Sage Journals: Discover world-class research

Abstract

The responses of 1,443 individuals to a questionnaire were used to test a cognitive-decision mediation model that predicts the expression of an opinion. Fear of Social Isolation and opinion congruence predicted a media surveillance motive that was mediated by attention, interpersonal discussions, and elaboration. In turn, perceived familiarity with an issue was mediated mainly by attention, opinion perspective diversity by discussions, and structural knowledge by elaboration. Also, information instrumentality (the perceived usefulness of information) mediated perceived familiarity and opinion perspective diversity, whereas presenting counter-arguments was mediated by structural knowledge. Finally, information instrumentality was the strongest predictor of opinion expression, followed by counter-arguing and cognitive rehearsal.

Keywords

Cognitive mediation model opinion expression spiral of silence structural equation model uses and gratifications

The Spiral of Silence (SOS) literature is dominated by the social control perspective, which suggests that individuals anticipate that social sanctions will be imposed on them for expressing a minority opinion, and an artificial, institutionalized form of agreement is the result. Individuals decide whether to speak out after searching for clues as to the state of public opinion. They observe the opinions that enjoy a society’s approval and those that may lead to a social isolation (Noelle-Neumann & Petersen, 2004). Since individuals’ expressions are determined by fear of other people’s disapproval, individual agency over expression is modest.

The current study attempts to explain the likelihood of opinion expression from a rational perspective, an approach that is less well represented in SOS literature (Scheufele & Moy, 2000). This study introduces a cognitive-decision mediation model (CDMM), based on Eveland’s Cognitive Mediation Model (CMM). The rational perspective is similar to CMM and suggests that people, exercising free will, consciously consume information to deliberate and make decisions, resulting in expressions of opinion that improve democracy (Noelle-Neumann & Petersen, 2004). CMM explains that an audience is instrumental in its usage of media and elaboration of their information (Eveland, 2001, 2002, 2004).

This study tests this rational perspective in Kuwait. Kuwait is an Arab and Muslim state, with a population of four million. Its political system is a constitutional monarchy and petroleum drives its economy (CIA Factbook, 2024). As a collectivist culture that resists change and is patriarchal in nature (Hofstede, 2022), Kuwait is a challenging context for testing a rational perspective because recent SOS studies from this country indicate the impact of social control on the expressions of individuals (Al-Kandari, Frederick, Alkazemi, & Sharif, 2022; Al-Kandari, Frederick, Hasanen, et al., 2022). For example, cultural norms of the Kuwaiti society accentuate collectivism and group loyalty over individuality; “the self is defined as an appendage of the collective, and an individual’s identity is associated with social affiliation to the family or tribe rather than to personal qualities or achievements” (Dwairy et al., 2006, p. 6). Expression of opinions that challenge social norms can be socially costly for individuals. In addition, fundamental Islamists label those who express opinions that do not accord with religious teachings as blasphemous (Alkazemi et al., 2018).

Using Kuwait as an example, this study can indicate a social and practical problem regarding the functioning of democracy in collectivist cultures or the fact that individuals in such cultures may abstain from expressing honest and transparent opinions due to fear of their social group, and not only out of fear of political repression. This study explores the way individuals use their cognitive internalization of news in order to escape the strong influence of a collectivist culture on individuals’ willingness of opinion expression. The results may make it possible to devise better ways for democracy to function by providing practical solutions and implications that can liberate individuals by countering the social constraints on free opinion expression and speech. The results of this study and its implications can be generalized to other similar collectivist countries that endeavor to adopt democracy where collectivism can hinder Western democratic practices. For example, Al-Kandari, Frederick, Al-Tourah, et al. (2022) indicate a manifold of sources of fear that are in effect in Arab and Muslim collectivist nations regarding opinion expression. In addition to fear of the political establishment and system, there are different forms of fear from the group. Those fears are related to expressing opinions that violate religious dogmas and social fear of negative labeling and ridicule (Al-Kandari, Frederick, Al-Tourah, et al., 2022).

The issue of non-citizen residents is examined as a controversial and moral question. The study is similar to other SOS studies that explore such issues (Scheufele & Moy, 2000). It is estimated that there are 100,000 non-citizen residents living in Kuwait (Hagagy, 2021). They have long been denied citizenship by a restrictive policy (Badry, 2021). Recently, activists have requested a policy change to allow those residents to obtain citizenship on the grounds that they deserve the basic human rights citizenship would provide. Opponents argue that such a policy change could lead to adverse repercussions. Recent polls indicate the issue is socially pressing as it tops Kuwaitis’ 10 priorities (Citizens’ Priorities Polls, 2022). This conflict-ridden issue provides an excellent context for an investigation of public opinion.

Outlining a CDMM for SOS

Eveland’s original CMM

Eveland’s (2001, 2002, 2004) original CMM examines how people learn about political news from the mass media. It operates on four premises: (1) a surveillance motivation to know drives people’s reception of media information that they (2) pay attention to and (3) cognitively elaborate on in order to (4) learn information. In addition to those premises, the current study extends CMM by adding (5) a decision stage to form a CDMM that is meant to (6) predict opinion expression as a behavioral outcome in the last stage (Figure 1).

Figure 1.

A proposed cognitive-decision mediation model for SOS.

In identifying the role of motivation in the process of learning from news media, Eveland borrowed from the Uses and Gratifications (U&G) literature. U&G argues that people consciously seek out media to gratify psychological and social needs (Al-Kandari, Frederick, Hasanen, et al., 2022). The influence the media exert is, therefore, predicated on the choices that audience members make as to which media to consume. The U&G literature identifies the audience member as an agent in the determination of the media’s influences rather than the object of their influences.

Eveland (2001) points out that certain types of information processing are the result of motivation. He also states that many psychologists contend that motivation to learn alone does not explain learning, but that specific information processing behaviors produce learning. He argues, therefore, that motivations “should have an indirect effect on learning through the information processing behaviors they instigate” (Eveland, 2001, p. 572). This presents motivations in the role of a conduit for information flow to a cognition system. The present study treats a surveillance motive as a mediator of SOS’s theoretical components.

In CMM, learning from mass media occurs through a cognitive process (Eveland, 2001); “Whether or not one intends to learn really does not matter. What matters is how one processes the material during its presentation” (As cited in Li, 2019). Eveland (2001) considers attention and elaboration the main cognitive processing routes because the U&G literature argues that “two types of information processing—attention and elaboration—should be employed by those who seek surveillance gratifications” (Eveland, 2001, p. 576). Attention is the level of mental focus on a media message and elaboration is a thinking process that mnemonically links information on an issue to previous information in memory. Eveland (2001) argues that research confirms that attention accounts for more variance in explaining media effects than does media exposure. Elaboration has the capacity to induce learning by attaching connotative and associative meanings to new information, producing inferences, and structuring information in memory (Eveland, 2002).

This study also examines a third cognitive process, interpersonal discussions. Like elaboration, it is “a reasoning behavior because exchanging opinions inherently entails mental elaboration” (Jung et al., 2011, p. 409). Deliberation in interpersonal discussions initiates a collective integrative reflection when ideas are exchanged with others. This results in weighing an argument’s pros and cons, comprehending issues, structuring ideas, processing complex concepts, and logically connecting them to conclusions, reasoned justifications, and rational arguments (Hahn, 2020).

Eveland (2001) initially operationalized learning as the acquiring of factual knowledge. In subsequent research he included a second dimension, the acquisition of structural knowledge (Eveland, 2003). He argued that linking elaboration only to factual knowledge underestimates elaboration, which has the capacity to produce multifaceted dimensions of knowledge. Other CMM research compared factual knowledge to perceived familiarity (Yang et al., 2017) and recognition to comprehension (Jensen, 2011). Yang et al. (2017) adopted perceived familiarity to explore whether knowing a basic idea, instead of only facts, was a better indicator of learning outcomes. Findahl (2001) argues, “The fact that so little specific information can be remembered from a news story does not mean that no learning takes place. The informative base from which conclusions are drawn can be forgotten while the conclusions remain” (p. 119). For this reason, perceived familiarity is adopted in this study as a measure of knowledge that lasts longer in memory.

This study also explores other knowledge dimensions. First, opinion diversity, which refers to the heterogeneity of opinions about an issue of which a person is aware. SOS research indicates that opinion heterogeneity stimulates a person to express an opinion (Al-Kandari, Frederick, Hasanen, et al., 2022). Second, structural knowledge indicates that an issue is related to other issues. Perceived knowledge represents factual information as cognitive nodes, while structural knowledge represents the links between those facts or nodes. For example, facts about American politics and those about international politics represent two different nodes and linking those different nodes together to find out how American politics influence international politics is structural knowledge. People with more structural knowledge enjoy greater cognitive complexity (Boukes et al., 2020) that enables them to generalize previous information to new situations (Eveland et al., 2002). Linking issues together helps people develop a universal and holistic view (Eveland et al., 2002), which allows individuals to develop and articulate a variety of arguments in support of their opinions. In the current study, perceived familiarity, opinion diversity, and structural knowledge are the storage components that indicate learning from media news and information when people store received and elaborated information in memory.

CMM extension into CDMM

This study contributes to CMM by adding a cognitive-decision stage. This stage mediates cognitive storage to predict opinion expression. It consists of information instrumentality, cognitive rehearsal, and counter-arguing. First, information instrumentality is an assessment of how confident and certain individuals are that their information can be used to change the opinions of others and defend their own viewpoints. Previous research examined similar concepts, such as political efficacy (Boulianne et al., 2023), communicative utility (Kperogi, 2018), and political information efficacy (Sweetser & Kaid, 2008). That research suggests self-confidence in personal abilities and knowledge encourages engagement in social behaviors.

Second, cognitive rehearsal is a personal visualization of a desirable reaction in imaginary situations (Jiang et al., 2021). Individuals rehearse how to perform well in situations to achieve objectives, such as adjusting behaviors, regulating and controlling settings, and leaving positive impressions (Park et al., 2023; Randall & Randall, 2021). In relation to SOS, cognitive rehearsal can reduce hesitation about speaking out.

Finally, counter-argumentation is a type of discourse consisting of justifications that challenge and refute an alternative argument. Counter-arguing is the process of exposing another argument’s inconsistency, invalidity, flaws, and weaknesses. Studies (Lin, 2022; Tso et al., 2022) have established that people who resist persuasive messages tend to generate counter-argument tactics to refute or at least neutralize other people’s arguments.

CDMM for the prediction of opinion expression

Eveland’s (2001, 2002, 2004) original CMM was meant to predict learning from mass media as the final output of the model. Recently, CMM researchers have applied the model not only to explore learning from the media, but also to assess whether that learning influences behavioral outcomes. Similarly, the current study contributes to the literature by extending CMM to CDMM, the latter being a model that will predict opinion expression as a behavioral outcome.

Researchers have used CMM as a basis to explore cognitive efforts as predictors of non-profit giving and volunteering (Liu & Nah, 2022), taking precautions in the H1 N1 pandemic (Ho et al., 2013), social media engagement (Guo & Chen, 2022), and conducting a breast cancer examination (Zhang & Yang, 2021). In their application of CMM to explore behavioral outcomes, Ho et al. (2013) explain that relating mental activities and knowledge to behavioral outcomes has been common in media research. The authors of this study argue that opinion expression is a behavioral outcome and apply CMM to explore the cognitive information processes that influence whether an individual expresses an opinion.

This study adopts an SOS perspective on the influences that affect individuals’ expression of opinion. An SOS process starts when a group of people perceive that their view on a public matter is a minority view. This group fears social isolation (FSI) from the perceived majority if they express their minority view. A reluctance to express an opinion means the minority withdraw from public discourse. This causes other minority opinion holders to refrain from expressing their view. As the minority’s opinion goes unexpressed, SOS expands (Noelle-Neumann & Petersen, 2004).

Hypotheses

SOS’s theoretical components

SOS incorporates two main components that determine whether individuals express their opinions or censor their views. The first is fear of being socially isolated for expressing minority or unpopular views. The second is the individual’s perception of the majority opinion. According to SOS, individuals who perceive their opinions as being incongruent with the perceived majority view will hesitate to express their opinions out of FSI that might result.

Fear of social isolation

FSI is fear of social rejection for expressing minority opinions that deviate from the perceived majority view. FSI, therefore, encourages the development of a social agreement which can be always safeguarded by an appropriate level of consensus on social values and objectives (Noelle-Neumann & Petersen, 2004). An eight-nation study by Hayes et al. (2013) confirms that “FSI does appear to motivate people to ascertain what the public thinks” (p. 404). Accordingly, this study’s first hypothesis is that: FSI predicts a surveillance motive (H1a).

Opinion congruence

FSI causes people to consume media information to find clues by which they can gauge public opinion. According to Noelle-Neumann and Petersen (2004), people spend time in observing their direct public opinion environment to eventually adopt popular opinions rather than risk being rejected for adopting unpopular opinions. The alternative, avoidance of media, may causes misestimation of public opinion distribution. Consequently, individuals who believe their opinions are incongruent with the perceived majority view will consume even more information to confirm the incongruity of their views or to gauge whether public opinion is shifting and their views are moving toward congruency. Thus, a person’s opinion congruence predicts a surveillance motive (H1b).

Exposure to media news and information

Surveillance motive

In his original publication on CMM, Eveland (2001) focused on a media use motivation he identified as the “surveillance” motive. With its “instrumental orientation” (Haberlin & Atkin, 2022), a surveillance motive is a “cognitive motivation” (Yin et al., 2023) that turns into an “impetus to action” (Wei & Lo, 2021) when people process media information for the purpose of making decisions. People consume media news and information that they translate into sources that they use in decisions about events impacting their lives (Diehl & Lee, 2022). The integration of media information into a person’s decision-making system is the culmination of many cognitive processes that start with paying attention. When people pay attention to media messages, they first reduce cognitive distractions to focus on incoming information, and then they direct mental effort to the message. More than other media content, information includes complicated cues that require attention to be adequately processed and fully understood (Diehl & Lee, 2022). As a result, a surveillance motive predicts attention (H2a).

People elaborate incoming news and information in conjunction with previously stored information to develop an understanding (Diehl & Lee, 2022). A greater motivation to absorb news and information means making more information available for the cognition to elaborate. Making more information available for elaboration means that the cognition has to retrieve more information previously stored in memory to elaborate them together with new incoming information (Eveland, 2001). When a person is fed little information, it will be elaborated on less. Thus, a surveillance motive predicts elaboration (H2b).

People with more information engage in more discussions with others because they have more things to say than people with little information. Research suggests that people who anticipate participation in future conversations tend to consume media news and information more frequently to feel secure and prepared to exchange opinions (Chan et al., 2012), convince others (Sweetser & Kaid, 2008) and defend their own opinions (Jones-Carmack, 2019). Thus, a surveillance motive predicts interpersonal discussions (H2c).

Cognitive processes

Attention

The more attention is given to a message, the more a person can absorb its details and features. Making more details and features available means retrieving more stored information from memory for elaborating the new details and features (Eveland, 2001). Thus, attention predicts elaboration (H3a).

Attention also predicts storage in memory when information bypasses the gatekeeping of elaboration processes. Often, people automatically and directly send information they receive to storage without elaboration, especially when the incoming news and information are repetitive and do not have many new features. When people reach elaboration saturation for an issue, they feel it is useless to elaborate more and end up storing incoming information in memory right away. Research suggests that attention often predicts knowledge more than it predicts elaboration (Eveland, 2002; Eveland et al., 2003). Therefore, attention predicts perceived familiarity (H3b), opinion diversity (H3c), and structural knowledge H3d).

Cognitive elaboration

Elaboration is the process of connecting an issue’s new information with older information in memory. This process produces new inferences, implications, conclusions, and rationales that “increase the strength of the memory storage and the number of mental pathways with which information can be accessed” (Ho et al., 2013, p. 775). Research confirms that elaboration increases perceived knowledge (Eveland et al., 2004; Ho et al., 2013).

People compare and weigh their own opinions with those of others. This requires more elaboration than merely elaborating factual knowledge because people need to justify and reinforce their own opinions to avoid cognitive dissonance when they are challenged by others’ views (Quinn & Wood, 2004). When exposed to diverse opinions, people elaborate the merits of their opinions as well as those of others. When exposed to diverse opinions, people also elaborate the morality of opinions and practicality and feasibility of opinions. Even if this activity is aimed at supporting a person’s own attitudes (Taber & Lodge, 2006), this comparative elaboration generates an awareness of different opinions on an issue.

Finally, research confirms that the more people elaborate their ideas, the more they will gain structural knowledge (Eveland et al., 2004; Lee et al., 2016). Eveland et al. (2004) argue that “Elaboration. . . should be most effective in increasing the integration or density of the structure of information in memory since elaboration is inherently about making mental connections between new information and existing information” (p. 214). Accordingly, elaboration predicts perceived familiarity (H3e), opinion diversity (H3f), and structural knowledge H3 g).

Interpersonal discussions

The Two-step Flow Model of mass communication, indicates that people receive information by discussing issues in their social networks. As a result, of discussions, people are exposed to more information than they would have been without discussion, which increases a person’s knowledge (Gil de Zúñiga et al., 2021). Also, unlike mass media communication, in which an individual can limit exposure to information by attending selectively to messages, people in interpersonal discussions are less able to control the information to which they are exposed. So, they end up hearing heterogeneous opinions on an issue, even those that contradict their views. Even if discussions include social networks of like-minded people, discussion partners bring up opposing opinions to criticize them and reinforce their own opinions. Finally, interpersonal discussions increase structural knowledge because people hear arguments relating one issue to others. Research confirms that interpersonal discussion increases structural knowledge (Eveland, 2004; Eveland & Hively, 2009; Lee et al., 2016) and that heterogeneity of networks on Facebook increases learning from Facebook discussions (M. Kim, Lu, & Lee, 2021; D. H. Kim, Weeks, et al., 2021). Therefore, interpersonal discussion predicts perceived familiarity (H3 h), opinion diversity (H3i), and structural knowledge H3j).

Cognitive storage

Perceived familiarity

People who are familiar with an issue’s details and background are more likely to perceive their information to be instrumental for opinion expression. In contrast, people who know little perceive their information to be insufficient for opinion expression. SOS research demonstrates that knowledgeable people tend to reflect and deliberate (Peacock, 2021). This boosts self-confidence and the ability to persuade and convince others (Chan et al., 2012; Sweetser & Kaid, 2008).

In studies in which people were given information and instructed to use it in discussions, participants rehearsed anticipated communications and behaviors for those discussions (Griffin, 2004; Koh, 2016). The researchers concluded that information in memory serves as a frame of reference and individuals retrieve information for rehearsal of future encounters (Griffin, 2004; Koh, 2016). In relation to opinion expression, people with a lot of information can retrieve more information to allow them to imagine different scenarios in which they respond to people’s questions, opinions, and reactions.

Finally, perceived familiarity with an issue allows for greater critical and analytic evaluation of the weaknesses and flaws in alternative arguments. Booth-Butterfield and Welbourne (2002) argue, “High levels of knowledge about a topic encourage greater elaboration on a persuasive message pertaining to that topic. Conversely, if a person has very little knowledge about a topic, thoughtful scrutiny of the arguments in a message might not be possible” (p. 160). Thus, perceived familiarity predicts information instrumentality (H4a), cognitive rehearsal (H4b), and counter-arguing (H4c).

Opinion diversity

People who know different viewpoints on an issue tend to perceive that their information is instrumental. In expressing an opinion, people who know more about heterogenous opinions on an issue know how others think, the kind of arguments they make, and the weaknesses and flaws of their opinions. Research confirms that people with diverse information can better anticipate other people’s ways of thinking (Sweetser & Kaid, 2008) and cope well with multiple dynamics in different discussion contexts (Brundidge, 2010). Thus, opinion diversity predicts information instrumentality.

If information facilitates cognitive rehearsal (Griffin, 2004; Koh, 2016), then awareness of other people’s opinions facilitates a broader rehearsal process. Conversations with like-minded people can be less effective for rehearsals, as opinions are not challenged. However, people expect engagement in incongruent opinion climates, in which a majority expresses an opinion that the individual opposes. Thus, a person who knows different opinions engages more in rehearsal to prepare responses and reactions to opposing views.

Finally, people who are aware of heterogenous opinions on an issue are better equipped to counter-argue. People who know little about others’ opinions, or retain only information that reinforces their own attitudes, generate few counter-arguments to defend their own position. People who are aware of other’s opinions can generate many counter-arguments. Research confirms that two-sided media messages with opposing perspectives allow audience members to assemble counter-arguments more often than those with one-sided arguments (Becker & Anderson, 2019). As a result, opinion diversity predicts information instrumentality (H4d), cognitive rehearsal (H4e), and counter-arguing (H4f).

Structural knowledge

People with more structural knowledge will be likely to believe their information to be instrumental for expressing opinions. SOS research suggests that exposure to diverse information and opinions leads to seeking news about other topics because topics are interconnected (Brundidge, 2010). This causes a person to link together multiple lines of news stories, and that generates cognitive complexity (Eveland & Schmitt, 2015). This complexity in turn produces a wide array of sophisticated reasoning, justifications, and rationales (Brundidge, 2010; Scheufele et al., 2004) that people can use to supplement their expressions of opinion (Shah et al., 2005). Having such a variety of reasoning, justifications, and rationales increases information instrumentality.

Also, cognitive rehearsal is enhanced by structural knowledge. Experimental studies on cognitive mapping, which links different parts of a subject’s thinking together, showed that subjects who were taught the structures of a behavioral process imagined exemplary behaviors in situations that they had never experienced. Those subjects transferred abstract structural knowledge to perform specific tasks and exhibit appropriate behaviors (Mark et al., 2020). Using their own structural knowledge, participants “infer the existence of paths they have never seen that conform to the structural form and make novel choices that are likely beneficial” (Mark et al., 2020, p. 1). In relation to expressing opinions, people use their structural knowledge to rehearse how to respond and react appropriately.

Finally, people with structural knowledge are in a better position to counter-argue and dispute what others say. Valenzuela et al. (2023) discuss that people’s thinking on an issue contributes to structural knowledge about the issue. That structural knowledge is later accessed when defending opinions from attack. Polk et al. (2009) argue that complex messages that link issues together encourage an audience to unpack contradictions and generate counter-arguments. Thus, structural knowledge predicts instrumentality (H4 g), rehearsal (H4 h), and counter-arguing (H4i).

Cognitive-decisions

Information instrumentality

Eveland et al. (2003), who discuss the concept “knowledge in use,” and Noelle-Neumann and Petersen (2004), who speaks about the media’s “articulation function,” both indicate the ways mass media instruct people on how to articulate themselves and apply their knowledge in discussions with others. The media feed people a variety of arguments, articulations, examples, justifications, and evidence. Such variety boosts an individual’s confidence in speaking out. Those with a variety of self-articulations have less reason to shy away from exchanges of opinion (Chan et al., 2012; Peacock, 2021). Thus, information instrumentality predicts opinion expression (H5a).

Cognitive rehearsal

Since cognitive rehearsal invol-ves a person acting and performing in imaginary situations, the frequency of rehearsing predicts opinion expression for many reasons. First, cognitive rehearsal provides expectations of what will happen in anticipated future situations. This reduces people’s surprise over mishaps. Second, people who rehearse are more motivated to express opinions because they know beforehand what to expect and anticipate the outcomes of expressing opinions. Third, people who rehearse mentally prepare their reactions and responses in advance. This preparation constructs mental schema, or mental manuals of instructions, that can be used in situations in which the expression of opinions is required. Research indicates that rehearsal increases an individual’s self-efficacy (Jiang et al., 2021) as well as communication skills (Aul et al., 2023), positive role modeling (Park et al., 2023), and communication strategies (Randall & Randall, 2021) that are needed in different situations. Thus, cognitive rehearsal predicts opinion expression (H5b).

Counter-arguing

People who counter-argue have greater confidence in their ability to express opinions for three reasons. First, by scrutinizing the flaws and weaknesses of other arguments, they indirectly reinforce the idea that their own attitudes are flawless and morally correct to defend. Second, people with stronger tendencies to counter-argue are equipped with analytical and critical thinking skills that make them less likely to accept other people’s arguments at face value. Therefore, they counter-argue. Third, because they retain many counter-arguments, they feel confident in responding to any allegation others make. In an SOS study, Lin (2022) found that people who thought of an issue to generate counter-arguments resisted other people’s opinions by speaking out rather than remaining silent. Another study found that individuals who generated counter-arguments were more likely to contest others’ arguments (e.g., Tso et al., 2022). Therefore, counter-arguing predicts opinion expression (H5c).

Methods

Sampling procedures and sample

This study utilized a combination of two sampling methods in an effort to achieve a more representative sample. First, a systematic random sampling technique was utilized to select university students at Kuwait University who would serve in a snowball sampling technique. For this systematic random sample of students, a list of all students was obtained from the university. The selected students were later asked to recruit close, first-degree family members of at least 18 years of age to also participate in the study. Students from Kuwait University were selected to participate in the snowball sample because Kuwait University is the first and largest public university with 30,000 enrolled students. Thus, its students reflect diverse social and economic backgrounds. Participants in the second and last technique, family members in the snowball sample, completed a self-administered questionnaire.

Recruiting participants for a survey in a gender-segregated, culturally Arab and Muslim conservative state such as Kuwait is challenging. Employing a network sampling approach was a novel approach selected to overcome this challenge. Allowing students to administer the questionnaire overcomes this cultural hurdle (Cohen & Arieli, 2011). Recruiting family members of students at Kuwait University, Kuwait’s only public university, ensured a diverse socio-economic representation in the sample. The study’s sample size was 1,443 from a country of four million people. Accordingly, this study’s sample size is adequate.

Of the 1,443 respondents, 653 (45.6%) were males and 778 (54.4%) were females. Also, 761 (52.8%) were aged 18 to 25, 252 (17.5%) 26 to 33, 139 (13.4%) 34 to 43, 175 (12.2%) 44 to 55, and 59 (4.1%) were more than 55 years of age.

Predictor variables

Fear of social isolation

FSI was measured with an index of three items adopted from Al-Kandari, Frederick, Al-Tourah, et al. (2022). The items were, “I worry about being isolated if people disagree with me,” “I avoid telling other people what I think when there is a risk that they will avoid me if they know my opinion,” and “I feel annoyed if nobody wants to be around me because of my personal opinions.” Response options were provided on a 5-point scale (5 = strongly agree, 1 = strongly disagree) (M = 2.51, SD = 1.24, Composite Reliability = 0.895, average variance extracted (AVE) = 0.74).

Opinion congruence

This index was adopted from Matthes (2015) and included two items: “My opinion about the non-citizen residents issue is similar to the opinion of the majority of people in our society” and “I have a similar opinion regarding the non-citizen residents issue to most people in our society.” Responses were provided on a 5-point Likert-type scale (5 = Strongly agree, 1 = Strongly disagree) (M = 3.30, SD = 1.07, composite reliability = 0.885, AVE = 0.793).

Mediating variables

Surveillance motive

This variable was created with three items. They were, “I often search for information about non-citizen residents,” “When I hear or see a news story about non-citizen residents, I go online to read more about it,” and “I very much seek information about the issue of non-citizen residents to know more about it.” Responses were given on a 5-point Likert-type scale (5 = Strongly agree, 1 = Strongly disagree) (M = 3.19, SD = 1.08, composite reliability = 0.909, AVE = 0.768).

Attention

Attention was measured using two questionnaire items. They were: “How much focus do you give to news and information about non-citizen residents?” and “How much attention do you pay to news and information about non-citizen residents?” Respondents used a 5-point Likert-type scale (5 = High focus/attention, 1 = Low focus/attention) (M = 3.08, SD = 1.11, composite reliability = 0.960, AVE = 0.923).

Information elaboration

This variable was composed of four items adopted from Eveland (2001). The items were: “I often think about the news I receive from media about the non-citizen residents issue,” “I think about how what I receive from the media about the non-citizen residents issue links to other things I know about this issue,” “I try to relate the news and information I receive from the media about the non-citizen residents issue to my own past experiences,” and “I frequently link new news about the non-citizen residents issue to my own prior personal experiences.” Respondents used a five-point scale (Strongly agree = 5, Strongly disagree = 1) (M = 3.21, SD = 1.10, composite reliability = 0.892, AVE = 0.673).

Interpersonal discussion

It was created using two questionnaire items. They were, “How much do you discuss the issue of non-citizen residents with other people?” and “How much do you converse with others about the non-citizen residents issue?.” Respondents provided answers using a 5-point Likert-type scale (5 = Very much, 1 = Very little) (M = 2.65, SD = 1.17, composite reliability = 0.953, AVE = 0.91).

Knowledge structure

The measure for knowledge structure was adopted from Lee et al. (2016) and included four items. Respondents indicated the level of connection between the non-citizen residents issue and “Kuwait’s political situation,” “the country’s economy,” “security issues,” and “cultural issues.” Respondents used a 5-point scale (Very closely related = 5, Not related at all = 1) (M = 3.55, SD = 1.45, composite reliability = 0.85, AVE = 0.586).

Perceived familiarity

Three questionnaire items, adopted from Yang et al. (2017) were used to assess perceived familiarity. They were: “How much background are you familiar with about the non-citizen residents issue?,” “How much understanding do you have about all aspects surrounding the issue of non-citizen residents?,” and “How much detail do you know about the non-citizen residents issue?” Respondents used a 5-point scale (Very much = 5, Very little = 1) (M = 3.23, SD = 1.08, composite reliability = 0.945, AVE = 0.851).

Information diversity

Information diversity was created using the following questionnaire items: “To what extent do you know about other people’s opinions, especially those that contradict your own opinion, on the issue of non-citizen residents?,” “To what extent are you familiar with different perspectives regarding the issue of non-citizen residents, especially those that differ from your own?,” and “To what extent are you aware of the other viewpoints that are different from yours about the non-citizen residents issue?.” Respondents provided answers on a 5-point scale (5 = Very great extent, 1 = Very little extent) (M = 3.11, SD = 0.99, composite reliability = 0.898, AVE = 0.746).

Cognitive rehearsal

The measure for rehearsal was created as an index of three statements: “Before getting in conversations about the issue of non-citizen residents, I try to predict what others will say to me to know how to respond to them,” “I think about my response to others in case I get in a discussion with them in the future about the non-citizen residents issue,” and “I always reflect about other people’s questions to know how I will respond to them in case I get into a conversation with them.” Response was on a 5-point interval scale (5 = Strongly agree, 1 = Strongly disagree) (M = 3.37, SD = 1.12, composite reliability = 0.887, AVE = 0.724).

Information instrumentality

This measure was created using three questionnaire items. They were: “To what extent do you trust that your information about the non-citizen residents issue will enable you to convince others about your viewpoint in conversations about this issue?,” “To what extent are you certain that the information you know about the issue of non-citizen residents will benefit you in your discussions with others about this issue?,” and “To what extent do you trust that your knowledge about the non-citizen residents issue will help you to exchange your opinions in conversations with others about this issue?.” The response was provided utilizing a 5-point scale (5 = Very great extent, 1 = Very little extent) (M = 3.23, SD = 1.10, composite reliability = 0.926, AVE = 0.808).

Counter-arguing

This measure included four items adopted from Moyer-Gusé and Nabi (2010). First, respondents were asked to envision being in a conversation in which people discuss the non-citizen residents issue and that the people in the discussion were presenting different opinions from that of the respondent. Then, they were asked how they would respond in such a situation by indicating how much they would agree with statements describing possible reactions. The reactions were: “I find myself wanting to respond to what is being discussed,” “I find myself thinking that the discussants offer inaccurate information,” “I find myself looking for flaws in the way information is discussed,” and “I find myself wanting to correct what is being said.” Response was recorded on a 5-point interval scale (5 = Strongly agree, 1 = Strongly disagree) (M = 3.47, SD = 1.05, composite reliability = 0.822, AVE = 0.539).

Criterion variable

Likelihood of expressing an opinion

Opinion expression was created with an index of three items. First, respondents were asked to imagine themselves being in a social gathering in which the issue of non-citizen residents was being discussed. Then, they were asked to estimate the likelihood of them getting involved in the discussion: “To express your opinion about the issue,” “To defend your view about the issue,” and “To convince others about your own opinion about the issue.” Responses were on a 5-point scale (Extremely likely = 5, Extremely unlikely = 1) (M = 3.63, SD = 1.39, composite reliability = 0.890, AVE = 0.730).

Results

Statistical analysis

A Partial Least Squares Structural Equation Modeling (PLS-SEM) was used. PLS-SEM is a multivariate statistical method that determines if different indicators (survey items) reflect different constructs by performing regression tests. PLS is a regression analysis and assesses modeling relationships between variables in high-dimensional data (Table 1).

Table 1.

Frequency individual item reliability statistics.

Survey items	Mean	SD	Loading	Cronbach’s alpha	Composite reliability	Average variance extracted (AVE)
FSI 1	2.50	1.22	0.901	0.827	0.895	0.74
FSI 2	2.50	1.26	0.895
FSI 3	2.53	1.25	0.78
Congruence 1	3.33	1.06	0.877	0.74	0.885	0.793
Congruence 2	3.27	1.08	0.904	0.74	0.885	0.793
Surveillance 1	3.18	1.02	0.886	0.849	0.909	0.768
Surveillance 2	3.24	1.11	0.865
Surveillance 3	3.15	1.10	0.879
Attention 1	3.10	1.10	0.961	0.917	0.960	0.923
Attention 2	3.06	1.12	0.96	0.917	0.960	0.923
Elaboration 1	3.13	1.09	0.78	0.838	0.892	0.673
Elaboration 2	3.21	1.11	0.85
Elaboration 3	3.23	1.11	0.827
Elaboration 4	3.28	1.07	0.824
Discussion 1	2.67	1.16	0.956	0.901	0.953	0.91
Discussion 2	2.63	1.17	0.952	0.901	0.953	0.91
Structure 1	3.90	1.31	0.792	0.769	0.85	0.586
Structure 2	3.39	1.43	0.815
Structure 3	3.57	1.47	0.742
Structure 4	3.34	1.45	0.711
Familiarity 1	3.32	1.08	0.925	0.912	0.945	0.851
Familiarity 2	3.19	1.06	0.927
Familiarity 3	3.17	1.10	0.914
Diversity 1	3.14	0.99	0.863	0.83	0.898	0.746
Diversity 2	3.09	0.99	0.877
Diversity 3	3.11	1.00	0.851
Rehearsal 1	3.49	1.13	0.768	0.81	0.887	0.724
Rehearsal 2	3.36	1.12	0.896
Rehearsal 3	3.28	1.12	0.882
Instrumentality 1	3.20	1.10	0.886	0.881	0.926	0.808
Instrumentality 2	3.26	1.10	0.923
Instrumentality 3	3.23	1.12	0.887
Counter-arguing 1	3.67	0.99	0.833	0.720	0.822	0.539
Counter-arguing 2	3.30	1.02	0.602
Counter-arguing 3	3.19	1.13	0.669
Counter-arguing 4	3.72	1.07	0.808
Expression 1	3.67	1.40	0.846	0.815	0.890	0.730
Expression 2	3.80	1.34	0.871
Expression 3	3.43	1.43	0.845

RMSEA: root mean square error absolute; CFI: comparative fit index; GFI: goodness-of-fit- index; AGFI: goodness-of-fit- index; NFI: normal fit index; TLI: tucker lewis index.

The structural equation model (SEM) demonstrates a good fit to the data, as indicated by several key fit indices. The Chi-square/df ratio of 2.778 is below the recommended threshold of 3, suggesting an acceptable model fit. The RMSEA value of 0.035 is well within the acceptable range, further supporting the model’s fit. The GFI (0.94) and AGFI (0.925) values are close to or above 0.90, indicating a good fit. In addition, the Standardized Root Mean Square Residual (SRMR) (0.033) is below the recommended 0.08 threshold, and the NFI (0.944), TLI (0.957), and CFI (0.963) are all above 0.90, signifying a strong fit of the model to the observed data. Overall, these results suggest that the model provides a robust representation of the underlying data structure.

Measurement model

Construct reliability

Construct reliability was measured using the values of Cronbach alpha and composite reliability. Those measurements assessed interconnection between items of a construct. Cronbach’s alpha scores were from 0.72 to 0.91 and composite reliability from 0.82 to 0.96. Both passed the acceptable threshold of 0.6 for Cronbach alpha and 0.7 for composite reliability according to DeVellis (2016) (Table 1).

Convergent validity

Convergent validity was assessed using the AVE, which reflects the strength of interrelation between indicators composing a construct, and factor loading, which tests if indicators significantly grouped together to create a construct and that no indicator was weak in its level of loading in comparison to others. Both had to pass the acceptable threshold levels (AVE ⩾ .50;|loading| ⩾ .60) (DeVellis, 2016). Values of AVE were from 0.54 and 0.92 and values of factor loadings were from 0.60 to 0.96 (Table 1).

Discriminant validity

Discriminant validity was assessed utilizing the Fornell-Larker criterion and the Heterotrait-Monotrait ratio (HTMT) (Hair et al., 2014) (Tables 2 and 3). Fornell-Larker’s acceptable criteria are recorded using the square root of AVE for a construct, which must be higher than the correlation between that construct and another. Table 2 shows that square roots of all AVEs were above their correlations with other constructs and they all met this criterion. As for an acceptable HTMT threshold ratio, the values of constructs must not exceed 0.85 (Hair et al., 2014). The highest construct value was 0.72.

Table 2.

Fornell-Larcker tests.

	Attention	Opinion congruence	Counter-arguing	Discussion	Elaboration	Expression	FSI	Info.	Diversity	Structural knowledge	Instrumentality	Familiarity	Rehearsal
Attention	0.961
Opinion congruence	0.216	0.891
Counter-arguing	0.461	0.182	0.734
Discussion	0.599	0.238	0.413	0.954
Elaboration	0.612	0.265	0.500	0.524	0.821
Expression	0.433	0.186	0.396	0.443	0.441	0.854
FSI	0.123	0.235	0.090	0.155	0.185	-0.009	0.860
Information	0.634	0.255	0.419	0.567	0.665	0.467	0.163	0.877
Diversity	0.403	0.167	0.398	0.454	0.405	0.371	0.090	0.371	0.864
Structural knowledge	0.236	0.094	0.265	0.199	0.256	0.237	0	0.218	0.215	0.766
Instrumentality	0.536	0.182	0.439	0.533	0.493	0.446	-0.008	0.459	0.579	0.233	0.899
Familiarity	0.543	0.216	0.384	0.507	0.47	0.403	0.031	0.466	0.564	0.277	0.648	0.922
Cognitive rehearsal	0.373	0.226	0.408	0.357	0.457	0.299	0.247	0.423	0.287	0.175	0.324	0.294	0.851

Bold values represent the square root of the Average Variance Extracted (AVE) for each construct. They assess discriminant validity, which ensures that a construct is distinct from other constructs in the model.

Table 3.

HTMT tests.

	Attention	Opinion congruence	Counter-arguing	Discussion	Elaboration	Expression	FSI	Surveillance	Diversity	Structural knowledge	Instrumentality	Familiarity
Attention
Opinion congruence	0.262
Counter-arguing	0.535	0.253
Discussion	0.659	0.291	0.486
Elaboration	0.692	0.332	0.62	0.598
Expression	0.499	0.239	0.472	0.516	0.53
FSI	0.135	0.299	0.145	0.172	0.212	0.067
Surveillance	0.719	0.32	0.512	0.647	0.781	0.559	0.188
Diversity	0.462	0.213	0.508	0.525	0.484	0.449	0.104	0.442
Structural knowledge	0.265	0.124	0.344	0.227	0.312	0.283	0.06	0.259	0.267
Instrumentality	0.597	0.224	0.52	0.598	0.573	0.524	0.049	0.53	0.677	0.274
Familiarity	0.594	0.26	0.457	0.559	0.536	0.466	0.051	0.529	0.649	0.329	0.723
Cognitive rehearsal	0.423	0.294	0.527	0.409	0.544	0.361	0.307	0.5	0.345	0.215	0.374	0.335

Table 4.

Bootstrap results of relationships.

	β	SD	T	95% lower confidence	95% upper confidence	p
FSI -> Surveillance	0.109	0.026	4.216	0.059	0.163	0.001***
Congruence -> Surveillance	0.229	0.028	8.097	0.172	0.284	0.001***
Surveillance -> Attention	0.634	0.019	34.085	0.595	0.669	0.001***
Surveillance -> Elaboration	0.464	0.028	16.695	0.409	0.519	0.001***
Surveillance -> Discussion	0.567	0.019	29.927	0.529	0.603	0.001***
Attention -> Elaboration	0.318	0.028	11.478	0.262	0.373	0.001***
Attention -> Diversity	0.121	0.035	3.445	0.054	0.19	0.001***
Attention -> Familiarity	0.3	0.033	9.091	0.234	0.363	0.001***
Attention -> Structure	0.106	0.038	2.823	0.034	0.181	0.005**
Elaboration -> Diversity	0.18	0.032	5.553	0.117	0.245	0.001***
Elaboration -> Familiarity	0.158	0.03	5.309	0.1	0.217	0.001***
Elaboration -> Structure	0.165	0.034	4.817	0.098	0.231	0.001***
Discussion -> Diversity	0.288	0.033	8.815	0.224	0.35	0.001***
Discussion -> Familiarity	0.245	0.03	8.082	0.185	0.305	0.001***
Discussion -> Structure	0.049	0.033	1.484	-0.016	0.113	0.138
Familiarity -> Instrumentality	0.464	0.027	17.118	0.41	0.515	0.001***
Familiarity -> Counter-arguing	0.199	0.03	6.589	0.139	0.257	0.001***
Familiarity -> Rehearsal	0.174	0.033	5.254	0.107	0.238	0.001***
Diversity -> Instrumentality	0.309	0.026	11.925	0.257	0.359	0.001***
Diversity -> Counter-arguing	0.252	0.031	8.144	0.189	0.314	0.001***
Diversity -> Rehearsal	0.169	0.032	5.315	0.108	0.233	0.001***
Structure -> Instrumentality	0.039	0.019	2.008	0.002	0.077	0.045*
Structure -> Counter-arguing	0.156	0.025	6.192	0.106	0.204	0.001***
Structure -> Rehearsal	0.091	0.028	3.242	0.034	0.144	0.001***
Instrumentality -> Expression	0.317	0.027	11.614	0.263	0.37	0.001***
Counter-arguing -> Expression	0.212	0.028	7.511	0.157	0.268	0.001***
Rehearsal -> Expression	0.109	0.027	4.106	0.058	0.161	0.001***

p < .05. **p < .01. ***p < .001.

Hypotheses testing

All the hypotheses were confirmed, except one (Figure 2). The results indicated that: FSI predicted a surveillance motive (H1a) ( β = 0.109, t = 4.216, p = 0.001), opinion congruence predicted a surveillance motive (H1b) (β = 0.229, t = 8.097, p = 0.001), a surveillance motive predicted attention (H2a) ( β = 0.634, t = 34.085, p = 0.001), elaboration (H2b) ( β = 0.464, t = 16.695, p = 0.001), and interpersonal discussions (H2c) ( β = 0.567, t = 29.927, p = 0.001).

Figure 2.

Node diagram.

Also, attention predicted elaboration (H3a) ( β = 0.318, t = 11.478, p = 0.001), perceived familiarity (H3b) ( β = 0.300, t = 9.091, p = 0.001), opinion diversity (H3c) ( β = 0.121, t = 3.445, p = 0.001), and structural knowledge (H3d) ( β = 0.106, t = 2.823, p = 0.005). Elaboration predicted perceived familiarity (H3e) ( β = 0.158, t = 5.309, p = 0.001), opinion diversity (H3f) ( β = 0.180, t = 5.553, p = 0.001), and structural knowledge (H3 g) ( β = 0.165, t = 4.817, p = 0.001). Interpersonal discussions predicted perceived familiarity (H3 h) ( β = 0.245, t = 8.082, p = 0.001), opinion diversity (H3i) ( β = 0.288, t = 8.815, p = 0.001), but failed to predict structural knowledge (H3j) (β = 0.049, t = 1.484, p = 0.138).

Perceived familiarity predicted information instrumentality (H4a) ( β = 0.464, t = 17.118, p = 0.001), cognitive rehearsal (H4b) (β = 0.174, t = 5.254, p = 0.001), and counter-arguing (H4c) ( β = 0.199, t = 6.589, p = 0.001). Information diversity predicted information instrumentality (H4d) ( β = 0.309., t = 11.925, p = 0.001), cognitive rehearsal (H4e) ( β = 0169., t = 5.315, p = 0.001), and counter-arguing (H4f) ( β = 0.252, t = 8.144, p = 0.001). Structural knowledge predicted information instrumentality (H4 g) ( β = 0.039, t = 2.008, p = 0.045), cognitive rehearsal (H4 h) ( β = 0.091, t = 3.242, p = 0.001), and counter-arguing (H4i) ( β = 0.156, t = 6.192, p = 0.001).

Finally, opinion expression was predicted by instrumentality (H5a) (β = 0.317, t = 11.614, p = 0.001), cognitive rehearsal (H5b) ( β = 0.109, t = 4.106, p = 0.001), and counter-arguing (H5c) ( β = 0.212, t = 7.511, p = 0.001).

Discussion

Discussion of results

Both opinion congruence and FSI predicted the media surveillance motive, but opinion congruence was a stronger predictor of the motive than FSI (Note: we discuss the results using T statistics. This is a statistic that is frequently used in SEM as an indicator of the strength of prediction, as reported by Hair et al. (2014). Being on “the winning side” is bandwagon appeal whereas FSI is a purely a “fear” appeal. Opinion congruence may fit with personal rationality, while FSI may fit a social control perspective. The indirect mediating paths for opinion congruence were greater in number and size of effect than those for FSI. One explanation is that people wanting to be on “the winning side” sought information and cognitively processed it to help in expressing socially approved opinions. As for FSI’s “fear” appeal, individuals fearing other people’s reactions did not process information as much as those wanting to confirm opinion congruence. Fearing others probably caused participants to process information less because they wanted to remain silent or express their opinions less. Individuals expressing an opinion that they thought congruent with the majority sought information and processed it to be able to join the bandwagon.

The surveillance motive was a stronger predictor of interpersonal discussions as a social form of reflection than it was for elaboration as a personal form of reflection. It is possible that people obtain information to share with others in discussions more than they use it as the basis of personal elaboration. Past studies report that people seek information if they are to contribute to social discussions later (Chan et al., 2012; Sweetser & Kaid, 2008). It is possible that people obtain information and hold on to it to elaborate it in interpersonal discussion later. There are several interpretations of this outcome. First, it is possible that not many people use personal elaboration, since elaboration requires great mental effort. Personal elaboration can be cognitively difficult and tiring for some people. Second, people may seek other people’s reflections and elaboration on an issue. Either they are too lazy to process information or think that other people are smarter or understand the issue better than they do.

The fact that attention was a stronger predictor of perceived familiarity than elaboration suggests that a lot of news and information is absorbed and directly stored in memory without much elaboration. This is in line with research that found attention predicted learning more than it predicted elaboration (Eveland, 2002; Eveland et al., 2003). When new information about an issue is repetitive or lacks detail, people do not necessarily elaborate on it. Another reason for attention being a stronger predictor of perceived familiarity than elaboration is that people receive information for reinforcement purposes rather than elaboration. Since their mind is already made up, they do not elaborate; they get new information and send it to memory as new evidence supporting their old attitude. However, the fact that attention was a stronger predictor of perceived familiarity while elaboration was a stronger predictor of opinion diversity and structural knowledge suggests that people pay attention and elaborate depending on the type of information. Perceived familiarity includes historical and background information. Many issues have been sufficiently elaborated on in the past, so people pay little attention to them and pass new information directly to storage with little elaboration. However, opinion diversity and structural knowledge continue to need elaboration. People hear new opinions and relate the issue to other issues all the time. Thus, elaboration is engaged.

While interpersonal discussion was a stronger predictor of perceived familiarity and opinion diversity than elaboration, elaboration was a stronger predictor of structural knowledge than interpersonal discussion. If people are left alone to think about an issue, their thinking leads to cognitive complexity and a holistic view. This is consistent with previous research (Eveland et al., 2004; Lee et al., 2016). Interpersonal discussion, on the other hand, is better in making people familiar with an issue and the different opinions people hold about it. But, interpersonal discussions are less effective in inducing structural knowledge than personal elaboration. Perhaps people discuss issues with like-minded people. When they do so they retain a single line of thought and seldom relate one issue to another.

Perceived familiarity and information diversity were both mediated first by instrumentality, then counter-arguing, and finally rehearsal. People who were familiar with an issue and those who had heterogeneous opinion perspectives felt their information was instrumental for opinion expression. This is in line with previous research linking information and opinion diversity to speaking out (Brundidge, 2010; Chan et al., 2012; Sweetser & Kaid, 2008). Interestingly, structural knowledge was mediated first by counter-arguing, then rehearsal and finally by instrumentality. This outcome suggests that structural knowledge can be dependent mainly on a personal cognitive complexity, cognitive efforts, and cognitive abilities. Linking issues together can produce counter-arguments, even though counter-arguing was predicted more strongly by opinion diversity, familiarity and structural knowledge. This suggests that having a wealth of information and opinion, if coupled with cognitive abilities, can predict counter-arguments and counter-arguing.

Theoretical contribution and significance

The SOS literature has been dominated by the social control perspective. This perspective considers that social sanctions that a majority imposes to institutionalize agreement and perceptions of approved views to be exchanged (Noelle-Neumann & Petersen, 2004) are the determinants of a person’s expressions of opinion. Accordingly, those individuals have little agency over internal decisions triggering them to speak out. The current study offers a rational perspective on opinion expression, which is a less researched perspective in the SOS literature. This study confirms that individuals process information, and their decisions influence their expressions of opinion.

This study also contributes to our knowledge about SOS by exploring cognitive-decision predictors as predictors of opinion expression. Those cognitive-decision predictors are information instrumentality, counter-arguing, and rehearsal. Those concepts are rarely examined in the SOS literature. They explore how an individual’s self-assessment contributes to opinion expression. The SOS literature is full of research that explores outer, social factors and attitudinal and emotional personality factors that condition an individual’s expressions and opinions. This study is exceptional in that it explores SOS from the perspective of cognitive psychology.

This study adapts and tailors the model of CMM to incorporate opinion expression behavior. Other CMM studies have explored the model’s ability to predict behaviors (Guo & Chen, 2022; Ho et al., 2013; Jiang et al., 2021; Liu & Nah, 2022; Zhang & Yang, 2021), but this study is the first to explore CMM in relation to opinion expression as a behavior. Expressing an opinion is a socially progressive behavior. The fact that CMM predicted this behavior confirms the applicability of the model to explore various behaviors, and not only learning from mass media, as earlier CMM studies have supposed (Eveland, 2001, 2002, 2003).

CMM started when Eveland (2001) wanted to explore learning from mass media. He mainly utilized a surveillance motive, attention, and elaboration as factors in the learning process. Over the years, Eveland himself, and other researchers, explored various motivations, cognitive processes, and newer knowledge dimensions. Others explored CMM in relation to behavioral outcomes. The current study extends CMM by exploring cognitive-decision predictors, which this study demonstrates predict opinion expression. This study confirms that people’s actions and behaviors are not performed arbitrarily or because they have the knowledge that enables them to act in a particular way. Before involvement in a behavior, people weigh things and decide. This decision process is a personal calculation before people act. Including a cognitive-decision stage in CMM is a logical step for advancing the model.

Practical implications

Since this study explored its CDMM in relation to opinion expression, it will be valuable to provide practical implications linking the model to other fields of interest and to greater and wider social and political contexts. This study picks the issue of environmental volunteerism. This can demonstrate the model’s generalizability and researchers can infer implications that are linked to other fields of interests, behaviors, issues, and contexts.

Based on this study’s results, perceived familiarity was mainly mediated by attention, structural knowledge by elaboration, and information diversity by interpersonal discussions. These results suggest that people can reach out for different types of knowledge when employing different types of cognitive processes. In relation to environmental volunteerism, it is suggested that raising people’s awareness about the environment can start with media publicity. Paying attention to this publicity can increase people’s perceived knowledge about the background and history of environmental issues. In case issues of the environment are negatively perceived by the public, environmentalists can engage in interpersonal discussions with other people to correct their misperceptions, because discussions, according to this study’s results, can provide people with alternative information and opinions. Such an approach can be performed through tours and involvement in one-on-one offline and online discussions. Finally, if environmentalists want to justify taking severe stands that advance their agenda, they can relate environmental causes to other issues in the society. This will strengthen structural knowledge on the environment by showing other people its connection to mega-society issues.

These different types of knowledge can predict taking different decisions by people regarding the environment. For example, being very aware about environmental information, volunteers can employ their information instrumentality to discuss issues of the environment in the society. By holding a structural knowledge, the volunteers can counter-argue negative arguments. Finally, by applying cognitive rehearsal mechanisms, environmentalists can rehearse being role-models. Regarding environmental causes, they can rehearse how to better present themselves and control their manners when encountering negative reactions of other people.

In a wider political and social context, the CDMM can be useful in highlighting the fact that individuals with information can encounter social controls that are repressive of individuals’ expression of opinions in collectivist nations. This study indicates that individuals with richer information tend to see their information to be more instrumental in expressing their opinions. Consequently, an important step in establishing a robust democracy will be by providing information to people to judge things for themselves. Over time, those individuals will feel empowered and liberated to encounter social sanctions and thus become more expressive of their opinions. Practically however, this can happen by offering information in different terms, such as offering information as facts, diverse opinions and clashing information regarding a phenomenon, and the way information about an issue is interlinked to other issues in the society at a macro level. As a result, establishing democratic political rule in collectivist nations may not achieve its desired end unless the social context is made less restrictive on individuals when they wish to express their opinions. This can occur by empowering individuals with information to judge things for themselves and express themselves on issues the way they see them rather than letting others judge for them or tell them what and how to say things. Once information is provided to individuals, the cognitive processing and internalization will be effective eventually.

Future research directions

Future research should continue the endeavor of this study and build on the CDMM to predict opinion expression in different settings and in relation to different social, political and cultural issues. The CDMM may be more effective with political issues and not cultural or religiously oriented issues. In general, politics leave more room and freedom for people to elaborate on issues. On the other hand, cultural and religious issues are tenacious in nature and learned generationally to a greater extent than political issues. Thus, future studies should test the applicability of CDMM in different settings as well as on different issues.

The fact that elaboration was the third strongest mediator of a surveillance motive after attention and interpersonal discussion suggests that elaboration can be operationally divided into two parts; a general elaboration that links new information to old information in memory, and applied elaboration in which thinking happens to produce solutions. Indeed, early forms of elaboration scales that Eveland introduced included practical elaboration items that were later removed from the index in Eveland et al. (2003) and Eveland et al. (2004). For example, the elaboration index in Eveland and Dunwoody (2002) included the items: “I tried to think of the practical applications of what I read” and “I thought about what actions should be taken by policy-makers based on what I read.” While people can use a general elaboration to bypass day-to-day situations, applied elaboration involves more advanced reflection that contributes to resolving societal problems. The latter requires deeper thinking that is strongly predicted by a surveillance motive and is itself a strong predictor for learning and behavior. Future studies can explore such observations.

This study found that attention influenced perceived familiarity more than elaboration and discussion. Elaboration was more effective on structural knowledge than the other forms of cognitive processing. Finally, interpersonal discussions are more influential than the other factors in developing information diversity. Such findings are important and promising for future research. This finding suggests that each cognitive process (attention, elaboration, discussion) can influence cognitive learning differently (perceived familiarity, structural knowledge, information diversity). Future research needs particularly to examine if this effect persists using different samples and different issues. If the same finding is validated by future research, it can be widely integrated into different media studies.

In this regard, and to better connect interpersonal discussion to structural knowledge, future research should examine interpersonal discussion in relation to discussion contexts. It is very possible that discussion settings can induce greater levels of structural knowledge if discussions include people with diverse political orientations who can relate an issue to other issues. More educated people and people with higher professional status may share complex discussions linking an issue to other issues, contributing to greater structural knowledge.

Future research needs to relate the CDMM to other forms of communication research and not only SOS’s opinion expression. In relation to political communication, CDMM can be tailored to predict political efficacy and engagement in civic organizations, political parties and political campaigning. CDMM can be also tested in relation to cognitive aspects and theories such as cognitive dissonance and the knowledge gap and other public opinion theories. The model can be also adapted to health communication, environmental communication, and many other communication fields.

Finally, future studies could tailor the model to be less rational. This can be done by applying the processes of selective exposure, selective attention, and selective retention. Also, the concepts of a media reinforcement motive, opinion conformity, and homogeneity of discussion groups, can all fit a model tailored to less rational decision-making. Such a model could predict individuals who are “hardcore,” opinionated individuals in SOS. Others include those who already have made up their mind about certain phenomena. With such predictors and mediators, the media can serve to stabilize an opinion or attitude instead changing it.

The study’s limitations

Even though this study included a network (snowball) sample that can be suitable for reaching women participants in a collective and socially and religiously conservative society, future studies need to apply simple random sampling techniques. The model will also be better confirmed by multiple samples exploring different issues instead of only one as this study did. However, this study can be a start in the process of applying CDMM to different issues in different contexts.

While the intention of this study is to test CDMM, the exclusion of examining the effects of covariate (moderating) variables has to be noted as a limitation. Future studies should test different demographics as covariates to better explain the way that CDMM functions. By applying the model using more than one sample, testing different political and social issues, and considering covariates in statistical analyses, the CDMM will be developed more fully.

Footnotes

ORCID iD

Ali A Al-Kandari

Author biographies

Ali A Al-Kandari (PhD,the University of Southern Mississippi) is a Professor of Mass Communication at Gulf University for Science and Technology,Kuwait. His research interests include Uses and Gratifications,Arab media and effects.

Edward Frederick (PhD,University of Wisconsin-Madison) is an Associate Professor of Communication Specializing in Public Relations at the University of Wisconsin–Whitewater,USA. His research focuses on mass media effects on public opinion and public relations.

Ahmad Alsaber (PhD,University of Strathclyde,UK) is an Assistant Professor of Statistics and at the American University of Kuwait (AUK). His research interests include cognitive modeling and the effect of environmental aspects toward public health applications.

Husain A Murad (PhD,Howard University) is an Associate Professor in the Department of Mass Communications at Kuwait University. His research interests include TV-radio as well as media effects and public opinion.

Ahmad Alshallal (PhD,the University of Southern Mississippi) is a Lecturer in Mass Communication at Gulf University for Science and Technology,Kuwait. His research interests include new media effects and news writing.

Ali A Dashti is an Associate Professor of Mass Communication at the Gulf University for Science and Technology. His research interests include public relations and campaigning,political communications,and women in public relations.

References

Al-Kandari

A. A.

Frederick

Alkazemi

Sharif

(2022). Offline and online gender differences in opinion expression about women inclusion in the police force in the Arab Gulf state of Kuwait. Journal of Intercultural Communication Research, 51(1), 93–113. https://doi.org/10.1080/17475759.2021.1950810

Al-Kandari

A. A.

Frederick

Al-Tourah

A. F.

Assad

Al-Kandiri

R. A.

(2022). Sources of fear that inhibit the Kuwaiti youth’s free expression about women serving in the military. Media Asia, 49(4), 310–332. https://doi.org/10.1080/01296612.2022.2065095

Al-Kandari

A. A.

Frederick

Hasanen

M. M.

Dashti

Ibrahim

(2022). Online opinion expression about women serving as judges among university students in Egypt and Kuwait: An integrative study of the Spiral of Silence and Uses and Gratifications. Electronic News, 16(2), 104–138. https://doi.org/10.1177/19312431211052056

Alkazemi

Dashti

Kaposi

Duffy

(2018). Media law in Kuwait. Wolters Kluwer Publishers.

Aul

Ferguson

Russo

(2023). Strategy to enhance undergraduate nursing students’ assertiveness. Journal of Nursing Education, 62(4), 263–264. https://doi.org/10.3928/01484834-20230208-09

Badry

(2021). When the subalterns find their voice–The example of Kuwait’s Bidun. Rocznik Orientalistyczny, 74(2), 98–115. https://doi.org/10.24425/ro.2021.139548.

Becker

Anderson

A. A.

(2019). Using humor to engage the public on climate change: The effect of exposure to one-sided vs. two-sided satire on message discounting, elaboration and counterarguing. Journal of Science Communication, 18(4), A07.

Booth-Butterfield

Welbourne

(2002). The elaboration likelihood model: Its impact on persuasion theory and research. In Dillard

J. P.

Pfau

(Eds.), The persuasion handbook: Developments in theory and practice (pp. 155–173). Sage.

Boukes

van Esch

F. A. E. J.

Snellens

J. A.

Steenman

S. C.

Vliegenthart

(2020). Using cognitive mapping to study the relationship between news exposure and cognitive complexity. Public Opinion Quarterly, 84(3), 599–628. https://doi.org/10.1093/poq/nfaa040

10.

Boulianne

Oser

Hoffmann

C. P.

(2023). Powerless in the digital age? A systematic review and meta-analysis of political efficacy and digital media use. New Media & Society, 25(9), 2512–2536. https://doi.org/10.1177/14614448231176519

11.

Brundidge

(2010). Encountering “difference” in the contemporary public sphere: The contribution of the internet to the heterogeneity of political discussion networks. Journal of Communication, 60(4), 680–700. https://doi.org/10.1111/j.1460-2466.2010.01509.x

12.

Chan

Hao

Jin

(2012). Microblogging, online expression, and political efficacy among young Chinese citizens: The moderating role of information and entertainment needs in the use of Weibo. Cyberpsychology, Behavior, and Social Networking, 15(7), 345–349. https://doi.org/10.1089/cyber.2012.0109

13.

CIA Factbook. (2024). Kuwait. https://www.cia.gov/the-world-factbook/countries/kuwait/

14.

Citizens’ Priorities Polls. (2022). Kuwaiti citizens priorities. http://www.kna.kw/clt-html5/index.asp

15.

Cohen

Arieli

(2011). Field research in conflict environments: Methodological challenges and snowball sampling. Journal of Peace Research, 48(4), 423–435. https://doi.org/10.1177/0022343311405698

16.

DeVellis

R. F.

(2016). Scale development: Theory and applications (Vol. 26). Sage.

17.

Diehl

Lee

(2022). Testing the cognitive involvement hypothesis on social media: “News finds me” perceptions, partisanship, and fake news credibility. Computers in Human Behavior, 128, 107–121. https://doi.org/10.1016/j.chb.2021.107121

18.

Dwairy

Achoui

Abouserie

Farah

Sakhleh

A. A.

Fayad

Khan

H. K.

(2006). Parenting styles in Arab societies: A first cross-regional research study. Journal of Cross-Cultural Psychology, 37(3), 230–247. https://doi.org/10.1177/0022022106286922

19.

Eveland

W. P.

(2001). The Cognitive Mediation Model of learning from the news: Evidence from non-election, off-year election, and presidential election contexts. Communication Research, 28(5), 571–601. https://doi.org/10.1177/009365001028005001

20.

Eveland

W. P.

Jr. (2002). News information processing as mediator of the relationship between motivations and political knowledge. Journalism & Mass Communication Quarterly, 79(1), 26–40. https://doi.org/10.1177/107769900207900103

21.

Eveland

W. P.

Jr. (2003). A “mix of attributes” approach to the study of media effects and new communication technologies. Journal of Communication, 53(3), 395–410. https://doi.org/10.1111/j.1460-2466.2003.tb02598.x

22.

Eveland

W. P.

(2004). The effect of political discussion in producing informed citizens: The roles of information, motivation, and elaboration. Political Communication, 21(2), 177–193. https://doi.org/10.1080/10584600490443877

23.

Eveland

W. P.

Jr. Dunwoody

(2002). An investigation of elaboration and selective scanning as mediators of learning from the Web versus print. Journal of Broadcasting & Electronic Media, 46(1), 34–53. https://doi.org/10.1207/s15506878jobem4601_3

24.

Eveland

W. P.

Jr. Hively

M. H.

(2009). Political discussion frequency, network size, and “heterogeneity” of discussion as predictors of political knowledge and participation. Journal of Communication, 59(2), 205–224. https://doi.org/10.1111/j.1460-2466.2009.01412.x

25.

Eveland

W. P.

Jr. Marton

Seo

(2004). Moving beyond “just the facts” the influence of online news on the content and structure of public affairs knowledge. Communication Research, 31(1), 82–108. https://doi.org/10.1177/0093650203260203

26.

Eveland

W. P.

Schmitt

J. B.

(2015). Communication content and knowledge content matters: Integrating manipulation and observation in studying news and discussion learning effects. Journal of Communication, 65(1), 170–191. https://doi.org/10.1111/jcom.12138

27.

Eveland

W. P.

Jr. Seo

Marton

(2002). Learning from the news in campaign 2000: An experimental comparison of TV news, newspapers, and online news. Media Psychology, 4(4), 353–378. https://doi.org/10.1207/S1532785XMEP0404_03

28.

Eveland

W. P.

Jr. Shah

D. V.

Kwak

(2003). Assessing causality in the cognitive mediation model: A panel study of motivations, information processing, and learning during campaign 2000. Communication Research, 30(4), 359–386. https://doi.org/10.1177/0093650203253369

29.

Findahl

(2001). News in our minds. In Renckstorf

McQuail

Jankowski

(Eds.), Television news research: Recent European approaches and findings (pp. 111–128). Quintessence.

30.

Gil de Zúñiga

Ardèvol-Abreu

Casero-Ripollés

(2021). WhatsApp political discussion, conventional participation and activism: exploring direct, indirect and generational effects. Information, Communication & Society, 24(2), 201–218. https://doi.org/10.1080/1369118X.2019.1642933

31.

Griffin

(2004). Teaching cognitive rehearsal as a shield for lateral violence: An intervention for newly licensed nurses. The Journal of Continuing Education in Nursing, 35(6), 257–263.

32.

Guo

Chen

H. T.

(2022). How does multi-platform social media use lead to biased news engagement? Examining the role of counter-attitudinal incidental exposure, cognitive elaboration, and network homogeneity. Social Media + Society, 8(4), 20563051221129140.

33.

Haberlin

K. A.

Atkin

D. J.

(2022). Mobile gaming and Internet addiction: When is playing no longer just fun and games? Computers in Human Behavior, 126, 106989. https://doi.org/10.1016/j.chb.2021.106989

34.

Hagagy

(2021). Kuwait’s stateless bedoun in limbo as hundreds lose bank accounts. Reuters. https://www.justiceinitiative.org/voices/stateless-kuwait-who-are-bidoon

35.

Hahn

(2020). Argument quality in real world argumentation. Trends in Cognitive Sciences, 24(5), 363–374. https://doi.org/10.1016/j.tics.2020.01.004

36.

Hair

J. F.

Jr. Hult

G. T. M.

Ringle

C. M.

Sarstedt

(2014). A primer on partial least squares structural equation modeling (PLS-SEM). Sage.

37.

Hayes

A. F.

Matthes

Eveland

W. P.

Jr. (2013). Stimulating the quasi-statistical organ: Fear of social isolation motivates the quest for knowledge of the opinion climate. Communication Research, 40(4), 439–462.

38.

S. S.

Peh

Soh

V. W.

(2013). The cognitive mediation model: Factors influencing public knowledge of the H1N1 pandemic and intention to take precautionary behaviors. Journal of Health Communication, 18(7), 773–794. https://doi.org/10.1080/10810730.2012.743624

39.

Hofstede

(2022). The 6-D model of national culture. https://geerthofstede.com/culture-geert-hofstede-gert-jan-hofstede/6d-model-of-national-culture/

40.

Jensen

J. D.

(2011). Knowledge acquisition following exposure to cancer news articles: A test of the Cognitive Mediation Model. Journal of Communication, 61(3), 514–534. https://doi.org/10.1111/j.1460-2466.2011.01549.x

41.

Jiang

Zhang

Teo

(2021). Social use of fitness apps and physical activity knowledge: The roles of information elaboration and interpersonal communication. Journal of Broadcasting & Electronic Media, 65(4), 549–574. https://doi.org/10.1080/08838151.2021.1990295

42.

Jones-Carmack

(2019). Political communication apprehension: Toward productive political discourse online and face-to-face. Journal of Leadership Studies, 13(3), 73–78. https://doi.org/10.1002/jls.21656

43.

Jung

Kim

De Zúniga

H. G.

(2011). The mediating role of knowledge and efficacy in the effects of communication on political participation. Mass Communication and Society, 14(4), 407–430. https://doi.org/10.1080/15205436.2010.496135

44.

Kim

D. H.

Weeks

B. E.

Lane

D. S.

Hahn

L. B.

Kwak

(2021). Sharing and commenting facilitate political learning on Facebook: Evidence from a two-wave panel study. Social Media+ Society, 7(3), 20563051211047876.

45.

Kim

Lee

J. K.

(2021). Heterogeneity of Facebook friend network facilitates political learning: Evidence from a panel survey during the 2016 US presidential campaign. Communication Monographs, 88(4), 463–482. https://doi.org/10.1080/03637751.2021.1882683

46.

Koh

W. M. S.

(2016). Management of work place bullying in hospital: A review of the use of cognitive rehearsal as an alternative management strategy. International Journal of Nursing Sciences, 3(2), 213–222.

47.

Kperogi

F. A.

(2018). “Your English is suspect”: Language, communication, and the pathologization of Nigerian cyber identity through the stylistic imprints of Nigerian e-mail scams. Journal of Communication Inquiry, 42(3), 218–240. https://doi.org/10.1177/0196859918771351

48.

Lee

E. W.

Shin

Kawaja

S. S.

(2016). The augmented cognitive mediation model: Examining antecedents of factual and structural breast cancer knowledge among Singaporean women. Journal of Health Communication, 21(5), 583–592. https://doi.org/10.1080/10810730.2015.1114053

49.

(2019). The cognitive-emotional mediation model of learning about Haze from the news: Examining the roles of risk perception, cognitive processing and emotional responses. The Chinese University of Hong Kong.

50.

Lin

W. W. K.

(2022). Enhancing inoculation in the spiral of silence to promote resistance to attacks: Examining public opinion on Taiwan-PRC relations. Asian Journal for Public Opinion Research, 10(3), 149–177.

51.

Liu

Nah

(2022). Community attachment, communication mediation, and nonprofit participation: An integrated community communication approach. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 33, 269–283. https://doi.org/10.1007/s11266-020-00276-9

52.

Mark

Moran

Parr

Kennerley

S. W.

Behrens

T. E.

(2020). Transferring structural knowledge across cognitive maps in humans and models. Nature Communications, 11(1), 4783.

53.

Matthes

(2015). Observing the “spiral” in the spiral of silence. International Journal of Public Opinion Research, 27(2), 155–176.

54.

Moyer-Gusé

Nabi

R. L.

(2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26–52.

55.

Noelle-Neumann

Petersen

(2004). The spiral of silence and the social nature of man. In Kaid

(Ed.), Handbook of political communication research (pp. 357–374). Lawrence Erlbaum Publisher.

56.

Park

Cho

O. Y.

Choi

J. S.

(2023). A pilot study to examine the effects of a workplace cyberbullying cognitive rehearsal mobile learning program for head nurses: A quasi-experimental study. Healthcare, 11(14), 1–11. https://doi.org/10.3390/healthcare11142041

57.

Peacock

(2021). Diversity, disagreement, and expression across liberal, conservative, and mixed groups. Communication Studies, 72(1), 17–32. https://doi.org/10.1080/10510974.2020.1819843

58.

Polk

Young

D. G.

Holbert

R. L.

(2009). Humor complexity and political influence: An elaboration likelihood approach to the effects of humor type in The Daily Show with Jon Stewart. Atlantic Journal of Communication, 17(4), 202–219. https://doi.org/10.1080/15456870903210055

59.

Quinn

J. M.

Wood

(2004). Forewarnings of influence appeals: Inducing resistance and acceptance. In Knowles

E. S.

Linn

J. A.

(Eds.), Resistance and persuasion (pp. 193–213). Lawrence Erlbaum.

60.

Randall

C. S.

Randall

C. E.

(2021). Critical incident videos: Developing a cognitive strategy in simulation for faculty development. Journal of Nursing Education, 60(2), 81–89. https://doi.org/10.3928/01484834-20210120-05

61.

Scheufele

D. A.

Moy

(2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12(1), 3–28. https://doi.org/10.1093/ijpor/12.1.3

62.

Scheufele

D. A.

Nisbet

M. C.

Brossard

Nisbet

E. C.

(2004). Social structure and citizenship: Examining the impacts of social setting, network heterogeneity, and informational variables on political participation. Political Communication, 21(3), 315–338. https://doi.org/10.1080/10584600490481389

63.

Shah

Cho

Eveland

Jr. Kwak

(2005). Information and expression in a digital age: Modeling internet effects on civic participation. Communication Research, 32(5), 531–565. https://doi.org/10.1177/0093650205279209

64.

Sweetser

K. D.

Kaid

(2008). Stealth soapboxes: Political information efficacy, cynicism and uses of celebrity weblogs among readers. New Media & Society, 10(1), 67–91. https://doi.org/10.1177/1461444807085322

65.

Taber

C. S.

Lodge

(2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50(3), 755–769. https://www.jstor.org/stable/3694247

66.

Tso

S. H.

Lin

W. W. K.

Siu

Y. T.

Song

Chang

C. Y.

(2022). Exploring the “inoculated spiral of silence” effect on the formation of public opinion about controversial issues in Hong Kong. Advances in Journalism and Communication, 10(4), 475–493. https://doi.org/10.4236/ajc.2022.104029

67.

Valenzuela

Bachmann

Bargsted

(2023). The personal is the political? What do Whatsapp users share and how it matters for news knowledge, polarization and participation in Chile. In Digital journalism in Latin America (pp. 26–46). Routledge.

68.

Wei

V. H.

(2021). News in their pockets: A cross-city comparative study of mobile news consumption in Asia. Oxford University Press.

69.

Yang

Chuah

A. S.

Lee

E. W.

S. S.

(2017). Extending the cognitive mediation model: Examining factors associated with perceived familiarity and factual knowledge of nanotechnology. Mass Communication and Society, 20(3), 403–426. https://doi.org/10.1080/15205436.2016.1271436

70.

Yin

Huang

Kim

(2023). The effect of esports content attributes on viewing flow and well-being: A focus on the moderating effect of esports involvement. Sustainability, 15(16), 12207. https://doi.org/10.3390/su151612207

71.

Zhang

Yang

(2021). Linking risk perception to breast cancer examination intention in China: Examining an adapted Cognitive Mediation Model. Health Communication, 36(14), 1813–1824. https://doi.org/10.1080/10410236.2020.1796283

A cognitive-decision mediation model as a predictor of opinion expression

Abstract

Keywords

Outlining a CDMM for SOS

Eveland’s original CMM

CMM extension into CDMM

CDMM for the prediction of opinion expression

Hypotheses

SOS’s theoretical components

Fear of social isolation

Opinion congruence

Exposure to media news and information

Surveillance motive

Cognitive processes

Attention

Cognitive elaboration

Interpersonal discussions

Cognitive storage

Perceived familiarity

Opinion diversity

Structural knowledge

Cognitive-decisions

Information instrumentality

Cognitive rehearsal

Counter-arguing

Methods

Sampling procedures and sample

Predictor variables

Fear of social isolation

Opinion congruence

Mediating variables

Surveillance motive

Attention

Information elaboration

Interpersonal discussion

Knowledge structure

Perceived familiarity

Information diversity

Cognitive rehearsal

Information instrumentality

Counter-arguing

Criterion variable

Likelihood of expressing an opinion

Results

Statistical analysis

Measurement model

Construct reliability

Convergent validity

Discriminant validity

Hypotheses testing

Discussion

Discussion of results

Theoretical contribution and significance

Practical implications

Future research directions

The study’s limitations

Footnotes

ORCID iD

Author biographies

References