See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/360831758

The influence of visitor-based social contextual information on visitors’

museum experience

Article in PLoS ONE · May 2022

DOI: 10.1371/journal.pone.0266856

CITATIONS READS
0 71

4 authors, including:

Taeha Yi Ji-Hyun Lee

Hanyang University Korea Advanced Institute of Science and Technology
15 PUBLICATIONS 20 CITATIONS 113 PUBLICATIONS 799 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Digital Performance View project

Rhythm Dance Game (Korean) View project

All content following this page was uploaded by Taeha Yi on 31 May 2022.

The user has requested enhancement of the downloaded file.

PLOS ONE

RESEARCH ARTICLE

The influence of visitor-based social

contextual information on visitors’ museum
experience
Taeha Yi ID1, Hao-yun Lee ID2, Joosun Yum ID2, Ji-Hyun Lee ID2*

1 Department of Interior Architecture Design, Hanyang University, Seoul, The Republic of Korea, 2 Graduate
School of Culture Technology, KAIST, Daejeon, The Republic of Korea

a1111111111 * [email protected]
a1111111111
a1111111111
a1111111111 Abstract
a1111111111
Visitor-centered approaches have been widely discussed in the museum experience
research field. One notable approach was suggested by Falk and Dierking, who defined
museum visitor experience as having a physical, personal, and social context. Many studies
OPEN ACCESS have been conducted based on this approach, yet the interactions between personal and
Citation: Yi T, Lee H-y, Yum J, Lee J-H (2022) The social contexts have not been fully researched. Since previous studies related to these inter-
influence of visitor-based social contextual actions have focused on the face-to-face conversation of visitor groups, attempts to provide
information on visitors’ museum experience. PLoS the social information contributed by visitors have not progressed. To fill this gap, we exam-
ONE 17(5): e0266856. https://doi.org/10.1371/
ined such interactions in collaboration with the Lee-Ungno Art Museum in South Korea. Spe-
journal.pone.0266856
cifically, we investigated the influence of individual visitors’ social contextual information
Editor: Maurizio Naldi, LUMSA: Libera Universita
about their art museum experience. This data, which we call “visitor-based social contextual
Maria Santissima Assunta, ITALY
information” (VSCI), is the social information individuals provide—feedback, reactions, or
Received: November 17, 2021
behavioral data—that can be applied to facilitate interactions in a social context. The study
Accepted: March 28, 2022 included three stages: In Stage 1, we conducted an online survey for a preliminary investiga-
Published: May 24, 2022 tion of visitors’ requirements for VSCI. In Stage 2, we designed a mobile application proto-
Peer Review History: PLOS recognizes the type. Finally, in Stage 3, we used the prototype in an experiment to investigate the influence
benefits of transparency in the peer review of VSCI on museum experience based on visitors’ behaviors and reactions. Our results indi-
process; therefore, we enable the publication of cate that VSCI positively impacts visitors’ museum experiences. Using VSCI enables visi-
all of the content of peer review and author
tors to compare their thoughts with others and gain insights about art appreciation, thus
responses alongside final, published articles. The
editorial history of this article is available here: allowing them to experience the exhibition from new perspectives. The results of this novel
https://doi.org/10.1371/journal.pone.0266856 examination of a VSCI application suggest that it may be used to guide strategies for
Copyright: © 2022 Yi et al. This is an open access enhancing the experience of museum visitors.
article distributed under the terms of the Creative
Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in
any medium, provided the original author and
source are credited. Introduction
Data Availability Statement: All relevant data are According to the International Council of Museums (ICOM), an art museum is defined as a
within the paper and its Supporting Information space for public education, enjoyment, and the promotion of culture [1]. Museum visitors and
files. their experiences constitute a key component of art museum operations and management; as
Funding: This work was supported by a National such, the development of a visitor-centered approach is highly relevant. In applying this
Research Foundation of Korea (NRF) grant funded approach, attempts have been made to more fully understand the museum experiences of

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 1 / 25

PLOS ONE Visitor-based social contextual information

by the Korea government (MSIT) (No. NRF- visitors [2, 3]. Among the various studies, Falk and Dierking [4] proposed a noteworthy
2019R1A2C1007042). URL of NRF: https://www. approach suggesting that visitors’ museum experiences occur amid the interactions of several
nrf.re.kr/eng/index The funders had no role in study
contexts. Specifically, they defined visitor experience as occurring in the interactions of the
design, data collection and analysis, decision to
publish, or preparation of the manuscript. physical (i.e., environment, exhibitions, artwork labels, and guide media), personal (i.e., visit-
ing motivation, prior knowledge, and personal interest or choice), and social (i.e., interactions
Competing interests: The authors have declared
within a group of visitors and conversations with art museum officials or other people)
that no competing interests exist.
contexts.
Previous research has mainly focused on the interactions among the three contexts men-
tioned above [5, 6]. However, the interactions between personal and social contexts are not
fully researched. Some related studies have focused on communication between accompanying
visitors in terms of social contexts [7, 8]. These studies have concentrated on face-to-face con-
versations between visitors, and there has been insufficient research investigating social inter-
action from the various visitors to art museums. When forming an opinion in daily life, people
are affected by the opinions or reactions of others, which can cause them to change their opin-
ions. Such social information is essential data in almost all product, content, or service areas
[9–11]. Through this social process that considers the experiences of ourselves and others
together, we can identify the strengths and weaknesses of specific items, further create mean-
ing, and make decisions [12, 13]. However, there have been few attempts to provide supple-
mental information on visitors’ thoughts or reactions to exhibitions in museums and examine
the effectiveness of utilizing such information. Regarding supplemental information, the exist-
ing research mainly focuses on content written by experts [14, 15]. Thus, to fill the gap in the
literature, this study aimed to investigate the influence of the social information contributed
by individuals in art museums.
Looking at museums as a service product [16], social information can help to provide a bet-
ter visitor experience in a social context. To do so, we identified the categories and compo-
nents of social information via a literature review. Based on the identified social information,
we investigated the visitors’ needs and opinions to design the museum guidance application
and developed an app to deliver the social information. Additionally, we experimented with
the influence of visitor-contributed social information using a mobile eye-tracker. In sum-
mary, this study attempted to reveal the possibility of enhancing the visitor experience via
social information. Further, we proposed a method of providing social information to enrich
the visitor experience through museum guidance applications.

Social information for improving the visitor experience

Researchers have systematically applied advanced digital technology to capture the behavior
and reactions of museum visitors to understand their experience [17]. Such studies have
aimed to provide researchers or museum experts details about the relationship between visi-
tors’ behavior when viewing exhibitions and artworks and their satisfaction [18]. For example,
Lanir et al. [19] proposed a system that provides museum experts with visualized visitor behav-
ior information, and Rodriguez-Boerwinkle, Boerwinkle, and Silvia [20] developed the method
for art research by tracking visitors in a virtual museum environment. Even though visitor
behavior and reaction data have been collected, these collected data have been used as infor-
mation sources for researchers or museum operators only and have not been provided to
museum visitors. Contemporary museums have provided information that meets visitors’
desires by applying communicative technology to encourage museum visits [21]. Therefore,
we identified the types of data used in prior visitor studies and sought ways to present it as
social information that visitors can use to improve their experience in real-time.

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 2 / 25

PLOS ONE Visitor-based social contextual information

Displaying social information is a promising approach for improving certain types of group
participation [22], and it has been demonstrated that social information about the involvement
of others can be used to encourage user participation and augment their engagement in online
communities [23] and physical activities [24]. For example, when added to the context of a
player’s interactions within an online game, social information can help increase the player’s
activity [25]. In terms of museums, social information that reveals the presence of others can
also be used to enhance visitor participation and broaden visitors’ experiences [26].
When studying interactions between personal and social contexts, researchers usually focus
on visitor companionship, and social information is mainly based on their communication
with accompanying visitors [27]. Although some have tried to use personal information as
social information, such as by using SNS in a museum learning context [28] and social tagging
systems that gather visitors’ thoughts [29], few studies have examined the application of social
information data to enhance the visitor experience. To overcome these limitations, we pro-
posed using “visitor-based social contextual information” (VSCI), the social information indi-
viduals provide—feedback, reactions, or behavioral data—can be applied to facilitate
interactions in a social context. VSCI is a collation of visitors’ behaviors and reaction data in
museums and provides visitors with information to enhance their museum experience. There-
fore, our research questions are as follows:
• RQ 1. What are the components of social information from other visitors in art museums?
• RQ 2. Are visitors’ museum experiences enhanced when social information from others is
provided to visitors?

Conceptualization of visitor-based social contextual information

To answer the first research question, we identified the types of data collected from previous
visitor studies to determine the information elements constituting the VSCI. We reviewed the
previous visitor studies from interdisciplinary research to investigate visitors’ behaviors, expe-
riences, reactions, and thoughts systematically. From the literature review, we identified five
categories with detailed components. First, visitor evaluation was divided into level of satisfac-
tion with exhibitions and museum objects. Second, visitor behavioral data were collected, such
as museum object viewing time, visiting time, and visitor type based on the behavioral charac-
teristics. Third, many studies have paid attention to the visitors’ emotions during the art appre-
ciation process. Fourth, some studies focused on museum objects’ characteristics to which the
visitors reacted sensitively. Lastly, the effect of visitors’ comments on their museum experience
were analyzed. Subsequently, we established the VSCI with the commonly discussed factors
from the literature review, as illustrated in Fig 1.
Visitor evaluation. Visitor satisfaction is an important factor that many researchers con-
sider when evaluating museum visit experiences [2, 3, 30]. Han and Hyun [31] demonstrated
the importance of measuring museum visitors’ satisfaction through empirical studies. In art
museum research, the many ways to evaluate aesthetic experiences while viewing artwork are
also commonly discussed. Among them, art experience rating scales, such as those measuring
visitors’ enjoyment, interest, and comprehension, are used frequently [32–34].
Visitor behavior. Tracking and timing visitor behavior is the most useful and representa-
tive method for understanding how visitors interact with the various elements of an exhibition
[35, 36]. Thus, many methods have been used to study visitor behavior, such as capturing visi-
tor location [37, 38] and tracking eye movement [39, 40]. Additionally, valuable measurements
are frequently made by tracking and timing visitor behavior in art museums, such as viewing

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 3 / 25

PLOS ONE Visitor-based social contextual information

Fig 1. Conceptual framework of VSCI based on previous visitor studies.

https://doi.org/10.1371/journal.pone.0266856.g001

time [41, 42], the ratio of the number of visitors [19, 43], visiting time [44], and revisit counts
[45], for a specific artwork or exhibition.
Emotional response. Emotional response is considered a critical aspect of art experience
[46]. Visitors are motivated to find meaningful and personal connections when experiencing
emotional reactions to artworks [47]; such connections indicate an enhanced visiting experi-
ence. Many researchers have studied the aesthetic emotions of visitors while they appreciate
artworks [48]. In these studies, the number of aesthetic emotions noted varies from four [49]
to thirty-five items [50]. Additionally, well-known emotional models from Russell [51] or
Ekman [52] have often been applied [47, 53].
Art features noticed by visitors. The visual features of artwork are usually discussed in
aesthetic experience research. For example, Sartori [54] performed an affective analysis of
abstract art based on visual elements (color, shape, and texture), and previous studies have
demonstrated that the differences in visitors’ art appreciation relate to features such as the
color [40, 55], shape [56], and texture [57] of objects. Additionally, Locher [58] revealed that
visitors are sensitive to artworks with the right visual composition created by a skilled artist.
Similarly, Silvia and Barona [59] demonstrated that visitor preferences are affected by the com-
position of objects.
Visitor comments. According to Coffee [60], other visitors’ comments are important ele-
ments that contribute to an individual’s museum experience. Comment books can be an essen-
tial dialogic activity where social discourse takes shape [61], and they can even stimulate public
debate on the visitor experience [62]. Winter [63] also pointed out that although it has been
relatively ignored by most museums, the comments of other guests provide visitors with new
insights into museum exhibitions. In the field of visitor studies, attempts to analyze visitors’
opinions and needs are being made through online visitor reviews [64], which are becoming
increasingly important.

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 4 / 25

PLOS ONE Visitor-based social contextual information

Method
Study overview
To examine the influence of VSCI on how visitors experience art exhibitions (Fig 2), we defined
the elements of VSCI based on commonly discussed aspects of art experience from visitor stud-
ies, as follows: visitor evaluation, visitor behavior, emotional responses, art features, and com-
ments. In short, we set the behavior and reaction of museum exhibition visitors as elements of
VSCI. Based on those elements of the visitor experience, the research procedure was designed
with reference to prototype-based studies that have designed and developed apps or systems by
analyzing user requirements [65, 66]. Additionally, an experiment was planned to confirm the
differences in visitor experiences depending on whether VSCI was provided or not. Therefore,
we designed a research process with three stages: (1) insight research to collect visitors’ VSCI
needs and opinions; (2) museum application design and prototyping based on the derived
insights; and (3) visitor experiments to reveal the impact of VSCI on the visitor experience.
In Stage 1, we used an online survey to investigate visitors’ needs and interests to determine
the applicable VSCI components. In Stage 2, we developed a prototype of a mobile guidance
application to collect and display the VSCI during an individual’s visit. Similar mobile guid-
ance apps have become increasingly common for improving the visitor experience in art
museum visits [67, 68]; they encourage visitors to stay longer and have a more positive experi-
ence than traditional guidance media in cultural exhibitions [69]. Therefore, we conducted vis-
itor observations to collect VSCI on which to base a prototype of a mobile guidance app. In
Stage 3, we conducted an experiment with a post-experiment interview to compare the experi-
ences of visitors using the mobile app with VSCI while viewing the exhibition with the visitors’
experiences using a mobile app without VSCI. We also tested the differences in visitors’ art
expertise because previous studies have revealed that visitors’ art knowledge and interest influ-
ence visitors’ behaviors and responses to artworks [70, 71]. To ensure the bioethical integrity
of the study, the study design and all the processes of this research were approved by the
KAIST Institutional Review Board (NO: KH2019-040).

Fig 2. Overview of research methods. VSCI = visitor-based social contextual information; MET = mobile eye tracker.
https://doi.org/10.1371/journal.pone.0266856.g002

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 5 / 25

PLOS ONE Visitor-based social contextual information

Stage 1: Online visitor survey to gather feedback for VSCI

We conducted an online survey to understand the interest and needs of visitors for various
VSCI factors; the survey content was compiled based on previous visitor studies. An online
survey was chosen to obtain opinions from the various visitors who had experience visiting art
museums and using mobile guidance services. The survey responses were collected via Google
Forms from May 24 to June 8, 2021. We applied the survey data to calculate the importance
ranking according to visitors’ “degree of curiosity” about VSCI factors. Additionally, we con-
ducted k-means clustering to investigate grouping characteristics according to the ratings for
the VSCI elements. The survey results were then used to design a VSCI mobile application and
recruit participants for an experiment.
Participants. A total of 71 participants (female: 42, male: 29) took part in the survey, and
the average age was 31.02 (SD = 5.18) years. Those who participated in this survey received a
USD 2 reward. The participants included university students, office workers, and museum
workers recruited from multiple online communities in South Korea (e.g., ARA and Hongik-
in). Among the participants, 30 were art majors, and 41 were non-arts majors.
Materials. The survey we used, a copy of which is provided with additional notes in S1
Appendix, consisted of four sections. The first section included questions for collecting partici-
pants’ demographic information and asked participants if they were willing to participate in
future visitor experiments. The second section gathered information regarding what the par-
ticipants were curious about when viewing artworks and exhibitions. This section included a
total of 17 information elements, for which the degree of participant interest was measured
using a 7-point Likert scale (1: very negative to 7: very positive). They were also asked to write
down the reason for giving the score and any additional elements about which they were curi-
ous. The third and fourth sections related to the participants’ art interest [72] and art knowl-
edge (similar to Belke, Leder, and Augustin [73]); this information was intended to allow us to
determine if the distinction between art majors and non-majors makes a difference in their
levels of interest or knowledge.

Stage 2: Design and development of mobile guidance application prototype

We considered the importance level of each VSCI element based on the survey outcome and
the overall visit process to design the interaction flow, as illustrated in Fig 3. The importance
ranking of VSCI elements from the survey helped clarify the information hierarchy used in the
mobile applications interface design. According to Djamasbi and Hall-Phillips [74], an inter-
face with a clear hierarchy provides a better experience than one that has no difference in the
relative importance of page elements.
In addition to the importance ranking, it is necessary to define some of the contents of the
VSCI before designing the application, including emotional response and art features. Since
cognitive load is an important factor when designing interfaces [75], and working memory is
limited by the number of items [76], we minimized the number of items on the interface to
reduce the cognitive load. Specifically, in our prototype, we used Ekman’s model [52] with six
simple emotions to indicate the visitor’s emotional response to an artwork. The six basic cross-
cultural emotional responses included anger, disgust, fear, happiness, sadness, and surprise.
An additional item, “indifference,” was added to represent neutral responses [47]. Moreover,
based on the abovementioned art features from previous research, four components were set
as options for the visitor to select the visual elements of the artwork they saw: color, composi-
tion, shape of the objects, and texture.
Furthermore, we used the average behavioral data to recognize the visitor types derived by
comparison with others. This comparison relies on the ability to track the behavior of visitors

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 6 / 25

PLOS ONE Visitor-based social contextual information

Fig 3. The process of mobile guidance application with visitor-based social contextual information (VSCI).
https://doi.org/10.1371/journal.pone.0266856.g003

in real-time. Thus, we determined our prototype would include a built-in function that records
and calculates the app log data from visitors. For example, the visiting time was measured
from the log-in time to the log-out time, and the viewing time was calculated between the
entering and exiting time points to and from the specific artwork pages. Additionally, the
“revisit count” for each artwork was calculated using the number of times each visitor
approached and temporarily remained in front of the artwork. Next, after identifying these fac-
tors as four levels based on quantiles, the sum of the levels for each visitor was calculated.
Finally, the application was designed to capture the visitor type with three equally sized groups,

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 7 / 25

PLOS ONE Visitor-based social contextual information

namely, diligent (high-rank), selective (middle-rank), and busy visitor (low-rank), referring to
Sparacino [77]. Additionally, we set up “most popular artworks” in the app to display all of the
works whose scores were within the top 30% for visitors’ “likes.” The remaining artworks were
considered less popular.
To develop a prototype mobile application, we selected artist Ung-no Lee, a master of South
Korean contemporary art, who creates modern abstract paintings using traditional Oriental
art ink. Since visitors tend to prefer works of famous artists and more frequently judge them as
beautiful [33], we planned an exhibition consisting of works of only one artist to eliminate the
effect of the artists’ fame. In collaboration with the Lee-Ungno Art Museum, we selected five
series representing all his artworks, with four artworks in each series. In this collaboration, we
designed a simple exhibition called “The Art World of Lee Ung-no” that introduced pieces
representative of the artist’s work. We developed the mobile application using the Python
(v.3.9) Flask web server platform (v.2.0.1), HTML, and JQuery.

VSCI preparation for the prototype

To make the prototype for the experiment, we needed to prepare real VSCI beforehand. There-
fore, we conducted visitor observations to collect data with participants recruited from those
who participated in our survey (randomly selected, n = 10). The observation simulated the
experiment’s main process and environment to collect the participants’ feedback, reactions,
and behavioral data as VSCI. The participants were asked to provide VSCI for every artwork
and exhibition (Table 1). However, VSCI from other visitors was not displayed to the individ-
ual visitor participants during the process.
To collect visitor behavioral data, we measured eye-tracking data through Pupil Capture
v.3.4 software. We set each surface (area of interest [AOI]) in advance by attaching markers on
four sides of the displayed artworks. The viewing time (VT) for each AOI was measured by cal-
culating the timestamps of all events where the gaze point of the participant entered (enter
time) and exited (exit time) the AOI. The visiting time was measured as the sum of overall
viewing time (VT_O) by adding the VT for a set of AOIs:
Pn
VT ¼ i¼1 ðExit timei End timei Þ in AOI ð1Þ

Pn
VT O ¼ i¼1 ðExit timei End timei Þ in set of AOI ð2Þ

Table 1. VSCI elements to which participants responded in the observation process.

Case Question
After viewing an artwork 1. Leave your comments about this artwork.
2. Which emotions did you feel were aroused by this artwork?
(1) Sadness; (2) Happiness; (3) Surprise; (4) Fear; (5) Anger; (6) Disgust; (7) Indifference
3. What did you pay attention to on this artwork?
(1) Color; (2) Form/Shape; (3) Composition; (4) Texture
4. Please rate this artwork. (from 1 star: negative to 5 stars: positive)
(1) Interest; (2) Liking; (3) Understanding
After viewing the 5. Please leave your comments about this exhibition.
exhibition 6. Please give an overall score for your opinion about this exhibition. (1 star: negative to 5
stars: positive)

VSCI represents “visitor-based social contextual information”.

https://doi.org/10.1371/journal.pone.0266856.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 8 / 25

PLOS ONE Visitor-based social contextual information

The revisit count for the artwork (RC_A) refers to the number of times a visitor returned
(at least once) to see a specific artwork. We also counted the number (CN_A) of cases where
the participant’s gaze moved to another work after viewing a specific work. When we collect
the CN_A for each artwork and exclude the first viewing situation, we can get the revisit count
as follows:
Pn
RC A ¼ i¼1 ðCN Ai 1Þ in AOI ð3Þ

In addition, when we divide the total number of visitors by the number of visitors who have
the value of CN_A, the proportion of visitors who viewed a specific artwork (PV_A) can be
measured, as below:
Number of visitors with CN A
PV A ¼ ð4Þ
Total number of visitors

Stage 3: Visitor experiment

We designed an experiment to determine the influence of providing VSCI to visitors by exam-
ining visitors’ behavior and reactions. For tracking and timing visitor behavior, indoor posi-
tioning technologies can be used; some examples are beacons [78], mobile sensors [79], and
cameras [80]. However, these methods make it difficult to accurately determine whether the
visitor is viewing the artwork or the mobile phone. In other words, it is difficult to distinguish
between and measure the amount of time spent looking at the mobile app while in front of
each exhibit. However, in this study, it was necessary to distinguish accurately when a visitor is
viewing an artwork or the mobile app; therefore, we used a mobile eye-tracker (Pupil Labs,
Pupil Core 120 Hz binocular) to collect the behavioral data of visitors. Mobile eye-tracking
(MET) technology is becoming a widely used tool for understanding visual processing and vis-
itor behavior [81]. It also provides rich data through a scene camera that allows users to obtain
information about the environment [82].
Participants. Forty participants (mean age = 27.29; SD = 5.02) were recruited for the
experiment and received a USD 10 reward. All participants had normal or corrected-to-nor-
mal vision and did not wear glasses for mobile eye-tracking. We contacted those who agreed
to participate in this experiment via the online survey. The participants were recruited based
on their gender (male:female = 19:21), major in art or not (art major:non-art major = 16:24),
and clusters derived from the online survey (Cluster 1:Cluster 2 = 20:20).
Materials. The two types of “Art World of Lee Ung-no” exhibitions were displayed in a
laboratory with three white walls, decorated to resemble the environment of a real museum.
Some studies have found differences in art appreciation depending on the size and location of
the artworks [83, 84]; therefore, all artworks displayed were similar in terms of size (fitted to
A3 size), spacing between artworks, and installation height on the wall. We also used high-
quality images of artwork provided by the Lee-Ungno Art Museum, given that existing
research indicates no significant difference in the appreciation of original artworks [85]. Sim-
ple descriptions of each artwork were similarly designed (i.e., word count: M = 24.85,
SD = 1.45) and content (series explanation and image interpretation).
Two curators whose majors were in art planning were consulted in preparing the descrip-
tions. To determine the impact of VSCI in a visitor experiment, we divided the art series
equally into two exhibitions so that a visitor could appreciate the two exhibitions (E1 and E2)
comprising different artworks. Next, we prepared apps for the four conditions by developing
apps with a 2 (with or without VSCI) × 2 (E1 or E2) design. Finally, to minimize the differ-
ences attributable to the mobile environment (e.g., screen size and speed performance), all visi-
tors were asked to use the same mobile phone (Samsung Galaxy Note 9).

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 9 / 25

PLOS ONE Visitor-based social contextual information

Fig 4. Experimental process for the three blocks. VSCI represents “visitor-based social contextual information”.
https://doi.org/10.1371/journal.pone.0266856.g004

Procedure. This experiment was conducted from August 6 to August 21, 2021, as three
blocks (Fig 4) and took about 90 mins, on average. In the first block (pre-experiment), the
experiment was introduced to the participants, and their written consent forms were collected.
Next, the participants were asked to wear the MET and look at the nine marker points for cali-
bration. Finally, to check the calibration accuracy, the experimenter drew a large circle with
their index finger in front of the participant and confirmed whether the participant’s gaze fol-
lowed the position of the fingertip correctly.
In the second block, the participants freely viewed the exhibition under two conditions
comprising different artworks (E1 or E2) and mobile apps (with or without VSCI) to eliminate
the effect of viewing the same materials. For example, if P01 viewed the first exhibition using a
mobile app with VSCI in the first session, they viewed the second exhibition using the mobile
app without VSCI. All participants were randomly assigned to a sequence of viewing cases,
and 20 cases were performed for each condition. After each viewing case was finished, the par-
ticipants were asked to evaluate their satisfaction with the current exhibition and the mobile
app.
In the third block, after finishing the free-viewing block, visitors were asked to move to an
interview space. In the interview, we asked questions about the participant’s overall experience
of viewing the exhibitions and of the mobile app (e.g., “Was there any difference in the exhibi-
tion experience between the two mobile guidance applications?” “Which of the two mobile
applications do you prefer? What is the reason?”). Using a 5-point Likert scale, we also col-
lected the preference level (“How much do you prefer this information element?”) and the par-
ticipants’ opinions (“Why do you prefer (or not) this information element?”) for each
component of the VSCI.
Data analysis. This study analyzed visitor behavior data and satisfaction with exhibitions
and mobile apps to determine whether there is a difference in their exhibition experience
depending on the VSCI. To analyze the visitor behavior data, we collected the viewing time of
AOI in the same ways as the VSCI preparation phase. In the experimental environment, to

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 10 / 25

PLOS ONE Visitor-based social contextual information

measure the viewing time of the artworks and the mobile app, the mobile phone was also set as
AOI by attaching markers, and the viewing time of the artwork (VT-A) and the mobile app
(VT-M) was measured separately. The visiting time was calculated by adding both the time to
view the artworks and the apps (VT-O = total of VT-A + total of VT-M).
Next, through the interview, we measured exhibition satisfaction using three items from
Han and Hyun [31] and the mobile app satisfaction using four items from Song et al. [86] on a
5-point Likert scale (see Table A in S2 Appendix). Lastly, we conducted multiple analyses of
variances (ANOVAs) on the difference between the participants’ behavioral data and satisfac-
tion according to the experimental conditions. For all measurements and statistical analyses
for this study, we used Python version 3.9 and the python libraries “statsmodels (v.0.13.0),”
“researchpy (v.0.3.2),” and “scipy (v.1.7.1)”.

Results
Online survey
In the survey, we asked about participants’ needs for information elements when viewing art-
works and viewing an exhibition (Table 2). We found that the survey participants were inter-
ested in the comments on artwork and emotions of others, as well as an artwork’s notable
features among the elements of VSCI. They were also interested in knowing which artworks
were most popular, the comments related to visitors’ impressions of the exhibition, and their
visitor type compared with others.
Differences by art expertise. In sections three and four of the survey, we checked whether
there was a difference between art majors and non-majors with regard to art knowledge (maxi-
mum 100 scores) and interest in art (maximum 77 scores). Regarding art knowledge, the aver-
age score for the art major group was 59.33 (SD = 18.13), and that for the non-art major group
was 26.71 (SD = 8.26). Regarding the degree of interest in art, the average score for the art
major group was 52.17 (SD = 14.46), and that for the non-art major group was 37.37
(SD = 13.48). There were statistically significant differences between the two groups in both

Table 2. Results of the ratings of participants on the elements of visitor-based social contextual information.

Factor Mean (SD)

Viewing Artwork Comments on artwork 5.25 (1.54)
Emotional response of others 5.00 (1.86)
Artwork features from others 4.97 (1.72)
Interest in artwork 4.54 (1.90)
Understanding of artwork 4.49 (1.89)
Like the artwork 4.35 (1.97)
Viewing time of artwork 3.90 (1.81)
Proportion of visitors who viewed artwork 3.62 (1.86)
Revisit count for artwork 3.44 (1.69)
Viewing exhibition Most popular artworks 5.38 (1.77)
Exhibition comments 4.97 (1.80)
Visitor type compared with others 4.80 (1.98)
Exhibition satisfaction 4.66 (1.77)
Less popular artworks 3.83 (2.08)
Average viewing time of artworks 3.77 (1.80)
Average proportion of visitors who viewed artworks 3.65 (1.90)
Average revisit counts for artworks 3.42 (1.70)

https://doi.org/10.1371/journal.pone.0266856.t002

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 11 / 25

PLOS ONE Visitor-based social contextual information

sections (art knowledge: t(37.85) = 9.182, p<0.001; art interest: t(60.01) = 4.385, p<0.001). In
other words, there was a clear difference in art expertise between art and non-art majors. Next,
we looked at the differences between the two groups for all VSI elements and found that the
groups did not differ significantly in all elements (all p>0.05). These results indicate no signifi-
cant difference in the needs for VSCI elements based on art expertise.
Visitor grouping through K-Means clustering. After confirming that there were no dif-
ferences in the VSCI needs based on art expertise, we examined whether differences can be
found among the groups when using the k-means clustering technique. The results from clus-
tering (k = 2) demonstrated differences in all the VSCI elements between the two clusters (Fig
5). The first cluster demonstrated a positive trend of over 4 points (4 points is neutral on the
7-point Likert scale) and high average scores in all VSCI elements compared with the second
cluster. For the second cluster, only one item (“comments about an artwork”) was over 4
points, while the others had a negative trend of fewer than 4 points. Next, an ANOVA was per-
formed to check the differences between the two clusters for all VSCI elements (all ps<0.001).
No differences were found in gender, age, and art expertise between the two clusters.
Considering the results, we examined the participants’ answers regarding their interest in
the VSCI information. For the first cluster (n = 45), the main reason participants found certain
elements interesting is captured in P13’s response: “I enjoy comparing other people’s ideas to
mine.” This opinion was shared by 66.7% of the participants. Meanwhile, for the second-larg-
est cluster (n = 26), was the group identifying the reason captured in P03’s response: “I do not
feel the need to compare my feelings and others because my personal feelings and others’ feel-
ings are a unique experience” was reflected in the responses of 69.2% of the participants. These
results might be interpreted as a difference in personal interest in the information of others. In
conclusion, we did not find any differences in information needs based on art expertise. How-
ever, a clear difference was found between those who were curious about others’ reactions and
those who were not.

Prototype of a mobile guidance application

Based on the survey outcome, we displayed each VSCI element considering the importance
level of the elements. Additionally, we completed the prototype by inputting the VSCI col-
lected through the data preparation phase into the mobile app, as illustrated in Fig 6. The
developed mobile guide application with VSCI comprised: (1) a simple description of the
ongoing exhibition with visitor satisfaction and comments; (2) indicators of popular artworks
such as star markings based on the “liking” scores of visitors; (3) visitor comments on an art-
work; (4) reaction tabs about the artwork, which contain the emotion, art features, ratings, and
behavioral data from visitors; and (5) visitor type through a comparison with the visitors’
behavioral data. The app without VSCI comprised exhibition information, locations of exhib-
ited works, and descriptions of artworks, similar to the existing guide app.

Visitor experiment
First, we confirmed whether there were statistical differences based on the demographic infor-
mation and experimental environments. There were no significant differences by gender
(male or female), major (art or non-art majors), experimental environments (first exhibition
[E1] or second exhibition [E2]), and exhibition appreciation sequence (first viewing or second
viewing) in terms of exhibition satisfaction, app satisfaction, and visiting time. Next, we ana-
lyzed the differences in the behavior and satisfaction of visitors according to the experimental
conditions (Table 3).

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 12 / 25

PLOS ONE Visitor-based social contextual information

Fig 5. Results of k-means clustering through an online survey analysis.

https://doi.org/10.1371/journal.pone.0266856.g005

To explore the effect of VSI, we conducted separate ANOVAs with post hoc Tukey honestly
significant difference (HSD) corrections for each factor (Fig 7). Visiting time differed signifi-
cantly between conditions (F(3,76) = 6.40; p<0.001). Compared with viewing exhibitions with-
out VSCI, participants viewing exhibitions with VSCI spent more time visiting the exhibition
(Meandiff(cond1, cond2) = 13.98 min; Meandiff(cond3, cond4) = 13.56 min.) There was also a
statistically significant difference in viewing time for the mobile app (F(3,76) = 5.32;
p = 0.002). However, in the case of viewing time for artwork, there was no significant differ-
ence (F(3,76) = 2.35; p = 0.08), and no differences among conditions were detected using a
post hoc Tukey test (Fig 7B). This implies that if VSCI is given, viewing time for mobile apps
could increase, but not the time of viewing artworks. This result is similar to Temme [87], who
indicated that although audiences want more information related to artwork, providing more
information does not increase their viewing time. In addition, Temme [87] demonstrated that
as the amount of information increases, the enjoyment of artworks decreases, and there is no

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 13 / 25

PLOS ONE Visitor-based social contextual information

Fig 6. Prototype of mobile guidance applications (two types of apps: With VSCI and without VSCI). VSCI represents “visitor-based social contextual
information”.
https://doi.org/10.1371/journal.pone.0266856.g006

Table 3. Mean (and standard deviations) in exhibition appreciation divided by conditions.

Factor Condition 1: E1 with VSCI Condition 2: E1 without Condition 3: E2 with VSCI Condition 4: E2 without
(n = 20) VSCI (n = 20) (n = 20) VSCI (n = 20)
Visitor Visiting time (min.) 38.92 (18.37) 24.94 (13.89) 37.54 (16.94) 23.98 (10.04)
behavior Viewing time for artwork 24.07 (11.90) 17.73 (12.12) 22.42 (14.71) 15.40 (6.95)
(min.)
Viewing time for mobile 14.85 (11.75) 7.21 (5.05) 15.12 (7.33) 8.58 (6.32)
app (min.)
Visitor Exhibition satisfaction 4.03 (0.53) 3.08 (0.56) 4.08 (0.39) 3.28 (0.73)
evaluation (5-Likert)
Mobile app satisfaction 3.76 (0.52) 2.80 (0.93) 3.90 (0.48) 2.86 (0.95)
(5-Likert)

E1 = First exhibition, E2 = Second exhibition, and VSCI = visitor-based social contextual information.

https://doi.org/10.1371/journal.pone.0266856.t003

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 14 / 25

PLOS ONE Visitor-based social contextual information

Fig 7. Box-and-whisker plots of the four conditions for visitor behavior (A to C) and satisfaction (D, E). The results of multiple comparison corrections by
Tukey HSD between groups are presented (� denotes p<0.05; �� , p<0.01, and ��� , p<0.001). E1 = first exhibition, E2 = second exhibition, and VSCI = visitor-
based social contextual information.
https://doi.org/10.1371/journal.pone.0266856.g007

difference in interest in art labels. On the contrary, when VSCI was presented, both the satis-
faction of the exhibition and mobile app increased. Moreover, the increase in visiting time
when VSCI was provided suggests the possibility of offering visitors an opportunity to enjoy
an exhibition for a longer time without shortening the time of viewing artworks.
Visitor satisfaction was high when VSCI was provided according to both indicators (exhibi-
tion satisfaction: Meandiff(cond1, cond2) = 0.95, Meandiff(cond3, cond4) = 0.8; mobile app satis-
faction: Meandiff(cond1, cond2) = 0.96, Meandiff(cond3, cond4) = 1.04), and the difference was
significant (exhibition satisfaction: F(3, 76) = 16.47, p<0.001; mobile app satisfaction: F(3, 76)
= 11.88, p<0.001). The result of the post hoc multiple comparisons demonstrated obvious dif-
ferences between four conditions (Fig 7D, 7E). In other words, the provision of VSCI to visi-
tors appeared to have a positive effect on exhibition appreciation. Moreover, in the post-
experiment interview, most of the participants (87.5%) answered that they preferred the appli-
cation with VSCI. They mentioned that the additional information provided by the app
increased their interest and understanding when viewing the artworks and exhibitions. Only
one visitor preferred the app without VSCI because they did not want their personal viewing

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 15 / 25

PLOS ONE Visitor-based social contextual information

to be affected by others’ reactions (P3, “I don’t like being influenced by other people’s opin-
ions”). The rest of the visitors (10.0%) mentioned that the preferred app differs depending on
the situation: (1) whether they have ample time (e.g., P37, “I think I will use the app with VSCI
if I have enough time to enjoy the exhibition”) and (2) whether visiting with a companion
(e.g., P26, “If I have a companion, I will probably use the app without others’ information”).
Additionally, a Pearson correlation coefficient was computed to assess the linear relation-
ship between exhibition satisfaction and mobile app satisfaction and demonstrated a high posi-
tive correlation between the two variables (r = 0.71, p<0.001). This result implies that visitors’
satisfaction with the mobile app affects their exhibition experience. In summary, the results
illustrate that in most cases, providing mobile guidance applications with VSCI has a positive
effect on the exhibition experience of visitors.
Evaluation of VSCI elements. The visitor evaluation of the detailed VSCI elements was
analyzed, as illustrated in Fig 8. Overall, “Visitor type (M = 4.25; SD = 0.81)” and “Comment
on artwork (M = 4.25; SD = 0.87)” were evaluated as the most positive factors, with few
responses in the negative region (1 to 2 points). By contrast, “Viewing time of artwork
(M = 3.03; SD = 1.27),” “Emotion of artwork (M = 3.0; SD = 1.1),” and “Proportion of visitors
to artwork (M = 2.98; SD = 1.25)” were rated relatively negatively, as the rating distribution
appeared evenly.
During the post-experiment interview session, we collected the opinions of the participants
on each VSCI element. We analyzed the interviews to identify the key features, among which
the comparison with others’ responses was most evident in “Visitor type” and “Comments on
artwork.” Regarding “Visitor type,” the participants enjoyed knowing their type and visiting
behavioral characteristics based on a comparison with others (e.g., P12, “I think it is fun to
show my type like MBTI. I think the comparison information with other people helped me
better understand myself”). Regarding the “comments on artworks,” most participants thought
that it was good to be able to compare their own thoughts with the opinions of other people. In
the process of comparison, they found pleasure in reading the comments that expressed
thoughts they shared (e.g., P27, “It was nice to see what people think and see if they have the
same thoughts as me”). They also gained a new perspective on an artwork from the comments
of others (e.g., P36, “I was looking at an artwork and thought it was not good because it was
too dark. However, after reading the reviews saying that it was like a star in the dark night sky,
the artwork looked new”). Through social comparisons, negative reactions to information that
were inconsistent with their thoughts arose (e.g., P18, “There were a lot of different opinions
from mine, so it didn’t really touch me”). Specifically, regarding the “Most popular artworks,”
negative opinions were expressed, and the participants were uncomfortable knowing that the
artworks they liked and the displayed artworks were different.
Next, the participants mentioned that the VSCI helped them understand the exhibition or
artworks and determine their viewing priorities regarding exhibition content. Regarding the
“Most popular artworks,” there were positive opinions, such as that VSCI gave some clues as
to which artworks to look at in detail and that they can appreciate thinking about what makes
a specific artwork popular. Moreover, some participants said that numerical information, such
as behavioral data, was helpful for judging the artworks objectively (e.g., P11, “I think numeri-
cal information is useful when deciding which artworks to see in detail”). This relates to Scre-
ven’s [88] results, who reported that it is necessary to provide information to address the
problem of visitors struggling to determine why the artworks exhibited in art museums are
important to them. In other words, it implies that VSCI, which represents the others’ art expe-
rience, can play a role in improving the understanding of artworks.
Lastly, for all VSCI elements, most of the differences were in the gap between those who
wanted to be provided with the information element and those who did not. As various VSCI

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 16 / 25

PLOS ONE Visitor-based social contextual information

Fig 8. Ridge plot of visitor ratings for VSCI elements (drawn in the order of average scores [highest to lowest],
depicted by the red lines). VSCI represents “visitor-based social contextual information”.
https://doi.org/10.1371/journal.pone.0266856.g008

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 17 / 25

PLOS ONE Visitor-based social contextual information

elements were provided in this study, the visitors’ needs were also widely revealed. Therefore,
to reflect the various needs of these visitors for VSCI, it is possible to apply a customization
option that allows individual visitors to manage information elements. For example, in the
case of a visitor who does not want their personal viewing to be disturbed by the VSCI (e.g.,
P19, “Information about how other people rated the exhibit was meaningless and disturbing”),
it is possible to design a control function on the preset page to stop the display of unwanted
information that is on the component list of the VSCI.
Visiting difference of clusters based on the preliminary survey. We also checked
whether there were visiting differences by cluster based on the preliminary survey regarding
the degree of interest in VSCI elements. Interestingly, we did not find any significant differ-
ences in either behavioral factors (visiting time: t(78) = 0.42, p = 0.43; viewing time for art-
work: t(78) = 0.32, p = 0.52; viewing time for app: t(78) = 0.35, p = 0.60) and satisfaction
(exhibition satisfaction: t(78) = -0.47, p = 0.64; app satisfaction: t(78) = 0.124, p = 0.902). While
the online survey showed notable differences in participant curiosity for all items, no differ-
ences in information preference were found for almost all VSCI factors. Unlike the partici-
pants in Cluster 2 who negatively evaluated the VSCI factors presented through the
questionnaire, they had a more positive evaluation than they expected when they viewed the
actual VSCI data in the exhibition environment (e.g., P17, “It was better to see information
from other visitors than I previously thought. I was able to think more deeply by looking at
other people’s thoughts and scores on the artwork”). However, the only difference was in
“Comments on artworks” (Meandiff(cluster1, cluster2) = 0.6; t(38) = 2.30; p = 0.03). The partici-
pants in Cluster 2 answered negatively regarding the visitors’ comments because this informa-
tion was of low quality, subjective, and unprofessional (e.g., P13, “I have doubts about the
professionalism of those who provided the comments. I think it would be good if the opinion
came from an expert, to have credibility.”) Although we did not get any significant or notable
findings, the result revealed that the level of professionalism conveyed by the information also
needs to be considered in audience-based information.

Discussion
The supplemental information provided by museums has been focused on delivering expert
knowledge about artists or artworks to visitors. In addition, the analysis of various visitor reac-
tions has been used for exhibition evaluation or research purposes. Going beyond these previ-
ous approaches, this study proposed a way to provide visitors with others’ responses to an art
exhibition via a mobile guidance application. Specifically, through a visitor experiment, we
revealed that providing VSCI has the potential to improve visitor exhibition experience in
terms of visitor behavior and satisfaction. From the interviews, we found that visitors who
compared their thoughts with others gained new insights on art appreciation through different
opinions and found key points in viewing the exhibition via VSCI.
We also identified visitors’ additional needs for VSCI. First, many participants look forward
to opportunities for interaction that are designed based on VSCI, such as sharing their visitor
type results on SNS, replying to or liking others’ feedback, having ranking games based on vis-
iting behavior, pairing people with similar visiting types to meet up for a museum trip, or
other personalized functions. Second, personalization was often mentioned, for example, hav-
ing a personal archive page for visitors’ feedback records or providing visit recommendations
based on individuals’ visit-related reactions. Third, participants also suggested that the options
for emotional responses could be more varied (e.g., P11, “In particular, the items of emotion
for artworks were strange. It would be nice if less extreme emotions, such as peaceful and joy-
ful, were taken into account”). In this study, we applied Ekman’s model, and it is necessary to

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 18 / 25

PLOS ONE Visitor-based social contextual information

consider the list of emotions presented in emotional studies related to viewing art, such as
those used by Rodriguez-Boerwinkle et al. [89]. In summary, we noticed an interest in
enriched interactive contents with VSCIs, indicating that future research on this topic is
warranted.
This study focused mainly on mobile guidance applications, and other applications that use
VSCI in various museum service areas should be considered for comparison. For instance, to
recall the audience’s visit, Petrelli et al. [90] suggested providing tangible data souvenirs by
capturing the personal visiting experience. If visitor-contributed information from other visi-
tors, such as VSCI, is combined with the function of generating data souvenirs, more diverse
interactive services can be created to extend the visitor experience in art museums. VSCI can
also be offered to online visitors in virtual museums. Walmsley [91] demonstrated that the dig-
ital engagement of visitors could enhance the opportunities to attend to artistic dialogue more
frequently and encourage empathy with others in terms of sociological roles. Sundar et al. [92]
revealed that applying communication technology (customization of the gallery, interactivity
through live chats, and 3D navigational tool) in virtual museums can positively support the
quality of visitors’ experience. Moreover, social connectivity between online visitors via VSCI
can be linked to the metaverse, which has received growing interest and represents a future
trend of virtual museums [93]. Therefore, it is necessary to investigate the use of VSCI in vari-
ous museum service areas.
Lastly, one of the impressive interview answers was that just giving various information ele-
ments made a visitor think more about the artwork (P21, “I liked the fact that I got to think
about my own thoughts, rather than the results of others’ responses while looking at the infor-
mation elements, for example, what emotions I felt, what features of artwork were good, and
why this artwork was good for me”). Regarding individual interpretations of artworks that are
important in art education [94], it is meaningful to identify information elements that help vis-
itors independently think and appreciate artworks (as in the VSCI), rather than simply using
elements that have been commonly discussed in visitor studies.

Limitations and future research

This study has several limitations that need to be interpreted carefully. First, as in previous
empirical studies [95], the experiment was conducted in a laboratory environment with a
small number of participants. However, because of the spread of COVID-19, it was difficult to
conduct experiments in a museum environment and recruit a large number of participants.
Second, even though the purpose of the online survey in Stage 1 was finding insights to design
a mobile guidance application with VSCI, the sample size was quite small (n = 71). In addition
to the purpose of understanding user demand, in-depth research is required in consideration
of various visitor characteristics (e.g., age differences [39], the composition of the visit groups
[96], and countries [97]) to derive standardized VSCI.
Third, a small number of artworks were installed in the experimental exhibition (ten art-
works for each exhibition). Meanwhile, large art museums usually take a relatively long time to
visit because of the huge number of displayed artworks. In such an environment, visitors can
experience museum fatigue [98, 99], and providing more information, such as VSCI, may
accelerate visitors’ fatigue. However, because VSCI captures artworks that visitors pay atten-
tion to, such social recommendations could reduce the museum fatigue phenomenon. Hence,
it is necessary to examine the effect of VSCI in a real museum environment on a large-scale
basis in future studies. Lastly, through the online survey, we found that the respondents can be
divided into two groups based on how they viewed information on other visitors’ reactions:
positive or negative. Therefore, we examined the characteristics of the two groups through an

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 19 / 25

PLOS ONE Visitor-based social contextual information

experiment, but no significant difference was found in all aspects of behavioral factors and sat-
isfaction. Based on previous studies on visitor classification, this topic may be further investi-
gated by grouping visitors based on behavioral characteristics in the exhibition [100, 101], visit
motivation [102], and anticipated experience [2].

Conclusions
This study investigated the effect of providing information from visitors via mobile guidance
applications on museum visitors’ experiences. The contributions of this research are twofold.
For academic interest, there is the need for an in-depth understanding of visitor-oriented
social information in the field of visitor studies. To do this, based on previous visitor studies,
we defined VSCI as encompassing the behaviors, emotions, and opinions of visitors. We then
developed a mobile application for VSCI and conducted a visitor experiment using mobile
eye-tracking technology. The experiment results demonstrated that when VSCI was provided,
the behavioral parameters of the visitors increased, and visitor satisfaction for the exhibition
and the app also improved. Moreover, the interview analysis indicated that the exhibition
experiences of visitors were enriched through VSCI. This means that the information elements
that previous studies have focused on can be used as data sources for researchers and museum
operators and as information sources for museum visitors. To our knowledge, this is the first
attempt to examine visitor-based social information to enhance visitor experience in art
museums.
For practical interest, we proposed a method of designing and developing a mobile museum
guidance application, including VSCI. This method could help museums use social informa-
tion as mobile app content. Despite the growing interest in mobile guidance applications,
some contemporary museum apps have failed to satisfy visitors’ expectations [103]. To address
this issue, this study proposed a method for improving visitor satisfaction by paying attention
to visitor-contributed information, which has been insufficiently considered in the past. In
summary, this study suggested one possible approach of providing social information for the
visitor-centered design of museum guidance applications, which contributes to the field of vis-
itor studies. We hope that our research will help enhance visitors’ museum experiences by pro-
viding them an opportunity to communicate their various opinions and reactions, which goes
beyond simply conveying information based on expert knowledge.

Supporting information
S1 Appendix. Explanation of online survey questionnaires.
(PDF)
S2 Appendix. Explanation of post-experiment interview.
(PDF)
S3 Appendix. Explanation of materials for visitor experiment.
(PDF)
S4 Appendix. Research data.
(XLSX)

Acknowledgments
We thank the Lee-Ungno Art Museum for providing valuable information and Jihye Oh for
assisting with the mobile app development.

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 20 / 25

PLOS ONE Visitor-based social contextual information

Author Contributions
Conceptualization: Taeha Yi, Hao-yun Lee, Ji-Hyun Lee.
Data curation: Taeha Yi.
Investigation: Taeha Yi, Joosun Yum.
Methodology: Taeha Yi, Joosun Yum.
Supervision: Ji-Hyun Lee.
Validation: Taeha Yi.
Visualization: Taeha Yi, Hao-yun Lee.
Writing – original draft: Taeha Yi, Hao-yun Lee.
Writing – review & editing: Taeha Yi, Hao-yun Lee, Ji-Hyun Lee.

References
1. ICOM (International Council of Museums. Museum definition. 2007. [Cited 2021 Nov 02]. Available
from: https://icom.museum/en/resources/standards-guidelines/museum-definition/.
2. Pekarik AJ, Doering ZD, Karns DA. Exploring satisfying experiences in museums. Curator: The
Museum Journal. 1999; 42: 152–173. https://doi.org/10.1111/j.2151-6952.1999.tb01137.x
3. de Rojas Mdel C, Camarero Mdel C. Experience and satisfaction of visitors to museums and cultural
exhibitions. Int Rev Public Nonprofit Mark. 2006. https://doi.org/10.1007/BF02893284
4. Falk JH, Dierking LD. The museum experience. New York: Routledge; 2011. https://doi.org/10.4324/
9781315417899
5. Chang E. Interactive experiences and contextual learning in museums. Stud Art Educ. 2006; 47: 170–
186. https://doi.org/10.1080/00393541.2006.11650492
6. Rainoldi M, Neuhofer B, Jooss M. Mobile Eyetracking of museum learning experiences. Information
and communication technologies in tourism. Cham: Springer; 2018. https://doi.org/10.1007/978-3-
319-72923-7_36
7. Christidou D. Social interaction in the art museum: Connecting to each other and the exhibits. The Int J
Soc Pol Commun Agendas Arts. 2016; 11: 27–38. https://doi.org/10.18848/2326-9960/CGP/v11i04/
27-38
8. Tröndle M, Wintzerith S, Wäspe R, Tschacher W. A museum for the twenty-first century: The influence
of “sociality” on art reception in museum space. Museum Manag Curatorship. 2012; 27: 461–486.
https://doi.org/10.1080/09647775.2012.737615
9. Tsao WC. Which type of online review is more persuasive? The influence of consumer reviews and
critic ratings on moviegoers. Electron Commer Res. 2014; 14: 559–583. https://doi.org/10.1007/
s10660-014-9160-5
10. Robinson R, Goh TT, Zhang R. Textual factors in online product reviews: A foundation for a more influ-
ential approach to opinion mining. Electron Commer Res. 2012; 12: 301–330. https://doi.org/10.1007/
s10660-012-9095-7
11. Dellarocas C. Strategic manipulation of internet opinion forums: Implications for consumers and firms.
Manag Sci. 2006; 52: 1577–1593. https://doi.org/10.1287/mnsc.1060.0567
12. Basu C, Hirsh H, Cohen W. Recommendation as classification: Using social and content-based infor-
mation in recommendation. In: Proc 15th Nat conf on Artificial Intelligence. Madison, WI; 1998. pp.
714–720.
13. Godes D, Silva JC. Sequential and temporal dynamics of online opinion. Mark Sci. 2012; 31: 448–473.
https://doi.org/10.1287/mksc.1110.0653
14. Bailey-Ross C, Beresford AM, Smith DT, Warwick C. Aesthetic appreciation and Spanish art: Insights
from eye-tracking. Digit Sch Humanit. 2019; 34: i17–i35. https://doi.org/10.1093/llc/fqz027
15. Serrell B. Exhibit labels: An interpretive approach. London: Rowman & Littlefield Publishing Group;
2015.
16. Gilmore A, Rentschler R. Changes in museum management: A custodial or marketing emphasis? J
Manag Dev. 2002; 21: 745–760. https://doi.org/10.1108/02621710210448020

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 21 / 25

PLOS ONE Visitor-based social contextual information

17. Yi T, Chang M, Hong S, Lee JH. Use of eye-tracking in artworks to understand information needs of
visitors. Int J Hum-Comput Int. 2021; 37: 220–233. https://doi.org/10.1080/10447318.2020.1818457
18. Kovavisaruch LO, Sanpechuda T, Chinda K, Sornlertlamvanich V, Kamonvej P. Museum content eval-
uation based on visitor behavior. In: Proc 13th intl conf on Electrical Engineering/electron, comput,
telecommunications and information technology (ECTI-CON); 2016. https://doi.org/10.1109/
ECTICon.2016.7561472
19. Lanir J, Kuflik T, Sheidin J, Yavin N, Leiderman K, Segal M. Visualizing museum visitors’ behavior:
Where do they go and what do they do there? Pers Ubiquit Comput. 2017; 21: 313–326. https://doi.
org/10.1007/s00779-016-0994-9
20. Rodriguez-Boerwinkle R, Boerwinkle M, Silvia P. Open gallery for arts research (OGAR): An open-
source tool for studying the psychology of virtual art museum visits. Psychiatry. 2021: ArXiv. https://
doi.org/10.31234/osf.io/nfwru
21. Vaz RI, Fernandes PO, Veiga AC. Interactive technologies in museums: How digital installations and
media are enhancing the visitors’ experience. In: Rodrigues JMF, Ramos CMQ, Cardoso PJS, Henri-
ques C, editors. Handbook of research on technological developments for cultural heritage and eTour-
ism applications. Hershey, Pennsylvania: IGI Global; 2017. pp. 30–53.
22. DiMicco JM, Pandolfo A, Bender W. Influencing group participation with a shared display. In: Proc
CSCW conf comput support coop work; 2004. https://doi.org/10.1145/1031607.1031713
23. Sun L, Vassileva J. Social visualization encouraging participation in online communities. In: Lecture
Notes in Computer Science. Proc 12th intl conf on Collaboration and Technology. Berlin, Heidelberg.
Springer. 2006: 349–363. https://doi.org/10.1007/11853862_28
24. Nakayama S, Tolbert TJ, Nov O, Porfiri M. Social information as a means to enhance engagement in
citizen science-based telerehabilitation. J Assoc Inf Sci Technol. 2019; 70: 587–595. https://doi.org/
10.1002/asi.24147
25. Kirman B, Lawson S, Linehan C, Martino F, Gamberini L, Gaggioli A. Improving social game engage-
ment on Facebook through enhanced socio-contextual information. In: Proc SIGCHI conf hum factor
comput syst; 2010. https://doi.org/10.1145/1753326.1753589
26. Boehner K, Thom-Santelli J, Zoss A, Gay G, Hall JS, Barrett T. Imprints of place: Creative expressions
of the museum experience. In: Proc SIGCHI conf hum factor comput syst (extended abstracts); 2005.
https://doi.org/10.1145/1056808.1056881
27. López Sintas J, Garcı́a Álvarez E, Pérez Rubiales E. E. Art museum visitors: Interaction strategies for
sharing experiences. Museum Manag Curatorship. 2014; 29: 241–259. https://doi.org/10.1080/
09647775.2014.919175
28. Charitonos K, Blake C, Scanlon E, Jones A. Museum learning via social and mobile technologies:
(How) can online interactions enhance the visitor experience? Br J Educ Technol. 2012; 43: 802–819.
https://doi.org/10.1111/j.1467-8535.2012.01360.x
29. Trant J, Wyman B. Investigating social tagging and folksonomy in art museums with steve.museum.
[Cited 2021 Nov 02]. Available from: https://www.ra.ethz.ch/cdstore/www2006/www.rawsugar.com/
www2006/4.pdf. In: Collab Web Tagging Workshop. Edinburgh, UK: ACM; 2006. p. WWW2006.
30. Packer J, Bond N. Museums as restorative environments. Curator (NY). Curator: The Museum Jour-
nal. 2010; 53: 421–436. https://doi.org/10.1111/j.2151-6952.2010.00044.x
31. Han H, Hyun SS. Key factors maximizing art museum visitors’ satisfaction, commitment, and post-pur-
chase intentions. Asia Pac J Tourism Res. 2017; 22: 834–849. https://doi.org/10.1080/10941665.
2017.1345771
32. Leder H, Carbon CC, Ripsas AL. Entitling art: Influence of title information on understanding and
appreciation of paintings. Acta Psychol (Amst). 2006; 121: 176–198. https://doi.org/10.1016/j.actpsy.
2005.08.005 PMID: 16289075
33. Mastandrea S, Crano WD. Peripheral factors affecting the evaluation of artworks. Empirical Stud Arts.
2019; 37: 82–91. https://doi.org/10.1177/0276237418790916
34. Brieber D, Nadal M, Leder H. In the white cube: Museum context enhances the valuation and memory
of art. Acta Psychol (Amst). 2015; 154: 36–42. https://doi.org/10.1016/j.actpsy.2014.11.004 PMID:
25481660
35. Yalowitz SS, Bronnenkant K. Timing and tracking: Unlocking visitor behavior. Visit Stud. 2009; 12: 47–
64. https://doi.org/10.1080/10645570902769134
36. Serrell B. The aggregation of tracking-and-timing visitor-use data of museum exhibitions for bench-
marks of “Thorough Use”. Visit Stud. 2020; 23: 1–17. https://doi.org/10.1080/10645578.2020.
1750830
37. Kosmopoulos D, Styliaras G. A survey on developing personalized content services in museums. Per-
vasive Mob Comput. 2018; 47: 54–77. https://doi.org/10.1016/j.pmcj.2018.05.002

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 22 / 25

PLOS ONE Visitor-based social contextual information

38. Yoshimura Y, Sobolevsky S, Ratti C, Girardin F, Carrascal JP, Blat J, et al. An analysis of visitors’
behavior in the Louvre museum: A study using Bluetooth data. Environ Plann B Plann Des. 2014; 41:
1113–1131. https://doi.org/10.1068/b130047p
39. Walker F, Bucker B, Anderson NC, Schreij D, Theeuwes J. Looking at paintings in the Vincent van
Gogh Museum: Eye movement patterns of children and adults. PLOS ONE. 2017; 12: e0178912.
https://doi.org/10.1371/journal.pone.0178912 PMID: 28636664
40. Massaro D, Savazzi F, Di Dio C, Freedberg D, Gallese V, Gilli G, et al. When art moves the eyes: A
behavioral and eye-tracking study. PLOS ONE. 2012; 7: e37285. https://doi.org/10.1371/journal.pone.
0037285 PMID: 22624007
41. Smith LF, Smith JK, Tinio PPL. Time spent viewing art and reading labels. Psychology of Aesthetics
Creativity and the Arts. 2017; 11: 77–85. https://doi.org/10.1037/aca0000049
42. Goller J, Mitrovic A, Leder H. Effects of liking on visual attention in faces and paintings. Acta Psychol
(Amst). 2019; 197: 115–123. https://doi.org/10.1016/j.actpsy.2019.05.008 PMID: 31146088
43. Serrell B. Are they watching? Visitors and Videos in Exhibitions. Curator: The Museum Journal. 2002;
45: 50–64. https://doi.org/10.1111/j.2151-6952.2002.tb00049.x
44. Cotter KN, Fekete A, Silvia PJ. Why do people visit art museums? Examining visitor motivations and
visit outcomes. Empirical Stud Arts. 2021. https://doi.org/10.1177/02762374211011740
45. Carbon CC. Art perception in the museum: How we spend time and space in art exhibitions. i-Percep-
tion. 2017; 8: 2041669517694184. https://doi.org/10.1177/2041669517694184 PMID: 28321289
46. Tinio PPL, Gartus A. Characterizing the emotional response to art beyond pleasure: Correspondence
between the emotional characteristics of artworks and viewers’ emotional responses. Prog Brain Res.
2018; 237: 319–342. https://doi.org/10.1016/bs.pbr.2018.03.005 PMID: 29779742
47. Alelis G, Bobrowicz A, Ang CS. Exhibiting emotion: Capturing visitors’ emotional responses to
museum artefacts. Lect Notes Comput Sci. 2013: 429–438. https://doi.org/10.1007/978-3-642-39238-
2_47
48. Schindler I, Hosoya G, Menninghaus W, Beermann U, Wagner V, Eid M, et al. Measuring aesthetic
emotions: A review of the literature and a new assessment tool. PLOS ONE. 2017; 12: e0178899.
https://doi.org/10.1371/journal.pone.0178899 PMID: 28582467
49. Van Paasschen J, Bacci F, Melcher DP. The influence of art expertise and training on emotion and
preference ratings for representational and abstract artworks. PLOS ONE. 2015; 10: e0134241.
https://doi.org/10.1371/journal.pone.0134241 PMID: 26244368
50. Hagtvedt H, Patrick VM, Hagtvedt R. The perception and evaluation of visual art. Empirical Stud Arts.
2008; 26: 197–218. https://doi.org/10.2190/EM.26.2.d
51. Russell JA, Mehrabian A. Evidence for a three-factor theory of emotions. J Res Pers. 1977; 11: 273–
294. https://doi.org/10.1016/0092-6566(77)90037-X
52. Ekman P. The argument and evidence about universals in facial expressions of emotion. In: Wagner
H, Manstead A, editors, Handbook of social psychophysiology. New York: Wiley; 1989. pp. 143–164.
53. Yang H, Han J, Min K. Distinguishing emotional responses to photographs and artwork using a deep
learning-based approach. Sensors (Basel). 2019;19. https://doi.org/10.3390/s19245533 PMID:
31847398
54. Sartori A. Affective analysis of abstract paintings using statistical analysis and art theory. In: Proc 16th
int conf multimodal InterAct; 2014. https://doi.org/10.1145/2663204.2666289
55. Braun DI, Doerschner K. Kandinsky or me? How free is the eye of the beholder in abstract art? Iper-
ception. 2019; 10: 2041669519867973. https://doi.org/10.1177/2041669519867973 PMID: 31565211
56. Savazzi F, Massaro D, Di Dio C, Gallese V, Gilli G, Marchetti A. Exploring responses to art in adoles-
cence: A behavioral and eye-tracking study. PLOS ONE. 2014; 9: e102888. https://doi.org/10.1371/
journal.pone.0102888 PMID: 25048813
57. Sartori A, Şenyazar B, Salah AAA, Salah AA, Sebe N. Emotions in abstract art: Does texture matter?
Lect Notes Comput Sci. 2015: 671–682. https://doi.org/10.1007/978-3-319-23231-7_60
58. Locher PJ. An empirical investigation of the visual rightness theory of picture perception. Acta Psychol
(Amst). 2003; 114: 147–164. https://doi.org/10.1016/j.actpsy.2003.07.001 PMID: 14529822
59. Silvia PJ, Barona CM. Do people prefer curved objects? Angularity, expertise, and aesthetic prefer-
ence. Empirical Stud Arts. 2009; 27: 25–42. https://doi.org/10.2190/EM.27.1.b
60. Coffee K. Visitor comments as dialogue. Curator. 2013; 56: 163–167. https://doi.org/10.1111/cura.
12017
61. Coffee K. Disturbing the eternal silence of the gallery: A site of dialogue about hunger in America.
Museum Manag Curatorship. 2011; 26: 11–26. https://doi.org/10.1080/09647775.2011.540124

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 23 / 25

PLOS ONE Visitor-based social contextual information

62. Magliacani M, Madeo E, Cerchiello P. From ‘listener’ to ‘speaker’ museum visitors: Guest book as a
means of dialogue. Museum Manag Curatorship. 2018; 33: 467–483. https://doi.org/10.1080/
09647775.2018.1506709
63. Winter M. Visitor perspectives on commenting in museums. Museum Manag Curatorship. 2018; 33:
484–505. https://doi.org/10.1080/09647775.2018.1496354
64. Stoleriu OM, Brochado A, Rusu A, Lupu C. Analyses of visitors’ experiences in a natural World Heri-
tage Site based on TripAdvisor reviews. Visit Stud. 2019; 22: 192–212. https://doi.org/10.1080/
10645578.2019.1665390
65. Kang B, Song B, Yang S, Lee J. A development technique for mobile applications program. Studies in
Computational Intelligence. Cham: Springer. 2017: (47–62). https://doi.org/10.1007/978-3-319-
51472-7_4
66. Liedtke C, Baedeker C, Hasselkuß M, Rohn H, Grinewitschus V. User-integrated innovation in Sus-
tainable LivingLabs: An experimental infrastructure for researching and developing sustainable prod-
uct service systems. J Clean Prod. 2015; 97: 106–116. https://doi.org/10.1016/j.jclepro.2014.04.070
67. Palumbo F, Dominici G, Basile G. Designing a mobile app for museums according to the drivers of visi-
tor satisfaction. [Cited 2021 Nov 02]. Available from: https://ssrn.com/abstract=2298361. In: Proc 1st
intl conf on Management, mark, tourism, retail, finance and computer applications; 2013.
68. Teixeira S, Angélico MJ, Abreu AJ, Camarinha AP. A longitudinal study About mobile applications in
the tourism sector. Smart Innov Syst Technol. 2021: 547–556. https://doi.org/10.1007/978-981-33-
4256-9_50
69. Petrie H, Othman MK, Power C. Smartphone guide technology in cultural spaces: Measuring visitor
experience with an iPhone multimedia guide in Shakespeare’s church. Int J Hum Comput Interact.
2017; 33: 973–983. https://doi.org/10.1080/10447318.2017.1304606
70. Winston AS, Cupchik GC. The evaluation of high art and popular art by naive and experienced view-
ers. Vis Arts Res. 1992; 18: 1–14.
71. Koide N, Kubo T, Nishida S, Shibata T, Ikeda K. Art expertise reduces influence of visual salience on
fixation in viewing abstract paintings. PLOS ONE. 2015; 10: e0117696. https://doi.org/10.1371/
journal.pone.0117696 PMID: 25658327
72. Specker E, Forster M, Brinkmann H, Boddy J, Pelowski M, Rosenberg R, et al. The Vienna Art Interest
and Art Knowledge Questionnaire (VAIAK): A unified and validated measure of art interest and art
knowledge. Psychology of Aesthetics Creativity and the Arts. 2020; 14: 172–185. https://doi.org/10.
1037/aca0000205
73. Belke B, Leder H, Augustin MD. Mastering style. Effects of explicit style-related information, art knowl-
edge and affective state on appreciation of abstract paintings. Psychol Sci. 2006; 48: 115–134.
74. Djamasbi S, Hall-Phillips A. Visual search. In: Bergstrom JR, Schall A, editors. Eye tracking in user
experience design. Amsterdam: Morgan Kaufmann Publishers; 2014. pp. 27–45. https://doi.org/10.
1016/B978-0-12-408138-3.00002–9
75. Drommi A, Ulferts GW. Interface design: A focus on cognitive science [Cited 2021 November 2]. Avail-
able from: https://proc.edsig.org/2001/02a/ISECON.2001.Drommi.pdf. In: Proc information systems
education conf.;18:§02a; 2001.
76. Miller GA. The magical number seven plus or minus two: Some limits on our capacity for processing
information. Psychol Rev. 1956; 63: 81–97. https://doi.org/10.1037/h0043158 PMID: 13310704
77. Sparacino F. The Museum Wearable: Real-time sensor-driven understanding of visitors’ interests for
personalized visually-augmented museum experiences. [Cited 2021 Nov 02]. Available from: https://
www.archimuse.com/mw2002/papers/sparacino/sparacino.html. In: Proc museums and the web;
2002.
78. He Z, Cui B, Zhou W, Yokoi S. A proposal of interaction system between visitor and collection in
museum hall by iBeacon. In: Proc 10th intl conf on Computer Science and Education; 2015. https://
doi.org/10.1109/ICCSE.2015.7250283
79. Martella C, Miraglia A, Frost J, Cattani M, van Steen M. Visualizing, clustering, and predicting the
behavior of museum visitors. Pervasive Mob Comput. 2017; 38: 430–443. https://doi.org/10.1016/j.
pmcj.2016.08.011
80. Hong S, Yi T, Yum J, Lee JH. Visitor-artwork network analysis using object detection with image-
retrieval technique. Adv Eng Inform; 48: 101307. https://doi.org/10.1016/j.aei.2021.101307
81. Garbutt M, East S, Spehar B, Estrada-Gonzalez V, Carson-Ewart B, Touma J. The embodied gaze:
Exploring applications for mobile eye tracking in the art museum. Visit Stud. 2020; 23: 82–100. https://
doi.org/10.1080/10645578.2020.1750271
82. Mayr E, Knipfer K, Wessel D. In-sights into mobile learning: An exploration of mobile eye tracking
methodology for learning in museums. In: Vavoula G, Pachler N, Kukulska-Hulme AG, editors.

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 24 / 25

PLOS ONE Visitor-based social contextual information

Researching mobile learning: Frameworks, tools and research designs. Oxford: Peter Lang Publish-
ing; 2009. pp. 189–204.
83. Silvera DH, Josephs RA, Giesler RB. Bigger is better: The influence of physical size on aesthetic pref-
erence judgments. J Behav Decis Making. 2002; 15: 189–202. https://doi.org/10.1002/bdm.410
84. Seidel A, Prinz J. Great works: A reciprocal relationship between spatial magnitudes and aesthetic
judgment. Psychology of Aesthetics Creativity and the Arts. 2018; 12: 2–10. https://doi.org/10.1037/
aca0000100
85. Grüner S, Specker E, Leder H. Effects of context and genuineness in the experience of art. Empirical
Stud Arts. 2019; 37: 138–152. https://doi.org/10.1177/0276237418822896
86. Song J, Kim J, Jones DR, Baker J, Chin WW. Application discoverability and user satisfaction in
mobile application stores: An environmental psychology perspective. Decis Support Syst. 2014; 59:
37–51. https://doi.org/10.1016/j.dss.2013.10.004
87. Temme JE. Amount and kind of information in museums: Its effects on visitors satisfaction and appre-
ciation of art. Vis Arts Res. 1992; 18: 28–36.
88. Screven CG. Motivating visitors to read labels. ILVS Rev J Visit Behav. 1992; 2: 183–211.
89. Rodriguez-Boerwinkle R, Fekete A, Silvia P, Cotter KN. The art of feeling different: Exploring the diver-
sity of emotions experienced During an art museum visit. Psychiatry. 2021: ArXiv. https://doi.org/10.
31234/osf.io/f7g5a
90. Petrelli D, Marshall MT, O’Brien S, McEntaggart P, Gwilt I. Tangible data souvenirs as a bridge
between a physical museum visit and online digital experience. Pers Ubiquit Comput. 2017; 21: 281–
295. https://doi.org/10.1007/s00779-016-0993-x
91. Walmsley B. From arts marketing to audience enrichment: How digital engagement can deepen and
democratize artistic exchange with audiences. Poetics. 2016; 58: 66–78. https://doi.org/10.1016/j.
poetic.2016.07.001
92. Sundar SS, Go E, Kim HS, Zhang B. Communicating art, virtually! Psychological effects of technologi-
cal affordances in a virtual museum. Int J Hum Comput Interact. 2015; 31: 385–401. https://doi.org/10.
1080/10447318.2015.1033912
93. Choi H, Kim S. A content service deployment plan for metaverse museum exhibitions—Centering on
the combination of beacons and HMDs. Int J Inf Manag. 2017; 37: 1519–1527. https://doi.org/10.1016/
j.ijinfomgt.2016.04.017
94. Barrett T. About art interpretation for art education. Stud Art Educ. 2000;42. https://doi.org/10.2307/
1320749
95. Brieber D, Nadal M, Leder H, Rosenberg R. Art in time and space: Context modulates the relation
between art experience and viewing time. PLOS ONE. 2014; 9: e99019. https://doi.org/10.1371/
journal.pone.0099019 PMID: 24892829
96. Sandifer C. Time-based behaviors at an interactive science museum: Exploring the differences
between weekday/weekend and family/nonfamily visitors. Sci Ed 1997; 81: 689–701. https://doi.org/
10.1002/(SICI)1098-237X(199711)81:6<689::AID-SCE6>3.0.CO;2-E
97. Rawlings D, Barrantes i Vidal N, Furnham A. Personality and aesthetic preference in Spain and
England: Two studies relating sensation seeking and openness to experience to liking for paintings
and music. Eur J Pers. 2000; 14: 553–576. https://doi.org/10.1002/1099-0984(200011/12)14:6<553::
AID-PER384>3.0.CO;2-H
98. Bitgood S. Museum fatigue: A critical review. Visit Stud. 2009; 12: 93–111. https://doi.org/10.1080/
10645570903203406
99. Falk JH, Koran JJ, Dierking LD, Dreblow L. Predicting visitor behavior. Curator: The Museum Journal.
1985; 28: 249–258. https://doi.org/10.1111/j.2151-6952.1985.tb01753.x
100. Zancanaro M, Kuflik T, Boger Z, Goren-Bar D. Analyzing museum visitors’ behavior patterns. In: Proc
11th intl conf on User Modeling; 2007. https://doi.org/10.1007/978-3-540-73078-1_27
101. Dean D. Museum exhibition. Theor Pract. London: Routledge. 1994. https://doi.org/10.4324/
9780203039366
102. Falk JH. Contextualizing Falk’s identity-related visitor motivation model. Visit Stud. 2011; 14: 141–157.
https://doi.org/10.1080/10645578.2011.608002
103. Hanussek B. Enhanced exhibitions? Discussing museum apps after a decade of development. Adv
Archaeol Pract. 2020; 8: 206–212. https://doi.org/10.1017/aap.2020.10

PLOS ONE | https://doi.org/10.1371/journal.pone.0266856 May 24, 2022 25 / 25

View publication stats