Using infographics to reduce the negative effects of jargon on intentions to vaccinate against COVID-19

Abstract

Guided by feelings-as-information theory, this experiment (N = 643), based in the United States, tested whether the use of jargon and infographics within messages designed to explain the COVID-19 mRNA vaccines affected behavioral intentions to vaccinate. The results revealed that the presence of jargon was associated with a difficult processing experience, message resistance, decreased perceptions of message credibility, and reduced intentions to get the COVID-19 vaccine. That said, when an infographic was integrated into the jargon message, these negative relationships went away and the presence of jargon no longer indirectly impacted intention to vaccinate. This experiment demonstrates that in contexts where jargon use exists, the use of an infographic can counteract some of the negative effects of a difficult processing experience.

Keywords

COVID-19 infographics jargon metacognition processing fluency science communication

In the domains of science and health communication, it is critical to design messages that effectively translate complex scientific information to the general public. In addition to accurately and clearly translating this information, these messages must also appear trustworthy and credible so that audiences accept the scientific recommendations advanced (Luo et al., 2020). Indeed, there is no clearer example of the critical role of science communication than during the COVID-19 pandemic. The newness of the virus as well as the scientific innovations continuously introduced to combat this disease required that the public engage with complex messaging and accept public health recommendations in order to make safe personal decisions. Despite the array of communication challenges presented during this time, this context also provides an ongoing opportunity for science communicators to both test and reflect upon how to best communicate scientific information within these difficult circumstances. With these ideas in mind, this investigation examines and compares how different message features affect individuals’ receptivity toward receiving a COVID-19 vaccination.

Specifically, the two message features under investigation include the presence or absence of jargon, defined as the presence of scientific, technical, or specialized words (Sharon and Baram-Tsabari, 2014; Merriam Webster, n.d.), and the use of infographics, defined as the graphic, visual presentation of complicated information designed to be processed in a fast and efficient manner (Bicen and Beheshti, 2017). The reason for studying these two message features is because a well-known convention in public science communication is to avoid the use of jargon (see Dean, 2009; Plaxco, 2010; Rakedzon et al., 2017; Willoughby et al., 2020). That said, an unprecedented public health crisis could lead to increases in jargon use due to the need to communicate novel, precise, and discerning scientific information to audiences unfamiliar with these scientific principles. As such, under these conditions, and guided by feelings-as-information theory (FIT; Schwarz, 2011), we experimentally test whether the use of an infographic can mitigate the negative effects of jargon by making it easier to process complex information on occasions where it tends to be delivered. By understanding the separate, and combined, effects of jargon and infographics on information processing, more could be learned about how to effectively and efficiently dispense critical public health information when the public needs it most.

The effects of jargon on processing fluency

In order to understand the negative effects of jargon, it first becomes necessary to explain the cognitive mechanisms associated with its use. First, because jargon is often used by experts to convey specialized and highly specific information, these words or terms are not readily recognized nor understood by members of the general public. Thus, the presence of jargon impairs message comprehension (e.g. Brooks, 2017; Krieger and Gallois, 2017; McDonnell et al., 2016; Markowitz and Hancock, 2016; Sharon and Baram-Tsabari, 2014).

Second, beyond comprehension, the presence of jargon has also been shown to impair, or disrupt, people’s experiences while processing information. These feelings of impairment, or subjective feelings of difficulty while processing information, describe the metacognition of processing fluency (Schwarz, 2010). FIT (Schwarz, 2011) can be used as a lens to understand how information processing experiences of ease (fluency) or difficulty (disfluency) affect judgments toward the topic under investigation. FIT proposes that individuals evaluate new information based on two dimensions: (1) the content of the information itself and (2) their feelings and thoughts about their information processing experience (metacognitions). Research in metacognition (for reviews see Alter and Oppenheimer, 2009; Petty et al., 2007; Schwarz, 2015) has consistently revealed that easy processing experiences are associated with positive feelings toward the subject matter, whereas difficult processing experiences are associated with negative feelings toward the subject matter. To offer an example, in an experiment by Song and Schwarz (2009), processing fluency was manipulated by using either easy or difficult to pronounce food additive names. When terms were easy to pronounce and, in turn, were processed more fluently, they were rated more favorably than when processing fluency was impaired due to difficult pronunciations. This experiment offers one example of how variance in processing fluency evokes variance in peoples’ affective responses toward the topic, even when the informational content itself was held constant.

Recent work in communication has furthered these ideas by linking jargon use to processing fluency (see Bullock et al., 2019; Krieger and Gallois, 2017; Shulman and Sweitzer, 2018; Shulman et al., 2020; Tolochko et al., 2019). These studies found that the presence of jargon impaired the fluent processing of information and, as a result, impacted judgments and decision-making in the ways expected by FIT (for a review of the importance of processing fluency for communication, see Shulman and Bullock, 2019). For example, one study from this line of research demonstrated that when jargon was used within political opinion questions (e.g. Super PAC), undergraduate students reported lower processing fluency and, in turn, reported lower levels of political interest and efficacy (Shulman and Sweitzer, 2018). Similarly, Tolochko et al. (2019) found that complex language within news articles led to lower reports of processing fluency and lower levels of self-reported knowledge gain. Taken together, guided by theory and past research, our first hypothesis proposes that jargon use within a public health message will negatively affect reports of processing fluency.

H1: There will be a negative relationship between the presence of jargon and processing fluency.

Infographics and processing fluency

It has thus far been argued that the use of jargon is associated with a difficult processing experience. That said, one of the complicating features of the COVID-19 pandemic is the novelty of the virus. This novelty has, regrettably, rendered it difficult at times to avoid the use of new, and unfamiliar, terms in public-facing communication (e.g. social distancing, respiratory droplets, mRNA, spike proteins, efficacy rates). Although not ideal from a public communication perspective, practically these jargon terms were likely used to provide complete, accurate, and distinct information about this new and evolving health topic (Hirst, 2003). Therefore, given the frequency of jargon use within this context, this work sought to test a plausible mitigation strategy—guided by FIT—to help combat the known negative effects of jargon on general audiences. We argue that the use of an infographic can help ease the burden of processing complex information.

To begin, it is important to define infographics and their use. Infographics are visual representations of information that include both image and text (Lee et al., 2021). Infographics can take many forms from explanatory, persuasive, to exploratory, as well as including or not including a visualization of data (Otten et al., 2015). The appeal of infographics is that they are visual; infographics strive to be visually appealing and intend to attract attention due to the use of images (Dunlap and Lowenthal, 2016). Furthermore, within the domains of health and science communication, infographics are often recommended because visualizing information can reduce its complexity (Bicen and Beheshti, 2017; Garcia-Retamero and Cokely, 2013; Lee et al., 2021; Otten et al., 2015) and promote elaborative processing (Lazard and Atkinson, 2015; Lee and Kim, 2016). In support of the intuitive benefits of infographics, the US Centers for Disease Control and Prevention (CDC) in their recommendations for clear writing include infographics as a strategy to simplify message presentation, particularly through social media (CDC, 2019). In sum, the inclusion of imagery through an infographic is thought to provide aesthetic and practical appeal, which should—presumably—improve message acceptance. Although best practices in health communication seem to support the use of infographics, to our knowledge, this conceit has not been formally tested using a metacognitive framework. As such, one contribution of this work is to examine whether the benefits of infographics can be explained through their positive impacts on processing fluency.

Specifically, we propose that visual presentation aids (i.e. infographics) are an ideal solution to the problem of impaired processing fluency due to jargon. This is because visualizing information should lessen the message’s complexity and enhance clarity by complementing semantic information with visual information. Moreover, an infographic could also make abstract concepts more concrete, which is a process that has been linked to gains in conceptual fluency (see Shulman and Bullock, 2019; Shulman and Sweitzer, 2018). Taken together, it is argued and tested here that under conditions of difficult processing, infographics have the potential to enhance one’s processing fluency experience. Under conditions of easy processing, however, we assume that no processing fluency enhancements will be needed. Guided by these expectations, we hypothesize that message format (text-only vs infographics) will moderate the strength of the relationship between jargon and processing fluency. Specifically, the relationship between the jargon condition and processing fluency should be stronger in the text-only format condition than in the infographic format condition:

H2: Message format will moderate the relationship between the jargon condition and processing fluency such that the effect of the jargon condition on processing fluency will be stronger in the text-only format condition compared to the infographic format condition.

Processing fluency and message resistance

The reason why it is important to consider the role of metacognitive feelings while processing information is because these metacognitive feelings, including processing fluency, have been linked to message resistance (Schwarz, 2006). According to Schwarz (2006), one’s metacognitive experiences not only influence positive and/or negative affect, but also impact message skepticism. This is because when ideas are processed fluently, people attribute these feelings of ease as indicative of familiarity, whereas when ideas are processed disfluently, people attribute these feelings of difficulty as indicative of novelty. Because people naively infer risk from novelty, and safety from familiarity, messages that are processed disfluently are also more likely to be perceived with skepticism and resistance.

In support of these ideas, an experiment by Bullock et al. (2019) found that jargon use in science messages led to message resistance about new technologies. In this experiment, a panel of MTurk workers were exposed to either a jargon or no-jargon message about a technological innovation (e.g. self-driving cars). Consistent with previous explanations, it was found that a disfluent experience, induced through the presence of jargon, increased resistance and skepticism toward the technology in question. This experiment linked reports of processing fluency with motivated resistance to persuasion (MRTP; Nisbet et al., 2015). MRTP describes the motivation to resist or oppose pressures to change. This concept is composed of two dimensions: (1) counterarguing: generating thoughts that undermine the message’s arguments and (2) reactance: a negative response when one perceives that a message is trying to persuade them. One of the goals of this study is to replicate the relationship between jargon use, fluency, and MRTP within a situation of practical importance: explaining how the new mRNA COVID-19 vaccines work in an effort to increase willingness to get the vaccine. Guided by theory and prior research (e.g. Bullock et al., 2019; Shulman and Bullock, 2020), it is expected that the presence of jargon will lead to a disfluent processing experience, which will in turn lead individuals to feel more critical of the message and less willing to accept the message’s arguments. These relationships are reflected in the following hypothesis:

H3a: There will be a negative relationship between processing fluency and MRTP.

In addition, as suggested by H2, if message format moderates the strength of the relationship between the jargon condition and processing fluency, we should see the impact of this effect throughout the entire causal model we are proposing (see Figure 1). Specifically, we assert that in the text-only format conditions, the effect of jargon on processing fluency should affect participants’ MRTP (the presence of jargon should increase reports of MRTP, through processing fluency). However, if the presence of an infographic mitigates the negative effects of jargon on fluency as we expect, then the indirect association between the jargon condition and MRTP should be considerably weaker (or non-existent) in the infographic format condition. This expectation is reflected in the following hypothesis:

H3b: There will be a positive indirect effect of jargon, for the text format condition only, on MRTP through processing fluency, such that the presence of jargon will reduce processing fluency which will, in turn, increase MRTP.

Figure 1.

Hypothesized relationships between concepts.

Motivated resistance to persuasion and message credibility

If variance in processing fluency affects message resistance, then we should observe evidence of this relationship through the association between MRTP and other message-related perceptions as well. The message-related perception of interest here is credibility. Credibility involves perceptions of expertise and trustworthiness (Pornpitakpan, 2004), which can be attributed to both source and message cues (for a review see Metzger et al., 2003). An abundance of research in persuasion and beyond has demonstrated that sources that are perceived as more credible are more likely to be persuasive than sources (or messages) perceived as less credible (Metzger et al., 2003; Petty et al., 1997). Therefore, one of the contributions of this work is to link credibility perceptions to message design and processing fluency through the mechanism of MRTP.

Conceptually, MRTP should decrease perceptions of credibility. The first subcomponent of MRTP, counterarguing, in particular should lead to the generation of cognitions that intend to undermine the message’s arguments, including source derogation (Byrne and Hart, 2009; Nisbet et al., 2015). In addition, FIT (Schwarz, 2011) proposes that when individuals feel that processing a certain message is difficult, they will attribute this negative feeling, brought about by a difficult experience, toward the message content. As a result, the message will be viewed less favorably. Taking these ideas one step further, we argue and test here whether motivation to resist a message will lead to reduced reports of message credibility. This expectation informs the next study hypothesis:

H4a: There will be a negative relationship between MRTP and message credibility.

Importantly, if the hypothesized relationship between MRTP and message credibility is supported, then the serial process we are building should be affected by this relationship as well. Specifically, if the presence or absence of jargon affects processing fluency, which in turn affects MRTP and then credibility, then this serial process should produce an indirect, and negative, relationship between the jargon condition and message credibility (but only within the text format conditions, not the infographic format conditions). One of the theoretical contributions of this work is considering the implications of message design on these theoretical metacognitive processes. Although some might expect that the presence of jargon would enhance perceptions of credibility, because the use of jargon should be associated with perceptions of expertise (Pornpitakpan, 2004), a metacognitive framework, such as FIT (Schwarz, 2011), would actually implicate the opposite pattern of results through the cognitive mediators of fluency and MRTP. These relationships are proposed in the following hypothesis:

H4b: There will be a negative indirect effect of the jargon condition, for the text-only format condition, on credibility through the serial mediators of processing fluency and MRTP.

Effects on vaccination intentions

Although contributing to and extending theory is always important, in the context of COVID-19, the more pressing goal for science communicators is to convince individuals to get vaccinated as soon as possible. Given this practical and urgent goal, we endeavored to link the aforementioned cognitive processes to COVID-19 vaccine behavioral intentions. Behavioral intentions can be defined as “motivational factors that influence a behavior” (p. 181, Ajzen, 1991). These intentions are important because it is widely considered that increasing intentions increases the likelihood of enacting the behavior in question (e.g. Ajzen, 2011; Webb and Sheeran, 2006). Thus, increasing intention to vaccinate should lead to actual vaccination behavior. Therefore, it is of critical importance to understand how message design can help increase vaccine intentions.

According to the theory of reasoned action (TRA; Fishbein and Ajzen, 1975) and theory of planned behavior (TPB; Azjen, 1991), there are three predictors of behavioral intention: social norms, attitudes toward the behavior, and perceived behavioral control. Relevant to this research is the effect of favorable attitudes on behavioral intentions, brought about through perceptions of credibility. Previous research has demonstrated that increasing message credibility can positively affect attitudes toward the message topic (see Pornpitakpan, 2004). As such, it is expected that credibility perceptions will be positively associated with vaccine intentions:

H5a: There will be a positive relationship between message credibility and vaccine intentions.

Once again, if vaccine intentions are linked to credibility, this association carries implications for the larger causal model we are testing. As such, our final hypothesis reflects our expectation that the jargon condition can set a process in motion that will, through the mediators of processing fluency, MRTP, and credibility, eventually influence intentions to vaccinate. If this is the case, then a negative, non-zero indirect effect should be observed between the jargon condition and intention to vaccinate (but only within the text format conditions, not the infographic format conditions).

H5b: There will be a negative indirect effect of jargon condition, for the text-only condition, on intention to vaccinate through the serial mediators of processing fluency, MRTP, and credibility, such that the presence of jargon will decrease intention to vaccinate.

And finally, in an effort to heed our own advice, Figure 1 visually depicts all of the relationships that will be tested here.

1. Method

Participants

Participants (N = 796) were recruited using TurkPrime, an online research platform for crowdsourcing participants, between 18 and 24 February 2021. Because, however, the outcome of interest in this study is vaccine intentions, participants who reported already receiving the vaccine were removed from analyses (n = 153, 19%) leaving a final sample size of 643. Of these participants, the sample was 62.5% male (n = 401) and ranged in age from 21 to 77 years old (M = 38.42, SD = 11.11). The racial breakdown of the sample was 72.7% White, 11.8% Black or African American, 0.6% American Indian or Alaska Native, 6.8% Asian, 0.2% Native Hawaiian or Pacific Islander, 4.8% Hispanic, and 2.8% reporting other or prefer not to say. To be eligible to participate, participants had to pass a CAPTCHA, be at least 18 years old, and provide consent. In addition, to improve data quality (see Chmielewski and Kucker, 2020), we required that participants have obtained at least a 95% completion rating on at least 500 Human Intelligence Tasks (HITs). In other words, each participant must have participated in at least 500 tasks, and at least 95% of this prior work must have been considered acceptable. Eligible participants were compensated $2.

Procedure

The study design was a 2 (jargon vs no-jargon) × 2 (message format: text-only vs infographic) between-subjects survey experiment hosted on Qualtrics. The primary message was a COVID-19 vaccine explainer that described the mechanics of how the mRNA vaccines worked. This topic was chosen for two reasons. The first was the contextual need for people to get vaccinated. The second reason was that because the vaccine used new technology, it was thought that vaccine hesitancy might be higher than usual. In response, it was commonplace around the time of this study for news sites to provide explainers for how the vaccine operated. Thus, using a vaccine “explainer” message for our primary manipulation was ecologically appropriate given the social context. Before beginning the experiment, participants completed an eligibility screening, provided consent, and began the procedure. Each participant was then randomly assigned to one version of the jargon manipulation and one version of the format manipulation. Across the four possible message condition combinations, participants were told that they would be shown an explanation of the COVID-19 vaccine and would then be asked their thoughts about the message and vaccine. In an effort to ensure that participants across all conditions had at least a baseline level of knowledge to comprehend the basics of the message, we provided the definitions of “mRNA,” “spike protein,” and “immune system” on the introductory screen (see Supplemental Material). By providing definitions for the message’s key terms, comprehension differences between conditions should be minimized and considered distinct from the impact of jargon on processing fluency (see Bullock et al., 2019). Following this introduction, participants were then randomly assigned to vaccine message condition. This message was held on-screen for 15 seconds. Only after this time elapsed were participants able to advance to the next page. This was done to nudge participants to read the information. Following exposure to message conditions, participants reported their processing fluency, MRTP, perceptions of message credibility, and intentions to vaccinate. In total, the entire procedure took an average of 20.70 minutes to complete (median = 16.80 minutes).

Stimuli

Jargon condition

Participants were randomly assigned to a jargon (n = 326) or no-jargon (n = 317) condition using Qualtrics software. Of practical importance, the jargon condition stimulus was adapted from a real news article published in The Washington Post (Johnson et al., 2021) that intended to explain how mRNA vaccines worked. The decision to use a real-world science message for the jargon condition was done to enhance the ecological validity of this experiment and to contextualize, and illustrate, that the use and impact of jargon is likely more pervasive than previously considered (see Baram-Tsabari et al., 2020). The jargon condition contained the jargon terms used in the actual The Washington Post article. In the no-jargon condition, technical terminology was replaced with simpler synonyms with guidance from public-facing science sources such as the Genetic Literacy Project, Vaccines.gov, and Live Science. Word count was held constant across both the jargon (WC = 127) and no-jargon conditions (WC = 128, see Supplemental Material). In addition, the Flesch–Kincaid grade level tool in Microsoft Word confirmed that the language in the no-jargon condition (grade 7) was easier than the language in the jargon condition (grade 10).

Message format

In addition to random assignment to the jargon conditions, participants were also randomly assigned to view either a text-only (n = 321) or infographic message format condition (n = 322) using Qualtrics software. Both versions of these messages explained the mechanisms underlying mRNA-based vaccines and included the same word-based information. The infographic stimulus was taken directly from the same The Washington Post article as the jargon text (Johnson et al., 2021). In the text-only format condition, this information was displayed in paragraph form. In the infographic format condition, the visual elements from the original image were recreated using Canva, an online graphic design platform and FireAlpaca, a free digital painting software (see Supplemental Material). Importantly, the information content displayed between the text and infographic conditions was identical. In the infographic condition, however, rather than present this information in paragraph form, sentences were split up and positioned alongside the infographic elements to complement and explain the material. Beyond the visual infographic model, font and text color were held constant across all conditions.

Measures

Unless otherwise stated, all responses were reported using a 7-point Likert-type scale wherein higher values indicate higher agreement with the concept being measured. All stimuli, survey items, data, and outputs for this research are made available on the Open Science Framework (OSF). Moreover, the descriptive statistics for all of the measures below are broken down by experimental condition in Table 1.

Table 1.

Descriptive statistics for each variable by experimental condition.

Message conditions		Model variables
Language	Format	Processing fluencyM (SD)	MRTPM (SD)	CredibilityM (SD)	Intention to vaccinateM (SD)
Jargon	Text	3.71 (1.42)	2.32 (1.12)	6.01 (0.92)	5.19 (1.48)
Jargon	Infographic	4.21 (1.43)	2.35 (1.16)	6.02 (1.07)	5.39 (1.48)
No jargon	Text	4.61 (1.23)	2.33 (1.06)	6.02 (0.85)	5.03 (1.66)
No jargon	Infographic	4.37 (1.28)	2.47 (1.09)	5.86 (1.07)	5.23 (1.42)
Total jargon		3.97 (1.44)	2.34 (1.14)	6.02 (1.00)	5.29 (1.48)
Total no jargon		4.50 (1.26)	2.40 (1.08)	5.95 (0.97)	5.13 (1.55)
Total text		4.17 (1.40)	2.33 (1.09)	6.02 (0.89)	5.11 (1.57)
Total infographic		4.29 (1.36)	2.41 (1.13)	5.95 (1.07)	5.31 (1.45)

MRTP: motivated resistance to persuasion.

Note: These statistics represent the direct effects of the experimental conditions on all of our study measures.

Processing fluency

Immediately after exposure to the vaccine message, participants responded to a six-item measure assessing their processing fluency (M = 4.23; SD = 1.38; α = .88; Kostyk et al., 2019; Shulman and Sweitzer, 2018). An example item includes, “It was easy for me to understand the information presented.”

MRTP

To assess participants’ MRTP, eight items were adapted from Nisbet et al. (2015; M = 2.37, SD = 1.11, α = .88). An example item includes, “The message I saw tried to pressure me to think a certain way.”

Credibility

Appelman and Sundar’s (2016) message credibility scale was adapted to assess participants’ perceptions of message credibility (M = 5.98, SD = 0.98, α = .93). The scale included four items that asked whether the vaccine message appeared accurate, authentic, believable, and trustworthy.

Intention to vaccinate

Participants’ intentions to vaccinate themselves against COVID-19 was measured with four items adapted from Gerend et al. (2008; M = 5.21, SD = 1.52, α = .83). An example item includes “In the future I will try to get vaccinated.”

2. Results

To test H1, H2, H3a, and H3b, PROCESS model 7 was used with 10,000 bootstrapped resamples (Hayes, 2018). PROCESS is a computational tool for modeling mediation, moderation, and conditional process analysis, whereas bootstrapping refers to the statistical technique of repeatedly taking random samples with replacement from a single data set. The current moderated mediation model assessed the indirect effect of jargon (X) on MRTP (Y) through processing fluency (M) with message format (W) moderating the path between jargon condition and processing fluency (see Figures 1 and 2). In support of H1, it was found that overall the presence of jargon negatively impacted reports of processing fluency (B = −0.90, SE = 0.15, t = −6.04, p < .001, R² = .06), as expected.

Figure 2.

Study results across all hypotheses.

In support of H2, it was found that the effect of jargon on processing fluency was moderated by message format (B = 0.74, SE = 0.21, p < .001, R² = .06). Specifically, in the text-only condition, the negative relationship between jargon and processing fluency was significant and negative (B = −0.90, SE = 0.15, p < .001). However, in the infographic condition, there was no significant association between jargon condition and processing fluency (B = −0.17, SE = 0.15, p = .27). Thus, the problematic effects of jargon on processing fluency were only apparent in the text-only format condition.

The third hypothesis set examined the relationship between processing fluency and MRTP (H3a) and the indirect effect of the jargon condition on MRTP (H3b). Consistent with H3a, a significant negative relationship between processing fluency and MRTP was observed (B = −0.20, SE = 0.03, t = −6.39, p < .001, R² = .06). Specifically, an easier processing experience was associated with less message resistance. Moreover, in support of H3b, the indirect effect of jargon on MRTP through processing fluency was non-zero and positive in the text-only condition (B = 0.18, SE = 0.04, 95% confidence interval (CI): (0.11, 0.26)) but not different from zero in the infographic condition (B = 0.03, SE = 0.03, 95% CI: (−0.03, 0.10)). Together, these results supported expectations (see Figure 2) such that the presence of jargon indirectly increased message resistance, but only in the text-only format condition.

To test H4a and H4b, PROCESS model 83 was used with 10,000 bootstrapped resamples (see Figures 1 and 2; Hayes, 2018). This model assessed the indirect effect of jargon (X) on message credibility (Y) through the serial mediators of processing fluency (M1) and MRTP (M2) and moderated by message format (W). Consistent with H4a, the relationship between MRTP and message credibility was significant and negative (B = −0.63, SE = 0.03, t = −25.05, p < .001, R² = .52). Specifically, an easier processing experience led to higher reports of message credibility and the strength of this effect could be considered quite large in size (Cohen, 1992). Furthermore, in support of H4b, the indirect effect of the jargon condition on message credibility through processing fluency and MRTP was also non-zero and negative within the text-only condition (B = −0.11, SE = 0.03, 95% CI: (−0.17, −0.07)). In the infographic conditions, the indirect relationship between jargon and credibility was not statistically different from zero, as hypothesized (B = −0.02, SE = 0.02, 95% CI: (−0.06, 0.02)), see Figure 2). Together, this model demonstrates not only that the presence of jargon undermined message credibility in the ways proposed in H4, but also that this effect was moderated by message format.

To test H5a and H5b, PROCESS model 83 was again used with 10,000 bootstrapped resamples (Hayes, 2018). This serial mediation model included the indirect effect of jargon (X) on intention to vaccinate (Y) through the mediators of processing fluency (M1), MRTP (M2), and credibility (M3) and moderated by message format (W). Consistent with H5a, message credibility was positively associated with intention to vaccinate (B = 0.58, SE = 0.08, t = 7.17, p < .001, R² = .16). Specifically, higher reports of message credibility were positively associated with vaccine intentions. In addition, and consistent with H5b, these effects were driven by the jargon condition. It was found that a non-zero and negative indirect effect of the jargon condition on intention to vaccinate through processing fluency, MRTP, and credibility was obtained in the text-only condition (B = −0.07, SE = 0.02, 95% CI: (−0.11, −0.03)). That said, this same effect was not obtained for those who viewed an infographic (B = −0.01, SE = 0.01, 95% CI: (−0.04, 0.01)). Thus, taken together, the presence of jargon indirectly decreased intention to vaccinate through the hypothesized mediators of processing fluency, MRTP, and message credibility (see Figure 2). It is also notable that although this effect was smaller than the model predicting credibility, it still can be considered medium to large in size (Cohen, 1992). In sum, all of the hypothesized relationships were supported by our data and are visually depicted in Figure 2. The complete set of results across all analyses can be found on our OSF page.

3. Discussion

The purpose of this study was to examine how communication could both facilitate and discourage COVID-19 vaccine intentions. To do so, we examined the role of jargon and message format on vaccine intentions through a variety of cognitive mechanisms that should either support or discourage vaccine behaviors. The results from this experiment revealed a few important trends. First, the results of this study replicated findings from past research (e.g. Bullock et al., 2019) that found that the presence of jargon impaired processing fluency (H1) and, as a result, led to message resistance (H3). Extending upon this relationship, this experiment further revealed that message resistance led to decreased perceptions of message credibility (H4), and, in turn, decreased intentions to receive the vaccine (H5). These findings support recommendations that science communicators should avoid the use of jargon (Baram-Tsabari et al., 2020; Bullock et al., 2019; Plain Language Action and Information Network, n.d.; Sharon and Baram-Tsabari, 2014). That said, in addition to replicating past relationships, this study also aimed to find a solution to lessen the negative effects of jargon because, regrettably, there are times, such as throughout the COVID-19 pandemic and exemplified in the public-facing science article used in this study, where the use of specialized terms was more common due to the novelty of the context and/or the need for specificity in reporting. In these instances, we found that the use of an infographic is a viable strategy to reduce, or remove, the negative effects of jargon on information processing (H2). In sum, this study sought to advance theoretical and practical implications for science and public health communicators who are tasked with designing effective messages under challenging circumstances.

In support of H3, this study provided evidence for the indirect relationship between jargon and MRTP through processing fluency. We found that the presence of jargon led to a more difficult processing experience, which in turn negatively impacted message acceptance. This finding is consistent with other work (e.g. Bullock et al., 2019; Shulman and Bullock, 2020; Song and Schwarz, 2009) that suggests a difficult processing experience will produce resistance and skepticism toward the topic under investigation. The fact that we were able to replicate these relationships using a salient health topic (COVID-19) offers theoretical support for FIT (Schwarz, 2011) and demonstrates the robustness of these processes.

The relationship between processing fluency and MRTP paved the way for one potentially counterintuitive finding from this research. We found that the presence of jargon reduced credibility, in support of the fourth set of hypotheses. One might assume that the presence of jargon would serve as a cue that enhances credibility by indicating expertise in a specific area. Indeed, previous research has demonstrated that source expertise is important for message acceptance (for review see Pornpitakpan, 2004). Despite these ideas, however, we offered a metacognitive argument, drawing on FIT (Schwarz, 2011), that suggests that difficult processing experiences lead to negative feelings about the message and more resistance that should, in turn, decrease, not increase, perceptions of credibility. In support of this metacognitive argument, we found that jargon indirectly decreased perceptions of credibility through processing fluency and MRTP. Notably, this effect was quite large in size (Cohen, 1992). For public communicators these results imply that although perceptions of source credibility and expertise play a key role in message acceptance and behavioral intentions, cuing this credibility through jargon use can backfire.

Importantly, this research went beyond illustrating the negative effects of jargon on message perceptions by investigating the effects of processing fluency on behavioral intentions. We found a negative, indirect, relationship between jargon and intention to vaccinate through the mechanisms of processing fluency, MRTP, and credibility. Thus, consistent with H5b, individuals reported lower intentions to receive the COVID-19 vaccine after exposure to a real vaccine explanation message that included jargon. And notably, this effect was medium to large in size (Cohen, 1992). Connecting the relationship between jargon and processing fluency to behavioral intentions has theoretical and practical value. Theoretically, this finding contributes to previous work by demonstrating that processing fluency can impact behavioral intentions in addition to message perceptions. Although work in this domain (for a meta-analysis see Alter and Oppenheimer, 2009) typically focuses more generally on issue perceptions, the fact that this study connected jargon use to behavioral intentions, through fluency, evinces the strategic potential of considering the role of processing fluency in message design (see Shulman and Bullock, 2019). If the goal of public health and science communication is to communicate in a way that will lead to message acceptance, then designing messages aimed at producing a fluent processing experience can exert a meaningful influence on whether individuals adopt the recommended behavior.

To this end, it was argued and tested here that one message format that could positively influence processing fluency was the use of an infographic. In support of this expectation, we found that the negative effects of jargon obtained in the text-only format condition disappeared when an infographic was used across all outcomes. Interestingly, although we found that message format moderated the effects of jargon, it is important to note that there was not a main effect of message format on processing fluency, t(641) = −1.10, p = .270. Furthermore, in the no-jargon condition, there was not a significant difference in processing fluency, t(315) = 1.69, p = .092, between the infographic and text-only conditions. Importantly, these findings do not suggest that infographics always work better than text but instead that visual aids can be uniquely suited to minimizing complexity where it exists. This is because it was only within the jargon condition that the unique, and beneficial, effects of the infographic were realized. This suggests that when public communicators know that the information they need to convey is difficult, complex, or unfamiliar, an infographic could be an effective presentational tool to combat the negative effects of a difficult processing experience.

Limitations

Despite consistent support for all study hypotheses, this experiment is not without its limitations. For one, the generalizability of these findings is limited by our decision to use only one message topic: a COVID-19 vaccine explanation. Although previous research has reliably demonstrated the robustness of findings regarding the detrimental effects of jargon on outcomes (e.g. Bullock et al., 2019; Shulman et al., 2020; Tolochko et al., 2019), this is one of the first studies to incorporate an infographic within these relationships. Although our experiment presented an infographic used in The Washington Post to enhance the realism and quality of our stimuli, future research should examine whether these findings replicate using a different topic, infographic, and persuasive target. In addition to limitations regarding topic and message generalizability, future work should also examine whether our findings generalize to other types of infographics. Infographics can differ on a variety of stylistic and informational dimensions. For example, an infographic may use more or less color, text, images, or numbers to describe the information. It could demonstrate a process, visualize statistics, or simply add visuals to key terms. All of these decisions may influence message effectiveness. In order to provide clear recommendations to message designers, in future work it will be critical to better understand the effects of specific design choices on outcomes of interest. Building from this research, future research could also consider other types of message perceptions likely to be affected by message design such as risk perceptions. Regarding behavioral intentions, we recognize that predicting vaccination behavior is a complex and difficult task as there are many reasons why individuals might decide whether to get vaccinated. Although we did find a statistically large indirect effect of jargon on vaccination intentions, we acknowledge that a one-shot message exposure is only one of many potential predictors. Thus, although this study provides insight into how language could impact decision-making within this context, future research is needed to understand how language might interact with various other factors as well. Finally, we acknowledge that the decision to use MTurk workers undermines the generalizability of our findings. MTurk workers are not representative of the general population and our sample of workers included more males than females (due to costs associated with sex quotas). Although our study prioritized internal validity and a timely, cost-effective, research design, future work should certainly assess whether the relationships obtained here hold for different samples and populations.

4. Conclusion

Although our experiment focused on the COVID-19 vaccine for important practical reasons, our findings could be applied to health and science communication more broadly. For example, in the medical context, infographics could help lessen the complexity of an explanation of a medical treatment, such as chemotherapy or hormone therapy. In the science context, for example, infographics could assist in visualizing the process by which carbon emissions negatively affect our climate. Furthermore, infographics are well-suited for communication via social media, where images tend to draw more attention (Frogers, 2014). As such, through the fast-paced and highly networked contexts of social media, the use of infographics could bolster health message dissemination.

In sum, the results from this experiment support the conventional wisdom that when the goal of science communication is to advance recommendations to the general public, message design should be used strategically to make information—even complex information—feel easier to process. That said, in instances where complex information is unavoidable, considering visual formats may help ease the burden of processing this information and promote message acceptance.

Supplemental Material

sj-docx-1-pus-10.1177_09636625221077385 – Supplemental material for Using infographics to reduce the negative effects of jargon on intentions to vaccinate against COVID-19

Supplemental material, sj-docx-1-pus-10.1177_09636625221077385 for Using infographics to reduce the negative effects of jargon on intentions to vaccinate against COVID-19 by Elizabeth E. Riggs, Hillary C. Shulman and Rachel Lopez in Public Understanding of Science

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Elizabeth E. Riggs

Hillary C. Shulman

Rachel Lopez

Supplemental material

Supplemental material for this article is available online.

Author biographies

Elizabeth E. Riggs (MA, Ohio State University) is a doctoral student in the School of Communication at Ohio State University. Her research considers how our information-processing experiences, both cognitive and metacognitive, affect our subsequent perceptions of health, science, and political information.

Hillary C. Shulman (PhD, Michigan State University) is an Assistant Professor in the School of Communication at Ohio State University. Her work focuses on how message design can increase engagement and persuasion for complicated topics.

Rachel Lopez (BA/BS, Trinity University) is a doctoral student in the School of Communication at the Ohio State University. Her current research focuses on social media effects and the role of race and culture in family health communication.

References

Ajzen

(1991) The theory of planned behavior. Organizational Behavior and Human Decision Processes 50(2): 179–211.

Ajzen

(2011) The theory of planned behaviour: Reactions and reflections. Psychology and Health 26(9): 1113–1127.

Alter

Oppenheimer

(2009) Uniting the tribes of fluency to form a metacognitive nation. Personality and Social Psychology Review 13(3): 219–235.

Appelman

Sundar

(2016) Measuring message credibility: Construction and validation of an exclusive scale. Journalism and Mass Communication Quarterly 93(1): 59–79.

Baram-Tsabari

Wolfson

Yosef

Chapnik

Brill

Segev

(2020) Jargon use in Public Understanding of Science papers over three decades. Public Understanding of Science 29(6): 644–654.

Bicen

Beheshti

(2017) The psychological impact of infographics in education. BRAIN. Broad Research in Artificial Intelligence and Neuroscience 8(4): 99–108.

Brooks

(2017) Student identity and aversions to science: A study of translation in higher education. Journal of Language and Social Psychology 36(1): 112–126.

Bullock

Colón Amill

Shulman

Dixon

(2019) Jargon as a barrier to effective science communication: Evidence from metacognition. Public Understanding of Science 28(7): 845–853.

Byrne

Hart

(2009) The boomerang effect: A synthesis of findings and a preliminary theoretical framework. Annals of the International Communication Association 33(1): 3–37.

10.

Centers for Disease Control and Prevention (CDC) (2019) Social Media at CDC. Available at: https://www.cdc.gov/socialmedia/tools/InfoGraphics.html (accessed 27 December 2020).

11.

Chmielewski

Kucker

(2020) An MTurk crisis? Shifts in data quality and the impact on study results. Social Psychological and Personality Science 11(4): 464–473

12.

Cohen

(1992) A power primer. Quantitative Methods in Psychology 112(1): 155–159.

13.

Dean

(2009) Am I Making Myself Clear? A Scientist’s Guide to Talking to the Public. Cambridge, MA: Harvard University Press.

14.

Dunlap

Lowenthal

(2016) Getting graphic about infographics: Design lessons learned from popular infographics. Journal of Visual Literacy 35(1): 42–59.

15.

Fishbein

Ajzen

(1975) Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Boston, MA: Addison-Wesley Publishing Company.

16.

Frogers

(2014) What Fuels a Tweet’s Engagement? Available at: https://blog.twitter.com/en_us/a/2014/what-fuels-a-tweets-engagement (accessed 24 September 2021).

17.

Garcia-Retamero

Cokely

(2013) Communicating health risks with visual aids. Current Directions in Psychological Science 22(5): 392–399.

18.

Gerend

Shepherd

Monday

(2008) Behavioral frequency moderates the effects of message framing on HPV vaccine acceptability. Annals of Behavioral Medicine 35(2): 221–229.

19.

Hayes

(2018) Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-based Approach (2nd ed.). New York, NY: Guilford Press.

20.

Hirst

(2003) Scientific jargon, good and bad. Journal of Technical Writing and Communication 33(3): 201–229.

21.

Johnson

Steckelberg

Sun

McGinley

Chiu

(2021) Your questions about coronavirus vaccines, answered. The Washington Post, 21 April, p. 2.

22.

Kostyk

Leonhardt

Niculescu

(2019) Processing fluency scale development for consumer research. International Journal of Market Research 63: 353–367.

23.

Krieger

Gallois

(2017) Translating science: Using the science of language to explicate the language of science. Journal of Language and Social Psychology 36(1): 3–13.

24.

Lazard

Atkinson

(2015) Putting environmental infographics center stage: The role of visuals at the elaboration likelihood model’s critical point of persuasion. Science Communication 37(1): 6–33.

25.

Lee

Kim

(2016) Effects of infographics on news elaboration, acquisition, and evaluation: Prior knowledge and issue involvement as moderators. New Media and Society 18(8): 1579–1598.

26.

Lee

Dockter

(2021) Effects of message presentation type on gm food risk perception, similarity judgment, and attitude. Health Communication 36: 1666–1676.

27.

Luo

Hancock

Markowitz

(2020) Credibility perceptions and detection accuracy of fake news headlines on social media: Effects of truth-bias and endorsement cues. Communication Research. Epub ahead of print 13 April. DOI: 10.1177/0093650220921321.

28.

McDonnell

Barker

Wieman

(2016) Concepts first, jargon second improves student articulation of understanding. Biochemistry and Molecular Biology Education 44(1): 12–19.

29.

Markowitz

Hancock

(2016) Linguistic obfuscation in fraudulent science. Journal of Language and Social Psychology 35(4): 435–445.

30.

Merriam Webster (n.d.) Jargon. Available at: https://www.merriam-webster.com/dictionary/jargon (accessed 11 March 2021).

31.

Metzger

Flanagin

Eyal

Lemus

Mccann

(2003) Credibility for the 21st century: Integrating perspectives on source, message, and media credibility in the contemporary media environment. Annals of the International Communication Association 27(1): 293–335.

32.

Nisbet

Cooper

Garrett

(2015) The partisan brain: How dissonant science messages lead conservatives and liberals to (dis)trust science. The Annals of the American Academy of Political and Social Science 658(1): 36–66.

33.

Otten

Cheng

Drewnowski

(2015) Infographics and public policy: Using data visualization to convey complex information. Health Affairs 34(11): 1901–1907.

34.

Petty

Briñol

Tormala

Wegener

(2007) The role of metacognition in social judgment. In: Higgins

Kruglanski

(eds) Social Psychology: Handbook of Basic Principles, 2nd edn. New York, NY: Guilford Press, pp. 254–284.

35.

Petty

Wegener

Fabrigar

(1997) Attitudes and attitude change. Annual Review of Psychology 48: 609–647.

36.

Plain Language Action and Information Network (n.d.) Plainlanguage.gov—Choose Your Words Carefully. Available at: https://www.plainlanguage.gov/guidelines/words/ (accessed 5 March 2021).

37.

Plaxco

(2010) The art of writing science. Protein Science : A Publication of the Protein Society 19(12): 2261–2266.

38.

Pornpitakpan

(2004) The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology 34(2): 243–281.

39.

Rakedzon

Segev

Chapnik

Yosef

Baram-Tsabari

(2017) Automatic jargon identifier for scientists engaging with the public and science communication educators. PLoS ONE 12(8): e0181742.

40.

Schwarz

(2006) Feelings, fit, and funny effects: A situated cognition perspective. Journal of Marketing Research 43(1): 20–23.

41.

Schwarz

(2010) Meaning in context: Metacognitive experiences. In: Mesquita

Barrett

Smith

(eds) The Mind in Context. New York, NY: Guilford Press, pp. 105–125.

42.

Schwarz

(2011) Feelings-as-information theory. In: Van Lange

Kruglanski

Higgins

(eds) Handbook of Theories of Social Psychology: Volume 1. Thousand Oaks, CA: SAGE, pp. 289–308.

43.

Schwarz

(2015) Metacognition. In: Mikulincer

Shaver

Borgida

Bargh

(eds) APA Handbook of Personality and Social Psychology: Vol. 1 Attitudes and Social Cognition. Washington, DC: American Psychological Association, pp. 203–229.

44.

Sharon

Baram-Tsabari

(2014) Measuring mumbo jumbo: A preliminary quantification of the use of jargon in science communication. Public Understanding of Science 23(5): 528–546.

45.

Shulman

Bullock

(2019) Using metacognitive cues to amplify message content: A new direction in strategic communication. Annals of the International Communication Association 43(1): 24–39.

46.

Shulman

Bullock

(2020) Don’t dumb it down: The effects of jargon in COVID-19 crisis communication. PLoS ONE 15(10): e0239524.

47.

Shulman

Sweitzer

(2018) Varying metacognition through public opinion questions: How language can affect political engagement. Journal of Language and Social Psychology 37(2): 224–237.

48.

Shulman

Dixon

Bullock

Colón Amill

(2020) The effects of jargon on processing fluency, self-perceptions, and scientific engagement. Journal of Language and Social Psychology 39(5–6): 579–597.

49.

Song

Schwarz

(2009) If it’s difficult to pronounce, it must be risky: Fluency, familiarity, and risk perception. Psychological Science 20(2): 135–138.

50.

Tolochko

Song

Boomgaarden

(2019) “That looks hard!”: Effects of objective and perceived textual complexity on factual and structural political knowledge. Political Communication 36(4): 609–628.

51.

Webb

Sheeran

(2006) Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin 132(2): 249–268.

52.

Willoughby

Johnson

Sterman

(2020) Quantifying scientific jargon. Public Understanding of Science 29(6): 634–643.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.72 MB