urlscan.io logo urlscan.io
SecurityTrails logo

www.tandfonline.com Open in urlscan Pro
2606:4700:4400::6812:277b  Public Scan

Back to summary
URL: https://www.tandfonline.com/doi/full/10.1080/08838151.2024.2377244
Submission: On July 16 via api from HK — Scanned from US

Form analysis 4 forms found in the DOM

Name: quickSearch — GET /action/doSearch

<form name="quickSearch" action="/action/doSearch" class="quickSearchForm" title="Quick Search" role="search" method="get" onsubmit="appendSearchFilters(this)" aria-label="Quick Search">
  <span class="quickSearchContainer">
    <span class="simpleSearchBoxContainer">
      <input name="AllField" class="searchText main-search-field autocomplete ui-autocomplete-input" value="" type="search" id="searchText-95bb198f-2314-4e18-995d-c13968636386" title="Type search term here" aria-label="Search"
        placeholder="Search Taylor &amp; Francis for journals, articles and special issues, by keywords, authors, DOI, etc." autocomplete="off" data-history-items-conf="3" data-publication-titles-conf="3" data-publication-items-conf="3"
        data-topics-conf="3" data-contributors-conf="3" data-fuzzy-suggester="false" data-auto-complete-target="title-auto-complete">
    </span>
    <div class="quick-search-btn">
      <button class="mainSearchButton searchButtons pointer" type="submit" value="" title="Search" aria-label="Search"></button>
    </div>
  </span>
  <button class="close-search" data-behaviour="close-dropdown" data-registered="">
    <span class="off-screen">Close search</span>
    <i aria-hidden="true" class="fa fa-times"></i>
  </button>
</form>

Name: quickSearch — GET /action/doSearch

<form name="quickSearch" action="/action/doSearch" class="quickSearchForm" title="Quick Search" role="search" method="get" onsubmit="appendSearchFilters(this)" aria-label="Quick Search">
  <span class="quickSearchContainer">
    <span class="simpleSearchBoxContainer">
      <input name="AllField" class="searchText main-search-field autocomplete ui-autocomplete-input" value="" type="search" id="searchText-" title="Type search term here" aria-label="Search" placeholder="Search journals, articles and special issues"
        autocomplete="off" data-history-items-conf="3" data-publication-titles-conf="3" data-publication-items-conf="3" data-topics-conf="3" data-contributors-conf="3" data-fuzzy-suggester="false" data-auto-complete-target="title-auto-complete">
    </span>
  </span>
  <div class="quick-search-btn">
    <button class="mainSearchButton searchButtons pointer" type="submit" value="" title="Search" aria-label="Search"></button>
  </div>
  <button class="close-search" data-behaviour="close-dropdown" data-registered="">
    <span class="off-screen">Close search</span>
    <i aria-hidden="true" class="fa fa-times"></i>
  </button>
</form>

Name: quickSearch — GET /action/doSearch

<form action="/action/doSearch" name="quickSearch" class="quickSearchForm " title="Quick Search" role="search" method="get" onsubmit="appendSearchFilters(this)" aria-label="Quick Search"><span class="simpleSearchBoxContainer">
    <input name="AllField" class="searchText main-search-field autocomplete ui-autocomplete-input" value="" type="search" id="searchText-d46e3260-1f5c-4802-821a-28a03a699c82" title="Type search term here" aria-label="Search"
      placeholder="Enter keywords, authors, DOI, etc" autocomplete="off" data-history-items-conf="3" data-publication-titles-conf="3" data-publication-items-conf="3" data-topics-conf="3" data-contributors-conf="3" data-fuzzy-suggester="false"
      data-auto-complete-target="title-auto-complete">
  </span>
  <span class="searchDropDownDivRight">
    <label for="searchInSelector-d46e3260-1f5c-4802-821a-28a03a699c82" class="visuallyhidden">Search in:</label>
    <select id="searchInSelector-d46e3260-1f5c-4802-821a-28a03a699c82" name="SeriesKey" class="js__searchInSelector">
      <option value="hbem20" data-search-in="thisJournal"> This Journal </option>
      <option value="" data-search-in="default"> Anywhere </option>
    </select>
  </span>
  <div class="quick-search-btn">
    <button class="mainSearchButton searchButtons pointer" type="submit" value="" title="Search" aria-label="Search"></button>
  </div>
</form>

POST /action/downloadCitation

<form action="/action/downloadCitation" method="post" id="dc_form">
  <h2>To cite this article:</h2>
  <div class="modal-content-body">
    <div role="radiogroup" aria-labelledby="referenceStyle" class="reference-style">
      <span id="referenceStyle">Reference style:</span>
      <label for="apaRadio"><input type="radio" value="apa" id="apaRadio" name="referenceStyle" checked="checked">APA</label>
      <label for="chicagoRadio"><input type="radio" value="chicago" id="chicagoRadio" name="referenceStyle">Chicago</label>
      <label for="harvardRadio"><input type="radio" value="harvard" id="harvardRadio" name="referenceStyle">Harvard</label>
    </div>
    <div class="csl-wrapper copy__text-wrapper">
      <p id="csl-response" class="csl-response copy__text" tabindex="-1"></p>
      <div class="copy-section">
        <div class="citation-copied" aria-label="Citation copied to clipboard" aria-live="polite">
          <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" fill="green" width="12" height="12">
            <path d="M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM369 209L241 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L335 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"></path>
          </svg> Citation copied to clipboard
        </div>
        <button class="btn btn-copy-citation" id="copyToClipBoard" aria-label="Copy citation to clipboard" tabindex="0">
          <i class="fa fa-clipboard" aria-hidden="true"></i> Copy citation to clipboard </button>
      </div>
    </div>
    <div><i>Reference styles above use APA (6th edition), Chicago (16th edition) &amp; Harvard (10th edition)</i></div>
  </div>
  <div class="modal-content-body">
    <h3 class="download-citation">Download citation</h3>
    <div>Download a citation file in RIS format that can be imported by citation management software including EndNote, ProCite, RefWorks and Reference Manager. </div>
    <div role="radiogroup" aria-labelledby="chooseFormat" class="format">
      <span id="chooseFormat">Choose format:</span>
      <label for="risRadio"><input type="radio" value="ris" id="risRadio" name="format" checked="checked">RIS</label>
      <label for="bibtexRadio"><input type="radio" value="bibtex" id="bibtexRadio" name="format">BibTex</label>
      <label for="refworksRadio"><input type="radio" value="refworks" id="refworksRadio" name="format">RefWorks Direct Export</label>
    </div>
    <div role="radiogroup" aria-labelledby="chooseOption" class="include">
      <span id="chooseOption">Choose options:</span>
      <label for="citRadio"><input type="radio" value="cit" id="citRadio" name="include" checked="checked">Citation</label>
      <label for="absRadio"><input type="radio" value="abs" id="absRadio" name="include">Citation &amp; abstract</label>
      <label for="refRadio"><input type="radio" value="ref" id="refRadio" name="include">Citation &amp; references</label>
    </div>
  </div>
  <div class="modal-content-footer">
    <div><button class="btn" id="btn-download-article-citations" tabindex="0" aria-live="polite">
        <i class="fa fa-download" aria-hidden="true"></i> Download citations </button></div>
  </div>
  <input name="doi" type="hidden" value="10.1080/08838151.2024.2377244" tabindex="-1">
  <input name="direct" value="true" type="hidden" tabindex="-1">
</form>

Text Content

Skip to Main Content
 * Browse
 * Search
   
   Close search
 * Publish
    * Find a journal
    * Search calls for papers
    * Journal Suggester
    * Open access publishing
   
   We’re here to help
   
   Find guidance on Author Services


Login  |  Register
Log in or Register
 * Login
 * Register

Cart Add to Cart
Search, Browse, or PublishClose Menu
 * Search
   
   Close search
 * Browse
 * Publish
    * Find a journal
    * Search calls for papers
    * Journal Suggester
    * Open access publishing
   
   We’re here to help
   
   Find guidance on Author Services

 1. Home
 2. All Journals
 3. Journal of Broadcasting & Electronic Media
 4. List of Issues
 5. Latest Articles
 6. Indirect Media Effects on the Adoption o ....

Search in: This Journal Anywhere

Advanced search
Journal of Broadcasting & Electronic Media Latest Articles
Submit an article Journal homepage

Open access

0
Views
0
CrossRef citations to date
0
Altmetric
Listen
Research Article


INDIRECT MEDIA EFFECTS ON THE ADOPTION OF ARTIFICIAL INTELLIGENCE: THE ROLES OF
PERCEIVED AND ACTUAL KNOWLEDGE IN THE INFLUENCE OF PRESUMED MEDIA INFLUENCE
MODEL

Zixi Lia School of Communication, Hong Kong Baptist University, Kowloon, Hong
KongView further author information
,
Jingyuan Shib Department of Interactive Media, Hong Kong Baptist University,
Kowloon, Hong Kong;c AI Media Centre, Hong Kong Baptist University, Kowloon,
Hong KongCorrespondencejolieshi@hkbu.edu.hk
View further author information
,
Yinqiao Zhaoa School of Communication, Hong Kong Baptist University, Kowloon,
Hong KongView further author information
,
Bohan Zhanga School of Communication, Hong Kong Baptist University, Kowloon,
Hong KongView further author information
&
Bu Zhongb Department of Interactive Media, Hong Kong Baptist University,
Kowloon, Hong Kong;c AI Media Centre, Hong Kong Baptist University, Kowloon,
Hong KongView further author information
Published online: 13 Jul 2024
 * Cite this article
 * https://doi.org/10.1080/08838151.2024.2377244
 * CrossMark


In this articleIn this article
 * ABSTRACT
 * Method
 * Results
 * Discussion
 * Conclusion
 * Supplemental material
 * Additional information
 * References

 * Full Article
 * Figures & data
 * References
 * Supplemental
 * Citations
 * Metrics
 * Licensing
 * Reprints & Permissions
 * View PDF PDF View EPUB EPUB


ABSTRACT

To examine the indirect media effects on the adoption of artificial intelligence
(AI), we employed the influence of presumed media influence (IPMI) model and its
extensions. The results of an online survey of adults (N = 1,360) in Hong Kong
revealed that both their media attention to AI-related information and
interpersonal communication about AI influenced presumed others’ media
attention. In turn, that dynamic shaped perceived descriptive norms and
injunctive norms of using AI, which influenced individuals’ intention to use AI
for work and/or learning. Furthermore, although persumed media influence was
stronger among individuals with high perceived knowlegdge than ones with low
perceived knowledge about AI, no difference emerged between individuals with
high versus low actual knowlege about AI.

Previous article View latest articles Next article

Artificial intelligence (AI) technology encompasses a wide range of computer
technology capable of performing tasks or making decisions that require human
intelligence (Zhang & Dafoe, Citation2019). Although the emergence and
prevalence of AI technology have brought substantive concerns, such as data
security and copyright infringement, the technology has also become imperative
in the development of society. For example, AI technology has created more jobs
than anticipated across various industries such as technology, financial
services, and health care (Loukides, Citation2022). Beyond that, AI is expected
to significantly impact economic growth. For instance, AI is expected to double
annual global economic growth rates by 2035 (Szczepański, Citation2019). Thus,
the adoption of AI technology, especially for learning and working purposes, is
crucial to society’s economic growth.

To facilitate the adoption of new technology, effective communication and the
diffusion of information are important (Rogers, Citation1995). Diffusion refers
to the process by which a novel technology spreads through different
communication channels, including mass communication, interpersonal
communication, and network communication within a social system (Rice,
Citation2009). This diffusion process is influenced by several underlying
factors, such as the extent to which actions, perceptions, communication
processes, social norms, and social structures jointly to alleviate adopters’
uncertainty regarding the new technology (Rice, Citation2009; Venkatesh & Davis,
Citation2000). In addition, at an early stage of diffusion, individuals may rely
on multiple communication channels to gather information and make informed
decisions regarding the adoption of such a technology. As of writing this paper,
OpenAI had launched ChatGPT 3.5 and subsequently ChatGPT 4.0, which made AI
technology unprecedently available and accessible among laypeople. Amid the
increasing popularity of ChatGPT, AI technology has become a trending topic.
This timing presents an ideal opportunity to study how communication influences
individuals’ technology adoption as well as its underlying mechanisms.

In this study, to understand how communication affects individuals’ intention to
adopt AI for learning and/or work, we have consulted the influence of the
presumed media influence (IPMI) model (Gunther & Storey, Citation2003) and its
extensions (Gunther et al., Citation2006; Shi et al., Citation2022). Those
frameworks aided us in particularly investigating how mass communication and
interpersonal communication shape individuals’ perceived norms via presumed
media attention from others, which subsequently influences their intentions. In
detail, mass communication, which refers to the dissemination of information
from one source to a large audience (Chaffee & Metzger, Citation2001; Napoli,
Citation2010), takes place through various channels. These channels encompass
traditional media such as television and newspapers, as well as new media
channels like social media platforms and news applications (Lang & Lang,
Citation2009; Napoli, Citation2010), which collectively represent the diverse
avenues available for mass communication in today’s media environment. In
addition to mass communication, interpersonal communication provides another
channel to diffuse information, both of which collaboratively shape presumed
media influence (Shi et al., Citation2022).

Furthermore, given that media effects are contingent (Valkenburg et al.,
Citation2016), we explored the roles of two types of knowledge – actual
knowledge and perceived knowledge – in the IPMI model as a means to examine how
such individual differences shape the IPMI. Our findings are expected both to
advance the IPMI model by identifying its boundary conditions and to inform
communication strategies for promoting the adoption of AI technology.


THE IPMI MODEL AND ITS EXTENSIONS

The IPMI model captures the multistep process illustrating the indirect media
effects (Sun, Citation2013). In that process, individuals who pay attention to
media content form basic initial impressions of a specific topic, and based on
their own level of attention, they estimate others’ level of attention to the
content as well. Subsequently, they assess how others are influenced by such
presumed attention, which further impacts their own cognitive, attitudinal, or
behavioral responses to the media content (Gunther & Storey, Citation2003).

Although the IPMI model originates from the theory of third-person effect
(Davison, Citation1983), it is considered to be a “more general” model with
“broader application” (Gunther & Storey, Citation2003, p. 201) because it
focuses on the presumed influence of others as the antecedent of the audience’s
reaction. In contrast to the theory of third-person effect, which emphasizes
self – other differences in media’s influence, the IPMI model’s core maintains
that an audience’s reactions depend on media content’s presumed influence on
others, not on media’s actual influence on self – other gaps in perception
(Gunther & Storey, Citation2003; Sun, Citation2013). In addition, the IPMI model
releases the constraint of “negative influence corollary” (p. 201), suggesting
that indirect media effects can be generalized to media content with a broader
range of presumed social influence, regardless of its valance (Gunther & Storey,
Citation2003).

The first proposition of the IPMI model contends that a positive association
exists between individuals’ personal media exposure and presumed others’ media
exposure (Gunther & Storey, Citation2003; Gunther et al., Citation2006). From an
audience-centered perspective, media exposure can be conscious or unconscious as
well as concentrated or unconcentrated. Based on the audience’s attention and
consciousness, the state of exposure can be categorized as passive exposure or
active attention (Potter, Citation2009). Whereas media exposure refers to
exposure to media content, media attention, meaning attention paid to specific
media content, involves processing information with certain cognitive efforts
(Potter, Citation2009). In other words, in a state of passive exposure,
audiences can perceive the element of the message but process it unconsciously,
whereas in active attention, audiences need to devote cognitive effort to
process the information, and thus media attention has been regarded as a better
proxy for cognitive effects than media exposure (Ho et al., Citation2020;
Potter, Citation2009). Therefore, in the current study, we opted to examine
media attention instead of media exposure. This allowed us to capture, to some
extent, valid cognitive efforts of the audience members in forming the basic
impressions of a given media content.

Based on their own media attention, individuals develop a subjective estimation
of others’ attention to the same media content, known as the “presumed reach”
(Gunther et al., Citation2006, p. 56). Then, individuals make persuasive
inferences about the influence of such media content on others’ responses,
assuming that it affects others’ attitudes and behaviors (Gunther,
Citation1998). In other words, based on their own media attention, individuals
infer the media attention of other audience members, as well as how others’
attitudes or behaviors are influenced by the same media content. In studies
examining how media content shapes individuals’ adoption of new technology, the
positive association between media attention and presumed others’ attention has
been documented. For instance, individuals who frequently paid attention to
messages about the benefits of plant-based meat or nano-enabled food presumed
that their peers paid frequent attention to those messages as well (Ho et al.,
Citation2020, Citation2022). Based on the first proposition of the IPMI model
and those empirical findings, we proposed Hypothesis 1 (H1):



H1:

Individuals’ media attention to information related to AI technology is
positively associated with presumed others’ attention to similar information.





The diversity of communication channels and the frequency of attention matter in
forming the impression of media content. Individuals are more likely to
passively or actively access information delivered through various channels when
an issue generates an information-rich environment (Avery, Citation2010).
Whether intended or not, individuals can engage in repetitive communication
about one specific topic through multiple communication channels to form a basic
impression about the specific media content (Jang & Park, Citation2018). By
contrast, if an issue emerges in an information-poor environment, then
individuals are less likely to access that information from any channel, which
leads to a weak impression of the media content.

To ensure that a message makes an impression, the literature addressing IPMI
model often presents messages as emerging in information-rich environments,
including political campaigns (e.g., Cohen & Tsfati, Citation2009) and public
health issues (e.g., Hong, Citation2021). Because AI technology is currently a
trending topic, both mass communication and interpersonal communication matter
for the diffusion of AI technology and its acceptance. In particular, as
information is diffused via mass communication channels, interpersonal
communication provides a supplementary channel to diffuse information from mass
media to those who pay little attention to the information on such media (Katz &
Lazarsfeld, Citation1955). In a recent study, interpersonal communication was
introduced and found to serve as the antecedent of presumed others’ media
attention within the context of COVID-19 pandemic (Shi et al., Citation2022).
That finding indicates that interpersonal communication matters for the
diffusion of information in information-rich environments and can consequently
shape individuals’ perceptions of others’ media attention and subsequently their
own behavioral intentions. Therefore, we proposed Hypothesis 2 (H2):



H2:

Individuals’ frequency of engaging in interpersonal communication about AI
technology is positively associated with presumed others’ media attention to
information related to AI technology.






NORMATIVE MECHANISMS IN THE IPMI MODEL

The presumed media influence on others refers to the subjective estimation of
how a given media content influences others (Gunther & Storey, Citation2003).
This presumed influence operates through two underlying mechanisms: ecological
influence and normative influence. In the mechanism of ecological influence,
individuals’ responses are driven by the consideration of the potential
influence of media content on others’ responses. It involves estimating the
magnitude of media influence on others or subjectively assessing how others may
behave in a certain way due to media effects (Sun, Citation2013). On the other
hand, normative influence assumes media to be a source of normative perceptions
and extends the concept of presumed influence to encompass normative
perceptions, including perceived descriptive norms and injunctive norms, which
further shape individuals’ responses (Gunther et al., Citation2006; Sun,
Citation2013).

In detail, perceived descriptive norms refer to the estimated prevalence of a
particular behavior, whereas perceived injunctive norms refer to the estimated
social approval of performing that behavior (Rimal & Lapinski, Citation2015).
Particularly, in an early stage of new technology diffusion, individuals may
lack direct experience with or be uncertain about the technology. People may
rely on the behaviors or opinions of others in their social environments to make
decisions regarding technology adoption (Rice, Citation2009; Venkatesh & Davis,
Citation2000). We thus focused on examining the normative mechanism of presumed
media influence in the current study. The positive association between presumed
others’ media attention and normative perceptions is thoroughly documented in
the literature addressing the IPMI across media content. For instance, in the
context of HIV prevention, individuals’ perceptions of the prevalence of the use
of pre-exposure prophylaxis prompted their intention to seek information about
using pre-exposure prophylaxis (Hong, Citation2021). Moreover, presumed
attention to pro-smoking and pro-drinking media content is positively associated
with perceived norms of the risky behavior, which in turn prompts behavioral
intentions (Gunther et al., Citation2006; Ho et al., Citation2014). Thus, we
proposed Hypotheses 3 and 4 (H3 and H4) as follows:



H3:

Presumed others’ media attention to AI technology is positively associated with
perceived descriptive norms of using AI technology for work and/or learning.



H4:

Presumed others’ attention paid to media content related to AI technology is
positively associated with perceived injunctive norms for work and/or learning.





Moreover, normative perceptions have been theorized as important antecedents of
behavioral intention in the IPMI (Gunther et al., Citation2006; Sun,
Citation2013). Additionally, the theory of technology acceptance holds that
individuals tend to rely on their normative perceptions in deciding whether to
use new technology at the early stage (Venkatesh & Davis, Citation2000).
Therefore, we also hypothesized:



H5:

Perceived descriptive norms of using AI technology for work and/or learning is
positively associated with the intention to use AI technology for work and/or
learning.



H6:

Perceived injunctive norms of using AI technology for work and/or learning is
positively associated with the intention to use AI technology for work and/or
learning.






KNOWLEDGE AS A BOUNDARY CONDITION IN THE IPMI MODEL

To investigate the boundary conditions for the indirect media effects
illustrated by the IPMI model, we followed the recommendation in Valkenburg et
al. (Citation2016) review that such indirect media effects can be enhanced or
reduced by individual differences. In particular, individuals differ in their
ability to process relevant media content (Krcmar, Citation2009), which
consequently prompts different patterns of media effects. In fact, the ability
to process media content can manifest as knowledge about the focal topic (Asghar
et al., Citation2022; Driscoll & Salwen, Citation1997).

Within the context of IPMI, when media attention has occurred, individuals
engage in the assessment and interpretation of media content (Krcmar,
Citation2009). This interpretation and assessment process of media information
is influenced by both subjective knowledge and actual knowledge. Indeed,
knowledge contributes to the variations in understanding and construal of the
media content under consideration (An, Citation2007; Krcmar, Citation2009).
Based on the metacognitive framework (Nelson & Narens, Citation1990), knowledge
can be understood at two underlying levels: the object level, which refers to
actual knowledge, and the meta level, which pertains to perceived knowledge. On
the one hand, actual knowledge is specific to a particular topic and serves as
an objective measure of competence regarding the topic (Park et al.,
Citation1988). By contrast, perceived knowledge refers to individuals’
self-assessment of their knowledge (Radecki & Jaccard, Citation1995), which
reflects the combination of actual knowledge and one’s self confidence (Raju et
al., Citation1995). In other words, individuals may exhibit discrepancies
between their actual and perceived knowledge.

In the literature on the media effects, perceived knowledge is often regarded to
be undesirable because it leads to the illusion of knowledge and could amplify
the presumed negative effects of media content on others (e.g., Schäfer,
Citation2020; Yang & Tian, Citation2021) or enlarge the self – other difference
in perception (e.g., Chen & Atkin, Citation2020; Driscoll & Salwen,
Citation1997). However, in McLeod et al. (Citation1997) study on rap music,
perceived knowledge about rap music was not found to be associated with the self
– other difference in perception. The findings regarding actual knowledge have
also been mixed. For instance, Driscoll and Salwen (Citation1997) found that
actual knowledge was not correlated to the third-person effect, whereas Huh and
Langteau (Citation2007) discovered that individuals with a high level of actual
knowledge tended to perceive the influence of direct-to-consumer prescription
drug advertising on others to be greater than individuals with a low level of
actual knowledge.

Although little research to date has investigated the role of perceived and/or
actual knowledge in the IPMI model, the mentioned finding on the third-person
effect indicates that the two types of knowledge could play a role in shaping
the perceived magnitude of media influence on others, which might further shape
the outcomes of the IPMI. In this study, drawing upon the IPMI model and taking
into account previous, mixed empirical findings, we asked the following research
questions (RQs):



RQ1:

Does actual knowledge (high vs. low) about AI moderate the pattern of the IPMI?
If so, then how does it moderate that pattern?



RQ2:

Does perceived knowledge (high vs. low) about AI moderate the pattern of the
IPMI? If so, then how does it moderate that pattern?






METHOD


PARTICIPANTS AND PROCEDURES

Using quota sampling, we administrated an online survey via Qualtrics with a
total of 1,360 citizens (677 women, 49.8%) in Hong Kong at least 18 years old
(See Table B1 in the Supplementary Materials). The survey took place from May 9
to May 19, 2023 (i.e., approximately 5 months after the launch of ChatGPT 3.5
and 5 days prior to the launch of ChatGPT 4), and the survey’s quota were set
with reference to the census data of adults in Hong Kong regarding age and
gender. After participants provided their consent, they reported their actual
and perceived knowledge about AI. Next, to ensure that participants had a
consistent definition of AI, we provided them with a definition of AI adapted
from Zhang and Dafoe (Citation2019) with several examples of AI applications to
facilitate their understanding. Thereafter, participants answered the questions
addressing the IPMI model’s variables in the context of AI in a randomized order
as a means to eliminate the potential order effects. The institutional research
ethics committee approved the questionnaire and procedure.

The participants had a mean age of 49 years old (SD = 15.66, range = 19–83) and
a median of monthly household income between HKD 50,001 to 60,000 (approx. USD
6,400 to 7,600). By level of education, 46% of participants (n = 750) had at
least a bachelor’s degree. Regarding their computer programming proficiency,
33.1% of participants (n = 450) reported having no basic knowledge of or
experience in programming.


MEASURES

ACTUAL KNOWLEDGE ABOUT AI

Participants answered 10 true-or-false statements (1 = true,
2 = false,3 = unknown) concerning their actual knowledge about AI (Roland
Berger, Citation2020), such as “The key technology of AI is based on machine
learning.” For each statement, the correct answer was assigned a score of 1,
whereas wrong or “unknown” answers were assigned a score of 0 (see Table A1 in
the Supplementary Materials). We totaled the scores for those 10 statements to
create the scale for actual knowledge about AI (M = 5.72, SD = 2.13,
range = 0–10).

PERCEIVED KNOWLEDGE ABOUT AI

We assessed participants’ perceived knowledge about AI using a single item (Akin
et al., Citation2020). Participants indicated the extent to which they believed
that they understood AI on a scale ranging from 0 (a complete lack of
understanding) to 100 (complete understanding everything related to AI). The
mean score was 58.78 (SD = 21.93, range = 0–100).

MEDIA ATTENTION TO AI-RELATED INFORMATION

Integrating the media typology proposed by Gil de Zúñiga et al. (Citation2012)
and practical media usage among citizens in Hong Kong (Kemp, Citation2023), we
assessed media attention to AI-related information using five items. Those items
covered the following media channels: (a) traditional media (e.g., television,
newspapers, and magazines), (b) social media for video sharing (e.g., YouTube),
(c) photo-sharing social media (e.g., Instagram and Xiaohongshu), (d) social
networking sites/applications (e.g., Facebook and WeChat), and (e) news
applications (e.g., HK01, Channel C HK, and TVB NEWS). Participants reported the
frequency with which they had paid attention to AI-related information via those
five media channels in the past month on a 7-point scale ranging from 1 (never)
to 7 (always). We calculated the average score of the five items to create the
scale for media attention (M = 4.46, SD = 1.29, α = .89).

INTERPERSONAL COMMUNICATION ABOUT AI

We assessed participants’ interpersonal communication about AI with three items
(Shi et al., Citation2022). Participants reported the frequency with which they
had discussed AI with (a) family members, (b) classmates and/or colleagues, and
(c) friends in the past month on a 7-point scale ranging from 1 (never) to 7
(always). We averaged the responses to those three items to create a scale for
interpersonal communication (M = 4.22, SD = 1.48, α = .89).

PRESUMED OTHERS’ MEDIA ATTENTION TO AI RELATED INFORMATION

We assessed participants’ presumption of others’ attention to media reporting
AI-related information with three items (Ho et al., Citation2020). Participants
reported the estimated frequency with which (a) family members, (b) friends, and
(c) the Hong Kong public paid attention to AI-related information in the past
month on a 7-point scale ranging from 1 (never) to 7 (always). We averaged the
responses for those three referent groups to create a scale for others’ presumed
media attention (M = 4.43, SD = 1.23, α = .81).

PERCEIVED DESCRIPTIVE NORMS

Participants rated their level of agreement with three statements on a 7-point
scale ranging from 1 (strongly disagree) to 7 (strongly agree): (a) most people
close to me, (b) most people I know, and (c) most people important to me use AI
technology for work and/or learning (Fishbein & Ajzen, Citation2010). We
averaged the responses for the three items to create the scale for perceived
descriptive norms (M = 4.62, SD = 1.27, α = .89).

PERCEIVED INJUNCTIVE NORMS

Participants rated the extent to which they thought the referent groups would
approve of their use of AI technology for work and/or learning on three items
using a 7-point scale ranging from 1 (strongly disapprove) to 7 (strongly
approve) adapted from Fishbein and Ajzen (Citation2010). The referent groups
included (a) people close to me, (b) people I know, and (c) people important to
me. We averaged the responses for the three items to create the scale for
perceived injunctive norms (M = 4.89, SD = 1.09, α = .87).

INTENTION TO USE AI TECHNOLOGY FOR WORK AND/OR LEARNING

We assessed participants’ likelihood of using AI technology for work and/or
learning in the following month using a 7-point Likert scale ranging from 1
(very unlikely) to 7 (very likely), M = 4.94, SD = 1.40.

Table A2 in Supplementary Materials reports the correlations between study
variables.


ANALYTICAL PROCEDURE

STRUCTURAL EQUATION MODELING FOR IPMI MODEL

We used the lavaan package (version 0.6–15) in R studio (version 4.2.2) to
perform structural equation modeling (SEM) with the maximum likelihood method
(MLM) estimator, which is robust with non-normal data. To test H1–H6, we
examined the relationships among the IPMI model’s variables using SEM (see
Figure 1). Age was introduced as a control variable because the use of AI and
perceptions of AI vary across age groups (Chien et al., Citation2019). The
overall model remains the same after introducing more demographic variables (See
Figure D1 in Supplementary Materials). In addition, all reverse models did not
fit the data well (See Table C1–C3 in Supplementary Materials).

Figure 1. The overall influence of the presumed media influence model.

N = 1,360. Age was introduced to the model as control variable.
***p < .001.
Display full size



MULTIGROUP SEM BETWEEN INDIVIDUALS WITH DIFFERENT LEVELS OF ACTUAL/PERCEIVED
KNOWLEDGE

To answer RQ1 and RQ2, we first employed multigroup CFA models for the latent
variables to examine the measurement invariance between individuals with
different levels of actual/perceived knowledge. In the multigroup CFA models,
configural invariance, metric invariance, and scalar invariance were tested in
three hierarchical steps.

After establishing the measurement invariance, we conducted multigroup
comparisons between people with low and high levels of actual/perceived
knowledge in SEM. The baseline model was established by restricting item
loadings, intercepts, and all of the path coefficients as being equal across the
low and high actual/perceived knowledge groups. Next, a series of multigroup
comparisons were performed in which we released the constraint of each path
coefficient one at a time to compare the difference in chi-squared values with
the baseline model and to examine whether the specific path significantly
differed between the low and high actual/perceived knowledge groups.


RESULTS


THE OVERALL IPMI MODEL

The SEM model for the overall IPMI model presented in Figure 1 revealed an
acceptable model fit: χ2/df = 6.52, CFI = .944, TLI = 0.933, RMSEA = .064, 90%
CI of RMSEA = [.060, .067], SRMR = .053.

The results of SEM showed that media attention to AI-related information was
positively associated with presumed others’ attention to AI-related information
(β = .249, p < .001), as was interpersonal communication(β = .730, p < .001).
Presumed others’ media attention to AI-related information was positively
associated with perceived descriptive norms (β = .769,p < .001) and perceived
injunctive norms (β = .764, p < .001). Meanwhile, perceived descriptive norms
(β = .344, p < .001) and perceived injunctive norms (β = .412, p < .001) were
positively associated with intention to use AI technology for work and/or
learning. Therefore, the data were consistent with H1 to H6.


MULTIGROUP SEM: HIGH VERSUS LOW ACTUAL KNOWLEDGE ABOUT AI

For actual knowledge, we used participants’ mean score for actual knowledge
about AI (M = 5.72, SD = 2.14) to categorize participants as having
low(M ≤ 5.72, n = 589, 43%) or high (M > 5.72, n = 771, 57%) actual knowledge
about AI. The results of multigroup CFA showed that measurement invariance was
established across the low and high actual knowledge groups(see Table A3 in
Supplementary Materials).

However, a series of chi-squared tests for the multigroup comparisons revealed
that the differences in the path coefficients of the IPMI model’s variables
across the low and high actual knowledge groups were non-significant (see Table
A4 in Supplementary Materials and Figure 2)

Figure 2. Multigroup analyses for low versus high actual knowledge groups (on
the left) and low versus high perceived knowledge groups (on the right).

The paired path coefficients in bold represent significantly different
coefficients between two groups. Age was introduced to the model as a control
variable.
***p < .001
Display full size




MULTIGROUP SEM: HIGH VERSUS LOW PERCEIVED KNOWLEDGE ABOUT AI

For perceived knowledge, we used participants’ mean score for perceived
knowledge about AI (M = 58.78, SD = 21.93) to categorize participants as having
low (M ≤ 58.78, n = 579, 43%) or high (M > 58.78, n = 781, 57%) perceived
knowledge about AI. The results of multigroup CFA showed that measurement
invariance was established across the low and high perceived knowledge groups
(see Table A3 in Supplementary Materials).

A series of chi-squared tests for the multigroup comparisons revealed that the
difference in the association between media attention and presumed others’ media
attention to AI-related information was significant, χ2(1) = 24.171, p < .001.
In particular, the positive influence of media attention on presumed others’
media attention was stronger in the high perceived knowledge group (β = .331,
p < .001) than in the low perceived knowledge group (β = .181, p < .001). The
results also showed a significant difference in the association between
interpersonal communication and presumed others’ media attention,
χ2(1) = 23.513, p < .001. The positive influence of interpersonal communication
on presumed others’ media attention was stronger in the high perceived knowledge
group (β = .773, p < .001) than in the low perceived knowledge group (β = .669,
p < .001). In addition, the differences in the association between presumed
others’ media attention with the perceived norms were significant,
χ2(1) = 6.558, p = .010 for perceived descriptive norms, and χ2(1) = 6.471,
p = .011 for perceived injunctive norms. In detail, the positive influence of
presumed others’ media attention on perceived descriptive norms was stronger in
the high perceived knowledge group (β = .794, p < .001) than in the low
perceived knowledge group (β = .680, p < .001). The positive influence of
presumed others’ media attention on perceived injunctive norms was also stronger
in the high perceived knowledge group (β = .775, p < .001) than in the low
perceived knowledge group (β = .677, p < .001). The difference in the remaining
paths between the IPMI model’s variables across the low and high perceived
knowledge groups was non-significant (see Table A5 in Supplementary Materials
and Figure 2).


DISCUSSION

By consulting the IPMI model and its extensions, our study revealed the indirect
media effects on individuals’ intention to use AI technology for work and/or
learning. Moreover, we advanced the IPMI model by distinguishing actual
knowledge from perceived knowledge and documented the latter as a boundary
condition for the IPMI model through a series of multigroup SEM analyses. In
detail, found that the model’s pattern did not manifest differently for
individuals with low versus high levels of actual knowledge about AI. However,
the pattern did differ between individuals with low and high levels of perceived
knowledge about AI. Those findings theoretically advance the IPMI model and
contribute to the literature about indirect and conditional media effects on the
adoption of AI technology.


THE IPMI MODEL AND ITS UNDERLYING MECHANISMS

Our findings, in replicating the application of the IPMI model and its
extensions, demonstrate that both media attention and interpersonal
communication are key antecedents of presuming others’ attention to focal media
content. Consistent with the IPMI model’s propositions, individuals’ media
attention was positively associated with presumed others’ media attention to the
same content across various media outlets (Gunther & Storey, Citation2003).
Moreover, interpersonal communication was positively associated with presumed
others’ media attention, which corroborates previous research involving the IPMI
model (Shi et al., Citation2022). However, contrary to Shi et al. (Citation2022)
findings, the effect size of interpersonal communication in our study was
greater than personal media attention’s, for both factors served as antecedents
of others’ presumed media attention. Those findings suggest that in the early
diffusion of a trending topic, whose information richness is less than that of
mainstream topics (e.g., COVID-19), individuals might rely more on interpersonal
communication to obtain information and consequently estimate others’ media
attention.

In terms of the normative mechanisms of the IPMI model, we found that presumed
others’ media attention was positively associated with perceived descriptive and
injunctive norms, which, in turn, positively influenced individuals’ intention
to use AI technology for work and/or learning. Those findings indicate that in
the early diffusion of technology adoption in a society, individuals estimate
others’ acceptance of the behavior under the influence of media content and
subsequently make decisions to engage in such behavior. Indeed, the literature
on social normative influences has long considered norms as communication
phenomena and that individuals can internalize normative influence through
various channels of communication (Rimal & Lapinski, Citation2015). In other
words, theories on social normative influences tend to emphasize the direct
influence of communication on individuals’ perceived norms about a certain
behavior. On the contrary, the IPMI model and its empirical evidence, including
our findings, indicate the indirect effects of communication on individuals’
perceived norms, mediated through one’s estimation of others’ exposure to the
information regarding the focal behavior. Thus, individuals’ normative
perceptions seem to stem from two sources: direct communication and their own
estimation of others’ exposure to the communication. Although testing that
assumption was beyond the scope of our research, we believe that it is valuable
to combine the IPMI model and the theory on social normative influences in
investigating how individuals construct and adjust their normative perceptions
via communication.


PERCEIVED KNOWLEDGE AS A BOUNDARY CONDITION

Our analytical technique (i.e., multigroup SEM technique) allowed comparing the
full IPMI model between individuals with high vs. low actual and perceived
knowledge, which enabled us to investigate the boundary condition for the
pattern of, instead of a single path within, the IPMI model. The findings showed
that perceived knowledge, not actual knowledge, served as the boundary condition
for the IPMI model. This suggests that the indirect media effects posited by
IPMI model were stronger among individuals with high perceived knowledge than
ones with low perceived knowledge.

Although actual knowledge and perceived knowledge are associated, they are
conceptually distinct from each other (Driscoll & Salwen, Citation1997; Radecki
& Jaccard, Citation1995). For one, perceived knowledge but not actual knowledge
has been regarded as a factor of metacognition, meaning one’s awareness of their
knowledge in a specific domain (Koriat & Levy-Sadot, Citation1999). Furthermore,
our findings suggest that perceived knowledge enhanced the associations between
media attention, presumed others’ media, and normative perceptions. However, the
relationship between normative perceptions and the intention to adopt AI
technology was not significantly different between the low vs. high perceived
knowledge groups. Such a consistent relationship between perceived norms and AI
use intention across groups echoed with the theoretical framework as well as
empirical studies about technology adoption (Venkatesh & Davis, Citation2000).
In addition, several meta-analyses have revealed that normative perceptions have
consistently shown moderate effect sizes on behavioral intentions to accept
technology (Chong et al., Citation2022; Schepers & Wetzels, Citation2007). These
findings align with the pattern observed in our study. It suggests that the
influence of normative perceptions on the intention to adopt technology may be
stable across groups with different levels of knowledge.

On the other hand, the pattern of the IPMI model was similar between individuals
with low and high levels of actual knowledge. This suggests that media effects
are contingent on subjective perceptions, such as perceived knowledge, rather
than solely relying on objective measures of competence. In other words, the
process of media effects on behavioral outcomes can be seen as a cognitive
procedure in which individuals are motivated to observe and analyze how reality
operates based on their subjective estimations, rather than hinging on objective
measures, such as actual knowledge.


PRACTICAL IMPLICATIONS

Our findings revealed that both media channels and interpersonal communication
matter in promoting the adoption of AI technology for working and learning.
Governments and other organizations could conduct large-scale media campaigns,
which could educate the public to ethically and responsibly use AI technology.
As media campaigns often induce interpersonal communication regarding the
campaign topic (Southwell & Yzer, Citation2007), such media campaigns about AI
would facilitate interpersonal discussion about AI. Both media campaigns and
induced interpersonal communication are expected to facilitate the AI adoption.
In addition, as the normative influence on AI adoption was found to be stable
across different levels of actual/perceived knowledge groups, it indicates that
social influence and environment are imperative in the adoption of AI for
working and learning purposes. To promote AI use for such purposes,
communication efforts could target at the organizational level, such as
companies and universities, instead of the individual-level, to nurture
supportive social norms to encourage AI use for personal development and
employability.


LIMITATIONS AND FUTURE RESEARCH

First, the data we collected is cross-sectional in nature, which offers limited
causal inferences. To address this limitation, we examined reverse models of our
study model, and the results suggested the reverse models did not fit the data
well. Nevertheless, we acknowledge that such statistical analyses still cannot
determine causality among the IPMI variables we examined. In fact, Sun
(Citation2013) pinpointed that the current formulation of the IPMI model
primarily serves as a depiction of how media effects occur, rather than
providing a comprehensive explanation for why these effects occur and
elucidating the underlying causality, despite containing causal propositions. To
advance the theory building, future research on the IPMI needs to investigate
the underlying causality between IPMI variables, such as through longitudinal
surveys or experiments.

Second, due to the scarcity of literature on the role of actual knowledge and
perceived knowledge in the IPMI framework, we investigated the role of the two
types of knowledge in the framework separately to address the research gap. It
also provides a parsimonious solution. However, these two types of knowledge are
not orthogonal with each other. Future research efforts can segment individuals
into four groups based on their levels (i.e., high and low) of these two types
of knowledge and examine how the IPMI pattern differs across the four groups.

Third, two-thirds of our participants reported having at least some basic
understanding of computer programming, which could be attributed to the
well-developed curriculum on computer programming implemented in Hong Kong since
2013 (Cremer, Citation2023). Such a prevalence of digital and information
literacy might be a practical reason why the normative mechanism of presumed
media influence shapes individuals’ intention to adopt AI. However, the
well-educated sample limits the generalizability of the current findings, making
them less applicable to the countries or regions with limited computer education
or underdeveloped information and communication technologies (ICTs). Future
research could explore how indirect media effects shape AI adoption in those
underdeveloped ICT areas.


CONCLUSION

In our study, we employed the IPMI model and its extension to examine the
indirect effects of media attention and interpersonal communication on the
adoption of AI technology through normative mechanisms. We discovered that both
media attention and interpersonal communication acted as antecedents of the
presumed media exposure of others, which influenced descriptive and injunctive
norms and subsequently individuals’ intention to use AI for work and/or
learning. Moreover, we identified that the indirect media effects posited in the
IPMI model were stronger among individuals with high perceived knowledge than
ones with low perceived knowledge. The findings theoretically advance the IPMI
model and offer insights to promote the adoption of AI technology.


Supplemental material



SUPPLYMENTARY_MATERIALS.DOCX

Download MS Word (346.2 KB)


DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).


DATA AVAILABILITY STATEMENT

The dataset, code for data analyses, and supplementary materials supporting the
findings of this study are available on the OSF webpage
(https://osf.io/h43cn/?view_only=aa677ec3f949445fa67ba57ff9dd4a43).


SUPPLEMENTARY MATERIAL

Supplemental data for this article can be accessed online at
https://doi.org/10.1080/08838151.2024.2377244


ADDITIONAL INFORMATION


NOTES ON CONTRIBUTORS

ZIXI LI

Zixi Li is a doctoral student in the School of Communication, Hong Kong Baptist
University. Her research interests include persuasion, health and environmental
communication, and meta-analysis.

JINGYUAN SHI

Jingyuan Shi is an Associate Professor at the Department of Interactive Media,
Hong Kong Baptist University. Her research interests converge at the
intersection of persuasion, health communication, and new communication
technology.

YINQIAO ZHAO

Yinqiao Zhao is a doctoral student in the School of Communication, Hong Kong
Baptist University. His research interests include journalism studies and media
effects.

BOHAN ZHANG

Bohan Zhang is a doctoral student in the School of Communication, Hong Kong
Baptist University. His research interests include media effects and mobile
communication.

BU ZHONG

Bu Zhong is a Professor at the Department of Interactive Media and the Dean of
the School of Communication, Hong Kong Baptist University. His research centers
on the convergence of communication, technology, and human behavior.




REFERENCES

 * Akin, H., Cacciatore, M. A., Yeo, S. K., Brossard, D., Scheufele, D. A., &
   Xenos, M. A. (2020). Publics’ support for novel and established science
   issues linked to perceived knowledge and deference to science. International
   Journal of Public Opinion Research, 33(2), 422–431.
   https://doi.org/10.1093/ijpor/edaa010
    Web of Science ®Google Scholar
 * An, S. (2007). Attitude toward direct-to-consumer advertising and drug
   inquiry intention: The moderating role of perceived knowledge. Journal of
   Health Communication, 12(6), 567–580.
   https://doi.org/10.1080/10810730701508633
    PubMed Web of Science ®Google Scholar
 * Asghar, M. Z., Barberà, E., Rasool, S. F., Seitamaa-Hakkarainen, P., &
   Mohelská, H. (2022). Adoption of social media-based knowledge-sharing
   behaviour and authentic leadership development: Evidence from the educational
   sector of Pakistan during COVID-19. Journal of Knowledge Management, 27(1),
   59–83. https://doi.org/10.1108/jkm-11-2021-0892
    Web of Science ®Google Scholar
 * Avery, E. J. (2010). Contextual and audience moderators of channel selection
   and message reception of public health information in routine and crisis
   situations. Journal of Public Relations Research, 22(4), 378–403.
   https://doi.org/10.1080/10627261003801404
    Web of Science ®Google Scholar
 * Chaffee, S. H., & Metzger, M. J. (2001). The end of mass communication? Mass
   Communication & Society, 4(4), 365–379.
   https://doi.org/10.1207/s15327825mcs0404_3
    Google Scholar
 * Chen, H., & Atkin, D. (2020). Understanding third-person perception about
   internet privacy risks. New Media & Society, 23(3), 419–437.
   https://doi.org/10.1177/1461444820902103
    Web of Science ®Google Scholar
 * Chien, S., Chu, L., Lee, H., Yang, C., Lin, F., Yang, P., Wang, T., & Yeh, S.
   (2019). Age difference in perceived ease of use, curiosity, and implicit
   negative attitude toward robots. ACM Transactions on Human-Robot Interaction,
   8(2), 1–19. https://doi.org/10.1145/3311788
    Web of Science ®Google Scholar
 * Chong, A. Y. L., Blut, M., & Zheng, S. (2022). Factors influencing the
   acceptance of healthcare information technologies: A meta-analysis.
   Information & Management, 59(3), 103604.
   https://doi.org/10.1016/j.im.2022.103604
    Web of Science ®Google Scholar
 * Cohen, J., & Tsfati, Y. (2009). The influence of presumed media influence on
   strategic voting. Communication Research, 36(3), 359–378.
   https://doi.org/10.1177/0093650209333026
    Web of Science ®Google Scholar
 * Cremer, J. (2023, January 20). Hong Kong experts highlight the usefulness of
   coding from a young age to enhance international school students’
   problem-solving skills. South China Morning Post.
   https://www.scmp.com/news/hong-kong/education/article/3206395/hong-kong-experts-highlight-usefulness-coding-young-age-enhance-international-school-students
    Google Scholar
 * Davison, W. P. (1983). The third-person effect in communication. Public
   Opinion Quarterly, 47(1), 1–15. https://doi.org/10.1086/268763
    Web of Science ®Google Scholar
 * Driscoll, P. D., & Salwen, M. B. (1997). Self-perceived knowledge of the O.J.
   Simpson Trial: Third-person perception and perceptions of guilt. Journalism &
   Mass Communication Quarterly, 74(3), 541–556.
   https://doi.org/10.1177/107769909707400308
    Web of Science ®Google Scholar
 * Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The
   reasoned action approach. Psychology Press.
   https://doi.org/10.4324/9780203838020
    Google Scholar
 * Gil de Zúñiga, H. G., Jung, N., & Valenzuela, S. (2012). Social media use for
   news and individuals’ social capital, civic engagement and political
   participation. Journal of Computer-Mediated Communication, 17(3), 319–336.
   https://doi.org/10.1111/j.1083-6101.2012.01574.x
    Web of Science ®Google Scholar
 * Gunther, A. C. (1998). The persuasive press inference: Effects of mass media
   on perceived public opinion. Communication Research, 25(5), 486–504.
   https://doi.org/10.1177/009365098025005002
    Web of Science ®Google Scholar
 * Gunther, A. C., Bolt, D. M., Borzekowski, D. L., Liebhart, J. L., & Dillard,
   J. P. (2006). Presumed influence on peer norms: How mass media indirectly
   affect adolescent smoking. Journal of Communication, 56(1), 52–68.
   https://doi.org/10.1111/j.1460-2466.2006.00002.x
    Web of Science ®Google Scholar
 * Gunther, A. C., & Storey, J. D. (2003). The influence of presumed influence.
   Journal of Communication, 53(2), 199–215.
   https://doi.org/10.1111/j.1460-2466.2003.tb02586.x
    Web of Science ®Google Scholar
 * Ho, S. S., Chuah, A. S. F., Koh, E. L. Q., Ong, L., & Kwan, V. Q. Y. (2022).
   Understanding public willingness to pay more for plant-based meat:
   Environmental and health consciousness as precursors to the influence of
   presumed media influence model. Environmental Communication, 16(4), 520–534.
   https://doi.org/10.1080/17524032.2022.2051576
    Google Scholar
 * Ho, S. S., Goh, T. J., Chuah, A. S. F., Leung, Y. W., Bekalu, M. A., &
   Viswanath, K. (2020). Past debates, fresh impact on nano-enabled food: A
   multigroup comparison of presumed media influence model based on spillover
   effects of attitude toward genetically modified food. Journal of
   Communication, 70(4), 598–621. https://doi.org/10.1093/joc/jqaa019
    Web of Science ®Google Scholar
 * Ho, S. S., Poorisat, T., Neo, R. L., & Detenber, B. H. (2014). Examining how
   presumed media influence affects social norms and adolescents’ attitudes and
   drinking behavior intentions in rural Thailand. Journal of Health
   Communication, 19(3), 282–302. https://doi.org/10.1080/10810730.2013.811329
    PubMed Web of Science ®Google Scholar
 * Hong, Y. (2021). Extending the influence of presumed influence hypothesis:
   Information seeking and prosocial behaviors for HIV prevention. Health
   Communication, 38(4), 765–778. https://doi.org/10.1080/10410236.2021.1975902
    PubMed Web of Science ®Google Scholar
 * Huh, J., & Langteau, R. (2007). Presumed influence of DTC prescription drug
   advertising: Do experts and novices think differently? Communication
   Research, 34(1), 25–52. https://doi.org/10.1177/0093650206296080
    Web of Science ®Google Scholar
 * Jang, K., & Park, N. (2018). The effects of repetitive information
   communication through multiple channels on prevention behavior during the
   2015 MERS outbreak in South Korea. Journal of Health Communication, 23(7),
   670–678. https://doi.org/10.1080/10810730.2018.1501440
    PubMed Web of Science ®Google Scholar
 * Katz, E., & Lazarsfeld, P. F. (1955). Personal influence: The part played by
   people in the flow of mass communications. Free Press.
    Google Scholar
 * Kemp, S. (2023). Digital 2023 global overview report. We Are Social &
   Meltwater.
   https://wearesocial.com/wp-content/uploads/2023/03/Digital-2023-Global-Overview-Report.pdf
    Google Scholar
 * Koriat, A., & Levy-Sadot, R. (1999). Processes underlying metacognitive
   judgments: Information-based and experience-based monitoring of one’s own
   knowledge. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social
   psychology (pp. 483–502). Guilford Publications.
    Google Scholar
 * Krcmar, M. (2009). Individual differences in media effects. In R. L. Nabi &
   M. B. Oliver (Eds.), The SAGE handbook of processes and media effects (1st
   ed., pp. 237–250). Sage.
    Google Scholar
 * Lang, K., & Lang, G. E. (2009). Mass society, mass culture, and mass
   communication: The meanings of mass. International Journal of Communication,
   3(20), 998–1024. https://ijoc.org/index.php/ijoc/article/view/597/407
    Google Scholar
 * Loukides, M. (2022, March 31). AI adoption in the enterprise 2022. O’Reilly
   Media. https://www.oreilly.com/radar/ai-adoption-in-the-enterprise-2022/
    Google Scholar
 * McLeod, D. M., Eveland, W. P., & Nathanson, A. I. (1997). Support for
   censorship of violent and misogynic rap lyrics. Communication Research,
   24(2), 153–174. https://doi.org/10.1177/009365097024002003
    Web of Science ®Google Scholar
 * Napoli, P. M. (2010). Revisiting ‘mass communication’ and the ‘work’ of the
   audience in the new media environment. Media Culture & Society, 32(3),
   505–516. https://doi.org/10.1177/0163443710361658
    Web of Science ®Google Scholar
 * Nelson, T. O., & Narens, L. (1990). Metamemory: A theoretical framework and
   new findings. In G. Bower (Ed.), The psychology of learning and motivation:
   Advances in research and theory (pp. 125–173). Academic Press.
   https://doi.org/10.1016/S0079-7421(08)60053-5
    Google Scholar
 * Park, C. W., Gardner, M. P., & Thukral, V. K. (1988). Self-perceived
   knowledge: Some effects on information processing for a choice task. The
   American Journal of Psychology, 101(3), 401–424.
   https://doi.org/10.2307/1423087
    Web of Science ®Google Scholar
 * Potter, W. J. (2009). Conceptualizing the audience. In R. L. Nabi & M. B.
   Oliver (Eds.), The SAGE handbook of media processes and effects (1st ed., pp.
   19–34). Sage.
    Google Scholar
 * Radecki, C. M., & Jaccard, J. (1995). Perceptions of knowledge, actual
   knowledge, and information search behavior. Journal of Experimental Social
   Psychology, 31(2), 107–138. https://doi.org/10.1006/jesp.1995.1006
    Web of Science ®Google Scholar
 * Raju, P. S., Lonial, S. C., & Glynn Mangold, W. (1995). Differential effects
   of subjective knowledge, objective knowledge, and usage experience on
   decision making: An exploratory investigation. Journal of Consumer
   Psychology, 4(2), 153–180. https://doi.org/10.1207/s15327663jcp0402_04
    Google Scholar
 * Rice, R. E. (2009). Diffusion of innovations: Theoretical extensions. In R.
   L. Nabi & M. B. Oliver (Eds.), Handbook of media effects (pp. 489–503). Sage.
    Google Scholar
 * Rimal, R. N., & Lapinski, M. K. (2015). A re-explication of social norms, ten
   years later. Communication Theory, 25(4), 393–409.
   https://doi.org/10.1111/comt.12080
    Web of Science ®Google Scholar
 * Rogers, E. M. (1995). Diffusion of innovations: Modifications of a model for
   telecommunications. In M.-W. Stoetzer & A. Mahler (Eds.), Die Diffusion von
   Innovationen in der Telekommunikation (pp. 25–38). Springer.
   https://doi.org/10.1007/978-3-642-79868-9_2
    Google Scholar
 * Roland Berger. (2020). Artificial intelligence quiz - test your knowledge.
   https://www.rolandberger.com/en/Insights/Publications/The-Artificial-Intelligence-Quiz-Test-Your-Knowledge.html
    Google Scholar
 * Schäfer, S. (2020). Illusion of knowledge through Facebook news? Effects of
   snack news in a news feed on perceived knowledge, attitude strength, and
   willingness for discussions. Computers in Human Behavior, 103, 1–12.
   https://doi.org/10.1016/j.chb.2019.08.031
    Web of Science ®Google Scholar
 * Schepers, J., & Wetzels, M. (2007). A meta-analysis of the technology
   acceptance model: Investigating subjective norm and moderation effects.
   Information & Management, 44(1), 90–103.
   https://doi.org/10.1016/j.im.2006.10.007
    Web of Science ®Google Scholar
 * Shi, J., Chen, L., & Tsang, S. J. (2022). Integrating interpersonal
   communication into the influence of presumed media influence model:
   Understanding intentions to censor and correct COVID-19 misinformation on
   social media. Journal of Broadcasting & Electronic Media, 66(3), 464–483.
   https://doi.org/10.1080/08838151.2022.2109638
    Web of Science ®Google Scholar
 * Southwell, B. G., & Yzer, M. C. (2007). The roles of interpersonal
   communication in mass media campaigns. Annals of the International
   Communication Association, 31(1), 420–462.
   https://doi.org/10.1080/23808985.2007.11679072
    Google Scholar
 * Sun, Y. (2013). When presumed influence turns real: An indirect route of
   media influence. In The Sage handbook of persuasion: Developments of theory
   and practice (2nd ed., pp. 371–387). Sage.
   https://doi.org/10.4135/9781452218410.n22
    Google Scholar
 * Szczepański, M. (2019). Economic impacts of artificial intelligence (AI) (PE
   637.967). European parliament.
   https://www.europarl.europa.eu/thinktank/en/document/EPRS_BRI(2019)637967
    Google Scholar
 * Valkenburg, P. M., Peter, J., & Walther, J. B. (2016). Media effects: Theory
   and research. Annual Review of Psychology, 67(1), 315–338.
   https://doi.org/10.1146/annurev-psych-122414-033608
    PubMed Web of Science ®Google Scholar
 * Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the
   technology acceptance model: Four longitudinal field studies. Management
   Science, 46(2), 186–204. https://doi.org/10.1287/mnsc.46.2.186.11926
    Web of Science ®Google Scholar
 * Yang, J., & Tian, Y. (2021). “Others are more vulnerable to fake news than I
   am”: Third-person effect of COVID-19 fake news on social media users.
   Computers in Human Behavior, 125, 106950.
   https://doi.org/10.1016/j.chb.2021.106950
    PubMed Web of Science ®Google Scholar
 * Zhang, B., & Dafoe, A. (2019). Artificial intelligence: American attitudes
   and trends. Social Science Research Network.
   https://doi.org/10.2139/ssrn.3312874
    PubMedGoogle Scholar


Download PDF
 * Share icon
   X Facebook LinkedIn Email Share
   
 * Back to Top




RELATED RESEARCH

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI
driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.

 * People also read
 * Recommended articles
 * Cited by

Integrating Interpersonal Communication into the Influence of Presumed Media
Influence Model: Understanding Intentions to Censor and Correct COVID-19
Misinformation on Social Media
Jingyuan Shi et al.
Journal of Broadcasting & Electronic Media
Published online: 8 Aug 2022
Examining public perceptions of cultivated meat in Singapore: food neophobia and
neophilia as precursors to the influence of presumed media influence model
Shirley S. Ho et al.
Asian Journal of Communication
Published online: 23 Apr 2024
Understanding Public Willingness to Pay More for Plant-based Meat: Environmental
and Health Consciousness as Precursors to the Influence of Presumed Media
Influence Model
Shirley S. Ho et al.
Environmental Communication
Published online: 4 Apr 2022
View more



TO CITE THIS ARTICLE:

Reference style: APA Chicago Harvard

Citation copied to clipboard
Copy citation to clipboard
Reference styles above use APA (6th edition), Chicago (16th edition) & Harvard
(10th edition)


DOWNLOAD CITATION

Download a citation file in RIS format that can be imported by citation
management software including EndNote, ProCite, RefWorks and Reference Manager.
Choose format: RIS BibTex RefWorks Direct Export
Choose options: Citation Citation & abstract Citation & references
Download citations


YOUR DOWNLOAD IS NOW IN PROGRESS AND YOU MAY CLOSE THIS WINDOW

Did you know that with a free Taylor & Francis Online account you can gain
access to the following benefits?
 * Choose new content alerts to be informed about new research of interest to
   you
 * Easy remote access to your institution's subscriptions on any device, from
   any location
 * Save your searches and schedule alerts to send you new results
 * Export your search results into a .csv file to support your research

Have an account?
Login now Don't have an account?
Register for free




LOGIN OR REGISTER TO ACCESS THIS FEATURE

Have an account?
Login now Don't have an account?
Register for free
Register a free Taylor & Francis Online account today to boost your research and
gain these benefits:
 * Choose new content alerts to be informed about new research of interest to
   you
 * Easy remote access to your institution's subscriptions on any device, from
   any location
 * Save your searches and schedule alerts to send you new results
 * Export your search results into a .csv file to support your research

Register now or learn more


INFORMATION FOR

 * Authors
 * R&D professionals
 * Editors
 * Librarians
 * Societies


OPEN ACCESS

 * Overview
 * Open journals
 * Open Select
 * Dove Medical Press
 * F1000Research


OPPORTUNITIES

 * Reprints and e-prints
 * Advertising solutions
 * Accelerated publication
 * Corporate access solutions


HELP AND INFORMATION

 * Help and contact
 * Newsroom
 * All journals
 * Books


KEEP UP TO DATE

Register to receive personalised research and resources by email
Sign me up
Taylor and Francis Group Facebook page
Taylor and Francis Group X Twitter page
Taylor and Francis Group Linkedin page

Taylor and Francis Group Youtube page
Taylor and Francis Group Weibo page



Copyright © 2024Informa UK Limited Privacy policy Cookies Terms & conditions
Accessibility

Registered in England & Wales No. 3099067
5 Howick Place | London | SW1P 1WG


×

View all notes




Cookies Button


ABOUT COOKIES ON THIS SITE

We and our partners use cookies to enhance your website experience, learn how
our site is used, offer personalised features, measure the effectiveness of our
services, and tailor content and ads to your interests while you navigate on the
web or interact with us across devices. By clicking "Continue" or continuing to
browse our site you are agreeing to our and our partners use of cookies. For
more information seePrivacy Policy
CONTINUE




COOKIE POLICY

When you visit any website, it may store or retrieve information on your
browser, mostly in the form of cookies. This information might be about you,
your preferences or your device and is mostly used to make the site work as you
expect it to. The information does not usually directly identify you, but it can
give you a more personalized web experience. Because we respect your right to
privacy, you can choose not to allow some types of cookies. Click on the
different category headings to find out more and change our default settings.
However, blocking some types of cookies may impact your experience of the site
and the services we are able to offer.
More information
Allow All


MANAGE CONSENT PREFERENCES

STRICTLY NECESSARY COOKIES

Always Active

These cookies are necessary for the website to function and cannot be switched
off in our systems. They are usually only set in response to actions made by you
which amount to a request for services, such as setting your privacy
preferences, logging in or filling in forms.    You can set your browser to
block or alert you about these cookies, but some parts of the site will not then
work. These cookies do not store any personally identifiable information.

PERFORMANCE COOKIES

Always Active

These cookies allow us to count visits and traffic sources so we can measure and
improve the performance of our site. They help us to know which pages are the
most and least popular and see how visitors move around the site.    All
information these cookies collect is aggregated and therefore anonymous. If you
do not allow these cookies we will not know when you have visited our site, and
will not be able to monitor its performance.

FUNCTIONAL COOKIES

Always Active

These cookies enable the website to provide enhanced functionality and
personalisation. They may be set by us or by third party providers whose
services we have added to our pages.    If you do not allow these cookies then
some or all of these services may not function properly.

TARGETING COOKIES

Always Active

These cookies may be set through our site by our advertising partners. They may
be used by those companies to build a profile of your interests and show you
relevant adverts on other sites.    They do not store directly personal
information, but are based on uniquely identifying your browser and internet
device. If you do not allow these cookies, you will experience less targeted
advertising.

Back Button


COOKIE LIST



Search Icon
Filter Icon

Clear
checkbox label label
Apply Cancel
Consent Leg.Interest
checkbox label label
checkbox label label
checkbox label label

Confirm My Choices

✓
Thanks for sharing!
AddToAny
More…