Exploring the Impact of Virtual Reality on Painting Appreciation

Abstract

This study aimed to explore the experience of painting appreciation through desktop virtual reality (VR) and a heads-up display (HTC Vive) and compare it with those of a physical painting and album. The study included 52 university students who viewed paintings in all four environments and answered a subjective feelings questionnaire. The results show that the physical paintings generated more positive emotions, such as pleasure, relaxation, and calmness, than virtual paintings. Additionally, the physical paintings were perceived and evaluated as more original, attractive, and valuable. Moreover, among the virtual paintings, the landscape paintings were more popular than abstract ones, whereas among the physical paintings, the abstract paintings were preferred. The head-mounted display made the participants feel uncertain about moving in the physical environment, and the transmission line connecting it to the computer limited their movements. Therefore, it is necessary to enhance the quality and ease of interactions in VR environments to make virtual paintings appear as impressive as physical ones.

1. Introduction

1.1. Virtual Reality (VR) in Art Exhibitions

VR is regarded as a significant next-generation technological innovation and has been employed in exhibitions, military, gaming, education, and various other fields [1,2,3]. A market research report has noted that the global VR market (including software and hardware) was valued at approximately USD 16 billion in 2022 and is poised to grow from USD 19.15 billion in 2023 to USD 91.54 billion by 2031 [4]. Many of the current VR devices, such as Oculus Rift, Samsung Galaxy Gear VR, and HTC Vive, were launched several years ago. In 2023, Apple Inc., released a new-generation VR device called the Apple Vision Pro, which has boosted the adoption of VR.

VR involves the use of computer simulations to generate a three-dimensional (3D) virtual world to provide users with the feeling of “immersion” [5]. VR display devices can be classified according to their degree of integration, from partial to complete, as desktop, projection-based, head-mounted display (HMD), and cave automatic, in that order, and their control devices also differ. For example, desktop VR only requires a keyboard and mouse, whereas projection-based VR requires a steering wheel or joystick and is often used for driving simulations. Moreover, an HMD can be equipped with a joystick and tactile gloves, whereas some devices, such as the Apple Vision Pro, can be controlled directly via gestures.

Although VR technologies have been used in various fields, their application to artworks has led to exciting discoveries [5,6,7]. Over the past few years, many famous museums and art galleries, such as the Helmut Newton Foundation, Louvre, and Tate Modern, have hosted VR exhibitions, providing audiences with unique and exciting experiences [8,9,10]. The VR design is crucial for the success and quality of these exhibitions [11]. Lee, Jung, tom Dieck, and Chung [7] assessed the role of VR in inducing the intention to visit museums and found that VR offered education, entertainment, and immersion; enhanced the overall experience; and aroused the intention to visit the physical museum. Jung, tom Dieck, Lee, and Chung [6] showed that VR can create positive experiences that increase the intention to visit museums. Dozio et al. [12] stated that VR technologies are popular because they reduce costs and enhance technological availability and user acceptance.

1.2. Impact of Exhibition Presentation Methods on Appreciation

The recent advancements in digital technologies have enabled the digitization of physical art; however, methods for using digital technology to fully interpret the connotations of an artwork and encourage the audience to delve deeper into them must be enhanced. Chen et al. [13] explored the differences between viewing physical exhibits and their photos, as well as theme display zones, on user experiences through subjective evaluations. The results showed that although the visitation experience was primarily affected by the viewing style, it had a lower impact on the overall experience provided by the exhibits or exhibition area. Physical exhibits provide a better experience owing to higher autonomy, freedom, and interactivity, whereas their photos allow for focusing on the themes within them, making it easier to highlight their value and enhance their evaluations. This study suggests that physical exhibits can be paired with brochures including their close-ups to provide visitors with a more comprehensive viewing experience.

Moreover, Horn et al. [14] compared the differences in audience interactivity between physical exhibits and screen images. The results showed that both media could make it easy for an audience to understand the themes of the exhibits. However, the audience generally thought that the physical exhibits were more attractive and liked them more, and they were willing to spend more time interacting with them. Hsu et al. [15] explored the impacts of four display types (3D images displayed via large-screen; picture and text billboards; VR; and audio–visual multimedia) on audience experience and found that VR and large-screen 3D displays were preferred. The higher preference and effectiveness indicate that, compared with static information or multimedia audio, 3D dynamic images are better at attracting the attention of the audience and enhancing their information acceptance. Higuera-Trujillo et al. [16] conducted a case study in a store to compare the effects of four presentation methods (physical field, field photos, 360° panoramas, and VR) on the psychological and physiological responses of participants. They found that the immersion effects of 360° panoramas and VR were better and induced a degree of viewing pleasure similar to that of a real scene and significantly higher than that when viewing photos.

Recently, Kuo et al. [17] reconstructed a modern painting in VR and asked participants to move around in its 3D world. The results showed that the VR rendering had a negative impacts on identifying objects within the painting and its interpretation but no impact on its analysis and assessment. However, students who viewed VR paintings exhibited greater interest and enjoyment than those who viewed 2D ones. Yang et al. [18] used a VR technology to construct immersive scenes with specific artistic styles and integrated avatars to allow for interactions with characters in artistic paintings. They noted that the feelings toward art viewed in the VR environment were more positive than for those viewed on a computer screen. In the painting activity that followed the art appreciation, participants in the VR group exhibited significantly higher attention levels than those in the computer-screen group. Chen [19] compared the effects of physical and VR exhibitions by analyzing the total visit time, participation, and interactive-experience levels. The results showed that VR exhibitions scored lower than physical exhibitions in color aesthetic perception, probably because ambient light affected the accurate perception of the colors in the paintings. In contrast, the stylistic–aesthetic scores of the VR exhibitions were higher than that of the physical exhibitions. The author noted that because virtual artwork can be manipulated, participants can observe its stylistic structure more comprehensively, whereas they are prohibited from touching the physical artwork. Lin et al. [20] explored the experience of art appreciation through desktop VR and mobile VR, comparing it to the appreciation of a physical painting. The findings indicated that there were no significant differences in painting evaluation and the emotions expressed during appreciation across the three conditions. This suggests that participants perceive paintings similarly, whether viewed through VR or in person. Moreover, the study noted that special attention should be paid to a device’s specifications and the interaction methods, which do not need to be considered in physical exhibitions but are key to audiences experiencing depth and satisfaction in VR systems.

1.3. Study Objectives

Some museums and galleries are already using VR to enhance visitor experiences by allowing them to immerse themselves in the artworks through unconventional methods. VR technologies allow visitors to view exhibitions remotely, providing those with physical disabilities who are unable to visit museums with opportunities to observe and interact with art. Remote participation also facilitates visitations during busy times, offering visitors more freedom in selecting their visitation times. However, digital art is not the original artwork but a digital image formed by converting the original work into an image after being photographed. Subsequently, the color, size, and three-dimensionality of the work can be changed, whether printed on paper or displayed on a screen. However, original works and their digital forms are likely to have some differences [21]. Additionally, it is worth investigating whether the viewing method employed in a VR environment affects the feelings induced by viewing an artwork and the preference for it. Accordingly, this study aimed to explore the impacts of the same paintings on the subjective feelings and preferences of visitors and provide design guidelines for displaying virtual paintings, with the aim of improving their viewing effects.

2. Materials and Methods

2.1. Paintings

The experiment employed 12 paintings created by Ms. Sandy Lee, as shown in Figure 1. They have dimensions of 41 × 31 cm without the outer frame and 62 × 52.5 cm with the frame. Paintings 1, 2, 4, 6, 10, and 12 are landscape paintings, whereas the remaining six (3, 5, 7, 8, 9, and 11) are abstract paintings.

2.2. Participants

Fifty-two university students (10 males and 42 females) participated in the experiment. The mean age of the males was 19.6 (SD = 0.7), whereas that of the females was 20.3 (SD = 2.5).

2.3. Experimental Design

2.3.1. Independent Variables

An experiment was designed to study the impact of the painting presentation method. For the evaluation of the experience of painting appreciation, a one-factor, four-level factorial design was employed. The independent variable was the painting presentation method, including the original physical painting, physical painting album, desktop VR, and HTC Vive. Further details are provided in the following sections.

• Physical Original Painting

The experimental site was 7 × 4.8 m; the 12 original paintings were hung on the walls in a sequence, and the other two walls were equipped with auxiliary visual hanging scrolls, as illustrated in Figure 2.

2.

Physical Painting Album

The paintings were scanned to create an album with an unfolded size of 60 × 30 cm. Each painting was presented as a panorama and partially enlarged, and its name and description were displayed in the lower-right or lower-left corner of the panorama. Figure 3 shows a panoramic view of Painting 6 in the painting album.

3.

Desktop VR

The desktop VR scenario is illustrated in Figure 4. This study adopted the technique of panoramic photography using a digital camera to capture multiple images, with the camera center as the rotation axis facing the surrounding environment, and then employed image processing. Subsequently, a difference was created by surrounding the captured images with a ring-shaped cylinder. Three observation points were established for the exhibition. Multimedia web technology was used to dynamically display the panoramic images and combine them into webpage structures. Once the user accessed the webpage, they could click the white arrow shown on the ground to move to a particular position or move the mouse to change the viewing direction. Each painting had a square pattern below it. When it was clicked, an enlarged image of the painting and its title were displayed, as shown in the bottom image in Figure 4.

4.

HTC Vive

In this condition, an HTC Vive COSMOS, New Taipei City, Taiwan, device was used. The VR environment was constructed using 3ds Max software (2022 version). The electronic files of the 12 paintings were then placed within it. UNITY software (2022 version) was used to integrate the scene with the HMD. The viewing method of this experimental condition was similar to that of the physical environment. The pictures viewed with the HMD moved in accordance with the physical movements of the participants, as shown in Figure 5.

2.3.2. Dependent Variables

The dependent variables were the answers to the subjective feelings questionnaire and viewing duration. The questionnaire comprised three sections. The first included basic information questions, such as gender, age, and latest academic title. The second section was on subjective feelings. Mehrabian and Russell [22] believe that the atmosphere of a venue affects the emotional response of an audience, so they proposed three emotional dimensions—pleasure, arousal, and dominance—to evaluate the emotional responses of an audience. Since there is no “dominance” process in viewing paintings, priority is given to pleasure and arousal in this study. The subjective feelings questions were adopted from Hsieh [23] and Lin, Chen, and Lin [20]. The first part of the questions was about the feelings of the viewing process, including “pleasure”, “relaxed”, “calmness”, and “having access to all the works I like”. The second part was an evaluation of the painting, including “originality”, “attractive”, and “valuable”. The third part was the overall feeling of the painting, including “the artworks make me feel the form or meaning of cloud”, “the title intensifies my interest in the painting”, “I would like to hang this painting in my house”, and “preference”. The answers were obtained on a five-point Likert scale with scores of 1 (strongly disagree), 2 (disagree), 3 (fair), 4 (agree), or 5 (strongly agree). The third section involved selecting three favorite paintings. The starting point of the viewing duration was considered as the start of the viewing condition, and the end point was when they had finished viewing all 12 paintings; the period between the starting and end points was the total viewing duration.

2.4. Experiment Procedure

First, the purpose and procedure of the experiment were explained to the participants and then the order of the experimental conditions was randomly determined for each participant. The experimental procedures for the four painting presentation methods are shown in Figure 6. When physically viewing the original paintings, the participants could move freely within the exhibition space and independently decide the viewing order, distance, and time spent on the paintings (Figure 6a). While viewing a physical painting album, they sat on a chair and flipped through the album on a desk (Figure 6b). In the desktop VR condition, they were shown the paintings on a 22 in screen (VA2248M-LED, ViewSonic) and could control the forward, backward, and opening actions using a mouse (Figure 6c). The viewing method with the HTC Vive was similar to that for the real environment, as follows; the HMD showed the painting that the participant was facing. The painting enlarged as they got closer it, whereas it shrunk as they moved away (Figure 6d). Note that the viewing time was not limited, and the participants could decide the length of time they wanted to spend viewing each painting based on their personal preferences. After experiencing each condition, they were allowed to rest for 5 min. Once they had viewed the paintings in the four conditions, they were asked to answer a subjective questionnaire.

2.5. Statistical Analysis

First, a one-way analysis of variance (ANOVA) was used to assess whether the painting presentation method factor of each question was significant (significance level = 0.05). Those that reached the significance level were tested using Duncan’s multiple comparison method for significant differences among the four averages. Additionally, Pearson’s correlation coefficient was used to determine the intervariable relationships (significance level = 0.05). Finally, a multiple regression analysis was conducted using sex and questionnaire items as independent variables to establish preference regression models for the four presentation methods.

3. Results and Discussion

3.1. Descriptive Statistics and ANOVA Analysis

Table 1. Descriptive statistics and ANOVA analysis.

Subjective Questionnaire (1–5 Points)	Physical Paintings		Virtual Paintings		Significance
	Original Paintings	Painting Album	Desktop VR	HTC Vive
1. Pleasure	4.3 a	3.9 b	3.4 c	3.6 c	***
2. Relaxed	4.4 a	4.3 a	3.9 b	3.2 c	***
3. Calmness	4.3 a	4.3 a	3.8 b	3.3 c	***
4. Having access to all works I like	4.4 a	4.4 a	4.1 ab	3.8 b	***
5. Originality	3.9	3.7	3.7	3.6
6. Attractive	4.1 a	3.7 b	3.7 b	3.6 b	*
7. Valuable	4.1 a	3.8 b	3.4 c	3.4 c	***
8. Meaning of cloud	4.3 a	3.7 b	3.6 b	3.7 b	***
9. Title encouragement	3.9 a	3.8 a	3.7 a	3.3 b	**
10. Hang it in my house	4.0 a	3.6 b	3.5 b	3.6 b	**
11. Preference	3.9 a	3.6 b	3.6 b	3.6 b	**
Viewing duration (s)	137.3 c	182.4 a	138.1 c	157.4 b	***

* p < 0.05, ** p < 0.01, and *** p < 0.001. ^{a b c} Duncan ex-post test grouping results. The original description of “meaning of cloud” is “the artworks make me feel the form or meaning of cloud”; the original description of “title encouragement” is “the title intensifies my interest in the painting”; and the original description of “hang it in my house” is “I would like to hang this painting in my house”.

presents the mean and ANOVA results for each variable. In terms of subjective feelings, except for the item “The work is original”, which did not exhibit statistically significant differences, all other questions exhibited statistical significance (p ≤ 0.05). Although Higuera-Trujillo, López-Tarruella Maldonado, and Llinares Millán [16] found that similar pleasure levels are involved in both physical and VR environments, we obtained different results. The scores for mood and “having access to all the works I like” were significantly higher for the conditions of physically viewing the original paintings and the album than for viewing them through the HTC Vive, which may be due to the limitations of the HMD. When using an HMD, users are unaware of the external environment and have concerns about moving in the physical environment, such as bumping into walls and tripping over signal transmission lines, making them less relaxed. The participants’ responses also indicated that the HMD has other limitations, as follows: it is fixed on the head, which causes discomfort; it weighs 700 g, which causes neck muscle fatigue; and the transmission line is connected to the computer host, resulting in limited movements. Higuera-Trujillo, López-Tarruella Maldonado, and Llinares Millán [16] also noted that owing to the limitations of the VR equipment and to ensure user safety, when using an HMD, users have less control than in a physical environment. Dozio, Marcolin, Scurati, Ulrich, Nonis, Vezzetti, Marsocci, La Rosa, and Ferrise [12] stated that any detail, asset, or interactivity within a VR environment influences the user experience. Therefore, compared with regular viewing methods, wherein users are not required to wear any equipment and can perceive the external environment, the subjective feelings induced by viewing through an HTC Vive are low.

The average and grouping results for “the title intensifies my interest in the painting” show that scores for viewing the original paintings (3.9 points), album (3.8 points), and through desktop VR (3.7 points) were significantly higher than that for viewing through the HTC Vive (3.3 points). Lin, Chen, and Lin [20] noted that the text size affects the reading experience in a VR environment. However, this study used an HMD and the participants could adjust the text size by moving closer to or away from the painting description board. Therefore, instead of the text size, the effect may be related to the manner in which the description board is viewed. In this study, the description boards for the original paintings and album were printed on paper, which does not offer a high resolution but also makes it easy to read the text even if the participants are not directly in front of the description boards. The desktop VR description boards were presented together with the paintings, which were conducive to the participants’ reading experience and mutual comparisons. However, while using HTC Vive, they were required to lower their heads and shift their gaze to directly in front of the description board to see the text clearly. This may have reduced their willingness to view the description board and shortened the reading time. Consequently, the description boards shown via the HTC Vive were not as effective as those in the other three viewing conditions.

Additionally, the viewing durations of the four painting presentation methods differed significantly (p < 0.001). Among them, the album involved the longest viewing time (182.4 s), followed by the HTC Vive (157.4 s), whereas the desktop VR and original paintings had the shortest viewing times of approximately 137.7 s. These differences may be attributed to the participants’ viewing behaviors. We found two differences in the viewing behaviors for the original paintings and album. First, when viewing an original painting, the participants spent more time looking at the painting itself and less on its description, whereas when viewing the album, they spent more time reading the descriptions. Second, the painting album provided a tactile sensation, and the participants could touch the paper with their fingers while browsing. By contrast, the longer viewing times with the HTC Vive may be attributed to the fact that the HMD prevents the participants from perceiving the external environment, resulting in an unconscious slow-down of their movements. Moreover, they moved slower owing to the transmission lines of the HMDs, thus increasing their overall viewing times. Therefore, the natural habit of reading books and the need to touch the paper extended the viewing times for the album, whereas those with the HTC Vive were longer because of the limitations of the HMD.

By contrast, the participants moved faster under the desktop VR conditions. After accessing the webpage, they could change the viewing direction using the mouse and click white arrows on the ground to move quickly in a particular direction. Additionally, they could click on the square pattern below the painting to display it and its description. In the physical environment, they could view the originals of the different paintings by walking back and forth in the exhibition space. Additionally, they could carefully see detailed information of a painting when they were close to it and a wide range of paintings when they were farther way. These factors shortened the overall viewing durations with the desktop VR and when physically viewing the original paintings.

3.2. Correlations among Measurements

The results of the Pearson’s correlation analysis are presented in

Table 2. Pearson correlation coefficients for significant values measured.

	1	2	3	4	5	6	7	8	9	10	11	12
1. Pleasure	1
2. Relaxed	0.66	1
3. Calmness	0.56	0.79	1
4. Having access to all works I like	0.42	0.51	0.51	1
5. Originality	0.37	0.38	0.38	0.39	1
6. Attractive	0.51	0.42	0.41	0.44	0.66	1
7. Valuable	0.54	0.40	0.44	0.44	0.60	0.74	1
8. Meaning of cloud	0.29	0.23	0.17	0.30	0.42	0.43	0.51	1
9. Title encouragement	0.39	0.34	0.37	0.31	0.43	0.40	0.52	0.23	1
10. Hang it in my house	0.42	0.39	0.38	0.37	0.36	0.50	0.52	0.36	0.28	1
11. Preference	0.47	0.35	0.33	0.26	0.44	0.54	0.47	0.32	0.33	0.62	1
12. Viewing duration												1

Items with high correlations (0.69 < r < 0.99) are indicated with a light purple background, those with moderate correlations (0.39 < r < 0.7) with a light red background, and those with low correlations (r < 0.4) with a light blue background. The original description of “meaning of cloud” is “the artworks make me feel the form or meaning of cloud”; the original description of “title encouragement” is “the title intensifies my interest in the painting”; and the original description of “hang it in my house” is “I would like to hang this painting in my house”.

. Except for the absence of a significant correlation between the viewing duration and other variables, all other variables showed significant positive correlations (p < 0.05 and above). To improve the readability, items with high correlations (0.69 < r < 0.99) are indicated with a light purple background, those with moderate correlations (0.39 < r < 0.7) with a light red background, and those with low correlations (r < 0.4) with a light blue background. There was a significant positive correlation between “pleasure”, “relaxed”, and “calmness” moods, indicating that they influenced each other.

The results also show that there were moderate positive correlations between the three moods and “the work is attractive” and “the work is valuable” items, indicating that the positive emotions generated while viewing also allowed for inducing a positive affirmation for the painting in the viewers. Simultaneously, the paintings imparted higher senses of value and were more attractive. Additionally, the degrees of preference for the paintings were significantly positively correlated with all items, except “viewing duration”. Among them, pleasant emotions while viewing; originality, value, and attractiveness of the paintings; and the willingness to hang them at home were highly relevant. These findings suggest that enhancing these feelings can increase the fondness for a painting. Additionally, no significant positive correlation between viewing duration and preference indicates that the participants did not necessarily look longer at a painting because they liked it, and the viewing duration may be longer owing to factors such as their inability to understand its content or evaluation of the painting skills.

3.3. Multiple Regression Analysis

For the multiple regression analysis, the respondent’s sex and 10 questionnaire items were used as the input variables, and preference was used as the response variable. A stepwise multiple regression analysis was used to establish a regression model of preference for the four presentation methods, and the following results were obtained:

• Original paintings:
  Preference = 2.23 + 0.41 (the work is attractive), adjusted R² = 0.36, p < 0.001

2.

Albums:

Preference = 0.20

                    +0.41 (I would like to hang this painting in my house)

                    +0.22 (the work is original)

                    +0.25 (mood = relaxed), adjusted R² = 0.53, p < 0.001

3.

Desktop VR:

Preference = 1.55

                    +0.39 (I would like to hang this painting in my house)

                    +0.19 (the work is original)

                    −0.35 (sex: male = 1, female = 0), adjusted R² = 0.51, p < 0.001

4.

HTC Vive:

Preference = 1.82

                    +0.39 (I would like to hang this painting in my house)

                    +0.15 (mood = pleasure), adjusted R² = 0.40, p < 0.001

The regression results indicate that when viewing the original painting, the preference was primarily affected by its attractiveness, whereas in the other three presentation methods, it was affected by the willingness of the participant to hang it at home. Thus, the preference for a painting is positively affected when it makes viewers feel that they want to hang it at home. Additionally, when paintings are presented through albums or desktop VR, viewers can discern their originality, which enhances their preference. Therefore, to promote preference, the presentation method of an HTC Vive must focus on the pleasure imparted by the viewing process.

3.4. Three Favorite Paintings

To further understand whether the participants’ preferences for paintings differed depending on the presentation method, they were asked to select their three favorite paintings in each experimental condition. The results show that among the original paintings, Paintings 5, 11, and 6 were the most popular as they were selected by 69, 46, and 37% of the participants, respectively. For the album condition, Paintings 11 (58%), 5 (54%), 6 (29%), and 3 (29%) were the most liked, whereas for the desktop VR condition, Paintings 11 (46%), 5 (42%), and 12 (38%) were the most liked. Finally, when viewing through an HTC Vive, Paintings 5 (60%), 6 (56%), and 12 (50%) were the most liked. These results indicate that nearly the same paintings were liked across all presentation methods. Therefore, the physical paintings liked by viewers were also liked when presented virtually.

As shown in

Table 3. Among the four painting presentation methods, the proportions of the three paintings selected as favorites that belong to landscape paintings or abstract paintings.

	Physical Painting		Virtual Painting
	Original	Album	Desktop VR	HTC Vive
Landscape paintings (No. 1, 2, 4, 6, 10, and 12)	23%	21%	27%	31%
Abstract paintings (No. 3, 5, 7, 8, 9, and 11)	27%	29%	23%	19%

, the 12 paintings were categorized as landscape (Nos. 1, 2, 4, 6, 10, and 12) or abstract (Nos. 3, 5, 7, 8, 9, and 11). The landscape and abstract paintings presented in the physical environments (originals and albums) were liked by 22% (8–37%) and 28% (6–69%) of the participants on average, respectively. However, when presented virtually (desktop VR and HTC Vive), 29% (19–56%) and 21% (8–60%) of the participants liked the landscape and abstract paintings, respectively. Thus, the landscape paintings were more popular than the abstract paintings in VR, whereas the abstract paintings were more popular in the physical environments. This finding is similar to those of Lin, Chen, and Lin [20], who also found that people like landscape paintings more than abstract ones in mobile VR environments. The reason for the lower popularity of abstract paintings in VR is related to the image quality. An abstract painting is composed of dots and lines, its composition has no context, and the resolution of its image impacts its viewing effect. Although this study used a screen and VR glasses with resolutions exceeding two million pixels, they could not sufficiently present the details of the abstract paintings. Therefore, using higher-resolution VR glasses may mitigate this issue.

4. Conclusions

This study explored the impact of four painting presentation methods, two physical (original paintings and album), and two virtual (desktop VR and HTC Vive) on the emotions induced by their viewing processes, painting evaluations and preferences, and viewing durations. The results indicate that physical paintings generate more positive emotions, such as pleasure, relaxation, and calmness, than virtual ones. The evaluation and preferences for originality, attractiveness, and value of the physical paintings were also higher than those for the virtual ones. Further analysis demonstrated that the landscape paintings were liked more in VR environments, whereas abstract paintings were liked more in physical environments. Moreover, the results of the correlation analysis indicate that, in addition to the paintings themselves, emotions and behavioral freedom during the viewing process, as well as the provision of additional information (such as titles and descriptions), affect viewers’ preferences. Collectively, the results of this study show that the viewing experience is not as good if the viewing method of a real environment is simulated using VR. Therefore, for presenting paintings in VR, we recommend adding multimedia or converting static paintings into dynamic images. Additionally, a wireless HMD should be used to minimize the weight, and a controller handle can be employed to move in the VR environment, instead of moving the entire body. Thus, VR can be leveraged to enhance painting appreciation.

This study had two limitations. Firstly, in order to avoid the impact of audience characteristics on the viewing experience, this study recruited young people of similar age as subjects. In future studies, the research team plans to assess the experiences of elderly individuals and other ethnic groups when viewing paintings in a VR environment. This will help to better understand the differences in how various audiences use VR systems. Second, to minimize the influence of the painting style on the viewing experience, this study used the same paintings with different viewing media. The paintings in this study were biased toward landscapes and abstract paintings, and paintings in other styles may produce different results.