Association for Information Systems

AIS Electronic Library (AISeL)

ECIS 2022 Research-in-Progress Papers ECIS 2022 Proceedings

6-18-2022

Trust Me, I’m an Influencer! - A Comparison of Perceived Trust in

Human and Virtual Influencers
Lennart Hofeditz
University of Duisburg-Essen, lennart.hofeditz@uni-due.de

Anika Nissen
Institute for Computer Science and Business Information Systems, anika.nissen@icb.uni-due.de

Reinhard Schütte
Institute for Computer Science and Business Information Systems, reinhard.schuette@icb.uni-due.de

Milad Mirbabaie
Paderborn University, milad.mirbabaie@uni-paderborn.de

Follow this and additional works at: https://aisel.aisnet.org/ecis2022_rip

Recommended Citation
Hofeditz, Lennart; Nissen, Anika; Schütte, Reinhard; and Mirbabaie, Milad, "Trust Me, I’m an Influencer! - A
Comparison of Perceived Trust in Human and Virtual Influencers" (2022). ECIS 2022 Research-in-Progress
Papers. 27.
https://aisel.aisnet.org/ecis2022_rip/27

This material is brought to you by the ECIS 2022 Proceedings at AIS Electronic Library (AISeL). It has been
accepted for inclusion in ECIS 2022 Research-in-Progress Papers by an authorized administrator of AIS Electronic
Library (AISeL). For more information, please contact elibrary@aisnet.org.
 TRUST ME, I’M AN INFLUENCER! - A COMPARISON OF
PERCEIVED TRUST IN HUMAN AND VIRTUAL
INFLUENCERS

Research in Progress

Lennart Hofeditz, University of Duisburg-Essen, Duisburg, Germany, lennart.hofeditz@uni-

due.de
Anika Nissen, University of Duisburg-Essen, Essen, Germany, anika.nissen@uni-due.de
Reinhard Schütte, University of Duisburg-Essen, Essen, Germany, reinhard.schuette@uni-
due.de
Milad Mirbabaie, Paderborn University, Paderborn, Germany, milad.mirbabaie@uni-
paderborn.de

Abstract
Influencers in social media are often perceived as a trusted source for many people which is why
companies increasingly promote their products through them. However, influencers can also cause
reputational damage for a brand. Virtual (computer-generated) influencers can be used to minimize
these risks and to better tailor content to a target group of a company. As trust is one success factor of
online marketing, we examine differences in the perception of trust in human and virtual influencers. In
a first online survey study, we presented N = 112 participants the content of human and virtual
influencers, published on Instagram. Preliminary findings reveal that although participants were often
unsure whether the presented influencer was human or computer-generated, perceived trust, social
presence, and humanness was consistently rated higher for human influencers. To gain deeper insights
into potential, unconscious decision conflicts which can determine trust evaluations, a follow-up
neuroimaging study is discussed.

Keywords: Virtual Influencer, Digital Influencer, Trust, Social Media, Transparency, NeuroIS.

1 Motivation
Miquela is a 19-year-old influencer living in Los Angeles with more than 3 million subscribers on
Instagram and more than 3.2 million followers on TikTok. She loves fashion and shows solidarity with
the Black Lives Matter movement. In 2018, TIME magazine named her as one of the 25 most influential
people in the Internet. However, Miquela is not a human being. She is a computer-generated, virtual
influencer which is described as 19-year-old robot in the profile description (Robinson, 2020).
Influencers are mostly defined as individuals with a significant number of social media followers
(Kadekova & Holienciova, 2018). For influencers, trust and authenticity are important elements to
generate and retain followers in order to be visible to a certain target group (Batista & Chimenti, 2021).
Virtual influencers are most often described as computer-generated, animated characters controlled by
a startup or an agency with their own social media accounts and a large number of followers. They are
used to promote products and to substitute human influencers in social commerce. Virtual influencers
inspire millions of users in social networks and are engaged by companies to market their products
(Moustakas et al., 2020). Some virtual humanoid robots (another synonym for virtual influencers) are
used by organizations to support campaigns for global brands such as Prada, Samsung, Porsche,

Thirtieth European Conference on Information Systems (ECIS 2022), Timișoara, Romania 1

Hofeditz et al. / Trust me, I’m an Influencer

Unilever, Calvin Klein, Ikea, and KFC (Batista & Chimenti, 2021). However, just like human
influencers, they not only promote products, but also endorse political messages, such as Black Lives
Matter or the support for transgender rights (Robinson, 2020).
Research on virtual influencers in general is still rather in its infancy. In information systems (IS)
research, only a hand full of studies exists that examined this new phenomenon of virtual influencers or
related topics which all claim for more in-depth IS research on this phenomenon (Arsenyan & Mirowska,
2021; Batista & Chimenti, 2021; Breves, 2021; Robinson, 2020; Seymour, Riemer, et al., 2021;
Seymour, Yuan, et al., 2021). Virtual Influencers raise ethical issues such as a decreased transparency
about who is responsible for the content and whose and which moral values are being espoused
(Robinson, 2020). As influencers are increasingly used as a highly effective tool by brands and
organizations to market products and influence the lives of adolescents’ and adults, it becomes necessary
to examine how virtual influencers are perceived in comparison to human influencers. Batista and
Chimenti (2021) conducted a systematic literature review and additional interviews with experts in
social media and influencer marketing in order to understand how virtual influencers affect online
marketing communication. They identified factors how virtual influencers can facilitate management
decisions and derived guidelines for organizations based on the success factors scalability,
anthropomorphism, authenticity, controllability, and attractiveness. Previous studies indicated that
virtual influencers may lack authenticity and reliability in comparison to human influencers (Ingham,
2018). Therefore, to provide a starting point and to address ethical issues, the question of how and
whether people trust virtual influencers needs to be addressed first. Trust is a complex concept, which
follows different premises depending on the context and discipline in which it is discussed (Rousseau
et al., 1998). Trust is a consciously chosen state, a relationship and cooperation between two parties
where one expects the best possible from the other party, although one is not sure about it (Dunn, 2000).
Recent IS research in the context of computer-generated media content revealed that more disclosure
and transparency on the creator of social media content does not necessarily result in an increase in
credibility and trust in the a brand or the associated organization (Hofeditz et al., 2021). However, trust
can mediate the effect of media content on the purchase intention (Ganguly et al., 2009), and
trustworthiness is often being considered an overarching ethical principle (Shneiderman, 2020). In
sociology, trust is initially intended solely as a concept between two humans (Robbins, 2016). By means
of certain mechanisms, this trust towards persons and their products can then also turn into trust towards
institutions and their products. In the case of virtual influencers, who are not actually people, it has not
yet been researched how trust in virtual influencers and their products can develop into trust in the
promoted institutions and their products. As virtual influencers are an emerging phenomenon with a
potentially high impact on societies and organizations, the role of trust and transparency therefore needs
to be newly discussed in this context. We therefore derived the following research question:
RQ: To what extend does perceived trust in virtual influencers differ from trust in human influencers?
By addressing this research question, we contribute to IS research by providing knowledge on how trust
needs to be considered when examining, designing or applying virtual influencers for promoting a
message, a brand or a product. We present the results of a first online survey study that provide insights
on how the perception of virtual influencers by social media users is intertwined with trust and
transparency. Our first findings indicate that the majority of users cannot distinguish between human
and virtual influencers. These results are used to derive a follow-up neuroimaging study to receive
deeper insights into and conditions of trust towards influencers and how these differ between virtual and
human influencer. Having seen that there are differences in trust evaluations, a neuroimaging study is a
critical next step to gain deeper insights into the context of the decision of trust paired with transparency
in virtual influencers.

2 Related Work and Hypotheses

As the first well-known virtual influencer, Miquela launched in 2016 (Tiffany, 2019). Quickly, the first
research articles on virtual influencers appeared in communication focused journals (Kadekova &
Holienciova, 2018). Until that point, research had only examined the phenomenon of digital (human)

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 2

Hofeditz et al. / Trust me, I’m an Influencer

influencers. Digital influencers “have an effect on the members of particular communities gathered
around similar interests” (Uzunoǧlu & Misci Kip, 2014, p. 1). Furthermore, they have a high number of
followers on their social media accounts, resulting in a significant degree of power over others (Wang
et al., 2020). They are humans, influencing behavioral intention regarding promoted brands, and
generate engagement for companies (Jiménez-Castillo & Sánchez-Fernández, 2019). One important
characteristic is their authenticity and their identity which make their opinions respected and trusted by
their followers (Hu et al., 2020). Although, they are digital influencers, it is possible to meet them in
person and their human character can sometimes lead to political incorrectness or boycotts of brands or
companies. Such negative consequences of working with a human influencer are less likely when
working with a virtual influencer as their content can be better tailored to a certain target group
(Robinson, 2020). Virtual influencers are defined as humanoid robots (Batista & Chimenti, 2021) that
are computer-generated with a large amount of followers on Instagram (Moustakas et al., 2020). As they
are less likely to be involved in scandals or harm advertisers and partners due to the high amount of
control (Robinson, 2020), this raises ethical and societal questions about potential misuse by brands or
organizations.
Although virtual influencers can have a high impact on social media marketing (Batista & Chimenti,
2021), the challenges accompanied with the rise of virtual influencers are not well addressed by IS
scholars. To generate virtual influencers, neural rendering, as a novel technology for generating and
animating human like faces can be used (Seymour, Riemer, et al., 2021). One use case for neural
rendering is also the creation of deep fakes. With creating artificial, digital human artefacts such as
virtual influencers, questions regarding the achieved anthropomorphism of the created artefact and
therewith, its uncanniness quickly arise (Seymour, Yuan, et al., 2021).
To avoid uncanniness and a negative perception of virtual influencer and thereby possibly also of a
promoted brand, trust is especially relevant because it constitutes dimensions of credibility and
benevolence (Benbasat et al., 2010; Gefen & Straub, 1997; Mcknight et al., 2011). Trust is tied to the
relationship between trustee and trustor. Trust gives discretion, and despite uncertainty and
vulnerability, the trusting person (trustor) needs to assume that the trusted person (trustee) is guided by
his or her goodwill, or at least not by malice (Dunn, 2000). The perceived trustworthiness of the
influencer is further necessary to establish parasocial relationships between influencer and consumer,
resulting in a positive perception of the promoted brands/ products (Duffett et al., 2019; Haobin Ye et
al., 2021; Xiang et al., 2016). Having stated that certainty is a significant component of trust, the origin
and what is behind virtual influencers such as Miquela is often unclear. Paired with potential higher
uncanniness due to high but not perfect human-likeness (Cheetham, 2011; Mathur et al., 2020; Mathur
& Reichling, 2008; Mori et al., 2012), we hypothesize that virtual influencers will be rated lower in trust
as comparative human counterparts:
Hypothesis 1: Human influencers are perceived as more trustworthy than virtual influencers.
By existing on different channels in videos and also in reality, it can also be assumed that human
influencers generate a higher social presence than virtual influencers who are limited to certain
computer-generated content and virtuality (Moustakas et al., 2020). Social presence is a construct which
is closely related to initial trust formation in social media and its operationalization, whose definition,
however, is particularly difficult (Lu et al., 2016; Ogonowski et al., 2014; Short et al., 1976). In
psychology, social presence is closely related to intimacy and psychological closeness (Short et al.,
1976). However, social presence can be defined as the perceived closeness and warmth towards a
computer system but also as the perceived level of awareness of other human users through a computer
system (e.g., in an online community) (Lu et al., 2016). Social presence has already been initially
examined in the context of virtual influencers (Arsenyan & Mirowska, 2021). In our study, we also refer
to social presence as the perceived closeness and warmth towards human and virtual influencers (Gefen
& Straub, 1997) and assume that human influencers achieve a higher level of social presence due to a
higher level of interactivity (e.g., more video content or live interactions with users). We hypothesize:
Hypothesis 2: Human influencers deliver a higher level of social presence than virtual influencers.

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 3

Hofeditz et al. / Trust me, I’m an Influencer

Also, because virtual influencers are tied to specially created computer-generated content and must be
directed by humans, it can be assumed that they are also perceived as less human due to their robotic
attributes (Carpinella et al., 2017). Previous studies also indicated that more human-like virtual
influencers have been retrieved less positive reactions than less human-like virtual influencers
(Arsenyan & Mirowska, 2021). In this paper, we aim to measure this human-likeness with humanness
(Holtgraves & Han, 2007) rather than anthropomorphism. Further studies also indicated that it is already
difficult for some individuals to distinguish between human and virtual influencers (Batista & Chimenti,
2021; Robinson, 2020). However, we hypothesize that the majority of social media users rate human
social media accounts as more human and therefore hypothesize the following:
Hypothesis 3: Human influencers are perceived more human-like than virtual influencers.
These hypotheses will be tested in a research design consisting of two studies. In the following, we will
describe our procedure in the first study, and in section 5, we will outline the second study.

3 Method
In our first study, we conducted an online survey, in which participants were presented with screenshots
of various human and virtual influencers on Instagram. Although virtual influencers are present on
various social media platforms, in this study we considered Instagram as one of the most popular
platforms in Western societies, with a majority focus on images and less on video content. As we plan
to further investigate our findings in a neuroimaging study, images are much easier to experimentally
control and compare.
Study Design and Stimulus Materials. After a short briefing, in which we informed the participants
about the general topic, data processing and information on the course of the study, they were asked to
provide demographic information such as age, gender, educational attainment, and main professional
activity to be able to determine influences of these factors later. In addition, we inquired about the
frequency of Instagram use, whether the participants follow influencers, and which Instagram features
they use regularly. Subsequently, participants were asked to answer a questionnaire on Personal Trust
in Technology (Mcknight et al., 2011). Subsequently, participants were presented with a total of eight
different influencers, four of which were human and four of which were virtual. To achieve better
comparability of human and virtual influencers, we first researched highly influential virtual influencers
with many followers, posts, and high engagement. In doing so, we selected representatives from the four
ethnic groups Asian, (Northern) European, Latin (American), and African. The influencers were
selected by a consensus of four independent coders. The search was conducted via websites that listed
various prominent virtual influencers and searches on Instagram, which were based on the networks of
already well-known virtual influencers (e.g., Miquela). The selection was also based on a clearly
identifiable ethnic group and known similar-looking human influencers. The coders then independently
researched similar-looking human influencers for each of the four virtual influencers and then discussed
to what extent they were comparable. We decided to focus on female influencers to avoid effects of
gender. A list of all influencers’ accounts presented can be found in Table 1.
To mitigate effects of color schemes, we chose influencers who were showed in similar poses and
presented them as mobile Instagram screenshots in black and white. The use of black and white materials
is also of high importance, as different colors could distort our results, especially in the subsequent
neuroimaging study. In addition, we ensured that no products were promoted in the screenshots to avoid
effects such as familiarity with products and brands. All screenshots belong to real Instagram accounts
of real virtual and human influencers (example for an included post see Figure 1).

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 4

Hofeditz et al. / Trust me, I’m an Influencer

Follower count on
Account name Ethnical group Influencer type
March 10th 2022
bermudaisbae (Northern) European Virtual influencer > 276 thousand
dagibee (Northern) European Human influencer > 6.5 million
shudu.gram African Virtual influencer > 226 thousand
anokyai African Human influencer > 709 thousand
zoedvir Latin (American) Virtual influencer > 27 thousand
marinadnery Latin (American) Human influencer > 46 thousand
rozy.gram Asian Virtual influencer > 123 thousand
jenndeugikim Asian Human influencer > 405 thousand

Table 1. Instagram accounts used as stimuli in the online survey

The participants were presented with one post from each influencer in randomized order, after which
they were asked to answer a number of questions for the constructs of trust in influencers (D. Y. Kim &
Kim, 2021), social presence (Gefen & Straub, 1997), and humanness (Holtgraves & Han, 2007).
Furthermore, also social attributes (Carpinella et al., 2017) and uncanniness (Tinwell & Sloan, 2014)
were assessed, but will not be analyzed in the frame of this work in progress. Finally, participants were
asked whether they already knew the influencers and whether they believed that each influencer was a
human being. Because the participants were not told which of the influencers are human and which are
virtual, we debriefed our participants about the real purpose of the study and provided more information
about virtual influencers and contact information. The procedure is summarized in Figure 2.

Figure 1. Exemplary screenshots of a virtual influencer (left) and human influencer (right) presented
in the online survey study

Figure 2. Overview of the procedure in study 1.

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 5

Hofeditz et al. / Trust me, I’m an Influencer

Sample. Overall, we recruited a sample of 163 participants via social media and the clickworker
platform, from which N = 112 filled out the questionnaire completely and have been used for further
analyses. From the sample, 43.8% are female (all remaining are male) and the age ranged from 20 to 66
years (M = 35.5, SD = 11.1). About one fourth of the sample holds an advanced university degree
(master or above) (27.7%), 23.2% hold a bachelor’s degree, 25.9% a general qualification for university
entrance, 10.7% are currently studying without having achieved a degree yet, and the remaining 12.5%
have a secondary or high school diploma. Regarding their social media activity, 3.6% stated that they
do not use Instagram, 11.6% assume they use it several times per month or less, 15.2% use it at least
once a week, and 69.6% state that they use Instagram at least once a day, with the majority using it
several times per day. Questions for the general actions taken on Instagram reveal that the majority of
our sample are passive users (not content-creating). Regarding their interest in influencers, 29.5% of the
sample stated that they do not follow any influencers on Instagram, 64.3% stated that they follow some
(semi-professional) influencers, and only 6.3% stated that they would follow many influencers on
Instagram. As we are concerned with virtual influencers in this paper, we also evaluated the personal
trust in technology of our sample which resulted in an average of M = 5.04 (SD = .75). Given that this
lies above the average of the 7-point Likert scale, it can be assumed that the sample is not too curious
of new technology, but also not overly confident and trusting.

4 First Results
We included a manipulation check for each influencer in which participants had to evaluated whether
the shown influencer was computer-generated or human. In the following Table 2, the part of the sample
who believed that the shown influencer is human is provided. We can see here that at least 50% of the
sample thought of the presented influencer post to be human, even if it was computer-generated.
Consequently, at least half of our participants were not able to distinguish between virtual and human
influencer. The difference between the virtual and human Asian influencer was most diminishing as the
same amount of the sample thought of both as being human. The Friedmann test show that although at
least half of the sample mistake the virtual influencers as humans, the number of right answers for the
human influencers was significantly higher (except for the Asian influencers).
Virtual* Human* c2** p
Latin 63.40% 80.40% 11.8 <.001
African 50% 74.10% 22.3 <.001
European 65.20% 92% 28.1 <.001
Asian 65.20% 65.20% 0.026 .873
*% of sample that believes shown Influencer is a human **Friedmann Test

Table 2. Manipulation Check "Is this influencer human?"

Nevertheless, to test whether participants did evaluate the human and virtual influencers differently
regarding the hypothesized constructs of trust, social presence, and humanness, repeated-measures one-
way ANOVAs were calculated. Results reveal that although a great part of participants could not
correctly distinguish between virtual and human influencer, the virtual influencers were rated
significantly lower in trust (F (1, 111) = 5.63, p = .019, hp2 = .048). Further, the virtual influencers were
also rated significantly lower in their social presence (F (1, 111) = 4.84, p = .03, hp2 = .042), and their
humanness (F (1, 111) = 9.181, p = .003, hp2 = .076).
F p hp2 MD (virtual - human)
Trust 5.63 .019 0.048 -0.123
Social Presence 4.84 .03 0.042 -0.157
Humanness 9.181 .003 0.076 -0.141

Table 3. ANOVA Results

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 6

Hofeditz et al. / Trust me, I’m an Influencer

5 Discussion and Next Steps

Through this first study, we have uncovered new findings that contribute to IS research. Even though a
great share of our participants could not correctly identify whether the virtual influencers were human
or not, and the majority even suggested that they are human, the perceived trust, social presence, and
humanness was still rated significantly lower. As one next step, we will further consider whether our
control variables, such as social media use or participant age, moderate the lower ratings of trust, social
presence, and humanness. Prior research already suggested that these control variables may have an
impact on the credibility of computer-generated content (Hofeditz et al., 2021).
Another finding is that for the human influencers, more participants believed that they are human, but
there was still a small part of the sample that was curious (at least 8%). This indicates that although the
participants perceived the virtual influencers less trustworthy, they were probably unable to determine
exactly why this was the case. However, further research is needed to really verify this. Additionally,
while at least 93% of participants were not familiar with the included influencers, there was an exception
for the European human influencer which was known by 41.1%. This might bias the results of the
manipulation check and therefore, should be considered as a limitation of this work.
In other words, virtual influencers seem to be already at such a high level of anthropomorphism that
they cannot be distinguished from real humans anymore – at least not on social media. This also means
that companies can promote their products and brand in a, potentially, more controlled way because
virtual influencers profiles are often run by businesses with which agreements can be made and which
are less too prone to “human error”. Given their high anthropomorphism, the use of virtual influencers
as online marketing measure seems to generate as much social engagement as that of human influencers.
Therefore, we assume that there lies great potential in the use of virtual influencers in online marketing.
However, as we have stated above, the high anthropomorphism of virtual influencers and the inability
of users to distinguish them from human influencers (if not declared in their profile), raises several
ethical concerns (Arsenyan & Mirowska, 2021; Robinson, 2020). In relation to previous IS research
such as Robinson's (2020) ethical issues of non-transparency and unclear moral responsibility, it is worth
discussing whether virtual influencers should be clearly declared as such? In line with this, and as
suggested by Seymor et al. (2021), it is now important to further explore how perceptions of virtual
influencers change when they are labeled as such. The lack of distinctiveness between virtual and human
influencers may become even more problematic in the future if artificial intelligence is used to further
enhance the richness, individuality, and autonomy of virtual influencers (Batista & Chimenti, 2021).
Therefore, adding self-disclosure as virtual influencer to the profile may not only help to comply with
ethical standards, but it may also maintain the credibility of promoted products and brands. Yet, research
in this field is still scarce and it stands to question how social media users react to virtual influencers
which are declared as such and those that are not. Furthermore, it also stands to question how this
perception further impacts behavioral intentions towards the influencer and the products or brands
promoted by him/her. There is also the question of whether people trust virtual influencers less or more
when they know they are virtual influencers.
Therefore, we plan to conduct a follow-up study in which we aim to investigate these questions. In line
with the prior argumentation, and since some virtual influencers disclose themselves as computer-
generated while others do not, it stands to question in how far transparency might be an impact factor
here. To uncover potential decision biases which might be partly unconscious (as participants could not
even say which influencer was virtual and which not), we plan to conduct a neuroIS study. In the frame
of neuroscience and neuroIS literature, first studies have already investigated differences in cortical,
prefrontal brain activity in response to decision conflicts in fake news (Moravec et al., 2019), as well as
regarding trust evaluations in humans and avatars, or social robots (Cheetham, 2011; R. Riedl et al.,
2014; Rosenthal-von der Pütten et al., 2019; Saygin, A.P. et al., 2010). Across these studies, it can be
identified that most processes relating to social trust evaluations and detection of decision bias happen
in the medial parts of the prefrontal cortex (PFC) (Moravec et al., 2019; Y. Riedl & Beetz, 2019;
Rosenthal-von der Pütten et al., 2019). Therefore, we plan a neuroimaging study in which neural activity
in the PFC is assessed in addition to self-reported measures (questionnaire) to also uncover neural

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 7

Hofeditz et al. / Trust me, I’m an Influencer

responses to virtual influencers compared to human influencers. With the follow-up study, we aim to
measure not only the influence of transparency by providing background information on influencers as
another influencing factor for trust. We also aim to verify and better explain the results of the first study,
which are based on participants' self-reports, by effects of neural activity. We will also contribute to IS
research by combining a traditional IS method, such as an online questionnaire study with a more
atypical method from the field of psychology and neuroIS, namely functional near-infrared spectroscopy
(fNIRS). fNIRS allows us to capture the neural activity on the PFC of participants while being a mobile,
non-invasive, and robust neuroimaging method (H. Y. Kim et al., 2017; Maior et al., 2015; Pinti et al.,
2019, 2020). The great advantage of fNIRS lies in is high mobility while still offering good spatial
localization of observed neural activation (Hirshfield et al., 2009; H. Y. Kim et al., 2017; Pinti et al.,
2019, 2020). With these characteristics, fNIRS is currently the only neuroimaging method that offers
high spatial localization with this level of mobility and non-invasiveness. Needless to say, fNIRS has
already been successfully applied in related IS research and has proven to be a valid method for neuroIS
studies (Gefen et al., 2014; Krampe et al., 2017; Nissen et al., 2019). Conducting such neuroIS study
will guide scholars to better understand the emerging phenomenon of virtual influencers.

References
Arsenyan, J., & Mirowska, A. (2021). Almost human? A comparative case study on the social media
presence of virtual influencers. International Journal of Human Computer Studies, 155(June),
102694. https://doi.org/10.1016/j.ijhcs.2021.102694
Batista, A., & Chimenti, P. (2021). “ Humanized Robots ”: A Proposition of Categories to Understand
Virtual Influencers. Australasian Journal of Information Systems, 25, 1–27.
Benbasat, I., Dimoka, A., Pavlou, P. A., & Qiu, L. (2010). Incorporating social presence in the design
of the anthropomorphic interface of recommendation agents: Insights from an fmri study. 31st
International Conference on Information Systems, 1–22.
http://aisel.aisnet.org/icis2010_submissions/228
Breves, P. (2021). Biased by being there: The persuasive impact of spatial presence on cognitive
processing. Computers in Human Behavior, 119(August 2020), 106723.
https://doi.org/10.1016/j.chb.2021.106723
Carpinella, C. M., Wyman, A. B., Perez, M. A., & Stroessner, S. J. (2017). The Robotic Social Attributes
Scale (RoSAS). Proceedings of the 2017 ACM/IEEE International Conference on Human-Robot
Interaction, 254–262. https://doi.org/10.1145/2909824.3020208
Cheetham, M. (2011). The human likeness dimension of the “uncanny valley hypothesis”: behavioral
and functional MRI findings. Frontiers in Human Neuroscience, 5(November), 1–14.
https://doi.org/10.3389/fnhum.2011.00126
Duffett, R., Petroşanu, D. M., Negricea, I. C., & Edu, T. (2019). Effect of YouTube marketing
communication on converting brand liking into preference among millennials regarding brands in
general and sustainable offers in particular. Evidence from South Africa and Romania.
Sustainability (Switzerland), 11(3), 1–24. https://doi.org/10.3390/su11030604
Dunn, J. (2000). Trust and Political Agency. In D. Gambetta (Ed.), Trust: Making and Breaking
Cooperative Relations (electronic, pp. 73–93). Department of Sociology, University of Oxford.
https://doi.org/10.1515/9781400861453.26
Ganguly, B., Dash, S. B., & Cyr, D. (2009). Website characteristics, Trust and purchase intention in
online stores: -An Empirical study in the Indian context. Journal of Information Science and
Technology, 6(2).
Gefen, D., Ayaz, H., & Onaral, B. (2014). Applying Functional Near Infrared (fNIR) Spectroscopy to
Enhance MIS Research. AIS Transactions on Human-Computer Interaction, 6(3), 55–73.
Gefen, D., & Straub, D. W. (1997). Gender differences in the perception and use of e-mail: An extension

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 8

Hofeditz et al. / Trust me, I’m an Influencer

to the technology acceptance model. MIS Quarterly: Management Information Systems, 21(4),
389–400. https://doi.org/10.2307/249720
Haobin Ye, B., Fong, L. H. N., & Luo, J. M. (2021). Parasocial interaction on tourism companies’ social
media sites: antecedents and consequences. Current Issues in Tourism, 24(8), 1093–1108.
https://doi.org/10.1080/13683500.2020.1764915
Hirshfield, L. M., Chauncey, K., Gulotta, R., Girouard, A., Solovey, E. T., Jacob, R. J. K., Sassaroli, A.,
& Fantini, S. (2009). Combining Electroencephalograph and Near Infrared Spectroscopy to
Explore Users’ Instantaneous and Continuous Mental Workload States. HCI International 2009
13th International Conference on Human-Computer Interaction, 239–247.
http://www.cs.tufts.edu/~jacob/papers/hciintl.09.pdf
Hofeditz, L., Mirbabaie, Mi., Stieglitz, S., & Holstein, J. (2021). Do You Trust An AI-Journalist? A
Credibility Analysis Of News Content With AI-Authorship. European Conference on Information
Systems (ECIS 2021). Marrakech, Morocco.
Holtgraves, T., & Han, T. L. (2007). A procedure for studying online conversational processing using a
chat bot. Behavior Research Methods, 39(1), 156–163. https://doi.org/10.3758/BF03192855
Hu, L., Min, Q., Han, S., & Liu, Z. (2020). Understanding followers’ stickiness to digital influencers:
The effect of psychological responses. International Journal of Information Management, 54(May
2019). https://doi.org/10.1016/j.ijinfomgt.2020.102169
Ingham, L. (2018). Virtual influencers: Lil Miquela and Shudu Gram are the vanguard of online
marketing’s future. Verdict. https://www.verdict.co.uk/virtual-influencers-lil-miquela-shudu-
gram/ (Retrieved March 23, 2022)
Jiménez-Castillo, D., & Sánchez-Fernández, R. (2019). The role of digital influencers in brand
recommendation: Examining their impact on engagement, expected value and purchase intention.
International Journal of Information Management, 49(February), 366–376.
https://doi.org/10.1016/j.ijinfomgt.2019.07.009
Kadekova, Z., & Holienciova, M. (2018). Influencer Marketing As a Modern Phenomenon Creating a
New Frontier of Virtual Opportunities. Communication Today, 9(2), 90–105. https://www.
Kim, D. Y., & Kim, H. Y. (2021). Trust me, trust me not: A nuanced view of influencer marketing on
social media. Journal of Business Research, 134(November 2019), 223–232.
https://doi.org/10.1016/j.jbusres.2021.05.024
Kim, H. Y., Seo, K., Jeon, H. J., Lee, U., & Lee, H. (2017). Application of functional near-infrared
spectroscopy to the study of brain function in humans and animal models. Molecules and Cells,
40(8), 523–532. https://doi.org/10.14348/molcells.2017.0153
Krampe, C., Gier, N., & Kenning, P. (2017). Beyond traditional neuroimaging: Can mobile fNIRS add
to NeuroIS? Lecture Notes in Information Systems and Organisation, 25, 151–157.
https://doi.org/10.1007/978-3-319-67431-5_17
Lu, B., Fan, W., & Zhou, M. (2016). Social presence, trust, and social commerce purchase intention: An
empirical research. Computers in Human Behavior, 56, 225–237.
https://doi.org/10.1016/j.chb.2015.11.057
Maior, H. A., Pike, M., Sharples, S., & Wilson, M. L. (2015). Examining the reliability of using fNIRS
in realistic HCI settings for spatial and verbal tasks. Conference on Human Factors in Computing
Systems - Proceedings, 3039–3042. https://doi.org/10.1145/2702123.2702315
Mathur, M. B., & Reichling, D. B. (2008). An uncanny game of trust: Social trustworthiness of robots
inferred from subtle anthropomorphic facial cues. Proceedings of the 4th ACM/IEEE International
Conference on Human-Robot Interaction, HRI’09, 313–314.
https://doi.org/10.1145/1514095.1514192
Mathur, M. B., Reichling, D. B., Lunardini, F., Geminiani, A., Antonietti, A., Ruijten, P. A. M., Levitan,
C. A., Nave, G., Manfredi, D., Bessette-Symons, B., Szuts, A., & Aczel, B. (2020). Uncanny but
not confusing: Multisite study of perceptual category confusion in the Uncanny Valley. Computers

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 9

Hofeditz et al. / Trust me, I’m an Influencer

in Human Behavior, 103(June 2019), 21–30. https://doi.org/10.1016/j.chb.2019.08.029

Mcknight, D. H., Carter, M., Thatcher, J. B., & Clay, P. F. (2011). Trust in a specific technology: An
investigation of its components and measures. ACM Transactions on Management Information
Systems, 2(2). https://doi.org/10.1145/1985347.1985353
Moravec, P. L., Minas, R. K., & Dennis, A. R. (2019). Fake News on Social Media: People Believe
What They Want to Believe When It Makes No Sense at All. MIS Quarterly: Management
Information Systems, 43(4), 1343–1360.
Mori, M., MacDorman, K. F., & Kageki, N. (2012). The Uncanny Valley. IEEE Robotics & Automation
Magazine, 19(2), 98–100. https://doi.org/10.1109/MRA.2012.2192811
Moustakas, E., Lamba, N., Mahmoud, D., & Ranganathan, C. (2020). Blurring lines between fiction and
reality: Perspectives of experts on marketing effectiveness of virtual influencers. International
Conference on Cyber Security and Protection of Digital Services, Cyber Security 2020.
https://doi.org/10.1109/CyberSecurity49315.2020.9138861
Nissen, A., Krampe, C., Kenning, P., & Schütte, R. (2019). Utilizing Mobile FNIRS to Investigate
Neural Correlates of the TAM in ECommerce. International Conference on Information Systems
(ICIS), 1–9.
Ogonowski, A., Montandon, A., Botha, E., & Reyneke, M. (2014). Should new online stores invest in
social presence elements? The effect of social presence on initial trust formation. Journal of
Retailing and Consumer Services, 21(4), 482–491.
https://doi.org/10.1016/j.jretconser.2014.03.004
Pinti, P., Scholkmann, F., Hamilton, A., Burgess, P., & Tachtsidis, I. (2019). Current Status and Issues
Regarding Pre-processing of fNIRS Neuroimaging Data: An Investigation of Diverse Signal
Filtering Methods Within a General Linear Model Framework. Frontiers in Human Neuroscience,
12. https://doi.org/10.3389/fnhum.2018.00505
Pinti, P., Tachtsidis, I., Hamilton, A., Hirsch, J., Aichelburg, C., Gilbert, S., & Burgess, P. W. (2020).
The present and future use of functional near-infrared spectroscopy (Fnirs) for cognitive
neuroscience. Annals of the New York Academy of Sciences, 1464(1), 5–29.
https://doi.org/10.1111/nyas.13948
Riedl, R., Mohr, P., Kenning, P., Davis, F., & Heekeren, H. (2014). Trusting humans and avatars: A
brain imaging study based on evolution theory. Journal of Management Information Systems,
30(4), 83–114. https://doi.org/10.2753/MIS0742-1222300404
Riedl, Y., & Beetz, K. R. (2019). Robotic Process Automation : Developing a Multi-Criteria Evaluation
Model for the Selection of Automatable Business Processes. Twenty-Fifth Americas Conference
on Information Systems, 1–10.
Robbins, B. G. (2016). From the general to the specific: How social trust motivates relational trust.
Social Science Research, 55, 16–30. https://doi.org/10.1016/j.ssresearch.2015.09.004
Robinson, B. (2020). Towards an ontology and ethics of virtual influencers. Australasian Journal of
Information Systems, 24, 1–8. https://doi.org/10.3127/AJIS.V24I0.2807
Rosenthal-von der Pütten, A. M., Krämer, N. C., Maderwald, S., Matthias Brand, & Grabenhorst, F.
(2019). Neural Mechanisms for Accepting and Rejecting Artificial Social Partners in the Uncanny
Valley. J. Neurosci, 2956(18). https://doi.org/10.1523/JNEUROSCI.2956-18.2019
Rousseau, D. M., Sitkin, S., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross
discipline view of trust. Academy of Management Review, 23(3), 393–404.
Saygin, A.P., Chaminade, T., & Ishiguro, H. (2010). The Perception of Humans and Robots: Uncanny
Hills in Parietal Cortex. Proceedings of the 32nd Annual Conference of the Cognitive Science
Society (Pp. 2716-2720). Austin, TX: Cognitive Science Society.
Seymour, M., Riemer, K., Yuan, L., & Dennis, A. R. (2021). Beyond deep fakes: Conceptual framework
and research agenda for neural rendering of realistic digital faces. Proceedings of the 54rd Hawaii
International Conference on System Sciences, 4859–4868.

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 10

Hofeditz et al. / Trust me, I’m an Influencer

Seymour, M., Yuan, L., Dennis, A. R., & Riemer, K. (2021). Have we crossed the uncanny valley?
Understanding affinity, trustworthiness, and preference for realistic digital humans in immersive
environments. Journal of the Association for Information Systems, 22(3), 591–617.
https://doi.org/10.17705/1jais.00674
Shneiderman, B. (2020). Bridging the gap between ethics and practice: Guidelines for reliable, safe, and
trustworthy human-centered AI systems. ACM Transactions on Interactive Intelligent Systems,
10(4). https://doi.org/10.1145/3419764
Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunications. John Wiley
& Sons Ltd.
Tiffany, K. (2019). Lil Miquela and the virtual influencer hype, explained. VOX.Com.
https://www.vox.com/the-goods/2019/6/3/18647626/instagram-virtual-influencers-lil-miquela-ai-
startups (Retrieved March 23, 2022)
Tinwell, A., & Sloan, R. J. S. (2014). Children’s perception of uncanny human-like virtual characters.
Computers in Human Behavior, 36, 286–296. https://doi.org/10.1016/j.chb.2014.03.073
Uzunoǧlu, E., & Misci Kip, S. (2014). Brand communication through digital influencers: Leveraging
blogger engagement. International Journal of Information Management, 34(5), 592–602.
https://doi.org/10.1016/j.ijinfomgt.2014.04.007
Wang, P., Huang, Q., & Davison, R. M. (2020). How do digital influencers affect social commerce
intention? The roles of social power and satisfaction. Information Technology and People, 34(3),
1065–1086. https://doi.org/10.1108/ITP-09-2019-0490
Xiang, L., Zheng, X., Lee, M. K. O., & Zhao, D. (2016). Exploring consumers’ impulse buying behavior
on social commerce platform: The role of parasocial interaction. International Journal of
Information Management, 36(3), 333–347. https://doi.org/10.1016/j.ijinfomgt.2015.11.002

Thirtieth European Conference on Information Systems (ECIS 2022), Timisoara, Romania 11