Initial Attitudes

Consistent with the medical (or biological) inoculation analogy, McGuire believed that inoculation should be used to protect healthy (previously unchallenged) beliefs (1964). Medical (or biological) inoculations are used to protect healthy individuals from succumbing to a viral infection. As Ivanov and colleagues pointed out, “McGuire’s (1964) use of the biological analogy was not merely stylistic, it was explanatory” (Ivanov et al., 2015, p. 220). As such, attitudinal inoculation should be used to protect individuals with healthy (or established) attitudes from succumbing to a counter-attitudinal challenge. Thus, the boundary condition here-imposed by the medical (or biological) analogy suggests that the attitude in place must be healthy (already established) for inoculation to be an appropriate strategy. Consequently, the application of inoculation is theoretically (or analogically) limited to protecting already-established attitudes (beliefs, behaviors, values, etc.). That is not to suggest that this strategy is ineffective as an approach to change or shape attitudes. In fact, studies have shown that inoculation messages can have a desired effect when used for individuals with neutral and opposed (or unhealthy) attitudes as well (Ivanov et al., 2017; Wood, 2007). However, in such cases, inoculation simply functions as a two-sided attitude formation or attitude change persuasive strategy, rather than as a preemptive attitude protection resistance strategy (Ivanov et al., 2017). If rigidly bound to the medical (or biological) analogy, inoculation is a preemptive attitude protection strategy, which requires that the attitude is established before it can be protected.

Involvement

Pfau and colleagues believed that “involvement holds the key to inoculation’s terrain” (1997, p. 210). In their study, these authors discovered that inoculation works best with moderately-involving issues. They reasoned that low-involving issues would fail to generate a threat—the requisite component of inoculation (Compton & Pfau, 2005)—because people will not care enough about these issues to be properly motivated to protect them. On the other hand, highly-involving issues should be so important to the individuals that they would not necessitate additional motivation. Given the high importance of the issue, these individuals have prior (or inherent) motivation to defend the issue and likely have already prepared their defenses in anticipation of attitudinal challenge. Thus, according to Pfau and colleagues (1997), inoculation would have the greatest contribution to resistance with moderately-involving issues. However, although the results of Banas and Rains’ (2010) meta-analysis showed patterns consistent with Pfau et al.’s findings, the results were not statistically significant; thus no curvilinear relationship for involvement was detected. Banas and Rains speculated that their findings could be influenced by a lack of power associated with the test performed. Alternatively, they suggested, the lack of significant findings could be the result of how involvement has been operationalized in previous inoculation studies, which mainly has consisted of embedding involvement in the topic rather than experimentally manipulating it in the messages. As a result, the notion that issue involvement creates a boundary condition for inoculation has to be considered with reservation.

Cross-Protection

One of inoculation’s main advantages, further discussed in the next section, is the ability to motivate individuals to defend their attitudes (McGuire, 1964). The motivation ensures that individuals do not solely rely on the content of the inoculation message for protection against forthcoming attitudinal challenges. Instead, motivated individuals go a step further by identifying new challenges they proceed to refute and by increasing their ability, via counter-attitudinal practice, to defend against challenges not previously encountered. Stated differently, inoculation forms an umbrella (or blanket) of protection against all arguments within an issue domain (Banas & Rains, 2010; Compton & Pfau, 2005; McGuire, 1964).

Recently, however, Parker and colleagues (2012, 2016) argued that the umbrella (or blanket) of protection may span much wider than originally thought. The authors suggested that inoculation used to protect a single attitude may provide cross-protection for different but related attitudes, as well. The support for their prediction comes from the use of vaccinations in medical settings where a vaccine used to generate protection against one virus can create protection against multiple viruses (Clemens et al., 1988). The results of a few recent studies provided early support for this prediction (Ivanov et al., 2015; Parker et al., 2012; Parker, Rains, & Ivanov, 2016), thus extending the boundaries of the theory.

This section of the chapter identified the terrain in which inoculation operates. The next section focuses on the mechanisms that drive the inoculation-elicited process of resistance.

Threat

In McGuire’s work, threat—the shock value of the message—was generated by presenting individuals with weakened counter-arguments displaying the potential vulnerability of the attitude. Hence, threat represented a realization of attitudinal vulnerability (Compton & Pfau, 2005). In McGuire’s (e.g., McGuire, 1961) initial work, threat was ostensibly generated simply due to the presentation of weakened counter-attitudinal arguments. McGuire termed this implicitly generated shock value as “inherent threat” (McGuire, 1970, p. 63). In his subsequent work (e.g., McGuire & Papageorgis, 1962), he introduced an explicit form of threat as well by using a forewarning in the message. The forewarning, rather explicitly, instructed individuals that their attitudes were vulnerable to forthcoming challenges. Consequently, the combination of explicit and implicit (or inherent) threat, McGuire believed, was responsible for unlocking defense motivation in the individual, which was pivotal in inspiring defense building and practice. As such, McGuire believed that threat is a “requisite construct within the process of inoculation” (Miller et al., 2013, p. 128; but also see Compton & Pfau, 2005). More contemporary inoculation scholars agree with McGuire’s assessment of threat’s importance in inoculation-based resistance (Compton, 2013; Compton & Pfau, 2005; Ivanov et al., 2012b; Miller et al., 2013; Pfau, 1995); this sentiment was punctuated by Pfau’s assertion that threat is “the most distinguished feature of inoculation” (1997, p. 137).

Without threat, the effect of inoculation would be limited to the one prescribed by Lumsdaine and Janis (1953). To remind, these authors believed that the inoculating success of two-sided messages rests, in part, with the familiarity of the forthcoming counter-attitudinal arguments, due to their prior encounter in the two-sided message. In that context, the individual would be more likely to ignore or discount the assailing challenge as one that has already been dealt with and overcome. Thus, the effectiveness of inoculation messages would be limited to scenarios in which the individual faces the same counter-attitudinal arguments previously encountered. However, the effect would be much less pronounced, if at all present, when the counter-attitudinal arguments are novel i.e., have not been faced before. McGuire (1964) argued that threat motivates individuals to shore up their defenses and not only rely on the counter-attitudinal arguments previously encountered in the inoculation message. To show that inoculation’s threat-generated motivation, rather than content, is primarily responsible for the effectiveness of inoculation messages, McGuire (1964) rendered inoculated individuals to either the same counter-attitudinal challenges encountered in the inoculation message or a set of different ones. His general findings and conclusions, supported by a recent meta-analysis (Banas & Rains, 2010), confirmed McGuire’s suspicion as no appreciable difference in the effectiveness of inoculation was discovered based on whether attacks featured the same or novel content. Thus, it seems, the motivation generated by threat, rather than the content of the inoculation message, is largely responsible for the effectiveness of inoculation. This finding has practical significance as well because it suggests that in order to use inoculation messages, one does not have to counter all possible arguments on the issue at hand because inoculation creates an umbrella protection that is capable of “safeguarding against both those counterarguments addressed . . . [by the inoculation message] . . . and those not addressed” (Pfau, 1995, p. 101; also see Banas & Rains, 2010).

Yet, in spite of its ascribed importance, threat remained a primitive construct until the late 1970s when a scale to directly measure threat was developed (Burgoon, Cohen, Miller, & Montgomery, 1978). Numerous subsequent inoculation studies have confirmed the role of threat in the process of resistance (e.g., Ivanov et al., 2012b; Miller et al., 2013; Pfau et al., 2004); however, a recent meta-analysis by Banas and Rains (2010) failed to validate these findings even though the results were in the expected direction. The authors speculated that low power of the test might be responsible for the results. Another possible reason for this inconsistency may be tied to the measure used to assess the presence of threat. A closer examination of previous inoculation studies assessing threat suggests a limited range of threat scores clustered around the midpoint (Compton, 2013). The question inoculation scholars are presently attempting to answer is whether this result is an artifact of the measure (i.e., lack of sensitivity) or the ability of scholars to enhance the level of generated threat. For example, Pfau and colleagues (2010) incorporated components of vested interest (Crano & Prislin, 1995) in their attempt to portray the threat as more “personally relevant” (p. 4). Their attempt was not successful as their enhanced threat manipulation (M = 3.39) was statistically undifferentiated from the regular threat manipulation (M = 3.41) on a seven-point scale. Based on an earlier work by Burgoon and colleagues (1976), Ivanov and colleagues (2013b) manipulated threat by varying the certainty with which the attack is likely to occur. Although the authors had some manipulation success, they were still unable to raise the level of threat beyond the midpoint on a seven-point scale (highest M = 3.88). Finally, Miller and colleagues (2013), borrowing from the theory of psychological reactance (Brehm, 1966), manipulated threat as a restriction to the freedom of individuals to hold the desired attitudes. Although the study was successful in manipulating threat, the highest level of generated threat did not reach the midpoint mark (M = 3.88) on the seven-point traditional scale.

Threat may indeed be a requisite for inoculation-based resistance (Compton & Pfau, 2005; but see Banas & Rains, 2010); however, scholars have much work to do to better understand how to significantly manipulate and better assess this variable. If threat motivates defense preparation (Compton, 2009; Ivanov et al., 2009b; McGuire, 1964), then enhancing threat may be an important factor in eliciting significant motivation.

Yet, the relationship between threat and motivation in the inoculation-elicited resistance is theoretical. The notion that enhanced threat leads to greater motivation is assumed, but unconfirmed. For example, Compton and Ivanov acknowledged this limitation in their study suggesting the “assumption is that threat . . . elicits motivation to engage in resistance processes, such as counterarguing . . .; however, [they] did not measure motivation in this study” (2012, p. 11). Thus, even though motivation seems to be identified as a key mediator of generated resistance, motivation has never been explicitly established nor incorporated into the inoculation model. Its presence has been inferred by inoculation scholars as being elicited by threat and contributing to defense bolstering. For example, McGuire suggested that “any extrinsic threat . . . should increase . . . motivation to assimilate the material and hence enhance its immunizing effectiveness.” (McGuire, 1964, p. 210, italics added for emphasis). Compton added that “. . . both inherent and explicit threat affects motivation to attend to the information presented in inoculation messages” (Compton, 2009, italics added for emphasis). Ivanov and colleagues stated that “the threat element warns the individual about the vulnerability of the attitude currently in place. This realized vulnerability acts as a motivator for the individual to seek out information that would strengthen his or her current attitude” (Ivanov et al., 2009b, p. 48, italics added for emphasis). Thus, motivation is implicitly assumed to be the catalyst for defense bolstering, but could motivation be generated independent of threat? How much of the variance in motivation is explained by threat compared to other mechanisms (e.g., anger, involvement, etc.)?

Threat continues to be a focal point in inoculation research (see Compton, 2013). However, much more remains to be learned about this variable and its interaction with motivation and other components of the inoculation-based process of resistance.

Counterarguing

The second original mechanism of inoculation-elicited resistance, according to McGuire (1964), is counterarguing. Consistent with the biological analogy, counterarguing against opposing counter-attitudinal challenges is analogous to a human body fending off a viral infection by producing antibodies. Compton defines this concept as a process that includes “the collective generation of counterarguments and refutations, [and] post-inoculation pretreatments” (2013, p. 222). McGuire believed that inoculation messages generate threat, which in turn motivates the process of defense bolstering, which he believed to be in the form of counterarguing. As such, McGuire’s vision of inoculation differs from that of Lumsdaine and Janis (1953). McGuire, believed that motivated counterarguing of opposing viewpoints is the mediating process of inoculation resistance, rather than the discounting or ignoring of opposing arguments or viewpoints. This process, according to inoculation scholars (e.g., Pfau et al., 2004), was generally assumed to be an intrapersonal or internal “subvocal, psychological process” (Brandt, 1979, p. 324). However, more recent work has suggested that the process of counterarguing may also be interpersonal as inoculated individuals may turn to others within their social networks, seeking attitudinal reassurance or practicing attitudinal advocacy (Compton & Pfau, 2004, 2009; Ivanov et al., 2012a, 2015).

Unlike threat, which remained as a primitive term in McGuire’s studies, counterarguing received a single assessment as participants were asked to list all of the arguments that came to mind in favor of the belief (see Papageorgis & McGuire, 1961). The quality of the arguments were subsequently rated by a trained outside professional; however, no differences were discovered between the inoculation and control groups, thus failing to confirm the process of inoculation as envisioned by McGuire. Smith suggested that “McGuire provided no evidence that his motivational and informational pretreatments . . . increased the ability and willingness of people to counterargue persuasive attacks” (1982, pp. 294–295). Compton and Pfau (2005) also conceded that the evidence on counterarguing was not as definitive, which is consistent with studies discovering inoculation-generated resistance in absence of significant counterarguing effect (e.g., Compton & Ivanov, 2012). Ivanov, Parker, and Dillingham attributed the equivocality of the findings, at least in part, to the methods used to capture counterarguing (2013a). After pointing out the strengths and weaknesses of several techniques used in measuring counterarguing, the authors suggested using a combination of measures to capture counterarguing by pointing to successful studies that have applied this approach (Banas & Bessarabova, 2009; Miller et al., 2013).

While, as suggested by Compton and Pfau (2005), the evidence may not be definitive, that is not to suggest that evidence of counterarguing mediating the inoculation process or resistance is lacking (e.g., Pfau et al., 2005). In fact, a significant number of studies have found support for McGuire’s theoretical mechanisms regardless whether counterarguing was treated and assessed as a subvocal (e.g., Ivanov et al., 2009a; Pfau et al., 1997, 2001, 2004) or a vocal (e.g., Ivanov et al., 2015) process.

Treating counterarguing as a vocal, in addition to a subvocal, process, a relative novelty in inoculation research, was first suggested in Compton and Pfau’s (2009) seminal article on inoculation and word of mouth communication. In their theoretical essay, the authors proposed that inoculated individuals, in addition to engaging in the process of subvocal counterarguing as prescribed by McGuire (1964), may also engage in conversations with other individuals for the purpose of reassuring their current position on the topic and/or advocating their position to others. Empirical tests by Ivanov and colleagues (2012a, 2015) have subsequently confirmed the reassurance and advocacy functions of post-inoculation word-of-mouth communication, termed as post-inoculation talk (PIT) (Ivanov et al., 2012a, 2015). As a collection, their findings showed that PIT, boosted by inoculation-elicited threat and anger, enhanced attitudinal resistance. A study by Dillingham and Ivanov (2016) partitioned the effects of vocal and subvocal counterarguing and added an additional dependent variable, attitude certainty. The subsequent results revealed a more nuanced interplay between these processes (vocal vs. subvocal counterarguing) and variables (attitude vs. attitude certainty) than originally conceptualized. The process of subvocal counterarguing, when decoupled from vocal counterarguing by restricting the participants from engaging in PIT, yielded familiar findings i.e., inoculation worked in generating attitudinal resistance. However, inconsistent with previous findings (e.g., Pfau et al., 2004), inoculation had no effect on attitudinal certainty. The addition of the vocal (PIT) to the subvocal process of counterarguing, while having no appreciable impact on attitudinal resistance, did have a positive impact on attitudinal certainty. As a result, the authors concluded that, while subvocal counterarguing may be responsible for generating attitudinal resistance, PIT plays an important role in the process of inoculation-elicited resistance by bolstering the certainty with which the attitudes are held, possibly as function of reassurance. An additional benefit of vocal counterarguing or PIT is the ability of the message to be diffused over social networks to individuals not exposed to the original inoculation; thus submitting others to messages of reassurance and/or advocacy. Hence, the effect of PIT in the inoculation process may be quite significant. Yet, while PIT seems to have a positive inoculative effect on the original message recipients, the impact of PIT on those receiving the inoculation message via social diffusion remains unknown and in need for more systematic research.

In general, inoculation works. Over 50 years of research has provided an overall support for the inoculation-elicited process of resistance mediated by the variables—threat and counterarguing—identified by its original author (but see Banas & Rains, 2010). Yet, subsequent inoculation scholarship has identified additional variables (moderating and mediating) impacting the inoculation process of resistance. The next section addresses some of the most relevant complementing and/or competing variables associated with the inoculation-based process(es).

Involvement

Issue involvement plays a significant role in the process of inoculation-generated resistance as it can moderate, mediate, and function independently of the inoculation process. Defined as the salience or importance for the receiver of an attitude object (Pfau et al., 1997), issue involvement may impact the effectiveness of inoculation. Pfau and colleagues (1997) discovered that moderate levels of issue involvement may produce most pronounced inoculation effects; yet these findings have not received meta-analytic support (Banas & Rains, 2010).

In a different study, Pfau et al. (2010) examined the effects of alternative types of involvement (outcome-relevant, value-relevant, and impression-relevant). They hypothesized that outcome-relevant involvement, which they likened to issue involvement, would confer resistance by influencing the original mechanisms of inoculation-based resistance, threat and counterarguing. In contrast, the authors expected that value-relevant (i.e., an attitude integrally connected to the person’s values) and impression-relevant involvement (i.e., a desire to express socially-supported attitudes by significant others) would bypass the original inoculation mechanisms and generate resistance independent of threat and counterarguing (Pfau et al., 2010). No statistically significant results were discovered for impression-relevant involvement, but both outcome-relevant and value-relevant involvement impacted resistance by bypassing threat and counterarguing. These findings are consistent with the earlier work of Pfau and colleagues (1997) who also discovered issue involvement to impact resistance independent of the traditional inoculation mechanism at each level of involvement: high, moderate, and low. In addition to the independent influence on resistance, moderate level of issue (or outcome-relevant) involvement also impacted threat and counterarguing as hypothesized in the Pfau and colleagues’ (2010) study. Thus, involvement seems to play a significant role in the process of resistance independently and by moderating the effect of inoculation.

Issue involvement also mediates the inoculation-elicited process of resistance. In their study, Pfau and colleagues (2004) discovered that inoculation generated threat, which elicited involvement. The elicited involvement then strengthened attitudinal certainty and accessibility and increased counterarguing level. In turn, counterarguing and attitude accessibility directly and certainty indirectly (through attitude accessibility), impacted resistance.

In conclusion, issue involvement has an instrumental role in the process of inoculation-generated resistance. Initial level of involvement, in general, seems to impact the process of resistance independently of the traditional inoculation processes; however, at moderate levels, it also has a direct positive impact on threat and counterarguing. Issue involvement is also elicited by the inoculation message and, as such, it increases the level of resistance via counterarguing, attitude certainty, and attitude accessibility.

Self-Efficacy

Similar to issue involvement, self-efficacy also plays multiple roles in the inoculation-elicited process of resistance as a moderator and mediator (Compton, 2013). In general, self-efficacy is expected to lower the level of experienced threat as increase in self-efficacy should increase the confidence of the individuals in their abilities to defend the challenged attitude (Pfau et al., 2001). Yet the first study to measure self-efficacy (Pfau et al., 2001) did not discover any definitive impact of self-efficacy on threat. However, it did show self-efficacy to be positively related to happiness and anger, both of which enhanced resistance (2001).

Additional studies tested self-efficacy as a product of inoculation and discovered that inoculation can enhance self-efficacy (Farchi & Gidron, 2010; Ivanov et al., 2016; Jackson et al., 2015; Pfau et al., 2009); however, none of these studies directly assessed the impact of elicited self-efficacy on the attitudes. Thus, much more remains to be learned about the mediating role of self-efficacy in the process of inoculation-elicited resistance.

Attitude Accessibility, Certainty, and Associative Networks

Roskos-Ewoldsen and Fazio (1997) suggested that beliefs that are more readily accessible from memory are more salient and more likely to be considered when confronting an issue. Inoculation enhances the accessibility of these beliefs (Pfau et al., 2003, 2004) and accessibility in turn enhances resistance (Pfau et al., 2004). This process works independently of the original mechanisms of the theory—threat and counterarguing—as well as collaboratively through threat, as threat increases attitude accessibility, which leads to greater resistance (Pfau et al., 2004).

The certainty with which individuals hold the attitude also plays an important role in the inoculation process of resistance (Pfau et al., 2004, 2005). Inoculation treatments directly boost attitudinal certainty, which in turn enhances resistance (Pfau et al., 2004, 2005). In addition, certainty can be boosted by threat and elicited involvement and can impact resistance both directly and indirectly by enhancing attitude accessibility (Pfau et al., 2004).

Besides bolstering attitude accessibility and certainty, inoculation modifies the structure of individuals’ associative networks by adding additional nodes and increasing the number of linkages among nodes (Pfau et al., 2005; but see Pfau et al., 2009). The associative network structures consist of affective and cognitive nodes located in long-term memory that are connected by associative pathways (or linkages). More complex structures consist of more interconnected nodes and, as such, are less vulnerable to challenges (2005). As Pfau and colleagues (2005) discovered, inoculation has the ability to enhance individuals’ associative networks, which increases resistance.

Affect

For many years, inoculation theory was considered to elicit primarily a cognitive process of resistance (Compton & Pfau, 2005). Recent inoculation scholarship has recognized the importance of affect in the process of inoculation-elicited resistance (e.g., Compton & Ivanov, 2014; Ivanov et al., 2012a, 2012b; Miller et al., 2013). For example, studies have shown that inoculation can elicit anger (e.g., Ivanov et al., 2012a) or happiness (Pfau et al., 2001), both of which had a positive impact on resistance. But, inoculation can also impact emotions, such as enhancing pride (Pfau et al., 2006) or reducing fear (Pfau et al., 2009). The precise moderating, mediating, and outcome roles specific emotions and moods play in the inoculation process of resistance is not yet clear. Much remains to be learned about the impact of affect on the process of inoculation.

Message Source and Gender

The message source plays a relevant role in the inoculation process. An and Pfau (2004) discovered that more positive perceptions of the credibility of the inoculation source leads to more effective inoculation. Thus, the credibility of the source can moderate the effectiveness of the inoculation message. In addition, a source can have a mediating role in the inoculation process. Miller and colleagues (2013), using the principles of psychological reactance, were able to use inoculation to elicit reactance against the source of the attack message, which lead to greater source derogation. Pfau and Kenski (1990), on the other hand, provided evidence that inoculation may be capable of protecting a message source i.e., the character of a political candidate.

Stone (1969), for the most part, did not find significant difference in the effect of inoculation messages based on gender. What he did find were females to be more influenced by the source of the message compared to males, a finding confirmed by Pfau and Kenski (1990).

Message Type

Over the years, inoculation researchers have tested different types of messages designed to maximize the inoculation’s effectiveness (Ivanov, 2012). For example, Banas and Miller (2013) successfully employed fact-based and logic-based inoculation messages. The fact-based inoculation treatments provided specific refutations of factual claims, while the logic-based messages refuted the reasoning process used in the counterarguments. Pfau and colleagues (2001), on the other hand, used cognitive, affective-happiness, and affective-anger messages to generate resistance. The cognitive messages used arguments, facts, and evidence. The affective-happiness and affective-anger messages relied on content that emphasized either how the desired goal would be facilitated by protecting the present attitude (affective-happiness messages) or how the goal attainment would be obstructed by the counter-attitudinal challenges and challengers (affective-anger messages). All of the different message types showed promise; however, Pfau and colleagues suggested that “resistance is optimized when the refutational preemption component of inoculation messages feature arguments supported by hard evidence in addition to the use of affect triggers which signal that goals may be thwarted” (2009, p. 93).

More recent research has also considered the role of psychological reactance in the message effectiveness. Szabo and Pfau (2001) discovered that inoculation messages can produce reactance, thus recent inoculation studies have added language and postscripts at the end of the inoculation message to ensure that receivers’ freedom of choice is not restricted by the message (e.g., Miller et al., 2013). Furthermore, Miller and colleagues structured the inoculation messages in a manner that elicited reactance against the counter-attitudinal challengers, thus increasing inoculation’s effectiveness (2013; also Niederdeppe, Heley, & Berry, 2015).

Modality

Although inoculation researchers have tinkered with the message content and structure, the effectiveness of the message has persisted (Ivanov, 2012). Its robust nature also crosses different communication modalities (e.g., Banas & Miller, 2013). To date, inoculation messages have been primarily presented in video (e.g., Nabi, 2003; Pfau, Van Bockern, & Kang, 1992) or print format (e.g., Ivanov et al., 2009b; Miller et al., 2013). Pfau and colleagues (2000) conducted a direct comparison of inoculation’s effectiveness presented in both modes and found the effects to be undifferentiated. Dillingham and Ivanov (in press) found similar results when manipulating the modality of the attack message. Inoculation was equally effective regardless of whether the attack was in a print or video format, thus illustrating the robustness of inoculation across different communication modalities.

Perhaps the greatest test of inoculation’s robustness was conducted by Banas and Miller (2013) who diluted the effects of the short print-based inoculation messages by using additional inoculation messages to forewarn participants of the effects of inoculation messages. In essence, they inoculate participants against the effects of inoculation, or metainoculation as termed by the authors. Subsequently, they presented participants with a 40-minute attack video supported by music, narration, and imagery in addition to the impressive length. Yet, inoculation messages still generated attitude protection, thus highlighting the robust nature of this message strategy.

Outcomes

The target outcomes of inoculation messages have varied and included established beliefs (e.g., McGuire, 1964), attitudes (e.g., Ivanov et al., 2009b), values (Bernard, Maio, & Olson, 2003), and behavioral intentions (e.g., Pfau et al., 2001). However, as previously discussed, inoculation has targeted a number of additional outcomes that play a role in the inoculation process such as self-efficacy, attitude accessibility, and attitude certainty, to name a few.

Overall, inoculation-elicited resistance is intricate and complex. Some processes are consistent, others complement (e.g., attitude certainty), and yet others compete (e.g., attitude accessibility) with McGuire’s originally identified process mechanisms: threat and counterarguing. Whereas inoculation research has increased our understanding of these processes, many questions and unknowns still remain. Throughout all of the research, one theme, however, has persisted. Inoculation works (see Banas & Rains, 2010). It is a robust strategy, which success has warranted application of inoculation in multiple contexts reviewed in the next section.

Relevant Application Contexts Outside of Health and Risk Communication

Relevant Application Contexts in Health and Risk Communication

Danger of Making Counterarguments Salient

A common concern with using inoculation messages is rooted in their two-sided nature. Inoculation messages provide opposing arguments, thus making the very existence of these counterarguments salient to message receivers. As a result, practitioners may fear that the introduction of counterarguments may destabilize the attitude in absence of an attack, by highlighting its vulnerability (see Dillingham & Ivanov, in press). However, recent studies have shown that inoculation does not produce attitudinal harm in the absence of an attack (Dillingham & Ivanov, in press; Ivanov et al., 2017; Wigley & Pfau, 2010). Instead, studies have found inoculation messages to have a strengthening attitudinal effect in the absence of an attack, rather than compromising the attitude (Ivanov et al., 2017; Wigley & Pfau, 2010). Thus, the danger of making counterarguments salient to message receivers does not seem to present a major strategic concern.

Ability of Inoculation to Generate Reactance

Inoculation messages are capable of generating psychological reactance (Brehm, 1966) that can obstruct, rather than facilitate, inoculation message acceptance. As previous research has shown (Szabo & Pfau, 2001), if not designed with care and accounting for the potential threat of eliciting reactance, inoculation messages are indeed capable of generating unintended reactance against the inoculation message itself. Thus, it is imperative that inoculation messages are designed using language that does not restrict the perceived freedoms of message receivers to self-determine. As the example in this chapter demonstrates, this can be accomplished by avoiding the use of controlling language, embedding restoration postscripts at the end of the message, and pre-testing the messages (see Ivanov et al., 2016; Miller et al., 2013).

On the other hand, as aforementioned, inoculation could be used strategically to effectively elicit reactance targeted at the attacking source and/or message and thus enhance resistance to attitudinal pressures (Miller et al., 2013). In addition, inoculation could be used to preempt the experience of message reactance and thus facilitate persuasion, or message acceptance, in situations when persuasion, rather than resistance, is the goal (Richards & Banas, 2015).

Inoculation messages are capable of creating reactance that can thwart, rather than facilitate, the intended communication goal. However, careful message design can avoid the negative effect of undesired elicitation of psychological reactance. In addition, inoculation can also be used to exploit the positive outcomes of psychological reactance.

The Impact of Metainoculation

In a very innovative study, Banas and Miller (2013) introduced an antidote to inoculation in the form of metainoculation. In their study, the authors preempted the preemptive strategy of inoculation by cuing individuals to the purpose of inoculation messages and how they work in an effort to prepare these individuals to counter the effects of inoculation. The metainoculation strategy worked, as it lessened the effect of the subsequent inoculation messages, thus helping facilitate, rather than reduce, persuasion.

Consequently, metainoculation can inhibit the full effect of inoculation messages. However, the authors also noted that although the inoculation effect was compromised by the metainoculation message that preceded the presentation of the inoculation message, the effect was not nullified. The results indicated that inoculated individuals who were pretreated with a metainoculation message were more resistant to conspiracy theory propaganda messages compared to individuals who received neither the metainoculation nor the inoculation messages.

As a result of the above, two important takeaway points need to be stated. First, although the inoculation’s effectiveness may be diluted, its effect does not appear to be eliminated by metainoculation. Thus, inoculation messages could be successful even in situations when preempted by metainoculation. Second, Banas and Miller (2013) attributed the success of metainoculation to heuristic processing. The authors used loaded terms to prime individuals by suggesting that they should avoid being manipulated like “sheep” by the inoculation messages and instead, should be “independent thinker[s]” who should make “up their own mind[s]” (Banas & Miller, 2013, p. 199). As a result, metainoculation seems to produce similar outcomes as psychological reactance. More specifically, with metainoculation, message receivers may perceive the freedoms to make up their own minds as being restricted by the inoculation message. As such, using the same aforementioned strategies to combat psychological reactance may help facilitate resistance and blunt the effect of metainoculation.

Relatively Small Effect Sizes

The effect sizes of inoculation messages could be classified as small. More specifically, Banas and Rains’s (2010) meta-analysis produced a mean effect size of d = .43. According to Cohen (1988), this would be a small magnitude effect, with moderate effects starting at d = .50. Hence, the small inoculation effect sizes could generate hesitance on the part of risk and health message strategists in regards to using this strategy. However, Banas and Rains suggest that the effect sizes of inoculation messages should be considered in context (2010). As these authors explain, in health campaigns aimed at increasing young adults’ resistance from engaging in unhealthy behaviors, even small increases in likelihood of resisting the risky persuasive appeals are of “great value” (Banas & Rains, 2010, p. 302).

As mentioned in this chapter, an ongoing inoculation-based intervention study is presently exploring the efficacy of the strategy in protecting recently sober addicts living in a recovery home from experiencing relapse. Prior to honoring the recovery home’s request to perform the inoculation-based intervention, the researchers, in the spirit of full disclosure, forewarned the home’s management that any positive effects experienced were likely to be modest. The response from the home’s management was instructive and poignant, as it suggested that even if one person was saved, the strategy would have been a success and well-worth the effort and investment, thus reiterating Pfau and colleagues’ assertion that in “the context of resistance research . . . small effect sizes are common and are meaningful” (Pfau, Haigh, Sims, & Wigley, 2007, p. 212, emphasis in original).