Abstract
Studies found that genuine artworks viewed in a museum receive higher appreciation ratings than reproductions in the laboratory. Due to the mutual variation of context and genuineness, these studies were not able to disentangle these factors. A study designed by Brieber, Leder, and Nadal to systematically differentiate between these two variables did not find an effect of context or genuineness. To substantiate these results, we setup a conceptual replication by using the same 2 (museum/laboratory) by 2 (genuine/reproduction) between-subjects design with improved manipulations of context and genuineness. We found an effect of context, as artworks presented in a museum were liked more and rated more interesting than in the laboratory. We did not find effects of genuineness. Exploratively, we found that art style had a big impact on how artworks were rated regardless of context and genuineness, indicating that this may be a more important factor for aesthetic experience.
When thinking about the fascination people have with art, famous museums and highly priced single pieces come to mind. But, explaining why art really touches us seems difficult. Researchers suggest that other factors, besides the artwork itself, influence our art experience. Two such factors are context and genuineness, the location in which the artwork is presented, and the fact that in a museum we see the real, unique artwork. However, in empirical studies, up until now, these effects have often been confounded. Typically, genuine artworks are viewed in a museum context, while in the context of comparison, the laboratory, reproductions of artworks are presented. Therefore, it is one of the greatest challenges for research in this field to disentangle these two factors in order to examine the impact of them separately as well as combined.
This article aims to address this issue and investigates the effects of context and genuineness in a way that differentiates between these two variables. We first give a short review of the literature on context and genuineness and consequently discuss this study.
Context
Previous research has established that visiting an exhibition in a museum is generally pleasurable (Smith & Wolf, 1996). Museums are places where people take time to reflect and enjoy cultural heritage (Smith, 2014). Most research to date has investigated art experience of the different stimuli in only one type of context (Locher & Dolese, 2004; Locher, Smith, & Smith, 1999, 2001). This consequently does not allow for a direct assessment of the influence of context, for this one would need at least two different types of context. Some research has investigated the influence of museum context directly, finding enhanced aesthetic experience and higher ratings in aesthetic value, liking, and interest in museums (Brieber, Nadal, & Leder, 2015; Specker, Tinio, & van Elk, 2017) as compared with laboratories. In addition, people look longer at artworks in a museum context than in a laboratory context (Brieber, Nadal, Leder & Rosenberg, 2014). Artworks, as well as added information concerning them, were remembered better and longer when they were viewed in a museum context (Brieber, Nadal, et al., 2015; Specker et al., 2017). Together, the presented studies draw a consistent picture of the positive effect of museum contexts on art experience.
Genuineness
Different theories exist in the literature on how genuineness can impact art experience. The majority of the literature on genuineness pays particular attention to ideal aspects, for example, Benjamin (2010), when discussing the aura of an artwork as a standalone value. Seeing it from a different angle, Newman and Bloom (2012) explained the impact of genuineness through two aspects: An artwork is understood first as a result of a performance and second as a source of physical contact with the artist. Despite the theoretical importance of genuineness, empirical work is scarce, probably due to practical concerns such as limited access to genuine artworks, which leads to difficulties in running scientific experiments. Coping successfully with this challenge, Locher et al. (1999, 2001) showed that only a quarter of their stimuli set of genuine artworks was rated as more interesting and more pleasant than replications (slides and computer; Locher et al., 1999), and that only art-trained people rated genuine artworks differently from reproductions (Locher et al., 2001). Art-untrained people did not rate genuine artworks differently from their reproductions.
In reaction to these unexpected findings, Locher et al. (2001) proposed the process of facsimile accommodation as an explanation: If people are aware that the object they are evaluating is a reproduction and if this reproduction is of high quality, they will automatically adjust their evaluation of an artwork as if they were looking at the original artwork. This hypothesis may be an explanation for the lack of a genuineness effect in previous studies. Locher et al. (2001) attributed importance to the special role of reproduction quality because reproductions are typically of a lower quality than the original and have a reduced level of artistic finesse. Size, relief structures, deepness of color, and so on are just some artistic aspects that can for the most part not be reproduced with the media used. Therefore, it is an open question whether and how the quality of reproductions plays a role in the effects of genuineness.
To summarize, context and genuineness usually are highly cofounded variables in empirical research. Therefore, none of the discussed studies can clearly determine whether the found effects are due to context or to genuineness. The only study up to now that aimed to address this issue and disentangle these two factors is a study by Brieber, Leder, and Nadal (2015). They used a 2 (laboratory vs. gallery) by 2 (genuine vs. reproduction) between-subjects design. The study was conducted in two rooms of a gallery (gallery condition) and a laboratory room (laboratory condition). Contemporary photographs were used as stimuli, which were either hanged on the wall (genuine condition) or shown in a computer presentation (reproduction condition). Participants reported their experience in terms of liking, interest, arousal, valence, and understanding. In contrast to Brieber, Leder, et al.’s (2015) expectations, no main effects (either of context or genuineness) were found, nor did they find interaction effects. These findings are in strong contrast to the existing literature. This contrast creates a necessity of further research to gain a better understanding of the (lack of) effects of context and genuineness reported by Brieber, Leder, et al. (2015). This study aims to fill this need.
The present study
This study aims to assess the effect of context and the effect of genuineness further. To do this, we carried out a conceptual replication of Brieber, Leder, et al. (2015). We used the same 2 × 2 design to allow for comparability of the findings but improved the manipulation of context and genuineness, as we are critically questioning the effectiveness of their manipulation. In addition, we exploratively investigated the influence of art style on art experience. Next, we shortly discuss the improvements made and our motivation for them, first regarding context, then regarding genuineness.
Context
Brieber, Leder, et al. (2015) addressed the potential of an inverse white cube effect. The white cube effect refers to the fact that a museum context can transform any kind of ordinary object into an art object or artwork. However, the contrary may also occur: Not only can a museum context turn things into art but also artworks hanging in a random physical context can turn this context into a museum. Brieber, Leder, et al. (2015) later supposed that the photographs could have turned the laboratory into a gallery. So do we, and critically question the effectiveness of the context manipulation as well. Participants might have felt as if they were in a gallery context rather than in a laboratory. In this study, we used separated places that represented real museum or real laboratory contexts. As Brieber, Leder, et al. (2015) did, we made sure that both contexts were located in different buildings. In contrast to Brieber, Leder, et al. (2015), we additionally used a museum rather than a commercial gallery context. As a museum of art is an even more stereotypical art exhibition context than a commercial gallery, especially for laypeople, we aimed to enhance the manipulation of context. In addition, as a commercial gallery space is not only a place where art is exhibited but also a place where it is sold, it could be that these monetary aspects of the gallery had an influence on the effect. Using a museum space controlled for these potential confounds. Especially because the museum in which we conducted the study is a city museum, which has a nonprofit educational task and free entrance for all visitors. Furthermore, we kept room size, lighting, as well as distance to the artworks in both contexts constant, as these are well-known cofounding variables (Griswold, Mangione, & McDonnell, 2013, p. 352; Pelowski, Forster, Tinio, Scholl, & Leder, 2017).
This leads to improved control of confounding variables like physical characteristics or expectancy of the participants (Cupchik, 2002; Locher, Overbeeke & Wensveen, 2010; Scherer, 2005). The participants’ expectations were and were not fulfilled in either of the two conditions, as people expect genuine artworks in a museum and reproductions in a laboratory but not vice versa.
Genuineness
Brieber, Leder, et al. (2015) discussed weaknesses of the used medium (photography). Due to the widespread digital presentation of photographs, digital reproductions may have a similar level of assumed genuineness as photographic prints, especially by lay viewers. Therefore, in this study, we only used artworks painted on canvas in the genuine condition and high-quality pictures in the computer presentation for the reproduction condition. We postulated that, especially for a lay sample, paintings would be seen as more genuine than photographs (Dutton, 2009) and would crucially have a higher level of assumed genuineness than their digital reproductions.
For a satisfying comparison, various features have to be matched. As a central point, reproductions were shown in exactly the same size as the genuine artworks to exclude size as a confound (Pelowski et al., 2017). In keeping this feature constant between conditions, this study seems to be a first fair test of art in different contexts.
In addition, we did not provide an information about any of the artworks. In the study by Brieber, Leder, et al. (2015), additional information was given for some of the artworks. This information was mostly technical data about the photographs. This led Brieber, Leder et al. (2015) to suggest that perhaps the participants did not feel the artwork. They did not focus on the content of the artwork but more on the explained abstract aspects of the medium photography. As information about artworks is typically offered in a museum context, omitting information may seem rather unnatural in this context. However, we feel that the added gains of comparability and exclusion of potential effects of information outweighed the potential drawbacks.
As an explorative extension, this study investigates the impact of art style on art experience in direct comparison with context and genuineness. Effects of art style have been consistently found with reproductions in the laboratory (Leder, Gerger, Dressler, & Schabmann, 2012) as well as with genuine artworks in the museum (Mastandrea et al., 2018; Tröndle & Tschacher, 2012). But evidence is lacking that the effect of art style is not influenced by context or genuineness themselves. To analyze the impact of art style in this study, we showed five artworks in abstract style and five in figurative style.
In sum, these improvements to the experimental design aim to create the optimal conditions for finding the effects of context and genuineness. As Brieber, Leder, et al. (2015), we hypothesized that both context and genuineness positively affect art experiences. Specifically, we hypothesize that art experience is enhanced in a museum context (vs. a laboratory context) as well as when looking at a genuine artwork (vs. a reproduction).
Method
Participants
Our total sample consisted of 155 participants, aged between 18 and 49 years (mean age 24 years, 108 women). Psychology students from the University of Vienna were rewarded with course credits for their participation. Six participants had an educational background in art. In view of age, sex, and art education, the participants in this study and the one of Brieber, Leder, et al. (2015) were highly similar.
Materials
Context
There were two different types of context: A museum context (the MUSA: Museum Start Gallery Artothek, Vienna) and a laboratory context (a laboratory room of the University of Vienna). The MUSA was established by the City of Vienna’s Department of Culture and exhibits the contemporary art collection of the City of Vienna. The laboratory room is a typical laboratory space that is used for empirical experiments and as computer room on daily basis.
Genuineness
People either saw real artworks (paintings) hanging on the walls (genuine condition) or reproductions of these artworks in a power-point presentation on high-resolution monitors (reproduction condition).
Stimuli
Ten prerated artworks (five oil painting, three mixed technique, one tempera, and one gouache paint) were used; nine from the collection of MUSA and one from the private collection of the third author. To select paintings, we ran an online prestudy in which 20 paintings were rated regarding arousal, complexity, and valence. Based on this, we selected 10 artworks (5 figuratives and 5 abstracts). Furthermore, the 10 paintings were made by nine different artists to keep the influence of individual painting style and specific character of the artist low. In the reproduction condition, artworks were shown in their original size in order to control for a potential confounding effect of artwork size between the two different conditions. In both contexts, the artworks hung at the same height (upper image border—182 cm) and with the same distance to each other (center of the artwork—89 cm).
Art experience
To assess the experience of artworks, we measured arousal, valence, liking, interest, and understanding. Using these scales enables the measurement of affective (valence and arousal) and cognitive (liking, interest, and understanding) aspects of art experience and further enables the comparison with previous studies (Brieber, Leder, et al., 2015; Brieber, Nadal, et al., 2015; Brieber et al., 2014). The operationalization was as follows: Arousal (How aroused do you feel when looking at this artwork?), valence (How does this artwork make you feel?), liking (How much do you like this artwork?), interest (How interesting do you find this artwork?), and understanding (How much do you have a sense of understanding this artwork?). All were rated on 7-point Likert-type scales from 1 (very little) to 7 (a lot), except for valence (1 = very negative to 7 = very positive) and arousal (1 = very calm to 7 = very excited).
Art interest and art knowledge
As in Brieber, Leder, et al. (2015), we measured individual differences in art interest and art knowledge. These individual differences are variables that can play a measurable role in art experience (Brieber, Nadal, et al., 2015; Leder, Belke, Oeberst, & Augustin, 2004; Locher et al., 2001). We used the validated Vienna Art Interest Art Knowledge Questionnaire (Specker et al., 2018).
Procedure
As in the study of Brieber, Leder, et al. (2015), participants were asked to freely walk around to view the artworks hanging on the walls (genuine condition) or to take place to view the high-quality digital reproductions on the screen (reproduction condition). Due to the analog way of presenting artworks in the genuine condition, it was not possible to fully randomize the presentation order of the artworks. Therefore, we created three different artwork orders. Participants were assigned to one of the three orders based on the day of participation (order 1: 50 participants, order 2: 57 participants, and order 3: 48 participants). Participants viewed the artworks either in the MUSA (museum condition) or in the laboratory (laboratory condition). Figure 1 provides an overview of the experimental design. There were never more than two participants in the same room. Following the viewing period, participants were given written instructions and were asked to rate the artworks, and the scale order was randomized. The artworks were rated individually; participants of every condition were urged to return to the artwork while rating it in the various scales (by selecting the artwork in the presentation [genuine condition] or physically moving back to the artwork [museum condition]). Afterward, people filled out questionnaires assessing art interest and art knowledge as well as demographic questions.

Photographs of the experimental design (top-left: museum/reproduction, top-right: museum/genuine, bottom-left: laboratory/reproduction, and bottom-right: laboratory/genuine).
Results
Participants did not differ in art interest or art knowledge between conditions, art interest: F (3,151) = .07, p = .974; art knowledge: F (3,151) = .95, p = .416. In general, our sample had a moderate interest in art (M = 43.92 on the scale, with a maximum score of 77, standard deviation [SD] = 14.64) and low knowledge in art (M = 12.51 on the scale, with a maximum score of 26, SD = 5.76).
To analyze the effects of context and genuineness, we ran a multivariate analysis of covariance (MANCOVA) with context (museum/laboratory) and genuineness (genuine/reproduction) as independent variables, and the rating scales (measured arousal, valence, liking, interest, and understanding) as dependent variables. To ensure comparability with the findings of Brieber, Leder, et al. (2015), we included art interest as a covariate. Means and SDs divided into condition are given in Table 1.
Mean (and Standard Deviations) in Art Experience Divided by Conditions.
There was a main effect of context, explaining 11% of variance in the model, Wilks’ Λ = .89, F(5, 146) = 3.56, p = .005,

Mean liking rating shown for context and genuineness, with error bars.

Mean interest rating shown for context and genuineness, with error bars.
Nonetheless, there was no main effect of genuineness, Wilks’ Λ = .96, F(5, 146) = 1.31, p = .265,
To test for the robustness of the effect, we conducted follow-up analyses. In a first step, we computed a multivariate analysis of variance without art interest as covariate. This was done because, in contrast with Brieber, Leder, et al. (2015), we did not observe a difference in art interest between conditions. This analysis did not change the reported effects. Second, we repeated the initial MANCOVA with exclusion of one artwork that was not presented in the correct size in the reproduction condition; our results did not change when this artwork was excluded.
Explorative Analysis: Art Style
In addition, we wanted to explore the effect of art style. By design, we included both abstract and figurative artworks to enable us to investigate the role of style in art experience in different conditions of art presentation (museum/laboratory as well as genuine/reproduction). To analyze the role of style, we used a mixed-effect MANCOVA with art style as a within-subject factor, context (museum/laboratory) and genuineness (genuine/reproduction) as between-subjects factors, and the rating scales (measured arousal, valence, liking, interest, and understanding) as dependent variables. To ensure comparability, we included art interest as a covariate.
There was a main effect of art style, explaining 44% of variance in the model, Wilks’ Λ = .57, F(5, 146) = 22.46, p < .001,
No interaction was observed between art style and context or genuineness.
Discussion
In contrast with the study of Brieber, Leder, et al. (2015), but in line with the previous literature (Brieber, Nadal, et al., 2015; Brieber et al., 2014; Locher & Dolese, 2004; Locher et al., 1999, 2001), this study found an effect of context on art experience. Our main finding was that people found artworks more interesting and liked artworks more in a museum context than in a laboratory context.
This suggests that the improvements made regarding the manipulation of context were effective. We worked in a real museum, instead of a commercial gallery, to control for cofounding variables like physical characteristics or expectancy of the participants (Cupchik, 2002; Locher et al., 2010; Scherer, 2005). This shows that for the effects of context to occur, the manipulation of context should be strong enough. When a museum just does not feel like a museum, or more applicable to the discussion of Brieber, Leder, et al. (2015), when participants still have the sense of being in a gallery context rather than in a real laboratory (inverse-with-cube effect), the effect will not appear.
In line with the study by Brieber, Leder, et al. (2015), we did not find an effect of genuineness. This differs from assumptions regarding the positive effect of genuineness, as seeing a genuine artwork is known as one of the main motivations, which drives people to visit a museum in the first place (Mastandrea, Bartoli, & Bove, 2007, 2009). In this study, we aimed to enhance the manipulation of genuineness by using artworks painted on canvas rather than photographs under the assumption that this difference in medium would highlight the genuineness (or nongenuineness) of the artworks.
One possibility is that our manipulation was still not effective. A potential explanation for this is that a potential effect of genuineness is subverted by the high quality of the reproductions. The reproductions were professional photographs shown on a high-resolution 4k monitor with 40-in. diagonal size. Due to the presentation of all artworks in their original size, presentation sometimes took up the full size of the screen. This high quality may have therefore reduced the difference between seeing the reproduction or the genuine or even artificially increased the pleasure in the art experience of the reproductions.
Another possibility for the lack of an effect is the facsimile-accommodation hypothesis as discussed in the introduction of this article. Locher et al. (2001) argue that if people are aware that they are looking at a reproduction rather than a genuine artwork, they will attempt to adjust their experience accordingly. This hypothesis may be an explanation of the lack of a genuineness effect in general but may be particularly relevant in our case. In this study, participants in the reproduction condition reported that they assumed the artworks to be larger in real life than presented on the screen, even though they were explicitly told that this was not the case and that the presented artworks were shown in their original size. This implicit assumption was probably created by the fact that reproductions are usually smaller than genuine artworks. Due to monitor size, one would need a pretty large monitor to be able to show a relatively small canvas artwork in its original size, leading to reproductions consistently being presented as smaller. This is the case not only for digital reproductions but can also be observed in analog reproductions such as postcards. However, if the facsimile accommodation is correct, our participants will have tried to accommodate for the fact that they were seeing a reproduction, specifically, participants imagined the artworks larger. This fault in accommodation (the genuine artworks were actually not larger in real life) may have obscured the effects of genuineness. This is possible, as artwork size can influence the art experience (Pelowski et al., 2017).
Finally, the lack of an effect of genuineness may be due to our sample consisting mainly of laypeople. The results of Locher et al. (2001) suggest that only art-trained people rated genuine artworks differently than reproductions. It may be the case that genuineness matters for people with a high knowledge of art but not for laymen. If this is the case, then any effect of genuineness can only be found in an art-trained sample.
Nonetheless, it is worth asking if this potential effect disappears when genuine artworks are compared with high-quality reproductions. It is likely that trained observers pick up on subtle visual cues (such as brushstrokes) that are in genuine artworks but not normally visible in low-quality reproductions due to the medium or limitations in technical aspects (e.g., display resolution). The presence (or lack) of these visual cues may be the underlying cause of the effects of genuineness. If this is the case, progress in technology and the digital display of images will lead to the ability to create reproductions that have the same subtle visual cues as the genuine artwork. This technical progress has the potential to change the way we experience art.
Furthermore, it is possible that effects of genuineness depend on the artistic quality of the artwork. In the case of famous artworks, such as the Mona Lisa, people will have explicit expectations, knowledge-based emotions, or other valuations, which make them differently and especially valued by people when seeing them in real life. As our sample of artworks included neither well-known artworks nor well-known artists, it may be that the effects of genuineness only come into play when people are looking at famous artworks or artworks by famous artists.
Nonetheless, it is crucial to discuss the possibility of the absence of a genuineness effect, in general, which is suggested by the findings of Brieber, Leder, et al. (2015) and this present study. Perhaps previous genuineness effects may have been caused by confounds.
As a new finding and extension of the previous literature, our study showed a strong effect of art style. Art style explained 44% of variance in the model, illustrating the importance of this factor. Interestingly, style did not interact with conditions of context or genuineness. This result indicates that style effects are present regardless of context or genuineness. This suggests that style effects previously found in a laboratory setting using reproductions are generalizable to people engaging with genuine artworks in a museum. This knowledge supports further investigations on style in laboratory designs. An interesting direction for the future would be to assess how important art style is for people with training in art (Leder & Nadal, 2014). Laypeople tend to consistently rate abstract art lower than figurative art on a variety of measures (Locher et al., 2001), which may result in art style being a more important aspect to laypeople than to experts, especially because art style mainly influences to what extent people understand the artworks. Figurative art may be easier to understand for laypeople because the representation of figurative scenes can be easier linked to scenes of everyday life (see “Cognitive Mastering” in Leder et al., 2004). Due to their low level of knowledge about art, the link between art and everyday life may act as an important resource for laypeople to understand art. In contrast, it seems likely that people with art training do not need to rely on figurative content as much, as their higher level of knowledge of art can act as an alternative resource for understanding art.
In sum, our study emphasizes the importance of art style in art experience. In addition, it provides further insight into the (lack of) effects of genuineness of artworks. Finally, it supports previous research by replicating a positive effect of the museum context. Our results show that artworks are liked more and are seen as more interesting in a museum context rather than a laboratory context. These findings show the importance of a museum as institution and physical space that create an optimal environment for the experience of art.
Footnotes
Acknowledgments
The authors thank the MUSA for allowing the use of its exhibition space, genuine artworks, as well as their professional photographs as reproductions. A special thanks goes to the former director of the MUSA Dr. Berthold Ecker and the photographer Heimo Watzlik for their support and highly valuable expert advices. Finally, the authors would like to thank the anonymous reviewers for their constructive comments and suggestions that helped to improve this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
