The powerful artist : the effect of context on influence inferences about painters

The Powerful Artist: The Effect of Context on Influence Inferences about

Painters

Abstract

Paintings tell a story, but they can also tell us about the painter. Here, we investigated how the style of a context wherein a painting is presented influences how people think about its artist. Whereas norm transgressions have previously been researched to explain a contrast in style between context and focal paintings, we proposed prototypicality judgments as a factor in perceiving this contrast. After implicitly creating a new prototype by exposure to four same style context paintings, a different style focal painting should appear less prototypical. We expect this non-prototypical painting to signal higher eccentricity and volitional capacity for the artist, which will lead to higher inferred influence. 96 Psychology students participated. After they were exposed to the context paintings, we presented the focal painting. We manipulated context and focal painting styles to either coincide or differentiate (contrast). Measurements after manipulation included influence, personality factors, prototypicality judgments, and norm transgression. The artist of the same focal painting was inferred as more influential in case of a contrast. The painter did not appear more eccentric whenever there was a contrast. Eccentricity did not mediate the relationship between condition and influence. Prototypicality ratings did differ according to context, but in a direction contrary to our hypothesis. A contrast increased influence of the artist. Eccentricity did not explain this relationship.

The Powerful Artist: The Effect of Context on Influence Inferences about Painters A few people are born into a power position, while most others have to earn it. The former has been happening for ages; from a child that is born in an influential family in the Roman Empire to George W. Bush being born as the son of ex-president George Bush sr. Apart from the lucky few, most have to work hard to achieve a legitimate power position. Once achieved, people can infer power of the beholder through a job title mentioned on their LinkedIn profile or business card. However, it appears that one does not need innate nor legitimate power to be considered powerful. Easily performable norm transgressive behaviors, like putting one’s feet on a table, can count as cues that signal power (Van Kleef, Homan, Finkenauer, Gündemir, & Stamkou, 2011). The norm transgressor is seen to act according to his or her own volition.

To see how far this effect stretches we can wonder if these cues can be signaled by overt behavior only. Or, could other constituents of behavior count as cues that signal power as well? The focus of this study lies on a product of behavior: paintings. Artists express themselves through their art. The painting not only displays the object that an artist wants to showcase, but the canvas portrays the artist’s personality as well. The constant changes in art styles were instigated by artists doing what they felt like doing. Without artists doing so, one might imagine the art field to have stayed the same as has been the norm for decades

(Alvarez, Mazza, Pedersen, & Svejenova, 2005). We aim to uncover the mechanisms that lead to awarding power to artists, when cues of norm transgressive behavior can only be inferred through a contrast of their art to the context of other artists’ art.

In a recent study, Stamkou et al. (in preparation) presented participants with different paintings of the same style in order to instill this style as the norm. Participants’ inferences about the painter of the painting of interest (focal painting) were expected to vary as a function of the newly instilled normative painting style (context paintings). Painters were

perceived to be more influential when there was a contrast between the context painting style and the focal painting style. However, it is questionable whether this unnatural exposure during the experiment actually instilled new norms. People may already possess steady norms about how paintings should be made, which could be constructed through naturalistic

exposure (Cialdini & Trost, 1998; Cutting, 2003). This norm may be especially strong for people living in countries that have a rich artistic culture, such as The Netherlands.

To find an alternative to predicting how the context influences the power awarded to artists, one might uncover this closer at the start of the perceptual experience. Processing one object in comparison to a category (e.g. the category of the context) has been found to activate brain areas that are involved in early visual processing (Reber, Stark, & Squire, 1998). The model of aesthetic appreciation and aesthetic judgments (Leder, Belke, Oeberst, & Augustin, 2004) outlines several stages that are passed through during an aesthetic experience. The first two early processing stages consist of the perceptual analysis followed by the implicit

memory integration. Prototypicality judgments occur during the implicit memory integration

stage. Leder et al. (2004) define prototypicality as “the amount to which an object is representative of a class of objects” (p. 496). For new information that is encountered the prototype serves as an anchor (Cottam, Dietz-Uhler, Mastors, & Preston, 2010).

A new prototype can be created through mere exposure without participants having to be consciously aware of this (Reber et al., 1998). Prototype learning can occur with certain dot patterns (Aizenstein et al., 2000; Winkielman, Halberstadt, Fazendeiro, & Catty, 2006), grammatical rules (Newell & Bright, 2001; Gordon & Holyaok, 1983; experiment 1) and complex visual patterns (Gordon & Holyaok, 1983; experiment 2). In the current research we are trying to extend the effect of prototype learning to visual patterns even more complex: paintings.

The norm is analogous to a prototype in the way that it is a standard to which other instances can be compared. Can this new prototype be the basis against which the focal painting is judged? Instead of creating a norm by showing context paintings in a certain style, a new prototype of this style could be implicitly integrated in the participants’ memory through mere exposure. The new prototype serves as an anchor for new instances. When a next instance is not complying with the rules that construct the prototype, processing speed will be slowed down (Buchner, 1994) and more effort will be required in order to process it (Reber et al., 1998).

In Western cultures someone or something that stands out is positively valued (Kim & Markus, 1999). In the case of paintings, uniqueness of the artist could be signaled if the focal painting deviates from the prototype that has been created by the context. Standing out by being non-prototypical might activate the artist as being an unique individual, doing things his or her own way. If an artist is unique, people could perceive him or her to be eccentric.

Perceived eccentricity in artists activates the stereotype of artists behaving in unconventional ways and deviating from the norm (Van Tilburg & Igou, 2014). This perceived eccentricity of artists was found to influence evaluations of their art. This has only yet been found to apply to unconventional instead of conventional sculptures. If one’s painting is non-prototypical due to the contrast in style to the context, this could lead to the painter being perceived as the

stereotypical eccentric artist. Through the contrasting painting the eccentricity stereotype is activated, illustrating an eccentric artist that follows his or her own conscience without being constrained by the norm. This eccentric artist will be considered more powerful and able to influence others.

During the experiment we exposed participants to context paintings that complied to a certain art style in order to create a new prototype. Directly hereafter, we presented

expected the focal painting to be judged by the newly acquired prototype. A different style focal painting does not adhere to the rules of the acquired prototype, hence it was expected to be processed less fluently and rated less prototypical. The non-prototypical focal painting will increase how eccentric the artist is perceived. Eccentric artists will be considered more

influential.

Method Participants

One hundred and eight participants (Mage=20.40; SD=3.42; 74 females) participated in

our experiment. Participants consisted of students recruited from the Psychology Department of the University of Amsterdam. They received course credit for participating. All participants were randomly assigned to one of four conditions of a 2 (Focal painting style: Figurative vs. Abstract) X 2 (Context painting style: Figurative vs. Abstract) between-subjects factorial design.

Materials

Participants were successively exposed to five paintings on a computer screen through the online program Qualtrics. The prototype was created by exposing participants to the first four paintings (context paintings) of the same artistic style, either figurative or abstract. The fifth and last painting (focal painting) was also either figurative or abstract. We manipulated to focal paintings to be of either the same (focal-same-as-context conditions) or different style than the context paintings (focal-different-from-context conditions) (example in Figure A1, see Appendix A). In order to prevent disrupted prototype learning all figurative paintings followed the realism style and all abstract paintings followed the cubist style. The paintings that we used were retrieved from http://www.wikiart.org and http://wikipedia.org. All

paintings were portraits of males made in the late 19th or 20th century. We edited the artworks to be black and white, with the figure facing left. They were cropped so that the upper part of the shoulders and the entire head were shown. Further selection of paintings was based on these exclusion criteria: (a) young adults or elderly people, (b) facial hair, (c) headwear, (d) long or short hair (baldness), (e) extremely attractive or unattractive facial features, and (f) unrecognizability of the figure. Based on our pilot study we selected five realistic and five cubist paintings that were found to not stand out on various factors (e.g. prototypicality, transgression) from a larger pool of seven realistic and seven cubist paintings. The final collection can be found in Figure A2 (see Appendix A).

Measures

After the participants were exposed to one of four conditions, the dependent variables were measured. Items focused on the focal painting and were answered using a seven point Likert-scale varying from 1 (totally disagree) to 7 (totally agree) (unless mentioned

otherwise). Power has been used as an umbrella term. Since powerful individuals are able to give reward and punishment to others (Keltner, Gruenfeld, & Anderson, 2003), these

individuals have the capacity to influence other people’s behavior through material or social means. Influence will be used as a determinant of power awarded. Influence was measured with an influence inference scale, consisting of eight items (e.g. ‘I think this artist may have started an artistic movement?’). Two supplementary measurements of impact were added. These were a purchase intentions scale of three items (e.g. ‘Would you buy a book about this artist?’) and a preference count, asking participants to select which of the presented paintings they would want to have as a poster. To measure prototypicality judgments, we presented three prototypicality items (e.g. ‘To what extent is the depicted figure an accurate

Wieringen (1990) and Rosch (1975) and were translated to apply to our Dutch participants. In previous research, the item used by Rosch was found to have a split half correlation of .97. Additionally, we measured processing fluency by timing how many seconds participants spent looking at the paintings. Finally, four attraction perception items (e.g. ‘I think this artwork is ugly vs. beautiful’), four differentiation perception items (e.g. ‘I think the last painting is unique compared to the rest of the paintings?’), three transgression perception items (e.g. ‘I think this artwork is traditional vs. progressive’), four competence perception items (e.g. ‘I think the artist is competent’), six volition inferences items (e.g. ‘I think that the artist feels free to do what he wants?’), one complexity perception (‘I think this artwork is easy vs. hard to understand’) and four perceived eccentricity items (e.g. ‘I think the artist is eccentric?’) were presented. Lastly, demographics were reported and art familiarity was measured with three items (e.g. ‘Do you like visiting museums?’).

Procedure

Participants entered the experiment online through their own computers. They were redirected to the Qualtrics test page. Participants read the information brochure and provided informed consent. Instructions were brief, asking the participants to look at the different paintings that they will be presented with, which were made by different artists in the first part of the 20th century. They were told to press the ‘Next’ button as soon as they had taken a good look at the painting and that the questions that will follow concern one of the presented

paintings. Qualtrics randomly assigned each participant to one of four conditions. After exposure to the paintings, responses on the previously mentioned scales were collected. After complementary parts of the experiment, demographic and art familiarity items were

presented. Finally, participants entered their email address for debriefing and they were thanked for participating. The entire experiment took about 25 minutes.

Results

The initial dataset contained responses of 113 participants. We directly excluded five of these responses because the participants were not assigned to a condition by Qualtrics. These participants either quit the program upon opening or they did not agree with the instructions. Of the resulting 108 participants, data from 12 participants were excluded from the analyses because they did not completely finish the part of the experiment relevant for this study, they participated on their smartphone or they did not pass our attention measure. The final dataset contained responses of 96 participants.

Because of the insignificant MANOVA on the variables related to individual differences we were able to conclude that there were no significant differences between conditions on socio-economic status, individualistic-collectivistic values and art familiarity.

Factor analyses revealed that all but one item loaded on their intended scales. This item was the complexity item, which loaded on the transgression scale. Since removing it would decrease the scale’s Alpha, we decided to retain this item. After reverse scoring the appropriate items, we calculated the remaining reliabilities. All scales had a Cronbach’s Alpha of .72 or higher, implying good internal consistency. We decided to delete two items. For conceptual reasons the item measuring how charismatic the painter is perceived was deleted from the influence scale. The scale’s Alpha slightly increased to .92. For conceptual and reliability related reasons the item measuring how normal the painter was perceived to be was deleted from the eccentricity scale. The Alpha improved from .80 to .90. Excluding any further items would not significantly improve the reliabilities.

To check whether or not participants were able to successfully differentiate between the focal painting and the context paintings, we conducted a two-way independent ANOVA on the differentiation scale with focal painting style and context painting style as the between-subjects factors. A main effect of context painting style was found, F(1, 92)=8.39, p=.005,

ηp2=.08. More importantly, the interaction effect was significant, F(1, 92)=176.68, p<.001,

ηp2=.66. Focal paintings were perceived to be more different from the context whenever they

were presented in a focal-different-from-context than in a focal-same-as-context condition. Simple main effects showed that these differences were significant, such that the abstract focal was perceived significantly more different in a figurative context (M=6.04, SD=0.89) than in an abstract context (M=2.73, SD=0.89), F(1, 92)=134.06, p<.001, ηp2=.48. The

figurative focal was perceived significantly more different in an abstract context (M=5.41,

SD=0.86) than in a figurative context (M=3.28, SD=1.33), F(1, 92)=52.71, p<.001, ηp2=.19.

All the following analyses were run using the aforementioned selection of

participants’ responses and scales. The means, standard deviations and test statistics for two of our measurements of impact, namely inferred influence and purchase intentions, and our personality variables are presented in Table B1 (see Appendix B).

We performed a two-way independent ANOVA on inferred influence and purchase intentions with focal painting style and context painting style as the between-subjects factors. A main effect of focal painting style was found on both inferred influence and purchase intentions. Most importantly, the interaction between focal and context painting style was significant for both variables. Focal paintings had a higher impact when they were presented in a focal-different-from-context condition than in a focal-same-as-context condition.Simple main effects on inferred influence showed that only the painter of the abstract focal painting appeared significantly more influential in a different style context than in a same style

context, F(1, 92)=10.97, p=.001, ηp2=.08. Simple main effects on purchase intentions showed

that only for the figurative focal painting the purchase intentions were significantly higher, but just marginally, in a different style context than in a same style context, F(1, 92)=3.54,

To analyze the impact variable preference a Chi-Square test was performed. Condition was significantly associated with preference, χ2(12)=62.60, p<.001. The expected count for the figurative focal different-from-context painting was 5.7, but it was preferred 0 times. The expected count for the figurative focal same-as-context painting was 5, but it was chosen 2 times. They were preferred 0% and 9.1% within their conditions, respectively. The abstract focal painting was preferred more often, with the focal-different-from-context being preferred 15 times compared to its expected count of 5.7. The abstract focal-same-as-context, which was expected to be preferred 5.5 times, was chosen 5 times. Within their conditions they were preferred 60% and 20.8%, respectively. The abstract painting was preferred and a contrast in style could increase the preference only for this painting.

To check whether the context was able to make both of our paintings’ styles appear as more norm transgressive, another two-way independent ANOVA on transgression perception was conducted. A main effect of focal painting style emerged, F(1, 92)=196.73, p<.001, ηp2=.68. The abstract focal painting (M=5.15, SD=0.75) was rated more transgressive than the

figurative focal painting (M=2.77, SD=0.92). The interaction was non-significant. Unexpectedly, there was a main effect of context painting style as well, F(1, 92)=5.18,

p=.025, ηp2=.05. Whenever the focal painting was presented among a figurative context this

painting was considered to be more transgressive (M=4.23, SD=1.45) than among an abstract context (M=3.75, SD=1.45). However, a contrast in style was not able to make either one painting style appear as more norm transgressive.

From a two-way independent ANOVA on prototypicality a main effect of focal painting style emerged, F(1, 92)=462.52, p<.001, ηp2=.83. The abstract focal painting

(M=2.47, SD=0.89) was considered less prototypical than the figurative focal painting (M=6.03, SD=0.74). The interaction term was marginally significant as well, F(1, 92)=2.87,

prototypical whenever they were presented in a focal-same-as-context condition than in a focal-different-from-context condition. Simple main effects revealed that only the figurative focal painting appeared significantly less prototypical when it was presented in the focal-same-as-context condition (M=5.76, SD=0.80) than in the focal-different-from-context condition, (M=6.27, SD=0.59), F(1, 92)=9.89, p=.002, ηp2=.02. A contrast made the focal

painting appear more prototypical instead of less prototypical. The context did not create the prototype which served as the anchor upon which the prototypicality judgment was based.

The marginally significant main effect of focal painting style on the processing fluency timer, which appeared from a two-way independent ANOVA, partially complement the main effect found on prototypicality ratings, F(1, 92)=3.01, p=.086, ηp2=.03. Participants

spent a longer time looking at the abstract painting (M=7.32, SD=6.76), which was rated less prototypical, than at the figurative painting (M=5.24, SD=4.94). Since we found no interaction effect, a contrast did not lead to the focal painting being processed less fluently.

For each of our three personality scales two-way independent ANOVA’s were carried out, of which the means, standard deviations and statistics are provided in Table B1 (see Appendix B). We found a main effect of focal painting style on eccentricity ratings. The artist of the abstract painting (M=5.21, SD=1.15) was considered more eccentric than the artist of the figurative painting (M=4.09, SD=1.38). However, focal and context painting style showed no significant interaction. A contrast in style did not lead to the focal painter being considered more eccentric. However unexpectedly, there was a main effect of context painting style as well. Whenever the focal painting was presented among a figurative context its painter was perceived to be more eccentric (M=5.07, SD=1.45) than when it was presented in an abstract context (M=4.27, SD=1.20).

Regarding artists’ volition inferences we found a main effect of focal painting style. Most important was the interaction between focal painting and context painting style. Painters

were perceived to act more according to their own volition when their art was presented in a focal-different-from-context condition than in a focal-same-as-context condition. The

significant simple main effect for the abstract focal painting confirmed that this painter is seen to act significantly more according to his or her volition when his or her painting was

presented in a figurative context than in an abstract context, F(1, 92)=6.91, p=.010, ηp2=.05.

The marginally significant simple main effect for the figurative focal painting confirmed that this painter is seen to act significantly more according to his or her volition when his or her painting was presented in an abstract context than in a figurative context, F(1, 92)=3.35,

p=.070, ηp2=.02.

Competence showed a significant interaction between focal painting and context painting style as well. It again appeared that focal painters were perceived to be more

competent when their art was presented in a focal-different-from-context condition. However, the simple main effects were non-significant.

Since we were able to confirm the interacting effect of focal and context painting style on our two impact measurements, we set out to test if these effects were mediated by our personality scales. Neither eccentricity nor competence led to significant indirect effects. Volition appeared as a significant mediator in the relationship between the conditions and both inferred influence as purchase intentions (Figure B1 in Appendix B). The schematic presentation was based on Mast, Jonas, and Hall (2009). Sobel tests for inferred influence and purchase intentions were significant, Z=2.41, p=.02 and Z=2.26, p=.02, respectively. Since both direct effects became insignificant, we were able to conclude that volition inferences fully mediated the relationship between the conditions and both inferred influence and purchase intentions.

Discussion

In this study we set out to test if the product of someone’s behavior presented in different contexts could lead to differences in inferences about that person. Specifically, we looked at how a contrast between the style of a focal painting and the style of the context paintings affected how influential people think the focal painter is. Differences in inferred influence varied as a function of whether or not there was a contrast in style, without this contrast necessarily being considered more norm transgressive. The painter appeared more influential whenever his or her painting was presented in a different style context. Eccentricity did not explain how a contrast in style caused these differences in inferred influence. Volition inferences could account for this relationship. Prototypicality ratings differed as a function of the contrast. The direction of this effect was contrary to our hypothesis. Prototypicality judgments were not based on the prototypes created by exposure to the context. It is improbable that the proposed prototypicality judgment was at the perceptual basis of the effect of contrast on inferred influence of painters.

It appears that solely a contrast to a context is enough for people to think of the artist as more influential, replicating findings of Stamkou et al. (in preparation). This contrast effect was not attributable to an increased perceived norm transgression. A contrast did not make either one of the focal paintings’ styles appear as more transgressive. Transgression ratings pointed in the same direction whenever the context was changed. Apparently, our participants had a clear idea about art, which includes abstract art as transgressing some sort of art norm. It even appeared that presenting art in a figurative context makes a perceived norm transgression more apparent for both focal art styles. Nevertheless, the main effects on inferred influence and purchase intentions do agree with the fact that the more transgressive abstract art was perceived to be, the higher its impact.

Prototypicality judgments were not based on perception of the contrast we created between the newly acquired prototypes and the focal painting. Comparable to our

transgression ratings, it seemed that participants had clear ideas about prototypicality, which our context was not able to change. Although we expected that diverging from either of our context styles would make the focal painting appear as less prototypical, our findings contradict this: a different style context actually caused the focal painting to be perceived as more prototypical. While new prototypes seem to be open to unconscious learning through exposure (e.g. Reber et al., 1998), it appears that the prototypes we tried to create by the context during the implicit memory integration stage were not strong enough to serve as the anchor against which the focal painting was judged. The model of aesthetic appreciation and aesthetic judgment proposes that prototypicality judgments influence processing of art (Leder et al., 2004). However, it seems that our exposure was not strong enough to influence existing prototypes. Because one is exposed to male figures on a daily basis, it might be that this prototype is not easily altered. These results might very well be attributed to our three

prototypicality items. These items focused on absolute prototypicality, referring to how much the depicted figure looks like the prototype of a person that people already possess. This prototype is more easily visible in the figurative paintings, in which a male face can be more easily distinguished. Items tapping in to prototypicality of the focal painting style relative to the context painting style could be a powerful addition for future research.

Nevertheless, it is worth mentioning that the processing fluency timer points in another direction. In accordance with the prototypicality ratings, participants processed the more prototypical figurative painting more fluently than the abstract painting. However, the timer suggests that participants looked for a longer amount of time at the focal painting if it was different from the context painting style. This implies that these paintings were less fluently processed. Yet, with our sample this effect did not reach significance. For future

research other sophisticated time measurements, like measuring time to recognize the figure (Hekkert & Wieringen, 1990), could lead to significant results.

Eccentricity ratings were not increased by a contrast in style between the focal and context paintings. The artist of the abstract painting was perceived to be more eccentric than the artist of the figurative painting. A contrast in style between the context and focal painting could only make the abstract artist appear more eccentric, meaning that the effect of a contrast did not count for both focal painting styles. This extends the eccentricity findings on

unconventional versus conventional sculptures by Van Tilburg and Igou (2014) to paintings. Overall, the contrast did not signal the artist’s eccentricity that we hypothesized and this variable was not able to mediate the relationship between our conditions and influence.

Volition of the artist did however mediate this relationship. A contrast in style made the painter appear to act more according to his or her own volition. Painters that appeared to act more according to their own volition were considered more influential. Mediation analysis results are comparable to those found by Van Kleef et al. (2011). This study does add a different flavor to these results, suggesting that it is not necessarily a norm transgression that leads to someone being perceived to act volitionally and hence be perceived more powerful and influential, but that merely a contrast in style can lead to higher volition and influence ratings as well.

Important limitations of this research were first of all the relatively small number of participants. This could have been one of the reasons for our non-significant effects where we expected significant results and re-occurring marginal effects. Additionally, we used paintings as our form of art. Our findings might be specific to paintings, but not to art in general. Van Tilburg and Igou (2014) found evidence contradictory to the preference for abstract art we found, concluding that conventional instead of unconventional sculptures are preferred.

Finally, Qualtrics would not allow counterbalancing the order of the context paintings. Some effects could be due to the specific contrast of the fourth context painting to the focal painting.

Pointers for future research include a focus on exposing participants to more than four context paintings. Exposure to four paintings could have been an amount too small to create an influential new prototype. Presentation of these paintings should also be counterbalanced. Additionally, it would be worthwhile to include participants from countries where art is not emphasized as much as in the Netherlands. Such ‘blank slate’ participants will probably have a lower expertise regarding art, which could make them more open to prototype creation in the lab (Hekkert & Wieringen, 1990). Because it is less likely for abstract art that a hardwired prototype exists (Hekkert & Wieringen, 1990), only abstract art could also be used for

conceptual replication purposes. A contrast can then be created by using paintings from different abstract styles, like cubist style versus fauvism.

Altogether, a painting in a contrasting style to an artificially created context made an artist appear to be more influential. Generally, perception of art does not occur in a vacuum and the context can change just how art and the artist is evaluated. Hence, our findings can be useful for artists. Decisions about what art style to follow, or even following your own new art style, will influence how you will be perceived by the public. Museums and auction houses are also to pay attention to the style of the context they present their art in. It is worth thinking about context that other artworks up for auction provide, when you consider record-breaking art sales like the 176 million dollars sale of Picasso’s Women of Algiers (Hickey, 2015). On this scale, context can have such extreme influences as to make or break artists and auction houses. How prototypicality is related to this contrast and why solely a contrast can lead to higher volition inferences is still unknown, but prototypicality remains an interesting variable for future research.

References

Aizenstein, H. J., Macdonald, A. W., Stenger, V. A., Nebes, R. D., Larson, J. K., Ursu, S., & Carter, C. S. (2000). Complementary Category Learning Systems Identified Using Event-Related Functional MRI. Journal of Cognitive Neuroscience, 12(6), 977-987. Retrieved from http://www.mitpressjournals.org/loi/jocn

Alvarez, J. L., Mazza, C., Pedersen, J. S., & Svejenova, S. (2005). Shielding Idiosyncrasy from Isomorphic Pressures: Towards Optimal Distinctiveness in European

Filmmaking. Organization, 12(6), 863-888. doi: 10.1177/1350508405057474

Buchner, A. (1994). Indirect Effects of Synthetic Grammar Learning in an Identification Task. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(3), 550-566. doi: 10.1037/0278-7393.20.3.550

Cialdini, R. B., & Trost, M. R. (1998). Social Influence: Social Norms, Conformity, and Compliance. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of social

psychology (4th ed., Vol. 2, pp. 151-192). Boston: McGraw-Hill.

Cottam, M. L., Dietz-Uhler, B., Mastors, E., & Preston, T (2010). Introduction To Political

Psychology (2nd ed.). New York: Psychology Press.

Cutting, J. E. (2003). Gustave Caillebotte, French Impressionism, and mere exposure.

Psychonomic Bulletin & Review, 10(2), 319-343. Retrieved from

Gordon, P. C., & Holyoak, K. J. (1983). Implicit Learning and Generalization of the "Mere Exposure" Effect. Journal of Personality and Social Psychology, 45(3), 492-500. doi: 10.1037/0022-3514.45.3.492

Hekkert, P., & Wieringen, P. C. W. (1990). Complexity and prototypicality as determinants of the appraisal of cubist paintings. British Journal of Psychology, 81(4), 483-495. doi: 10.1111/j.2044-8295.1990.tb02374.x

Hickey, S. (2015, May 12). Picasso painting breaks record for most expensive artwork sold at auction. The Guardian. Retrieved from

http://www.theguardian.com/artanddesign/2015/may/12/pablo-picasso-work-sets-record-for-most-expensive-artwork-sold-at-auction

Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, Approach, and Inhibition.

Psychological Review, 110(2), 265-284. doi: 10.1037/0033-295X.110.2.265

Kim, H., & Markus, H. R. (1999). Deviance or Uniqueness, Harmony or Conformity? A Cultural Analysis. Journal of Personality and Social Psychology, 77(4), 785-800. doi: 10.1037/0022-3514.77.4.785

Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. British Journal of Psychology, 95(4), 489–508. doi:

Mast, M. S., Jonas, K., & Hall, J. A. (2009). Give a Person Power and He or She Will Show Interpersonal Sensitivity: The Phenomenon and Its Why and When. Journal of

Personality and Social Psychology, 97(5), 835-850. doi: 10.1037/a0016234

Newell, B. R., & Bright, J. E. H. (2001). The relationship between the structural mere exposure effect and the implicit learning process. Quarterly Journal of Experimental

Psychology, 54(4), 1087-1104. doi: 10.1080/713756009

Reber, P. J., Stark, C. E. L., & Squire, L. R. (1998). Cortical areas supporting category learning identified using functional MRI. Proceedings of the National Academy of

Sciences, 95, 747-750. Retrieved from http://www.pnas.org/

Rosch, E. (1975). Cognitive Representations of Semantic Categories. Journal of Experimental

Psychology: General, 104(3), 192-233. doi: 10.1037/0096-3445.104.3.192

Stamkou and colleagues (in preparation). The Impact of Transgressive Art.

Van Kleef, G. A., Homan, A. C., Finkenauer, C., Gündemir, S., & Stamkou, E. (2011). Breaking the Rules to Rise to Power: How Norm Violators Gain Power in the Eyes of Others. Social Psychological and Personality Science, 2(5), 500-507. doi:

10.1177/1948550611398416

Van Tilburg, W. A. P., & Igou, E. R. (2014). From Van Gogh to Lady Gaga: Artist

eccentricity increases perceived artistic skill and art appreciation. European Journal

Winkielman, P., Halberstadt, J., Fazendeiro, T., & Catty, S. (2006). Prototypes Are Attractive Because They Are Easy on the Mind. Psychological Science, 17(9), 799-806.

Retrieved from http://pss.sagepub.com/

http://www.wikiart.org

Appendix A

Figure A1. The Abstract context, Figurative focal painting condition

(focal-different-from-context).

Figure A2. The abstract painting collection (above) and figurative painting collection (below)

Appendix B

Table B1

Means, Standard Deviations (between Parentheses), Main and Interaction Effects of Focal and Context Paintings on the Impact and Personality Variables

Mean (SD) Main Effects

(partial eta-squared) Interaction (partial eta-squared) Same-as-Context Different-from-Context Variable Abstract Focal Figurative Focal Abstract Focal Figurative Focal Focal Painting Context Painting Focal*Context Inferred influence 4.64 (1.00) 3.66 (1.12) 5.61 (0.94) 3.98 (1.13) 36.57*** (.28) 2.28 (.02) 9.01*** (.09) Purchase intentions 2.92 (1.16) 1.73 (0.72) 3.21 (1.45) 2.44 (1.23) 16.53*** (.15) 0.74 (.01) 4.37** (.05) Eccentricity 4.64 (0.97) 4.29 (1.46) 5.76 (1.06) 3.91 (1.30) 19.99*** (.18) 9.28** * (.09) 2.25 (.02) Volition 5.22 (0.94) 3.89 (1.14) 5.96 (0.76) 4.53 (1.22) 43.21*** (.32) 0.06 (<.01) 10.86*** (.11) Competence 5.20 (0.80) 5.15 (1.35) 5.64 (0.89) 5.56 (1.04) 0.10 (<.01) 0.01 (<.01) 4.10** (.04)

Note. Degrees of freedom for each ANOVA were dfM=1 and dfR=92. *

Figure B1. The two mediation models showing that volition inferences fully mediated the

relationship between condition and inferred influence and between condition and purchase intentions. * p < .10. **p < .05. ***p < .01. Condition Volition Inferred influence Direct effect, b=0.07

Indirect effect, b=0.15**, BCa CI [0.05, 0.31] 0.46*** 0.33*** Condition Volition Purchase intentions 0.39*** 0.33*** Direct effect, b=0.18