We present a review of findings that have resulted in boomerang effects, broadly defined. An analysis of theoretical mechanisms for the effect eventuates in the proposal of two distinct paths to the boomerang. One path predicts that message receivers will process harmful elements in a message at the expense of those that were intended. The other path predicts that receivers will process the message as intended, but then resist complying with it. Finally, we offer a preliminary theoretical framework of boomerang effects.
Repetition of a pattern of television commercials caused wearout in viewers' evaluation of the commercials and the products being advertised. As predicted by an information processing view, wearout was not forestalled by strategies designed to enhance attention.
This study explored the collective impact of repetition and position on advertising effectiveness as evidenced through recognition and likeability of advertisements that were broadcast at different times in the Super Bowl. The findings indicate that brands advertised more in the first half and brands that appeared in both halves but shown more in one half than the other were better recognized than those equally promoted in both halves. Meanwhile, advertisements presented in both halves but repeated more in the second half were less favored than those evenly shown in both halves. The results support theories of repetition and primacy effects.
The impact of entertainment-education messages on beliefs, attitudes, and behavior is typically explained in terms of social cognitive theory principles. However , important additional insights regarding reasons why entertainment-education messages have effects can be derived from the processing of persuasive content in narrative messages. Elaboration likelihood approaches suggest that absorption in a narrative, and response to characters in a narrative, should enhance persuasive effects and suppress counterarguing if the implicit persuasive content is counterattitudinal. Also, persuasion mediators and moderators such as topic involvement should be reduced in importance. Evidence in support of these propositions are reviewed in this article. Research needed to extend application of these findings to entertainment-education contexts, to further develop theory in the area of persuasion and narrative, and to better account for other persuasive effects of entertainment narrative, such as those hypothesized in cultivation theory, are discussed.