UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl) UvA-DARE (Digital Academic Repository) The art of influence: When and why deviant artists gain impact Stamkou, E.; van Kleef, G.A.; Homan, A.C. DOI 10.1037/pspi0000131 Publication date 2018 Document Version Final published version Published in Journal of Personality and Social Psychology Link to publication Citation for published version (APA): Stamkou, E., van Kleef, G. A., & Homan, A. C. (2018). The art of influence: When and why deviant artists gain impact. Journal of Personality and Social Psychology, 115(2), 276-303. https://doi.org/10.1037/pspi0000131 General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. Download date:06 Apr 2021 https://doi.org/10.1037/pspi0000131 https://dare.uva.nl/personal/pure/en/publications/the-art-of-influence-when-and-why-deviant-artists-gain-impact(d60898ef-3483-43a7-84ea-30c766d1e31d).html https://doi.org/10.1037/pspi0000131 The Art of Influence: When and Why Deviant Artists Gain Impact Eftychia Stamkou, Gerben A. van Kleef, and Astrid C. Homan University of Amsterdam Some artists rise to fame, while others sink into oblivion. What determines whether artists make an impact? Considering deviance in its sociohistorical context, we propose that artists whose work deviates from their own previous style (intrapersonal deviance) and other artists’ styles (interpersonal deviance) gain greater impact than nondeviant artists, as long as deviance is directed toward a progressive style. A preliminary study showed that in western cultures nonrealistic styles are considered more progressive than realistic styles (Study 1). Five more studies provide evidence for the effects of the two types of artistic deviance on several aspects of impact (i.e., perceived influence of the artist, valuation of the artwork, and visual attention to the artwork). First, individuals considered artists who deviated from their previous style more impactful than artists who consistently followed a single style (Study 2), effects that were stronger when artists transitioned from a retrogressive style to a progressive one (Study 3). Second, artists who deviated from their contemporaries’ style were considered more impactful than artists who followed the predominant style, effects that were stronger when artists strayed from a predominant retrogressive style by using progressive means of expression (Studies 4 and 5). When the historical context prevented observers from inferring the progressiveness of the deviant artists’ expressive means, artistic deviance enhanced perceived impact regardless of the means by which the artists deviated (Study 6). Supporting our theoretical model, the effects of intrapersonal and interpersonal deviance on impact were mediated by perceived will-power (Studies 3, 5, and 6). Keywords: artistic impact, deviance, influence, norms Supplemental materials: http://dx.doi.org/10.1037/pspi0000131.supp The rue le Peletier is a road of disasters. After the fire at the Opera, there is now yet another disaster there. An exhibition has just been opened at Durand-Ruel which allegedly contains paintings. I enter and my horrified eyes behold something terrible. Five or six lunatics, among them a woman, have joined together and exhibited their works. I have seen people rock with laughter in front of their pictures but my heart bled when I saw them. These would-be artists call themselves revolutionaries, “Im- pressionists.” They take a piece of canvas, color and brush, daub a few patches of paint on it at random, and sign the whole thing with their name. It is a delusion of the same kind as if the inmates of Bedlam picked up stones from the wayside and imagined they had found diamonds. (Leroy, 1876, as cited by Gombrich, 1995, p. 519) Of the few humans who are remembered centuries after their death, a large proportion are artists. Painters like Claude Monet, the famous pioneer of Impressionism, are nowadays considered geniuses, but in their time they were fiercely criticized for violating prominent norms of beauty, as the above review mirthfully illustrates. From a psycho- logical point of view, deviations from prevailing norms pose a chal- lenge to the viewer because they disrupt fluency and render meaning extraction more effortful (Koffka, 1935; Landau, Greenberg, Solo- mon, Pyszczynski, & Martens, 2006; Reber, Schwarz, & Winkiel- man, 2004; Reber, Winkielman, & Schwarz, 1998; Rosch, 1975). Why, then, do visual artworks often deviate from the prevailing artistic norms of their era? And how do deviant artists gain recogni- tion and make a lasting impact? The theoretical model developed here seeks to explain how artists who deviate from prevailing norms gain impact. Our approach bridges theorizing and research in psychology with empirical and philosophical accounts of aesthetics to provide a comprehensive model that accounts for the role of deviance in shap- ing artists’ impact. How Deviance Shapes Artistic Impact How does deviance from prevailing artistic norms shape an artist’s impact? At first blush, one might think that artistic devi- ance undermines impact, because unexpected stimuli are more Eftychia Stamkou and Gerben A. van Kleef, Department of Social Psychology, University of Amsterdam; Astrid C. Homan, Department of Work and Organizational Psychology, University of Amsterdam. This research was supported by a research grant from the Public Welfare Foundation “Propondis” awarded to Eftychia Stamkou, and a research grant from the Netherlands Organisation for Scientific Re- search (NWO, 406-11-024) awarded to Gerben A. van Kleef and Eftychia Stamkou. We thank Merei Sandbrink, Mies Wegener Sleeswijk, Eva Specker, and Sam Steltman for their help in collecting the data for this project; the Cobra Museum, Amstelveen, for their kind permission to collect data in the museum café, and the Munch Museum, Oslo, for providing us with high-resolution images; the Hundertwasser Foundation and Pictoright for giving us permission to reproduce images of artworks; and the artists Graham White, Lena Levin, and Richard Whitney for their kind permission to use their artworks in our research. Correspondence concerning this article should be addressed to Efty- chia Stamkou, Department of Social Psychology, University of Am- sterdam, Nieuwe Achtergracht 129B, 1018 WT Amsterdam, the Neth- erlands. E-mail: e.stamkou@uva.nl T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. Journal of Personality and Social Psychology © 2018 American Psychological Association 2018, Vol. 115, No. 2, 276 –303 0022-3514/18/$12.00 http://dx.doi.org/10.1037/pspi0000131 276 http://dx.doi.org/10.1037/pspi0000131.supp mailto:e.stamkou@uva.nl http://dx.doi.org/10.1037/pspi0000131 difficult to process and comprehend (Koffka, 1935; Landau et al., 2006; Reber et al., 1998, 2004; Rosch, 1975). Indeed, theories of visual perception hold that perceiving predicted stimuli requires fewer cognitive resources and engenders more positive affect (de-Wit, Machilsen, & Putzeys, 2010). Preferences for predictable stimuli have been observed with regard to color (Martindale & Moore, 1988), furniture (Whitfield & Slatter, 1979), paintings (Farkas, 2002), and exemplars of semantic categories (Martindale, Moore, & West, 1988). In the social domain, too, people tend to prefer individuals who adhere to norms and rules rather than individuals who violate norms and rules, because the latter pose a potential threat to smoothly functioning groups and societies (Heerdink, Van Kleef, Homan, & Fischer, 2013; Jetten & Hornsey, 2014; Jonas et al., 2014; Proulx, Heine, & Vohs, 2010; Van Kleef, Wanders, Stamkou, & Homan, 2015). Individuals who violate behavioral norms are more likely to evoke negative emotions such as anger and blame (Helweg-Larsen & LoMonaco, 2008; Kam & Bond, 2009; Ohbuchi et al., 2004), to be punished (Boyd & Richerson, 1992; Zuckerman, 1999), to be considered uncommit- ted to the group (Feldman, 1984), and to lose their leadership position (Yukl, 2010). Furthermore, ideas that violate expectations and people who violate gender norms are often discouraged and treated with suspicion (Mueller, Melwani, & Goncalo, 2012; Okimoto & Brescoll, 2010). Based on this logic, one would expect people to generally prefer art that keeps with tradition to art that deviates from it, because the former contains more predictable patterns and conforms more to implicit norms and expectations. Despite the intuitive plausibility of these arguments, several strands of theorizing and research suggest that people may actually appreciate and reward deviance, both within and outside the artis- tic domain. When it comes to encountering deviant art, unpre- dicted patterns may pose less of a threat as compared with en- countering potentially order-undermining behavior of a deviant group member. In fact, in the safe as-if context of art, works that belie a predicted pattern may even lead to perceptual pleasure, as the wavering state of prediction error amplifies the subsequent positive affect of prediction confirmation by means of a contrast effect (Huron, 2006; Van de Cruys & Wagemans, 2011). Empirical evidence in favor of this account comes from studies investigating the aesthetic appreciation of music. For instance, Sloboda (1991) found that marked violations of expectations in music correlate with “shivers down the spine,” which are associated with increases in flow in reward- and euphoria-related regions of the brain (e.g., ventral striatum and orbitofrontal cortex; Blood & Zatorre, 2001). Other theories of visual perception suggest that incongruent and unfamiliar stimuli can cause interest (Berlyne, 1960), which can motivate the exploration of one’s environment and the learning of new knowledge, skills, and experiences (Silvia, 2006, 2008). Research outside the artistic domain also provides suggestive evidence that deviance can, under particular circumstances, con- tribute to influence. First, research on the perception of social targets has demonstrated that deviating from behavioral standards can bring about positive outcomes. Deviating from norms indicates that one experiences the leeway to act according to one’s own volition in spite of situational constraints and potential repercus- sions (Stamkou & Van Kleef, 2014). Given that social power is associated with lack of constraint (Galinsky, Gruenfeld, & Magee, 2003; Keltner, Gruenfeld, & Anderson, 2003), individuals whose behavior appears unconstrained by normative pressures may be perceived as powerful. Indeed, empirical studies have shown that individuals who violated prevailing norms were considered more powerful than individuals who complied with the norms (Stamkou, Van Kleef, Homan, & Galinsky, 2016; Van Kleef, Homan, Finke- nauer, Gündemir, & Stamkou, 2011). Moreover, norm violators whose behavior benefited others were more likely to be given a leadership role (Popa, Phillips, & Robertson, 2014; Van Kleef, Homan, Finkenauer, Blaker, & Heerdink, 2012). Other studies showed that individuals who entered a boutique wearing gym clothes rather than appropriate attire or who attended a black tie event wearing a red tie were ascribed higher status because they were considered autonomous (Bellezza, Gino, & Keinan, 2014). Of note, when the violator’s behavior was portrayed as uninten- tional, these effects were attenuated, which is consistent with the notion that inferences of power are contingent on the perception that violators have greater will-power. In short, deviant behavior can fuel perceptions of influence in social interactions as long as the actor’s behavior is considered willful. Second, deviation from default thinking styles has been associ- ated with creativity, which is the basis of an artist’s reputation and impact. For example, in one study the activation of counterstereo- typical thinking propelled the generation of creative ideas (Gocłowska, Crisp, & Labuschagne, 2013). Moreover, divergent cultural experiences, such as multiculturalism or living abroad, foster creative performance by diversifying one’s cognitive per- spectives (Maddux & Galinsky, 2009; Maddux, Adam, & Galin- sky, 2010; Tadmor, Galinsky, & Maddux, 2012). In other studies, participants who were primed with cues representing the concept of deviancy showed greater creative engagement than participants who were primed with conformity cues (Förster, Friedman, But- terbach, & Sassenberg, 2005). This association between divergent thinking and creativity may contribute to lay beliefs that maverick artists can create work of high impact (Feist, 1998; Van Tilburg & Igou, 2014). In sum, given that people who deviate from conventional be- haviors or thinking styles gain benefits that may translate into an elevated social position, we propose that artists who use uncon- ventional means of expression in their work may make a greater impact than artists who follow conventional means of expression. Thus, if we accept the premise that deviating from norms and expectations can elevate one’s impact in social hierarchies (see Van Kleef et al., 2015), the hypothesis follows that artists who deviate from a given artistic norm are perceived as more impactful than artists who follow the norm. Although we thus contend that, generally speaking, deviating from prevailing norms and expecta- tions can boost an artist’s impact, we acknowledge that artistic deviance is not a fixed notion—it is bound to the social and historical context that shapes the background against which devi- ance is considered. Considering Artistic Deviance in Its Social Context The social context refers to the immediate social setting in which artists develop their work. Here we focus on two different types of social context that we believe may shape perceivers’ responses to deviant art: the artist’s own previous work and the work of the artist’s contemporaries. Our definition of artistic deviance is derived from the notion of descriptive norms, which describe which behaviors are typically performed (Cialdini, Reno, T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 277ART OF INFLUENCE & Kallgren, 1990). Deviance from an artistic descriptive norm thus implies that artists deviate from a prevailing artistic style. Deviat- ing from one’s own previous style can be considered a form of intrapersonal artistic deviance, whereas deviating from the pre- dominant style of one’s contemporaries constitutes a form of interpersonal artistic deviance. In the first case, the focus is on how an artist’s work is judged against the context of his or her own previous style, whereas in the second case the focus is on how an artist’s work is judged against the style adopted by the majority of his or her contemporaries. Intrapersonal Deviance New art movements do not emerge out of a vacuum. Art mostly evolves in an incremental way, where the new is folded into the old (Gombrich, 1995), much alike biological and cultural evolution (Heine, 2015; Nunn, 2008; Voigtländer & Voth, 2012). Accord- ingly, artworks are often judged against the background of the artist’s previous work, and indeed many expositions are structured chronologically so as to highlight the artist’s development over time. This notion of contextual judgment is important, because it implies that the evaluation of a particular piece of art may depend on the type of art the artist made before. In other words, an artwork may be judged differently depending on whether the artist fol- lowed the same style before or whether he deviated from his previous style. According to idiosyncrasy credits theory (Hollander, 1958), one can only deviate from old practices after one has proven oneself capable of following them. Earlier conformity to old practices allows others to develop confidence in a person’s skills and com- mitment to the group, which in turn licenses deviance at a later stage (Bray, Johnson, & Chilstrom, 1982; Stone & Cooper, 2009). The operation of such mechanisms can be seen in the careers of famous artists. For instance, historical analyses of the career of the pop band The Beatles highlight how the early albums of The Beatles conformed to the norms of their time, and how after gaining credits by following these norms they began producing highly innovative music that skyrocketed their sales and fame (Inglis, 1996). Another example is found in the rap scene where Lena and Pachucki (2013) empirically demonstrated that rap artists gained status by first showing a repetition of practices that were understood as legitimate by their audience and then introducing novel artistic content that increased their popularity. In classical music too, Beethoven diligently studied and even copied parts of the works of his musical predecessors (e.g., Haydn, Mozart, Bach) before he pushed the boundaries of traditional compositional tech- nique to infuse his late works with unheard-of passion and drama that account for his international fame (Schonberg, 1997; Swaf- ford, 2014). Extending these ideas to the realm of the visual arts, we argue that it is important for artists to obtain sufficient idiosyncrasy credits by first practicing forms of art that are considered tradi- tional in a given era before allowing themselves the leeway to stray to innovative forms of art. If an artist sticks to the same style, observers may assume that this choice of style is dictated by lack of alternatives— because, for instance, the artist did not have the skills to adopt an alternative style—rather than by a deliberate choice. In contrast, if an artist’s portfolio integrates different styles, observers may assume that the artist is able to produce a certain artistic style, but deliberately chose to adopt another style. In this latter case, the artist’s choice to switch to another style indicates the artist’s will to develop an autonomous artistic path. Appreciating the artist’s course of action under the prism of will-power should in turn enhance perceptions of impact (Bellezza et al., 2014; Stamkou & Van Kleef, 2014). In light of these considerations, we hypothesize that artists who show a variety of styles by deviating from what they were making before (i.e., intrapersonal artistic deviance) are credited more for their work than artists who stay within a certain style by repro- ducing what they were making before. Interpersonal Deviance Existential and evolutionary accounts converge to suggest that humans have an enduring and universal need to distinguish them- selves from others (Burris & Rempel, 2004). Previous research has shown that information is better memorized when it distinguishes the self from others (Leyens, Yzerbyt, & Rogier, 1997). Feelings of extreme similarity to others are associated with negative affect (Fromkin, 1972), positive evaluation of scarce experiences (From- kin, 1970), and greater identification with distinctive groups (Brewer & Pickett, 1999). Accordingly, people generally describe themselves as less similar to others than others are to themselves (Codol, 1987). These findings indicate that individuals are motivated to establish and maintain a sense of differentiation from others by acting in ways that show their distinctiveness to others (Brewer, 1991). Research also indicates systematic differences in the particular ways in which feelings of distinctiveness can be achieved depending on one’s social, cultural, or professional milieu (Dutton, Roberts, & Bednar, 2010; Hornsey, Jetten, McAuliffe, & Hogg, 2006; Vignoles, Chryssochoou, & Breakwell, 2000). In the realm of art, the most straightforward way for artists to differentiate themselves is to depart from the style employed by most other artists in their era (Alvarez, Mazza, Pedersen, & Svejenova, 2005). Because making a difference entails comparison with others, judgments of artists’ work are influ- enced by their contemporaries’ work. In other words, contemporaries’ work constitutes the interpersonal context within which an artist’s work is evaluated (Barkow, Cosmides, & Tooby, 1992; Leder, Belke, Oeberst, & Augustin, 2004; Sammartino & Palmer, 2012). The significance of being different from one’s contemporaries is compatible with theories of aesthetics that see aesthetic apprecia- tion as the outcome of contrastive explanations that compare the respective values of a set of artworks (Stecker, 2003). Further- more, the interpersonal context of an artwork makes people pay attention to and inquire about the intentions of the artist (Bullot & Reber, 2013). For instance, a nonrealistic artwork usually contains features that are difficult to process, and this lack of processing fluency may undermine the viewers’ understanding and apprecia- tion of the artwork (Reber et al., 1998, 2004; Wiersema, Van der Schalk, & Van Kleef, 2012). A nonrealistic artwork that is pre- sented among realistic artworks, however, may prompt viewers to quest into the artist’s reasons for adopting a deviant style and thus to infer the artist’s will to propose novel means of expression. In other words, the context of the artwork may lead people to make sense of the deviance through their inferences about the artist’s will-power (Stamkou & Van Kleef, 2014). Suggestive empirical support for the role of context in shaping the appreciation of artworks comes from a study showing that a mismatch between the T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 278 STAMKOU, VAN KLEEF, AND HOMAN style of a focal design object and the style of contextual design objects increased the perceived value of the focal object (Blijlev- ens, Gemser, & Mugge, 2012). In keeping with these theoretical accounts and empirical find- ings, we propose that artists who deviate from their contempo- raries’ styles (i.e., interpersonal artistic deviance) are perceived as more impactful than artists who follow their contemporaries’ styles, because their decision to deviate indicates that their choice of style is dictated by their own will rather than by external influences. Considering Artistic Deviance in Its Historical Context A work of art is an artifact that is embedded in a historical context. Accordingly, modern theories of aesthetics appeal to beholders’ sensitivity to historical contexts and the evolution of such contexts to explain art appreciation (Bullot, 2009; Bullot & Reber, 2013; Davies, 2004; Levinson, 2007). Bullot and Reber (2013), for instance, introduced a psycho-historical framework of art appreciation that posits that individuals’ responses to artworks rely not only on the visible traces of the artwork but also on their knowledge about the historical context in which the artist worked. By the same token, historical contingencies play an essential role in what people consider deviant artistic work. Examples of deviant movements in different eras showcase that the definition of artistic deviance has changed through the ages. Dutch Masters, for instance, dared to show the imperfections of life by depicting their subjects the way they really looked rather than beautified; Cubists created the illusion of three-dimensional forms by depicting their subjects from a multitude of viewpoints rather than a single view- point; and Futurists captured the rush of industrialization by de- picting their subjects in a dynamic rather than static manner. Proponents of these movements were all considered deviants in their era because they led to an artistic development by straying away from tradition. These examples demonstrate that a proper understanding of deviant art requires that one consider the histor- ical evolution of art movements in a given culture (Bullot & Reber, 2013; Leder et al., 2004; Levinson, 2007). In Western cultures, popular styles have changed throughout the centuries, with realistic forms of art mostly being the norm till approximately the second half of the 19th century. This is not to say that painting before that time-point aimed exclusively and entirely at the imitation of reality, as there are examples of art- works that did not aim at a veridical representation of natural objects before that time (Gombrich, 1985). However, the link with nature provided some kind of anchorage in western painting up until the rise of nonrealistic movements.1 By the end of the 19th century most of the movements that rose to prominence rejected the study of natural appearances (Gombrich, 1995). Along other historical events, the shift from realistic to nonrealistic forms of art in western cultures coincides with the spread of photography, which was then seen as a rival to painting (Rosenblum, 1989). Painters of that era were therefore motivated to explore alterna- tives to the representation of nature, which spurred new artistic developments (Gombrich, 1995). The example of nonrealistic art illustrates that artistic deviance is effective when it leads to some kind of artistic progress. On the contrary, artistic deviance that does not move art forward or is backward-looking is less likely to prevail. This possibility reso- nates with evolutionary theories on the accumulation of cultural information. Given that innovations build on previously existing structures (Voigtländer & Voth, 2012), cultural information grows in complexity over time, and so cultural evolution moves like a ratchet—it only goes forward and never slips back (Tennie, Call, & Tomasello, 2009). Similarly, historical accounts of evolution suggest that deviant ideas are culture’s engines that are responsible for the dynamism of human species because they compel us to think, reevaluate, and criticize (Harari, 2015). Apparently the same happens with the evolution of art movements, because artistic development over time is the result of an unquenchable thirst for novelty (Martindale, 1990). As such, artists who embrace older art styles might be seen as retrogressive and therefore less impactful. Our historical analysis suggests that realistic art should be considered less progressive than nonrealistic art because realism appeared earlier than nonrealism, which only prevailed in the latest centuries. Given that deviance is valued when it is progressive, we expected that the hypothesized intrapersonal and interpersonal artistic deviance effects would occur only if the artist deviates toward a progressive movement, such as nonrealism. In other words, we expected that deviating from realism to nonrealism fuels stronger perceptions of impact than deviating from nonrealism to realism because of the historical evolution of art movements in Western cultures.2 Overview of Model and Hypotheses In the current research, deviance and impact are conceptualized within the domain of art, and our findings should thus be inter- preted in the context of artistic judgments. From now on, we therefore use the terms “intrapersonal deviance” and “interper- sonal deviance” to refer to deviance in the artistic domain. Based on the theorizing above, we advance five interrelated hypotheses, including a generic mechanism that drives the effects of artistic deviance. First, because artists are judged against the background of their own previous work, we hypothesize that artists who deviate from their previous means of expression by adopting distinct styles within their career are seen as more impactful than artists who consistently follow a certain style (intrapersonal deviance effect). 1 Artwork styles that aim to render a naturalistic representation of the external world (e.g., realism, symbolism, romanticism) are grouped under the term “realistic.” In contrast, artwork styles that depart from a natural- istic representation of the world (e.g., impressionism, cubism, abstract expressionism) are defined as “nonrealistic.” 2 Given that cultural evolution relies on the pursuit of novelty, artistic movements that do not move art forward (e.g., realism in present-day western cultures) might be seen as retrogressive and thereby less impactful (Tennie et al., 2009; Voigtländer & Voth, 2012). We therefore expected that artists who adopt a rather progressive style, such as nonrealism, will be seen as more impactful than artists who adopt a rather retrogressive style, such as realism. We tested this idea in three studies that we report in the Online Supplemental Material. First, we showed that artists who made nonrealistic artworks were considered more influential than artists who made realistic artworks because nonrealistic artists’ work was perceived to be more unconventional (Study S1). Second, we demonstrated that artworks made in an artist’s idiosyncratic period that is characterized by nonrealistic elements are valuated higher and attract people’s attention more than artworks made in an artist’s academic or transi- tional periods, which are mostly characterized by realistic elements. This effect was replicated among museum visitors (Study S2a) and university students (Study S2b). T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 279ART OF INFLUENCE http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1037/pspi0000131.supp Second, we hypothesize that the effect of intrapersonal deviance is more pronounced if the artist’s style transitioned toward a progressive style, for instance, from realism to nonrealism rather than the other way around (moderation of the intrapersonal devi- ance effect by progressiveness). Third, given that artists are evaluated in the context of their contemporaries’ work, we propose that artists who deviate from their contemporaries’ style are considered more impactful than artists who follow their contemporaries’ style (interpersonal devi- ance effect). Fourth, we hypothesize that the effect of interpersonal deviance is more pronounced when artists deviate toward a progressive style, for example, when they deviate from a predominant realistic style by using nonrealistic means of expression rather than when they deviate from a predominant nonrealistic style by using real- istic means of expression (moderation of the interpersonal devi- ance effect by progressiveness). Fifth, we propose that the effects of deviance on artistic impact are driven by observers’ inferences about the artists’ intentionality in their choice of style. Namely, deviant artists may be perceived as more impactful because they deviated from their previous style or other artists’ style out of personal will-power (mediation of the intrapersonal and interpersonal deviance effects through per- ceived will-power). In conjunction, we hypothesize that artists who deviate from the norms they established during their career (intrapersonal deviance) as well as the norms that their contemporaries have established (interpersonal deviance) will be perceived as more impactful than artists who follow the various types of norms. Importantly, artistic deviance at either level is more effective when it is imbued with progressiveness. Our theoretical model is visualized in Figure 1. Overview of Studies We conducted six studies to test our theoretical model. We operationalized artistic deviance by letting the focal artwork be similar to or different from its intrapersonal or interpersonal con- text. Artworks were either of realistic or nonrealistic styles. We used the discernible differences between realistic and nonrealistic art to visualize deviance. We assumed that the contrast between these two styles would make the operationalization of deviance accessible to our sample, which consisted of western individuals with no particular expertise in art. In most studies, we operation- alized progressiveness of deviance by letting the focal artwork be nonrealistic while context artworks were realistic. This operation- alization was premised on the assumption that nonrealistic styles would be considered more progressive than realistic styles. We tested this assumption in Study 1, in which we examined people’s beliefs about the progressiveness of realistic and nonrealistic styles. Next, Studies 2 and 3 focused on intrapersonal deviance (i.e., deviance from one’s own previous style), whether in the direction of realism or nonrealism. Studies 4 and 5 focused on interpersonal deviance (i.e., deviance from the predominant style of one’s contemporaries), whether in the direction of realism or nonrealism. To establish that the effects of artistic deviance are not confounded with abstractness that was common among all art- works of nonrealistic style used in Studies 2 to 5, we further examined interpersonal deviance in Study 6 by operationalizing the motif of the artwork (triangle vs. rectangular) while keeping the style of the artworks constant across conditions, that is, non- realistic. The deviance operationalization in Study 6 allowed us to address the alternative possibility that the predicted effect of artistic deviance is attributable to other positive qualities (rather than artistic deviance) that are associated with the artist’s expres- sive means. Across studies, we used different ecologically valid operation- alizations of artistic impact that were informed by prominent definitions of impact in the field of art (Schönfeld & Reinstaller, 2007). According to these definitions, impactful artists are influ- ential (influence), and produce artworks that are highly priced (valuation) and catch people’s attention (attention). We assessed the influence aspect by measuring the extent to which an artist is perceived to be influential across all studies that tested the effects of deviance (Studies 2, 3, 4, 5, and 6). We assessed the valuation aspect by measuring the estimated price of an artist’s work (Study 2) or intentions to purchase merchandise depicting the artist’s work (Study 5). Finally, we measured the attention aspect by investigating engagement of visual attention while looking at the artwork (Study 5). An overview of the artistic impact operation- alizations per study is presented in Table 1. All studies started with a brief introduction, after which partic- ipants indicated their participation consent. Participants were then presented with high-quality copies of original artworks, and they indicated whether they thought the artists would gain (or had gained) impact in terms of influence, valuation, and attention. We also estimated individuals’ beliefs about the artist’s will-power (Studies 3, 5, and 6) to examine whether inferences about the intentionality of the artist’s actions could explain the effects of deviance on the various indices of artistic impact. Additional measures that functioned as manipulation checks, control vari- ables, and alternative mediators are described in the method sec- tion of each study. The artworks used in each study were retrieved from an online visual art encyclopedia (www.wikiart.org). We used G�Power 3 (Faul, Erdfelder, Lang, & Buchner, 2007) to calculate the required sample size of our studies. In the power analyses we used the cumulative effect size of previous studies to compute the required sample of each follow-up study to achieve statistical power of at least .80, given alpha level of .05. When there were no previous studies available or there was no indication in the literature of the size of the hypothesized effect, we used contemporary conventions regarding number of participants per Figure 1. Theoretical model illustrating that deviating from the style previously employed by the artist (intrapersonal deviance) or the style currently employed by the artist’s contemporaries (interpersonal deviance) enhances an artist’s impact, especially when artistic deviance is directed toward a progressive style. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 280 STAMKOU, VAN KLEEF, AND HOMAN http://www.wikiart.org condition (Lakens & Evers, 2014). In studies where there were multiple artistic impact variables we calculated the average effect size across variables. Study 1 investigated progressiveness, and so it was not included in the estimation of the cumulative effect size of the rest of the studies that investigated artistic impact. Data collection stopped when we had reached at least the required sample size calculated in the power analysis. In all studies the actual sample size was slightly larger than the required sample size because of compensation for anticipated drop-out (Zhou & Fish- bach, 2016). Table 1 gives an overview of the actual and required sample size, as well as the effect size of each study. In the studies below we report all relevant measures and manipulations. Besides the studies reported in the current paper, we carried out five additional studies (S1, S2a, S2b, S3, and S4) that tested effects not focal to this paper. All of these studies replicated our main effect of interest. Information about these studies is provided in the Online Supplemental Material. Study 1 The main tenet of our model is that artistic deviance should enhance an artist’s impact, especially when the artist deviates by means of a progressive style. In light of the historical evolution of styles in Western cultures, we assumed that nonrealistic styles are thought of as more progressive than realistic styles. In Study 1 we set out to substantiate this assumption and lay the groundwork for the ensuing studies, in which progressiveness of deviance is op- erationalized as the artist’s choice for a nonrealistic style (i.e., Studies 3, 4, and 5). Method Because existing literature provided no indication of the size of the hypothesized effect we could expect in Study 1, we aimed for a large sample to ensure we would achieve sufficient statistical power. We recruited 124 participants through Prolific Academic (www.prolific.ac). Participants were citizens of the U.K. with an Asian (6), African American (3), or Caucasian (115) background. The study was administered as an online questionnaire and partic- ipants were compensated with money (see Table 1 for further sample details). Participants were asked to indicate the extent to which they find realistic and nonrealistic art progressive on 7-point bipolar scales ranging from 1 � retrogressive to 7 � progressive. Furthermore, participants were asked to estimate the time-point when painters started making realistic and nonrealistic artworks on 12-point Likert scales ranging from the 10th to the 21st century. Results A paired-samples t test with progressiveness ratings of realistic and nonrealistic styles as paired variables showed that nonrealistic styles (M � 5.16, SD � 1.45) are considered more progressive than realistic styles (M � 3.91, SD � 1.52), t(123) � �6.05, p � .001, d � �0.54. Another paired-samples t test showed that participants dated the beginning of realistic art (M � 5.37, SD � 3.01) earlier than the beginning of nonrealistic art (M � 8.25, SD � 2.96), t(123) � �8.96, p � .001, d � �0.80. Discussion These results demonstrate that participants generally consider nonrealistic styles more progressive than realistic styles. The find- ing that realistic art is dated earlier than nonrealistic art lends additional credence to the idea that nonrealistic art is perceived as more progressive than realistic art. The perceived progressiveness of nonrealistic styles was expected to moderate the effects of intrapersonal and interpersonal deviance on impact in Studies 3, 4, and 5, where the deviance operationalization entailed the adoption of a different style (either realistic or nonrealistic) than adopted previously by the focal artist (intrapersonal deviance) or by the artist’s contemporaries (interpersonal deviance). (In Study 6, where the deviance operationalization did not imply a choice for a Table 1 Sample Characteristics and Operationalization of Artistic Impact per Study Study Sample Age Gender Artistic impact operationalization Required Na Actual N f M (SD) Range Men Women Transgender Unknown 1 — 124 .27 36.09 (11.64) 19–72 55 69 0 0 — 2 96 116 .25 35.12 (15.98) 18–78 57 59 0 0 Perceived influence Price estimation 3 122 136 .18 23.00 (6.85) 18–65 41 95 0 0 Perceived influence 4 168 172 .19 35.88 (11.92) 18–75 95 76 1 0 Perceived influence 5 181 227 .18 20.46 (3.01) 18–44 75 138 1 13 Perceived influence Purchase intention Visual attention 6 181 218 .55 38.62 (12.64) 18–84 85 133 0 0 Perceived influence Note. In Study 5, age and gender data from 13 participants were not recorded. Perceived influence and price estimation correlated positively in Study 2, r(112) � .33, p � .001, and Study 5, r(227) � .33, p � .001. Visual attention correlated positively with purchase intention, r(227) � .15, p � .027, and showed no significant correlation with perceived influence, although their relationship was in the predicted direction, r(227) � .11, p � .090. a In the power analyses we used the cumulative effect size of previous studies to compute the required sample of each follow-up study. The required sample size of Study 2 was based on the cumulative effect size observed in Studies S1 (f � .31) and S2a (f � .50), whereas Study S2b’s effect size (f � .25) was not taken into account because the data for this study were collected in parallel to the Study 2 data. Furthermore, the effect size of Studies S3 (f � .25) and S4 (f � .20) were not included in the power analyses because these studies were conducted after the studies reported in the main text. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 281ART OF INFLUENCE http://dx.doi.org/10.1037/pspi0000131.supp http://www.prolific.ac nonrealistic style, deviant artists’ impact should not depend on the expressive means by which the artists deviated.) Study 2 The intrapersonal deviance hypothesis holds that artists who deviate from their previous style are seen as more impactful than artists who consistently follow a certain style. This implies that the evaluation of artists is influenced by whether they switch to an alternative style at some point in their career. The current study tested the intrapersonal deviance effect by investigating whether artists would gain greater impact when their portfolio contains artworks of both realistic and nonrealistic styles (intrapersonal deviance) than when their portfolio contains artworks of only one style (no intrapersonal deviance). We assessed the influence and valuation aspects of artistic impact by measuring perceived influ- ence of the artist and price estimation of the artist’s work in an auction. Method Sample. The required sample size of Study 2 (NR � 96) was based on the cumulative effect size observed in Studies S1 and S2a, whereas study S2b was not taken into account because the data for this study were collected in parallel to the Study 2 data. The actual sample consisted of 116 adults who were recruited in a central train station in Amsterdam and participated in the study on a voluntary basis. The study was administered as a paper-and- pencil questionnaire, which could be completed in English (4 participants) or Dutch. The original Dutch questionnaire was trans- lated into English following the back-translation procedure out- lined by Brislin (1986). Further sample details appear in Table 1. Design, materials, and procedure. Participants were pre- sented with a portfolio that consisted of four paintings and they were told that the artworks were made by the same artist. All artworks were made by Dutch artist Willem De Kooning, apart from one that was made by Graham White, an artist who was inspired by De Kooning’s style. We selected paintings from De Kooning because of the stark contrast between his academic period and his abstract expressionism period (Yard, 2007). Participants were randomly assigned to one of three experimental conditions, with the first two being nondeviant and the third one being deviant in terms of showing a departure from the artist’s earlier style. That is, the portfolio either included exclusively realistic style artworks (only realistic style), exclusively nonrealistic style artworks (only nonrealistic style), or an equal number of realistic and nonrealistic style artworks (mixed style). In the first condition we used three artworks from De Kooning’s academic period and one realistic- style artwork from Graham White, in the second condition we used four artworks from De Kooning’s abstract expressionism period, and in the third condition we showed two artworks from each of the previous conditions (see Appendix A and Table 2 for details about the artworks). The artworks were presented on separate pages of the questionnaire in a pseudorandomized order. All art- works were presented in greyscale. After seeing all artworks, participants rated the artist’s perceived influence on a scale that consisted of three items (“I think this artist . . . has influenced his contemporaries,” “. . . made a great contribution to art,” and “. . . is famous”). The items were an- swered on 7-point Likert scales ranging from 1 � strongly dis- agree to 7 � strongly agree (� � .83). They then estimated the artist’s work by filling out an amount in euro they thought one of the artist’s paintings would be auctioned for. Results We tested the effect of artist’s intrapersonal deviance on per- ceived influence and price estimation by means of ANOVAs. We also conducted a planned contrast to compare the difference be- tween the deviant condition (mixed style) and the nondeviant conditions (only realistic-style and only nonrealistic style). De- scriptives for perceived influence and price estimation are dis- played in Table 3, and test statistics are reported in text below. Perceived influence. ANOVA showed a main effect of art- ist’s intrapersonal deviance on perceived influence, F(2, 113) � 7.18, p � .001, �p2 � .11. Specifically, the artist was considered more influential when his portfolio included artworks of mixed styles than when his portfolio included artworks of a single style, t(113) � �3.55, p � .001, d � �0.67 (see left panel of Figure 2). For exploratory purposes we examined the difference between the deviance condition and each of the no-deviance conditions separately. These analyses showed that the artist was considered more influential when his portfolio included artworks of mixed styles than when his portfolio included exclusively realistic style artworks, t(113) � �3.73, p � .001, d � �0.70, or exclusively nonrealistic style artworks, t(113) � �2.39, p � .019, d � �0.45. Price estimation. Four participants did not fill out the price estimation question, leaving a sample of 112 participants. Because the distribution of price estimation scores was positively skewed (skewness � 10.40, SE � 0.23), we applied a logarithmic trans- formation that resulted in a normalized distribution (skewness � 0.75, SE � 0.23). ANOVA indicated that artist’s intrapersonal deviance influenced respondents’ price estimation, F(2, 109) � 3.33, p � .039, �p2 � .06. In line with the intrapersonal deviance hypothesis, participants estimated the artist’s work higher when his portfolio included artworks of mixed styles than when his portfolio included artworks of a single style, t(109) � �2.33, p � .021, d � �0.45 (see right panel of Figure 2). Again, we explored the difference between the deviance condi- tion and each of the no-deviance conditions. These analyses indi- cated that the artist’s work was valuated higher when his portfolio included artworks of mixed styles than when his portfolio included exclusively nonrealistic style artworks, t(109) � �2.57, p � .012, d � �0.70, but not when his portfolio included exclusively real- istic style artworks, t(109) � �1.45, p � .149, d � �0.28. Discussion The results of Study 2 indicate that artists are considered more influential and their work is valuated higher when they use both realistic and nonrealistic forms of expression than when they use only one form of expression. These findings provide initial evi- dence that deviant artists are better off if they also produce art- works of a less progressive style (i.e., realism), presumably be- cause these artworks readily testify that the artist’s choice to depart T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 282 STAMKOU, VAN KLEEF, AND HOMAN from a previous style was a product of the artist’s free will rather than lack of skill (Bray et al., 1982; Hollander, 1958). Although these findings are consistent with our theoretical framework, the conclusions that can be drawn from this study are limited by three interrelated issues. First, the results leave open an alternative interpretation in terms of perceived competence or versatility. That is, artists who deviate from their previous style by employing an alternative style may be perceived as more skillful or more versatile than artists who stick with one style.3 Second, the current study does not provide evidence for the underlying mechanism that drives the effect of intrapersonal deviance, which according to our theorizing relates to viewers’ inferences about the artist’s intentional shift to alternative means of ex- pression (rather than perceptions of competence or versatility). Third, the current results do not speak to the question of whether artists are better off endorsing a progressive style before or after a retrogressive one. Such temporal variations may have implications for how an artist’s deviance is inter- preted (e.g., to what extent it is seen as a sign of will-power). These limitations were addressed in Study 3. 3 Although artists who can master more than one style (intrapersonal deviance) would be naturally considered more competent than artists who master only one style (no intrapersonal deviance), perceived competence was not expected to explain the effects of intrapersonal deviance on impact because, according to our theory, the effect of deviance was driven by the inference that the artist made a willful choice for an alternative style. In Study 2 we assessed perceived competence of the artist to rule out the explanation that the effect of intrapersonal deviance on impact would be accounted for by competence. As expected, results showed that the artist was considered more competent when his portfolio included artworks of mixed styles than when his portfolio included artworks of a single style, but perceived competence did not mediate the effect of intrapersonal deviance on perceived influence. Table 2 Details About the Artworks Used in Studies 2 to 6 Artist name Artwork title Date of artwork production Study 2 Willem de Kooning Still Life (Bowl, Pitcher, and Jug)a 1921 Willem de Kooning Portrait of Elainea 1940–1941 Willem de Kooning Untitled (Still Life)a 1916 Graham White Drawings, Winter–Springb 2010 Willem de Kooning Untitled (The Cow Jumps Over the Moon)a 1937–1938 Willem de Kooning Charcoal Drawinga circa 1970–1980 Willem de Kooning Secretarya 1948 Willem de Kooning Figurea 1949 Study 3 Pablo Picasso Bull (plate II)a 1945 Pablo Picasso Bull (plate III)a 1945 Pablo Picasso Bull (plate VII)a 1946 Pablo Picasso Bull (plate IX)a 1946 Study 4 Andrew Wyeth Braids (Helga Terstorf)a 1979 Pablo Picasso Portrait of the Artist’s Mothera 1896 Boris Kustodiev Portrait of a Woman 1920 Salvador Dalí Portrait of Katharina Cornella 1951 Pablo Picasso Portrait of Jacqueline Roque With Her Hands Crosseda 1954 Francis Picabia Ridensa 1929 Study 5 Edvard Munch Self-Portrait 1881–1882 Paul Feeley Portrait of Samuel L. M. Barlow 1932 Jamie Wyeth Portrait of Andrew Wyeth 1969 Thomas Cowperthwait Eakins Portrait of Samuel Murray 1889 Richard Whitney Michael J. McGivney Unknown Juan Gris Portrait of Pablo Picasso 1912 Nadezhda Udaltsova Cubist Portrait 1915 Lyubov Popova Sketch for Portrait 1889 Pablo Picasso William Uhde 1910 Pablo Picasso Portrait of Daniel-Henry Kahnweiler 1910 Study 6 Unknown artists Untitled artworksc Unknown Note. All artworks reproduced in the article are reproduced with permission, unless the artist’s rights are in public domain (e.g., Boris Kustodiev). Copyright on works of visual artists affiliated to the International Confederation of Societies of Authors and Composers (CISAC) organization (i.e., Willem de Kooning, Pablo Picasso, Andrew Wyeth, Salvador Dalí, and Francis Picabia) has been arranged with Pictoright in Amsterdam. Copyright on Graham White’s work has been arranged with the artist himself. a Pictoright Amsterdam, 2018. b © Graham White, 2018. c Artist names, artwork titles, and year of artwork production are unknown because these artworks were derived from Internet service providers (e.g., Pinterest). T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 283ART OF INFLUENCE Study 3 The current study investigated a qualification of the intraper- sonal deviance hypothesis, namely that artists gain more impact when their work evolves from a less progressive style to a more progressive one rather than in the opposite order. The logic un- derlying this prediction is that artists who produce traditional artworks early on in their career and then stray away from tradi- tions demonstrate that their style change is driven by a deliberate choice to broaden the horizon of their artistic expression rather than by a shortage of technical skills. To test this idea, we crossed a manipulation of intrapersonal deviance similar to that employed in Study 2 with a manipulation of the artist’s early style, so that the artist who deviated either deviated in the direction of a progressive style (i.e., nonrealism) or in the direction of a retrogressive style (i.e., realism). Furthermore, the design of Study 3 allows us to examine whether perceived competence or versatility explain the interactive effect of intrapersonal deviance and progressiveness of deviance on artistic impact. Artists who deviate toward a progres- sive style should be seen as more willful than artists who deviate toward a retrogressive style, because the artists’ choice to switch to a progressive style indicates their will to develop an autonomous artistic path. At the same time, artists who deviate toward a progressive style should be considered equally competent and versatile as artists who deviate toward a retrogressive style, be- cause in both cases the artist masters the same type and number of styles. We also investigated whether the effect of intrapersonal deviance on artistic impact could be explained by individuals’ perceptions of the artist’s will-power to shape their individual artistic path. Finally, we included a manipulation check of intrap- ersonal deviance. We investigated the above predictions by focus- ing on the influence aspect of artistic impact. Method Sample. The required sample size of Study 3 (NR � 122) was based on the cumulative effect size observed in Studies S1, S2a, S2b, and 2. The actual sample consisted of 136 students from the University of Amsterdam who participated in the study in ex- change for course credits. Participants were recruited via the university’s online system (www.test.uva.nl). Further sample de- tails are provided in Table 1. Design, materials, and procedure. Participants were ran- domly assigned to a 2 (artist’s intrapersonal deviance: no vs. yes) � 2 (artist’s early style: realistic vs. nonrealistic) between- subjects experimental design. They were presented with a portfolio of two paintings made by Spanish artist Pablo Picasso. We selected two realistic and two nonrealistic artworks from Picasso’s bull series, which consists of 11 lithographs that show the development of an artwork from a realistic style to a nonrealistic style (Daix, Table 3 Descriptives of Perceived Influence and Price Estimation Across Conditions of Artist’s Intrapersonal Deviance in Study 2 Measure Artist’s intrapersonal deviance: no (only realistic style) Artist’s intrapersonal deviance: no (only non-realistic style) Artist’s intrapersonal deviance: yes (mixed style) Totala M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI Perceived influence 3.82 (1.22) [3.47, 4.16] 4.14 (1.05) [3.80, 4.48] 4.72 (0.91) [4.38, 5.05] 4.23 (1.12) [4.03, 4.42] Price estimation 3.68 (0.26) [3.53, 3.82] 3.57 (0.40) [3.43, 3.71] 3.83 (0.58) [3.69, 3.96] 3.69 (0.45) [3.61, 3.77] Note. CI � confidence interval. a Average descriptives of artist’s intrapersonal deviance conditions. Figure 2. Perceived influence and price estimation as a function of artist’s intrapersonal deviance in Study 2. Bars represent means with their associated standard errors. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 284 STAMKOU, VAN KLEEF, AND HOMAN http://www.test.uva.nl 1994). Participants saw two artworks next to each other, and they were told that the artist made the left-side artwork at an earlier stage of his career and the right-side artwork at a later stage. The artworks were presented in their original color shade (see Appen- dix B and Table 2 for details about the artworks). In the no- intrapersonal-deviance condition the two artworks represented the same style, whereas in the intrapersonal-deviance condition the two artworks represented different styles. We then measured the extent to which the artist was considered influential with the same scale used in Study 2 (� � .90). Next we measured the artist’s perceived will-power with the items “I think this artist . . . has a personal artistic vision,” “. . . thinks out-of- the-box,” and “. . . is open-minded” (� � .83). We also asked participants whether they perceived the later artwork’s style to deviate from the earlier artwork’s style to check the manipulation of intrapersonal deviance. Perceived intrapersonal deviance was measured with the items “These two paintings represent two different artistic styles” and the reverse-scored “These two paint- ings represent two similar artistic styles,” r(136) � .35, p � .001. All items were answered on 7-point Likert scales ranging from 1 � strongly disagree to 7 � strongly agree. Results We tested the effects of artist’s intrapersonal deviance and artist’s early style on perceived intrapersonal deviance, perceived influence, and perceived will-power with three two-way ANOVAs, which we followed up with simple effects analyses. Descriptives for perceived intrapersonal deviance, perceived influence, and perceived will-power are displayed in Table 4, and test statistics are reported in text below. Perceived intrapersonal deviance (manipulation check). There was a main effect of artist’s intrapersonal deviance, indicat- ing that the later artwork’s style was perceived to deviate from the earlier artwork’s style to a greater extent when the two artworks represented different styles than when they represented the same style, F(1, 132) � 71.85, p � .001, �p2 � .35. There was no main effect of artist’s early style, F(1, 132) � 1.63, p � .204, �p2 � .01. There was no significant interaction effect between artist’s intrap- ersonal deviance and artist’s early style, which indicated that the manipulation of artist’s intrapersonal deviance was orthogonal to the manipulation of artist’s early style, F(1, 132) � 0.68, p � .411, �p 2 � .01. Perceived influence. There was a main effect of artist’s in- trapersonal deviance, indicating that an artist who deviated from his previous style appeared more influential than an artist who followed his previous style, F(1, 132) � 4.51, p � .036, �p2 � .03. There was no main effect of artist’s early style, F(1, 132) � 0.15, p � .701, �p2 � .01. As expected, there was an interaction between artist’s intrapersonal deviance and artist’s early style, F(1, 132) � 9.57, p � .002, �p2 � .07. Probing the interaction showed that an artist who deviated from his previous style was considered more influential when his early work was realistic rather than nonreal- istic, F(1, 132) � 6.05, p � .015, �p2 � .05; perceived influence of an artist who did not deviate from his previous style did not depend on his early artwork style, F(1, 132) � 3.67, p � .058, �p2 � .03 (see left panel of Figure 3). Perceived will-power. There was a main effect of artist’s intrapersonal deviance, indicating that an artist who deviated from his previous style was perceived as having greater will-power than an artist who did not deviate from his previous style, F(1, 132) � 5.31, p � .023, �p2 � .04. There was no main effect of artist’s early style, F(1, 132) � 0.06, p � .802, �p2 � .01. There was an interaction effect between artist’s intrapersonal deviance and art- ist’s early style, F(1, 132) � 10.35, p � .002, �p2 � .07. Probing the interaction showed that an artist who deviated from his previ- ous style was thought of as having greater will-power when his earlier work was realistic than nonrealistic, F(1, 132) � 6.01, p � .016, �p2 � .05. An artist who consistently followed an earlier nonrealistic style was perceived as having greater will-power than an artist who consistently followed an earlier realistic style, F(1, 132) � 4.40, p � .038, �p2 � .03 (see right panel of Figure 3). This finding likely reflects the greater impact of progressive styles, such as nonrealism, as compared with retrogressive styles, such as realism (see Studies S1, S2a, and S2b in the supplemental mate- rial). Table 4 Descriptives of Perceived Intrapersonal Deviance, Perceived Influence, and Perceived Willpower Across Conditions of Artist’s Intrapersonal Deviance and Artist’s Early Style in Study 3 Measure Artist’s intrapersonal deviance: no Artist’s intrapersonal deviance: yes Totala M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI Perceived intrapersonal deviance Artist’s early style: realistic 3.81 (1.04) [3.42, 4.20] 5.31 (1.04) [4.92, 5.70] 4.56 (1.28) [4.29, 4.83] Artist’s early style: non-realistic 3.40 (1.28) [3.01, 3.79] 5.22 (1.19) [4.83, 5.61] 4.31 (1.53) [4.04, 4.58] Totalb 3.60 (1.18) [3.33, 3.88] 5.26 (1.11) [4.99, 5.54] 4.43 (1.41) [4.24, 4.63] Perceived influence Artist’s early style: realistic 3.78 (1.44) [3.31, 4.26] 5.05 (1.05) [4.57, 5.53] 4.42 (1.41) [4.08, 4.76] Artist’s early style: non-realistic 4.44 (1.49) [3.96, 4.92] 4.21 (1.61) [3.73, 4.69] 4.32 (1.54) [3.98, 4.66] Totalb 4.11 (1.49) [3.77, 4.45] 4.63 (1.41) [4.29, 4.97] 4.37 (1.47) [4.13, 4.61] Perceived willpower Artist’s early style: realistic 4.12 (1.42) [3.69, 4.54] 5.30 (1.17) [4.88, 5.73] 4.71 (1.42) [4.41, 5.01] Artist’s early style: non-realistic 4.75 (1.23) [4.33, 5.18] 4.56 (1.17) [4.13, 4.98] 4.66 (1.20) [4.36, 4.96] Totalb 4.44 (1.36) [4.14, 4.74] 4.93 (1.22) [4.63, 5.23] 4.68 (1.31) [4.47, 4.90] Note. CI � confidence interval. a Average descriptives of artist’s intrapersonal deviance conditions. b Average descriptives of artist’s early style conditions. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 285ART OF INFLUENCE http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1037/pspi0000131.supp Moderated mediation. To test whether the interactive effect of artist’s intrapersonal deviance and artist’s early style on per- ceived influence was mediated by perceived will-power, we spec- ified a bootstrapped moderated mediation model with artist’s early style and intrapersonal deviance as predictors, perceived will- power as mediator, and perceived influence as outcome variable (Hayes, 2012; Model 7 in PROCESS, 1000 reiterations). Artist’s early style was coded as �1 for realistic and 1 for nonrealistic, artist’s intrapersonal deviance was coded as �1 for nondeviant and 1 for deviant, and perceived will-power was centered at its mean. This model was significant with a point estimate of �0.56, SE � 0.17, 95% CI [�0.92, �0.23]. Analysis of the simple effects indicated that an artist who deviated from his previous style was perceived as having stronger will-power when his early artwork represented a realistic rather than a nonrealistic style, and this will-power perception consequently enhanced the artist’s per- ceived influence (�0.30, SE � 0.12, 95% CI [�0.55, �0.09]). Furthermore, an artist who followed his earlier nonrealistic style was perceived as having stronger will-power than an artist who followed his earlier realistic style, and this will-power perception enhanced the artist’s perceived influence (0.26, SE � 0.13, 95% CI [0.001, 0.52]). Discussion Study 3 indicates that artists who deviate from their past style are seen as more influential than artists who follow the same style throughout their careers, especially when artists depart from a rather retrogressive style (realism) by adopting a rather progressive style (nonrealism). The mediation analysis identified an underlying process that explains this effect: Artists who deviate in the direc- tion of a progressive style are seen as having stronger will-power, which enhances individuals’ perceptions of the artist’s influence. Additionally, the design and results of Study 3 refute the alterna- tive account that perceived competence or versatility explain the interactive effect of intrapersonal deviance and progressiveness of deviance on artistic impact, because the artist in both deviance conditions should be considered equally competent and versatile.4 In line with our hypothesis, Studies 2 and 3 jointly indicate that artistic impact depends on the deviance showcased in an artist’s portfolio and the direction in which an artist’s style develops during his or her career. These studies indicate that people evaluate deviant artistic work in the context of the artist’s previous work, which is a means of gaining insight into the artist’s personal history and intentions. As noted in the Introduction, judgments of artists’ work are not only influenced by their own artistic histories but also by their contemporaries’ work, which shapes what is considered the tradi- tion in a given era. In Studies 4 and 5 we investigated how artists who deviate from the traditional style of their era—what we termed interpersonal deviance—may gain impact. Study 4 The style endorsed by an artist’s contemporaries forms the context within which a deviant artwork is judged. The artwork’s context highlights the deviant artist’s intention to introduce new means of expression. We therefore expected that artists who de- viate from their contemporaries’ style would have more impact than artists who follow the predominant style. However, the inter- personal deviance effect should be more pronounced when artists 4 Given that our hypothesized intrapersonal deviance effects were mod- erated by progressiveness of artistic deviance in Study 3, we expected that perceived will-power—rather than perceived competence or versatility—is the mechanism underlying the effect of intrapersonal deviance on impact. We therefore assessed perceived will-power as well as perceived compe- tence and perceived versatility as alternative mediators. There were no main or interaction effects of intrapersonal deviance and contemporaries’ style on perceived competence. There was a main effect of intrapersonal deviance on perceived versatility indicating that artists who deviate from their previous style are seen as more versatile than artists who follow their previous style; there was no main effect of artist’s early style and no interaction effect between intrapersonal deviance and artist’s early style on perceived versatility. A moderated mediation model that included per- ceived competence, perceived will-power, and perceived versatility as competing mediators showed that only perceived will-power mediated the effects of intrapersonal deviance on perceived influence. Figure 3. Perceived influence and perceived will-power as a function of artist’s intrapersonal deviance and artist’s early style in Study 3. Bars represent means with their associated standard errors. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 286 STAMKOU, VAN KLEEF, AND HOMAN deviate from realism by making nonrealistic art than when they deviate from nonrealism by making realistic art, because the for- mer form of deviance reveals the artist’s will to introduce a style that could move art forward whereas the latter form of deviance may be seen as retrogressive (Tennie et al., 2009). The current study tested the interpersonal deviance hypothesis by investigating whether an artist would gain greater impact when his artwork appears among contemporaries’ artworks that represent a different style than among contemporaries’ artworks that represent the same style. Furthermore, contemporaries’ style was manipulated to be realistic or nonrealistic so that the focal artist’s deviance would be directed toward a progressive or a retrogressive style (i.e., nonre- alism or realism, respectively). We examined the above predic- tions by assessing the influence aspect of artistic impact. Method Sample. The required sample size of Study 4 (NR � 168) was based on the cumulative effect size observed in Studies S1, S2a, S2b, 2, and 3. The actual sample consisted of 172 American citizens who were recruited through Amazon’s mechanical Turk (www.mturk.com). The study was administered as an online ques- tionnaire and participants were compensated with money (see Table 1 for further sample details). Design, materials, and procedure. Participants were ran- domly assigned to a 2 (artist’s interpersonal deviance: no vs. yes) � 2 (contemporaries’ style: realistic vs. nonrealistic) between-subjects experimental design. Participants viewed three artworks of the same or different style, and they were told that all artworks dated from the beginning of the 20th century (see Appendix C and Table 2 for details about the artworks). We presented this information so that the various artists would be perceived as contemporaries. The artworks were presented in their original color shade. Participants were then asked to evaluate the second artwork and the artist who made it (the focal artist). All artworks depicted female portraits made by different artists, except for the focal artworks that were both made by the same artist (i.e., Picasso). After showing each artwork to participants, we mea- sured perceived influence with the same scale used in the previous studies (� � .80). Results We tested the effects of artist’s interpersonal deviance and contemporaries’ style on perceived influence with a 2-way ANOVA. We followed up these analyses with simple effects analyses to test the effect of contemporaries’ style on the percep- tion of deviant and nondeviant artists. Descriptives are displayed in Table 5, and test statistics are reported in text below. There was a main effect of artist’s interpersonal deviance, with deviant artists being perceived as more influential than nondeviant artists, F(1, 168) � 6.06, p � .015, �p2 � .04. There was no main effect of contemporaries’ style, F(1, 168) � 1.23, p � .270, �p2 � .01. As predicted, there was a significant interaction effect between artist’s interpersonal deviance and contemporaries’ style, F(1, 168) � 7.13, p � .008, �p2 � .04. Probing the interaction revealed that artists who deviated from a predominant realistic style by using nonrealistic means of expression were seen as more influ- ential than artists who deviated from a predominant nonrealistic style by using realistic means of expression, F(1, 168) � 7.35, p � .007, �p2 � .04. Artists who followed a predominant nonrealistic style did not differ from artists who followed a predominant realistic style, F(1, 168) � 1.31, p � .253, �p2 � .01 (see Figure 4). Discussion Study 4 demonstrated that artists who deviate from their con- temporaries’ style are seen as more influential than artists who follow the predominant style. The results additionally showed that deviating in the direction of a progressive style (e.g., nonrealism) is more effective than deviating in the direction of a retrogressive style (e.g., realism). These findings thus provide evidence for the interpersonal deviance effect and its magnifying conditions, but they provide no insight into the underlying processes that drive the effect of interpersonal deviance on perceived artistic impact. This limitation was addressed in Study 5, where we examined whether inferences about the artist’s will-power mediate the effect of interpersonal deviance on artistic impact. Study 5 Study 5 aimed to replicate and extend the findings of Study 4 in several ways. First, we investigated whether the effect of interper- sonal deviance on artistic impact operates through perceived will- power, which was the main mediator suggested in previous studies that examined the effects of norm violation on power and status (Bellezza et al., 2014; Van Kleef et al., 2011). Second, we used a greater number of contemporaries’ artworks compared with the previous study and changed the position of the focal artwork to see whether we would find similar effects. Third, we included a manipulation check of interpersonal deviance. Finally, we exam- ined whether the effect of interpersonal deviance generalizes to an indirect behavioral measurement of impact, namely visual atten- tion. Because indirect measures are unobtrusive, they can be very insightful as long as they are validated against other direct mea- sures (Palmer, Schloss, & Sammartino, 2013). We therefore also included perceived influence and purchase intention as direct mea- sures of the influence and valuation aspects of artistic impact, respectively. Method Sample. The required sample size of Study 5 (NR � 181) was based on the cumulative effect size observed in Studies S1, S2a, S2b, 2, 3, and 4. The actual sample consisted of 227 Dutch students from the University of Amsterdam who were recruited via an online platform (www.test.uva.nl) and participated in exchange for course credits (see Table 1 for further sample details). Design, materials, and procedure. Study 5 employed the same experimental design as Study 4. In Study 5, however, par- ticipants viewed five artworks that they were told had all been made at the beginning of the 20th century. Participants were asked to evaluate the last artwork and the artist who had made it (focal artist). All artworks were male portraits with the model facing left. We edited the original artworks to be greyscale, and controlled for luminosity and size to standardize the artworks’ visual features (see and Table 2 for details about the artworks). After participants had viewed each artwork, we measured their perceptions of the artist’s will-power, using an adjusted version of the T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 287ART OF INFLUENCE http://www.mturk.com http://www.test.uva.nl volitional capacity scale (Magee, 2009). This scale’s focus on inter- personal relations corresponds with the current study’s level of anal- ysis. This 6-item scale includes items such as “To what extent . . . does this artist feel free to do what s/he wants in his/her relations with others?” and the reverse-scored “. . . this artist’s behavior is driven by the wishes of other people?” (� � .91), which were answered on 7-point Likert scales ranging from 1 � strongly disagree to 7 � strongly agree. Participants then reported on the artist’s perceived influence by completing the same scale we used in Studies 2, 3, and 4 (� � .85). Next we evaluated purchase intention by asking partic- ipants to indicate the extent to which they would buy products depicting the artwork on a scale ranging from 1 � not at all to 7 � very much. Furthermore, we recorded the time participants spent looking at the focal artwork, which was included as an unobtrusive measure of participants’ attention to the artwork (Palmer et al., 2013). In the end, we asked participants how different they thought the style of the focal artwork was compared with the style of the other art- works, which we used as a manipulation check of interpersonal deviance. The perceived interpersonal deviance item was answered on a 7-point Likert scale ranging from 1 � strongly disagree to 7 � strongly agree. Results In a series of 2-way ANOVAs we tested the effects of artist’s interpersonal deviance and contemporaries’ style on perceived interpersonal deviance, perceived influence, purchase intention, visual attention, and perceived will-power. We followed up each ANOVA with simple effects analyses to investigate the effect of contemporaries’ style on the various measures of Table 5 Descriptives of Perceived Influence, Perceived Interpersonal Deviance, Purchase Intention, Visual Attention, and Perceived Willpower Across Conditions of Artist’s Interpersonal Deviance and Contemporaries’ Style in Studies 4, 5, and 6 Measure Artist’s interpersonal deviance: no Artist’s interpersonal deviance: yes Totala M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI Study 4 Perceived influence Contemporaries’ style: realistic 4.08 (1.13) [3.71, 4.46] 5.06 (1.19) [4.69, 5.43] 4.57 (1.26) [3.91, 4.84] Contemporaries’ style: non-realistic 4.38 (1.39) [4.00, 4.76] 4.34 (1.27) [3.96, 4.72] 4.36 (1.32) [4.09, 4.63] Totalb 4.23 (1.27) [3.97, 4.50] 4.71 (1.27) [4.44, 4.97] 4.47 (1.29) [4.28, 4.66] Study 5 Perceived interpersonal deviance Contemporaries’ style: realistic 3.92 (1.67) [3.51, 4.32] 5.71 (1.31) [5.30, 6.11] 4.80 (1.75) [4.53, 5.10] Contemporaries’ style: non-realistic 3.60 (1.68) [3.18, 4.02] 5.51 (1.60) [5.09, 5.93] 4.55 (1.89) [4.26, 4.85] Totalb 3.76 (1.68) [3.47, 4.05] 5.61 (1.45) [5.32, 5.90] 4.68 (1.82) [4.48, 4.89] Perceived influence Contemporaries’ style: realistic 3.75 (1.21) [3.44, 4.05] 5.29 (1.10) [4.98, 5.59] 4.51 (1.39) [4.30, 4.73] Contemporaries’ style: non-realistic 4.53 (1.25) [3.57, 4.19] 3.88 (1.15) [3.57, 4.19] 4.20 (1.24) [3.98, 4.43] Totalb 4.13 (1.29) [3.92, 4.36] 4.60 (1.33) [4.36, 4.80] 4.36 (1.33) [4.21, 4.52] Purchase intention Contemporaries’ style: realistic 2.03 (1.22) [1.53, 2.20] 3.47 (1.71) [2.65, 3.32] 2.74 (1.64) [2.47, 3.03] Contemporaries’ style: non-realistic 3.04 (1.69) [2.12, 2.90] 2.80 (1.57) [2.13, 2.82] 2.92 (1.63) [2.63, 3.21] Totalb 2.52 (1.54) [2.25, 2.82] 3.14 (1.67) [2.84, 3.42] 2.83 (1.63) [2.63, 3.04] Visual attention Contemporaries’ style: realistic 0.66 (0.26) [0.58, 0.74] 0.84 (0.36) [0.75, 0.92] 0.75 (0.32) [0.69, 0.81] Contemporaries’ style: non-realistic 0.64 (0.33) [0.55, 0.73] 0.61 (0.36) [0.52, 0.70] 0.62 (0.35) [0.56, 0.69] Totalb 0.65 (0.29) [0.59, 0.71] 0.73 (0.38) [0.66, 0.79] 0.69 (0.34) [0.64, 0.73] Perceived willpower Contemporaries’ style: realistic 3.90 (1.15) [3.62, 4.17] 5.70 (0.87) [5.42, 5.98] 4.79 (1.37) [4.60, 5.00] Contemporaries’ style: non-realistic 5.34 (0.90) [5.05, 5.62] 4.25 (1.33) [3.97, 4.54] 4.80 (1.26) [4.59, 5.00] Totalb 4.59 (1.26) [4.42, 4.82] 5.00 (1.33) [4.78, 5.18] 4.79 (1.31) [4.66, 4.94] Study 6 Perceived interpersonal deviance Contemporaries’ motif: triangle 2.03 (1.16) [1.68, 2.37] 3.94 (1.46) [3.59, 4.28] 2.99 (1.63) [2.74, 3.23] Contemporaries’ motif: rectangular 2.46 (1.18) [2.12, 2.81] 3.98 (1.34) [3.64, 4.33] 3.22 (1.47) [2.98, 3.47] Totalc 2.25 (1.19) [2.00, 2.49] 3.96 (1.39) [3.72, 4.20] 3.10 (1.55) [2.93, 3.28] Perceived influence Contemporaries’ motif: triangle 3.77 (1.35) [3.42, 4.12] 5.25 (1.18) [4.91, 5.60] 4.52 (1.47) [4.27, 4.76] Contemporaries’ motif: rectangular 4.10 (1.53) [3.75, 4.45] 5.46 (1.14) [5.11, 5.81] 4.77 (1.51) [4.53, 5.03] Totalc 3.94 (1.45) [3.69, 4.18] 5.35 (1.16) [5.11, 5.60] 4.65 (1.49) [4.47, 4.82] Perceived willpower Contemporaries’ motif: triangle 4.11 (1.34) [3.82, 4.40] 5.68 (0.93) [5.39, 5.96] 4.90 (1.39) [4.69, 5.10] Contemporaries’ motif: rectangular 4.28 (1.04) [3.99, 4.56] 5.57 (0.95) [5.29, 5.86] 4.92 (1.19) [4.72, 5.13] Totalc 4.19 (1.20) [3.99, 4.40] 5.63 (0.93) [5.42, 5.83] 4.91 (1.29) [4.77, 5.05] Note. CI � confidence interval. a Average descriptives of artist’s interpersonal deviance conditions. b Average descriptives of contemporaries’ style conditions. c Average descriptives of contemporaries’ motif conditions. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 288 STAMKOU, VAN KLEEF, AND HOMAN impact of deviant and nondeviant artists. Even though we had no predictions about the main effects of contemporaries’ style, we report below whether such effects were observed on each measure (details are provided in the supplemental material). Descriptives are displayed in Table 5, and test statistics are reported in text below. Perceived interpersonal deviance (manipulation check). As expected, there was a main effect of artist’s interpersonal deviance, with deviant artists’ style being perceived as more dis- tinct from their contemporaries’ style as compared with nondevi- ant artists’ style, F(1, 223) � 78.48, p � .001, �p2 � .26. There was no main effect of contemporaries’ style, F(1, 223) � 1.51, p � .221, �p2 � .01. There was also no interaction between artist’s interpersonal deviance and contemporaries’ style, which indicated that the manipulation of artist’s interpersonal deviance was orthog- onal to the manipulation of contemporaries’ style, F(1, 223) � 0.08, p � .779, �p2 � .01, as intended. Perceived influence. There was the predicted main effect of artist’s interpersonal deviance, with artists who employed a devi- ant style being perceived as more influential than artists who employed a nondeviant style, F(1, 223) � 8.01, p � .005, �p2 � .04. There was also an unpredicted main effect of contemporaries’ style on perceived influence, F(1, 223) � 3.92, p � .049, �p2 � .02 (see supplemental material). More importantly, we found the pre- dicted interaction effect between artist’s interpersonal deviance and contemporaries’ style, F(1, 223) � 49.08, p � .001, �p2 � .18. Probing the interaction revealed that artists who deviated from a predominant realistic style by using nonrealistic means of expres- sion were seen as more influential than artists who deviated from a predominant nonrealistic style by using realistic means of ex- pression, F(1, 223) � 40.93, p � .001, �p2 � .18. Artists who followed a predominant nonrealistic style were also seen as more influential than artists who followed a predominant realistic style, F(1, 223) � 13.12, p � .001, �p2 � .06 (see top left panel of Figure 5). Again, this finding likely reflects the greater impact of pro- gressive style, such as nonrealism, as compared with retrogressive styles, such as realism (see Studies S1, S2a, and S2b in the supplemental material). Purchase intention. There was a main effect of artist’s inter- personal deviance, which showed that participants were more willing to purchase products that depict artworks made by deviant artists than products that depict artworks made by nondeviant artists, F(1, 223) � 8.38, p � .004, �p2 � .04. There was no main effect of contemporaries’ style, F(1, 223) � 0.67, p � .415, �p2 � .01. We also observed the predicted interaction between artist’s interpersonal deviance and contemporaries’ style, F(1, 223) � 16.31, p � .001, �p2 � .07. Probing the interaction revealed that participants were more willing to purchase products depicting work made by artists who deviated in the direction of nonrealism than products depicting work made by artists who deviated in the direction of realism, F(1, 223) � 5.43, p � .021, �p2 � .02. Participants were also more willing to purchase products depicting work made by artists who followed a predominant nonrealistic style than products depicting work made by artists who followed a predominant realistic style, F(1, 223) � 13.12, p � .001, �p2 � .06 (see top right panel of Figure 5), which may reflect the greater impact of progressive (nonrealistic) art as compared with retro- gressive (realistic) art. Visual attention. Because the distribution of visual attention scores was positively skewed (skewness � 5.66, SE � 0.16), we applied a logarithmic transformation that resulted in a normalized distribution (skewness � 0.31, SE � 0.16). ANOVA showed a marginal main effect of artist’s interpersonal deviance, which indicates that artworks made by deviant artists tended to attract more attention than artworks made by nondeviant artists, F(1, 223) � 2.86, p � .092, �p2 � .01. It also showed an unpredicted main effect of contemporaries’ style, F(1, 223) � 8.05, p � .005, �p 2 � .04 (see supplemental material). More importantly, results showed the predicted interaction between artist’s interpersonal deviance and contemporaries’ style, F(1, 223) � 5.77, p � .017, �p 2 � .03. Simple effects analyses revealed that individuals looked longer at artworks made by artists who deviated toward nonrealism than at artworks made by artists who deviated toward realism, F(1, 223) � 13.90, p � .001, �p2 � .06. The time participants spent looking at artworks made by artists who followed a predominant realistic style did not differ from the time participants spent look- ing at artworks made by artists who followed a predominant nonrealistic style, F(1, 223) � 0.07, p � .791, �p2 � .01 (see bottom left panel of Figure 5). Perceived will-power. There was a main effect of artist’s interpersonal deviance, which showed that people inferred greater will-power when artists deviated from the predominant style than when they followed the predominant style, F(1, 223) � 6.40, p � .012, �p2 � .03. There was no main effect of contemporaries’ style on perceived will-power, F(1, 223) � 0.01, p � .976, �p2 � .01. As predicted, we did find an interaction effect between artist’s inter- personal deviance and contemporaries’ style, F(1, 223) � 101.23, p � .001, �p2 � .31. Probing the interaction revealed that individ- uals regarded artists who deviated toward nonrealism as having stronger will-power than artists who deviated toward realism, F(1, 223) � 51.45, p � .001, �p2 � .23. Individuals further considered artists who followed a predominant nonrealistic style as having stronger will-power than artists who followed a predominant re- alistic style, F(1, 223) � 51.50, p � .001, �p2 � .23 (see bottom right panel of Figure 5), which likely reflects the higher impact of progressive styles, like nonrealism, in comparison to retrogressive styles, like realism. Figure 4. Perceived influence as a function of artist’s interpersonal deviance and contemporaries’ style in Study 4. Bars represent means with their associated standard errors. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 289ART OF INFLUENCE http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1037/pspi0000131.supp Moderated mediation. Finally, we performed moderated me- diation analyses to test whether the interactive effect of artist’s inter- personal deviance and contemporaries’ style on each dependent vari- able (i.e., perceived influence, purchase intention, and visual attention) was mediated by perceived will-power. We specified three bootstrapped moderated mediation models with contemporaries’ style and artist’s interpersonal deviance as predictors, perceived will-power as mediator, and perceived influence (Model A), purchase intention (Model B), and visual attention (Model C) as outcome variables that were tested sequentially (Hayes, 2012; Model 7 in PROCESS, 1000 reiterations). Contemporaries’ style was coded as �1 for realistic and 1 for nonrealistic, artist’s interpersonal deviance was coded as �1 for nondeviant and 1 for deviant, and perceived will-power was centered at its mean. All three models were significant, providing evidence for moderated mediation (Model A: point estimate � �0.59, SE � 0.12, CI [�0.83, �0.36]; Model B: point estimate � �0.60, SE � 0.13, CI [�0.87, �0.36]; and Model C: point estimate � �0.06, SE � 0.03, CI [�0.12, �0.02]). Analysis of the simple effects indicated that, compared with artists who deviated in the direction of realism, artists who deviated in the direction of nonrealism were perceived as having stronger will-power, and this perception in turn increased the artist’s perceived influence (point estimate � �0.29, SE � 0.07, 95% CI [�0.44, �0.17]), purchase intentions for the artist’s work (point estimate � �0.30, SE � 0.07, 95% CI [�0.45, �0.17]), and attention paid to the artist’s work (point estimate � �0.03, SE � 0.01, 95% CI [�0.07, �0.01]). Artists who followed a dominant nonrealistic style were also thought of as having stronger will-power than artists who followed a dominant realistic style, and this inference in turn in- creased the artist’s perceived influence (point estimate � 0.29, SE � 0.07, 95% CI [0.17, 0.44]), purchase intentions for the artist’s work (point estimate � 0.30, SE � 0.07, 95% CI [0.16, 0.45]), and attention paid to the artist’s work (point estimate � 0.03, SE � 0.01, 95% CI [0.01, 0.06]). Discussion The results of the current study support the interpersonal devi- ance hypothesis, which holds that artists who deviate from their Figure 5. Perceived influence, purchase intention, visual attention, and perceived will-power as a function of artist’s interpersonal deviance and contemporaries’ style in Study 5. Bars represent means with their associated standard errors. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 290 STAMKOU, VAN KLEEF, AND HOMAN contemporaries’ style gain greater impact than artists who follow the predominant style, especially when artists deviate by employ- ing progressive means of expression (i.e., nonrealism). The medi- ation analyses further revealed an underlying mechanism that accounts for these effects. Consistent with our theoretical argu- ments, artists who deviate from their contemporaries’ style using nonrealistic means of expression are perceived to have stronger will-power, which in turn enhances individuals’ influence percep- tion, willingness to purchase the artists’ products, and attention to the artist’s work. These findings were robust across three measures of artistic impact, including an unobtrusive behavioral measure of attention. Study 6 Study 6 examined the interpersonal deviance effect by incorpo- rating several methodological improvements. First, in all previous studies where we operationalized artistic deviance, the artist would deviate by adopting a realistic or nonrealistic style. Nonrealistic art, however, features abstract elements, which can potentially introduce a confound between deviance and abstractness. In the current study we disentangled the effect of deviance from abstract- ness by keeping the style of the artworks constant across condi- tions (i.e., always nonrealistic) and instead manipulating the motif of the artworks: The focal artist would deviate by adopting a triangle-motif or a rectangular-motif in a nonrealistic artwork. Second, realistic and nonrealistic styles were born out of the Western culture, so our participants might have been familiar with the historical evolution and significance of these styles. For in- stance, because nonrealistic art has become increasingly prominent in the last few centuries, participants may expect artists to produce nonrealistic art nowadays. One might thus argue that in our pre- vious studies, people’s perceived impact of artists who deviated toward a nonrealistic style demonstrates an understanding of what is considered good art nowadays rather than a preference for deviance. To tackle this issue in the current study we used a cover story that neutralized the historical context by placing the artists on another planet with a unique culture and by temporally situating the production of the artworks in the current century. Participants were given no information about what is common and what is considered progressive or good art on this planet, so that they would have no preconceptions about the value of the expressive means used by the artists (i.e., triangle or rectangular motifs). Because the different motifs were thus freed from perceptions of progressiveness we expected that deviant artists would be consid- ered more impactful than nondeviant artists regardless of the expressive means by which they would deviate (i.e., no moderation of the interpersonal deviance effect). Finally, we included a mea- sure of will-power to see whether our proposed explanatory mech- anism would still operate after we stripped the design from ab- stractness (the primary feature of nonrealistic art) and discharged the historical context. This study focused on the influence aspect of impact, which was used across all previous studies that examined the effect of artistic deviance. Method Sample. The required sample size of Study 6 (NR � 181) was based on the cumulative effect size observed in Studies S1, S2a, S2b, 2, 3, 4, and 5. The actual sample consisted of 218 participants who were recruited through Prolific Academic (www.prolific.ac). Participants were citizens of Ireland (6), the United Kingdom (165), and the Unites States of America (44). Three participants did not report their nationality. The study was administered as an online questionnaire and participants were compensated with money (see Table 1 for further sample details). Design, materials, and procedure. Study 6 employed a sim- ilar experimental design to Studies 4 and 5. In Study 6 interper- sonal deviance was operationalized as a choice for a different or the same motif employed by the artist’s contemporaries (triangle vs. rectangular motif). We selected artworks of geometric abstrac- tion, which is a form of nonrealistic art based on the use of geometric forms depicted in 2-dimensional space. The artworks were presented in their original color shade (see Table 2 for details about the artworks). Images of the artworks used in Study 6 may be requested by the first author. Participants were instructed that they will read a story about a country from another planet, named “Aratartland,” where people have developed a culture that is different from human culture. We then presented participants with three artworks that represented the dominant artistic trend in Aratartland. The dominant trend was characterized by either a triangle motif or a rectangular motif. Next participants saw the focal artist’s painting, which employed either the same or a different motif than the other artists. Participants were told that all artworks had been made at the beginning of the century. We measured the focal artist’s will-power with the same scale we used in Study 5 (� � .87). We then measured perceived influence of the focal artist using the same scale we used in the previous studies (� � .91). We also measured perceived deviance of the artwork using the item “I think this artwork . . . is deviant compared with the rest of the artworks” and the reverse-scored item “. . . is similar to the rest of the artworks,” r(218) � .48, p � .001. Finally, participants rated the perceived abstractness of each artwork with a 7-point bipolar item ranging from 1 � realistic to 7 � abstract. Results We first compared the perceived abstractness of the triangle- motif artworks to the rectangular-motif artworks by averaging the abstractness ratings of the artworks in the no-deviance conditions, namely, the conditions where participants were presented with only triangle- or rectangular-motif artworks. This comparison showed than triangle-motif artworks (M � 5.67, SD � 1.43) did not differ from rectangular-motif artworks (M � 5.46, SD � 1.42) in terms of abstractness, F(1, 107) � 0.58, p � .448, which confirms that artworks of different motifs were perceived as equally abstract. Next, we carried out a series of 2-way ANOVAs to examine the effects of artist’s interpersonal deviance and contemporaries’ motif on perceived interpersonal deviance, perceived influence, and per- ceived will-power. Descriptives are displayed in Table 5, and test statistics are reported in text below. Perceived interpersonal deviance (manipulation check). As expected, there was a main effect of artist’s interpersonal deviance, with deviant artists’ motif being perceived as more deviant from their contemporaries’ motif as compared with nondeviant artists, F(1, 214) � 188.24, p � .001, �p2 � .47. There was no main effect of T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 291ART OF INFLUENCE http://www.prolific.ac contemporaries’ motif, F(1, 214) � 2.26, p � .134, �p2 � .01. Furthermore, there was no interaction between artist’s interpersonal deviance and contemporaries’ motif, which indicated that the manip- ulation of artist’s interpersonal deviance was orthogonal to the ma- nipulation of contemporaries’ motif, F(1, 214) � 2.39, p � .123, �p2 � .01, as intended. Perceived influence. There was the predicted main effect of artist’s interpersonal deviance, with artists who deviated from the dominant motif being perceived as more influential than artists who followed the dominant motif, F(1, 214) � 63.96, p � .001, �p 2 � .23. As expected, there was no main effect of contempo- raries’ motif, F(1, 214) � 2.20, p � .139, �p2 � .01, and no interaction effect, F(1, 214) � 0.12, p � .729, �p2 � .01 (see left panel of Figure 6). Perceived will-power. There was a main effect of artist’s interpersonal deviance, which showed that people inferred greater will-power when artists deviated from the dominant motif than when they followed the dominant motif, F(1, 214) � 96.55, p � .001, �p2 � .31. There was no main effect of contemporaries’ motif, F(1, 214) � 0.05, p � .829, �p2 � .01, and no interaction effect, F(1, 214) � 0.84, p � .362, �p2 � .01 (see right panel of Figure 6). Mediation. We performed a mediation analysis to test whether the effect of artist’s interpersonal deviance on perceived influence was mediated by perceived will-power. We specified a bootstrapped me- diation model with artist’s interpersonal deviance as predictor, per- ceived will-power as mediator, and perceived influence as outcome variable (Hayes, 2012; Model 4 in PROCESS, 1000 reiterations). Artist’s interpersonal deviance was coded as �1 for nondeviant and 1 for deviant and perceived will-power was centered at its mean. This model provided evidence for mediation (point estimate � 0.16, SE � 0.07, CI [0.03, 0.30]). This indicates that, compared with artists who follow the dominant motif of their era, artists who deviate from it are considered to have stronger will-power, which in turn enhances their perceived influence. Discussion Study 6 showed that artists who deviate from the dominant artistic trend of their era are perceived as more influential than artists who follow the dominant trend. Most important, the means by which the artists deviated did not moderate the effects of interpersonal deviance in this case: Artists who deviated using a triangle motif were consid- ered equally influential as artists who deviated using a rectangular motif. This anticipated nullification of the moderation effect that was consistently observed in the previous studies is consistent with our theoretical argument. In the current study, the use of nonrealistic artworks across conditions erased the condition of progressiveness that was introduced in the previous studies by the contrast between realistic and nonrealistic styles. The fact that the effect of deviance on impact was not moderated when this condition was raised signifies that progressiveness of deviance is an important boundary condition: Artistic deviance is effective when it leads to some sort of progress; when participants cannot make inferences about the progressiveness of the deviant artworks, the effect of deviance remains but it is not moderated. Furthermore, in all previous studies deviation from the contemporaries’ style entailed a choice for either a realistic or a nonrealistic style, which also differed in the perceived abstractness of each style. In the current study we controlled for this stylistic property by using artworks of the same style across all conditions (i.e., nonre- alistic) and by manipulating another element of the artwork, namely, the motif. Importantly, the category of triangle-motif artworks was perceived as equally abstract as the category of rectangular-motif artworks. The results thus indicate that the effect of interpersonal deviance on artistic impact occurs above and beyond abstractness. Moreover, the effect of deviance proved robust in a neutralized historical context, which implies that the observed effect cannot be accounted for by participants’ preconceptions or stereotypes about the value of the expressive means used by the artists since participants had no access to evaluative information of these means. Instead, in keeping with our theoretical argument, the effect of interpersonal deviance on perceived influence was explained by inferences of the deviant artist’s will-power. Meta-Analytic Synthesis of Findings Even though our studies provide rather consistent support for our hypotheses, we carried out meta-analyses that synthesized the findings of different studies to provide more reliable estimates of Figure 6. Perceived influence and perceived will-power as a function of artist’s interpersonal deviance and contemporaries’ motif in Study 6. Bars represent means with their associated standard errors. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 292 STAMKOU, VAN KLEEF, AND HOMAN the intrapersonal deviance main effect and the interpersonal devi- ance main and interaction effects. The intrapersonal deviance interaction effect was not meta-analyzed since it was only tested in one study (Study 3). Because some studies included more than one measure of artistic impact (i.e., Studies 2 and 5), we standardized the scores of each study’s variables to enable combining them into a single estimate of artistic impact per study. The use of composite estimates was necessary because in meta-analysis each effect size estimate has to be based on a unique sample, which means that only one effect size estimate per study can be included (Rauden- bush, 2009). Furthermore, we used standardized regression coef- ficients as effect size estimates because they could be computed in all studies, and we used a random-effects approach because of the variety of methodologies used across studies. Meta-analysis was performed using Comprehensive Meta-Analysis (Borenstein, Hedges, Higgins, & Rothstein, 2009). Meta-analytic results are commonly presented in a forest plot that depicts both the individual effects observed in each study and the overall effects estimated across studies (see Figures 7 to 9). The left part of the figure presents the standardized regression coefficients (beta) that express the difference between the condi- tions under comparison (e.g., Condition A vs. Condition B) as individual and overall effects. The right part of the figure graph- ically presents these effects with their confidence intervals within a range of �1 SD and relative to a reference line set at 0. The individual effects are represented with an empty rectangular, and the overall effects are represented with a solid diamond. When the confidence intervals of an effect fall on the left side of the refer- ence line, participants in Condition A scored higher on perceived artistic impact than participants in Condition B; when they fall on the right side, participants in Condition A scored lower than participants in Condition B; and when they fall in between, there was no significant difference in artistic impact scores between Conditions A and B. Intrapersonal Deviance Main Effect According to the intrapersonal deviance hypothesis, artists who deviate from their previous means of expression by adopting distinct artistic styles are considered more impactful than artists who consistently follow a single style throughout their career. The intrapersonal deviance main effect was tested in Studies 2 and 3, which we combined in the current meta-analysis. The heterogene- ity tests showed that the effect of artist’s intrapersonal deviance on artistic impact did not significantly vary across studies, Q(1) � 0.02, p � .891. The overall statistics showed that artists who deviated from their previous style were considered more impactful than artists who consistently followed a single style throughout their career, � 0.27, SE � 0.08, Z � 3.61, p � .001, 95% CI [0.12, 0.42] (see Figure 7). Interpersonal Deviance Main and Interaction Effects According to the interpersonal deviance hypothesis, artists who deviate from the style adopted by their contemporaries have more impact than artists who follow their contemporaries’ style. The interpersonal deviance main effect was tested in Studies 4, 5, and 6, which we combined in the current meta-analysis. The test of heterogeneity showed that the effect of artist’s interpersonal devi- ance on artistic impact significantly varied across studies, Q(2) � 16.71, p � .001. The difference in the magnitude of the effects is due to the operationalization of deviance (style vs. motif). The direction of the effects, however, was homogeneous, which allows us to interpret the overall statistics. The overall statistics showed that artists who deviated from their contemporaries’ style were considered more impactful than artists who followed their contem- poraries’ style, � 0.27, SE � 0.11, Z � 2.51, p � .012, 95% CI [0.06, 0.49] (see Figure 8). In light of the progressive development of artistic movements (Tennie et al., 2009), we expected that the interpersonal devi- ance effect would be more pronounced when artists deviate toward a progressive style than when they deviate toward a retrogressive style (moderation of the interpersonal deviance effect). For instance, artists who deviate from a predominant realistic style by adopting nonrealistic means of expression should be seen as more impactful than artists who deviate from a predominant nonrealistic style by adopting realistic means of expression, because nonrealism is considered a more progres- Figure 7. Meta-analytic effect of artist’s intrapersonal deviance (no vs. yes) on artistic impact across Studies 2 and 3. CI stands for confidence interval. Diamonds represent overall effects and rectangles represent effects of individual studies. When the CIs of an effect fall on the right [left] side of 0, deviant artists were considered more [less] impactful than nondeviant artists. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 293ART OF INFLUENCE sive movement. The interpersonal deviance interaction effect was expected to be found in Studies 4 and 5 where deviance was operationalized by means of realistic and nonrealistic artworks. This effect was therefore tested in another meta-analysis that included these two studies. The meta-analytic model examined the impact of artists who followed or deviated from a predom- inant realistic or nonrealistic style (as determined by their contemporaries’ style). As predicted, the analysis showed that perceived artistic impact differed depending on whether artists followed or deviated from their contemporaries’ style, Q(1) � 26.04, p � .001. The pattern of the moderation indicated that artists who deviated from their contemporaries’ style were considered more impactful when the predominant style was realistic rather than nonrealistic, � �0.32, SE � 0.07, 95% CI [�0.46, �0.19], Z � �4.71, p � .001 (see Figure 9). Thus, artists who deviated from their contemporaries’ style toward a progressive style (nonrealism) were more influential than artists who deviated toward a retrogressive style (realism). Further- more, artists who followed their contemporaries’ style were considered more impactful when the predominant style was nonrealistic rather than realistic, � 0.16, SE � 0.07, 95% CI [0.03, 0.29], Z � 2.47, p � .014. The greater impact of nonrealistic artists likely reflects participants’ belief that pro- gressive styles, such as nonrealism, are more impactful than retrogressive styles, such as realism (see Studies S1, S2a, and S2b in the Online Supplemental Material). Figure 8. Meta-analytic effect of artist’s interpersonal deviance (no vs. yes) on artistic impact across Studies 4, 5, and 6. CI stands for confidence interval. Diamonds represent overall effects and rectangles represent effects of individual studies. When the CIs of an effect fall on the right [left] side of 0, deviant artists were considered more [less] impactful than nondeviant artists. Figure 9. Meta-analytic effect of artist’s interpersonal deviance (no vs. yes) and contemporaries’ style (realistic vs. nonrealistic) on artistic impact across Studies 4, 5, and 6. CI stands for confidence interval. Diamonds represent overall effects and rectangles represent effects of individual studies. When the CIs of an effect fall on the right [left] side of 0, artists who followed contemporaries’ nonrealistic style were considered more [less] impactful than artists who followed contemporaries’ realistic style. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 294 STAMKOU, VAN KLEEF, AND HOMAN http://dx.doi.org/10.1037/pspi0000131.supp General Discussion The current research investigated whether deviating from the norms in the world of art enhances an artist’s potential to rise to fame. The results of six studies demonstrate that various forms of artistic deviance influence perceived artistic impact as long as deviance is directed toward a progressive movement. In line with our hypotheses, artists who deviate from their previous artistic style are considered more impactful than artists who consistently follow a single style throughout their career, especially when the artistic path of deviant artists features a transition toward a pro- gressive style, such as nonrealism (intrapersonal deviance). More- over, artists who deviate from their contemporaries’ style gain greater impact than artists who follow their contemporaries’ style, in particular when artists deviate from the predominant style by adopting a progressive style, like nonrealism (interpersonal devi- ance). Contributions and Implications Our findings make a number of contributions to the literature. First, although breaking the rules in art may seem to be the new rule, our research has revealed a crucial boundary condition to the effectiveness of artistic deviance in terms of gaining impact: the progressiveness of deviance. At the intrapersonal level, we saw that artists increase their appeal by first establishing their place within the conventions of their era and then straying away from conventions, whereas artists who start off with ground-breaking work or stick to old practices are less acclaimed. These findings indicate that people are not only interested in art as an end product, but also in the process that led to it. This insight can inform art education programs, which should emphasize not only the devel- opment of a distinct style but also the acquisition of traditional techniques. At the interpersonal level, we observed that artists increase their impact when they break with conventions by adopt- ing a style that is considered progressive, whereas artists that differentiate themselves by adopting an old-fashioned style gain less acclaim. These findings imply that artists need to differentiate themselves from others in such a way that their distinct style is seen as the evolution of previously existing artwork styles; a distinct style that has been employed in the past is seen as back- ward movement and is appreciated less. Apparently, people want to see how an artist’s work contributes to the progress of art over time. Deviant artists could thus benefit from pitching their work as a linkage between past and future artistic trends. Progressiveness, however, is not a prerequisite for the positive effects of artistic deviance when people are not familiar with the historical signifi- cance of the artist’s expressive means. Our last study showed that when art is stripped from the historical context, deviance works irrespective of the means by which the artist deviated. This might imply that for areas of art of which people have less knowledge (e.g., artworks from unfamiliar cultures), deviance in general might be enough. Second, we have shown that artists who break with conventions are more likely to flourish because the public perceives their actions to be intentional. The role of intentionality was established by the mediation analyses in Studies 3, 5, and 6. Study 3 demon- strated that artists who stray away from an earlier realistic style by adopting nonrealism are perceived to have stronger will-power and thereby greater impact potential than artists who stray away from an earlier nonrealistic style by adopting realism. Although realistic artworks are considered less progressive, when they appear at an early stage of an artist’s career they irrefutably prove that the artist’s later nonrealistic works were not a product of incompe- tence but an intentional choice to shape a personal artistic path. Studies 5 and 6 showed that artists who deviate from their con- temporaries’ style are also thought of as having stronger will- power and therefore greater impact than artists who follow their contemporaries’ style. Artists who are impervious to external influences evince that their style choice is dictated by their will to move art forward by introducing innovative means of expression. These findings are consistent with research on the perception of deviant social targets in nonartistic domains who were considered more powerful because of their perceived volition (Van Kleef et al., 2011) or autonomy (Bellezza et al., 2014). Third, there is ample research on the potential of social move- ments to gain influence, but only limited research on the potential of artistic movements to gain ground (but see Inglis, 1996, and Lena & Pachucki, 2013). It is currently unknown whether common social influence mechanisms that have been widely investigated in social psychology and marketing, among other fields, generalize to the domain of art appreciation. For instance, our research provides evidence that the accumulation of innovation credits through con- formity is important in art too (Hollander, 1958). Future research could explore the application of social influence theories in the field of art by taking into account that art appreciation is governed by a unique set of rules that relate to social perceptual processes as well as the historical context (Bullot & Reber, 2013). In fact, our research points out how different aspects of artistic production, such as the evolution of styles within an artist’s career or between artists of an era, interact to explain artistic impact. Finally, our research shows that deviance in art is a decisive factor for people’s aesthetic preferences. This finding builds upon and enriches previous research that aims to explain aesthetic preferences (Heinrichs & Cupchik, 1985; Leder et al., 2004; Lin- dell & Mueller, 2011) and opens up new research directions by highlighting the important role of beholders’ perception of an artwork as deviant. This finding is also consistent with visual perception theories that consider the resolution of expectancy violations a crucial aspect of aesthetic pleasure (Huron, 2006; Van de Cruys & Wagemans, 2011) as well as philosophical theories of aesthetic appreciation, which maintain that the contrast between the negative state of obstruction and the positive state of resolution is the working ingredient of catharsis, that is, the purification of emotions through art (Aristotle, 1965). Strengths, Limitations, and Future Directions We have advanced and tested a parsimonious research model that delineates the effects of artistic deviance on artistic impact at different levels of analysis, in which progressiveness of deviance was conceptualized as a boundary condition and will-power as an underlying mechanism. Across five studies, we demonstrated con- sistent effects of the previous work of the artist and the work of the artists’ contemporaries on three aspects of artistic impact (i.e., influence, valuation, and attention). Furthermore, we used differ- ent operationalizations of artistic deviance that allowed us to rule out the influence of confounding variables (e.g., the artwork’s perceived abstractness) and alternative explanations of our find- T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 295ART OF INFLUENCE ings (e.g., preference for deviant art derived from participants’ stereotypes about the value of nonrealistic styles). Our effects were tested in a large sample that spanned a broad age range and included people from diverse educational backgrounds. Our con- clusions were corroborated in a series of meta-analyses that attest to the robustness of the overall effects as well as the homogeneity of the effects of the individual studies. Despite the strengths of our research, one may be concerned about the fact that in our studies we used real-world paintings rather than standardized stimuli. Even though standardized stimuli allow greater control over confounding factors, we opted for real-world stimuli to increase the ecological validity of our stim- ulus set. This approach addresses one of the main criticisms of empirical aesthetics, namely that research on aesthetic experiences is often reduced to the study of visual stimuli devoid of artistic meaning and historical context (Currie, 2003). Additionally, the results of Study 4, which involved exact copies of original art- works, were replicated in Study 5, where the most prominent aspects of the artworks were standardized. We thus conclude that the artwork stimuli we used are both ecologically and methodolog- ically valid. Another consideration may stem from people’s inferences about the age of the artworks. One may argue that in the deviance conditions of Studies 4 and 5 participants might have considered realistic paintings less impactful than the focal nonrealistic paint- ing because the realistic paintings were thought of as older- looking. Participants might have further thought that older paint- ings have more time to accumulate impact, so if they now look unfamiliar it means that they did not stand the test of time. This implies that observers might have evaluated the nonrealistic artist who deviated from his or her contemporaries’ style as more impactful because they considered the contemporaries’ realistic artworks older and unfamiliar rather than the nonrealistic artwork more deviant. However, in both Studies 4 and 5 participants were told that all artworks were made in approximately the same time period to keep the temporal component constant. This means that participants in this study could not have thought that the realistic artworks are less impactful because they look older than the nonrealistic artworks. Furthermore, in Study 6, in which we kept both the artwork style and the temporal component constant by using nonrealistic artworks that were made at the beginning of the century, people’s judgments of impact could only be attributed to artistic deviance rather than artwork age. Although inferences about the artwork’s age cannot explain the pattern of findings in our data, we do believe that investigating the role of artwork age on impact in future studies could further our understanding of the processes that affect artistic impact.5 Our studies reflect how people react to deviance in the context of art, which may limit the generalizability of our findings to artistic judgments. Future studies could further examine whether our model generalizes to judgments in other types of creative industries, such as advertising, fashion, film making, and music (Baker & Faulkner, 1991; Howkins, 2001). For example, it would be interesting to investigate whether experimental music genres, like jazz and contemporary classical music, are appreciated more when the audience becomes aware that they are derived from more conventional music genres, that is, blues and classical music, respectively (Burkholder, Grout, & Palisca, 2006). Another inter- esting avenue for future research is to examine how people’s responses to deviance in the artistic domain compare with their responses in nonartistic domains. Some domains have a restricted range of acceptable behaviors and leave little room for individual discretion in determining behavior (e.g., business), whereas other domains are more ambiguously structured and place fewer external constraints on individuals (e.g., art). The extent to which domains afford or constrain behavioral options has been referred to as situational constraint (Price & Bouffard, 1974). Domains of higher situational constraint may be less tolerant of deviant behav- ior than domains of lower situational constraint. Future research could thus examine whether the constraint of the domain affect people’s reactions to deviance. In our research, artistic deviance was studied within a Western cultural context, which potentially limits the generalizability of our findings. One of the main features of Western societies is the importance given to the concept of the individual (Triandis, 1987). Individualistic values in Western cultures have been an important force that fundamentally changed the way people think about deviance (Baumeister, 1987). The rise of individualism enforced artists’ concerns about being authentic and “staying true to them- selves” (the ideal of authenticity; Trilling, 1971) and allowed the development of idiosyncratic styles in art as well as a constant renunciation of the customary (the notion of counterwill; Rank, 1932/1989). One could therefore assume that reactions to artistic deviance depend on the cultural context. Individuals in Western cultures may have more favorable reactions to artistic deviance because deviant artists fulfill the cultural ideal of being unique. On the contrary, individuals in East Asian cultures may have less favorable reactions to artistic deviance because deviant artists do not fulfill the cultural ideal of conforming to normative standards (Kim & Markus, 1999). Different cultural ideals may also explain why East Asian cultures are more likely to foster incremental innovations, whereas Western cultures encourage more break- through ones (Herbig & Palumbo, 1996). Furthermore, the way people perceive individuals who deviate from the norm and the cultural values people endorse could account for the differential evolution of art movements in different parts of the world. Spe- cifically, artists in East Asia might have been driven to learn basic skills through diligent imitation of old masters, whereas artists in the Western world might have been driven to produce original works, which may explain the rapid succession of different art movements in the West. Future research could thus investigate how the effect of artistic deviance on impact plays out in nonwest- ern populations. Three studies that examined perceived will-power as underlying mechanism suggested that deviant artists are acclaimed because their behavior is considered willful. In other words, artistic devi- ance begets impact because it is seen as a means to some aesthetic or expressive goal. For instance, artists who deviate toward a realistic style may be considered less acclaimed than artists who deviate toward a nonrealistic style, because the latter form of deviance serves the goal of artistic progress. However, the reasons why artists deviate may vary. Artist A might deviate in a certain 5 Study S3, reported in the Online Supplemental Material, shows that older artworks are perceived as more influential than younger artworks; however, artwork age does not moderate the effect of artistic deviance on impact. These findings suggest that the effect of artistic deviance on impact is independent from the effect of artwork age. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 296 STAMKOU, VAN KLEEF, AND HOMAN http://dx.doi.org/10.1037/pspi0000131.supp direction because she wants to give voice to suffering, whereas artist B might deviate in the same direction because he realizes that the deviant style is more profitable. This raises the question, is willful deviance valued even if it is for ignoble or immoral rea- sons? Philosophical accounts that examine what is considered valuable in aesthetics hold that the evaluation of an artwork is often influenced by concerns about the morality of a certain decision (Guyer, 2005). Accordingly, future research could inves- tigate whether deviant artists whose behavior is considered moral may gain more impact than artists who deviate by the same expressive means but whose behavior is considered immoral. Perceived morality may therefore constitute an interesting bound- ary condition of the current findings.6 Another interesting boundary condition to the effects of artistic deviance on impact relates to observers’ condition of reception while viewing an artwork. Deviance in art may get an edge over convention in situations where the target artwork is not pertinent to urgent goals, and thus when people have the luxury to engage it on their own time (Kant, 1952/2007). Previous research provides suggestive evidence for this account. A study that examined the role of personal need for structure (PNS)—a construct that cap- tures variability in the need to apprehend the world in clear-cut terms— on appreciation of modern art found that high-PNS par- ticipants rated modern paintings less favorably than low-PNS participants. The proposed argument explaining this finding was that high-PNS participants felt more threatened by the lack of direct meaning that modern art may create. This may explain why high-PNS participants liked a modern artwork more when it was imbued with meaning by providing a constructive title (Landau et al., 2006). Another study showed that individuals high in need for cognitive closure (NFC)—a construct that correlates with PNS— evaluated nonrealistic paintings less favorably than realistic paint- ings as compared with individuals low in NFC. Also, under time pressure, individuals preferred realistic paintings more than non- realistic paintings (Wiersema et al., 2012). Similarly, Silvia (2005) reported that participants found visual artworks more interesting when their perceived ability to understand them increased. These findings suggest that observers may appreciate deviant artworks more when their psychological states or traits (e.g., low-PNS, low-NFS, low time pressure, high self-appraised ability to under- stand) allow them to engage the artworks on their own time. Future studies could shed further light on the conditions that facilitate the appreciation of deviant artistic work. Conclusion Although the popular notion “there is no accounting for taste” may be perennial, the current research indicates that people’s aesthetic judgments are reliably predicted by the degree to which and the way in which artists’ works deviate from prevailing artistic norms. We demonstrated that whether artists deviate from their previous style and the conventions of their era may determine the extent to which their artistic ideas persist or perish. The early works of Claude Monet and his associates prove that they were adept at producing a realistic representation of the world, whereas their impressionistic works attest that they were determined to break away from the iron cage of realistic art forms embraced by their contemporaries. The art critic who coined the term “Impres- sionists” to satirically refer to these deviant artists could have never imagined that their artistic work and vision would echo down the ages. 6 Study S4, reported in the Online Supplemental Material, provides suggestive evidence that perceptions of intentionality enhance deviant artists’ impact only when the reason for deviance is perceived to be moral—when deviance is not considered moral, intentionality may back- fire. References Alvarez, J. L., Mazza, C., Pedersen, J. S., & Svejenova, S. (2005). Shield- ing idiosyncrasy from isomorphic pressures: Towards optimal distinc- tiveness in European filmmaking. Organization, 12, 863– 888. http://dx .doi.org/10.1177/1350508405057474 Aristotle. (1965). On the art of poetry. In Classical literary criticism (T. S. Dorsch, Trans. & Ed.). Baltimore, MD: Penguin. Baker, W. E., & Faulkner, R. R. (1991). Role as resource in the Hollywood film industry. American Journal of Sociology, 97, 279 –309. http://dx .doi.org/10.1086/229780 Barkow, J. H., Cosmides, L., & Tooby, J. (1992). The adapted mind: Evolutionary psychology and the generation of culture. New York, NY: Oxford University Press. Baumeister, R. F. (1987). How the self became a problem: A psychological review of historical research. Journal of Personality and Social Psychol- ogy, 52, 163–176. http://dx.doi.org/10.1037/0022-3514.52.1.163 Bellezza, S., Gino, F., & Keinan, A. (2014). The red sneakers effect: Inferring status and competence from signals of nonconformity. Journal of Consumer Research, 41, 35–54. http://dx.doi.org/10.1086/674870 Berlyne, D. E. (1960). Conflict, arousal, and curiosity. New York, NY: McGraw-Hill. http://dx.doi.org/10.1037/11164-000 Blijlevens, J., Gemser, G., & Mugge, R. (2012). The importance of being ‘well-placed’: The influence of context on perceived typicality and esthetic appraisal of product appearance. Acta Psychologica, 139, 178 – 186. http://dx.doi.org/10.1016/j.actpsy.2011.11.004 Blood, A. J., & Zatorre, R. J. (2001). Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proceedings of the National Academy of Sciences of the United States of America, 98, 11818 –11823. http://dx.doi.org/10.1073/pnas .191355898 Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta-analysis. Chichester, UK: Wiley. http://dx.doi.org/ 10.1002/9780470743386 Boyd, R., & Richerson, P. (1992). Punishment allows the evolution of cooperation (or anything else) in sizable groups. Ethology and Sociobi- ology, 13, 171–195. http://dx.doi.org/10.1016/0162-3095(92)90032-y Bray, R. M., Johnson, D., & Chilstrom, J. T. (1982). Social influence by group members with minority opinions: A comparison of Hollander and Moscovici. Journal of Personality and Social Psychology, 43, 78 – 88. http://dx.doi.org/10.1037/0022-3514.43.1.78 Brewer, M. B. (1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 17, 475– 482. http://dx.doi.org/10.1177/0146167291175001 Brewer, M. B., & Pickett, C. L. (1999). Distinctiveness motives as a source of the social self. In T. R. Tyler, R. M. Kramer, & O. P. John (Eds.), The psychology of the social self (pp. 71– 87). Mahwah, NJ: Erlbaum. Brislin, R. W. (1986). The wording and translation of research instruments. In W. J. Lonner & J. W. Berry (Eds.), Field methods in cross-cultural research (pp. 137–164). Newbury Park, CA: Sage. Bullot, N. J. (2009). Material anamnesis and the prompting of aesthetic worlds: The psycho-historical theory of artworks. Journal of Conscious- ness Studies, 16, 85–109. Retrieved from http://www.ingentaconnect .com/content/imp/jcs/2009/00000016/00000001/art00004 T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 297ART OF INFLUENCE http://dx.doi.org/10.1037/pspi0000131.supp http://dx.doi.org/10.1177/1350508405057474 http://dx.doi.org/10.1177/1350508405057474 http://dx.doi.org/10.1086/229780 http://dx.doi.org/10.1086/229780 http://dx.doi.org/10.1037/0022-3514.52.1.163 http://dx.doi.org/10.1086/674870 http://dx.doi.org/10.1037/11164-000 http://dx.doi.org/10.1016/j.actpsy.2011.11.004 http://dx.doi.org/10.1073/pnas.191355898 http://dx.doi.org/10.1073/pnas.191355898 http://dx.doi.org/10.1002/9780470743386 http://dx.doi.org/10.1002/9780470743386 http://dx.doi.org/10.1016/0162-3095%2892%2990032-y http://dx.doi.org/10.1037/0022-3514.43.1.78 http://dx.doi.org/10.1177/0146167291175001 http://www.ingentaconnect.com/content/imp/jcs/2009/00000016/00000001/art00004 http://www.ingentaconnect.com/content/imp/jcs/2009/00000016/00000001/art00004 Bullot, N. J., & Reber, R. (2013). The artful mind meets art history: Toward a psycho-historical framework for the science of art apprecia- tion. Behavioral and Brain Sciences, 36, 123–137. http://dx.doi.org/10 .1017/S0140525X12000489 Burkholder, J. P., Grout, D. J., & Palisca, C. V. (2006). A history of Western music. New York, NY: Norton. Burris, C. T., & Rempel, J. K. (2004). “It’s the end of the world as we know it”: Threat and the spatial-symbolic self. Journal of Personality and Social Psychology, 86, 19 – 42. http://dx.doi.org/10.1037/0022-3514 .86.1.19 Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015– 1026. http://dx.doi.org/10.1037/0022-3514.58.6.1015 Codol, J. P. (1987). Comparability and incomparability between oneself and others: Means of differentiation and comparison reference points. Cahiers de Psychologie Cognitive/Current Psychology of Cognition, 7, 1–19. Currie, G. (2003). Aesthetics and cognitive science. In J. Levinson (Ed.), The Oxford handbook of aesthetics (pp. 706 –721). New York, NY: Oxford University Press. Daix, P. (1994). Picasso: Life and art. New York, NY: Harper Collins. Davies, D. (2004). Art as performance. Oxford, UK: Blackwell. http://dx .doi.org/10.1002/9780470774922 de-Wit, L., Machilsen, B., & Putzeys, T. (2010). Predictive coding and the neural response to predictable stimuli. The Journal of Neuroscience, 30, 8702– 8703. http://dx.doi.org/10.1523/JNEUROSCI.2248-10.2010 Dutton, J. E., Roberts, L. M., & Bednar, J. S. (2010). Pathways for positive identity construction at work: Four types of positive identity and the building of social resources. The Academy of Management Review, 35, 265–293. http://dx.doi.org/10.5465/AMR.2010.48463334 Farkas, A. (2002). Prototypicality-effect in surrealist paintings. Empirical Studies of the Arts, 20, 127–136. http://dx.doi.org/10.2190/UD7Y-GN8P- Q0EV-Q13J Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G�Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191. http:// dx.doi.org/10.3758/BF03193146 Feist, G. J. (1998). A meta-analysis of personality in scientific and artistic creativity. Personality and Social Psychology Review, 2, 290 –309. http://dx.doi.org/10.1207/s15327957pspr0204_5 Feldman, D. C. (1984). The development and enforcement of group norms. The Academy of Management Review, 9, 47–53. Förster, J., Friedman, R., Butterbach, E. B., & Sassenberg, K. (2005). Automatic effects of deviance cues on creative cognition. European Journal of Social Psychology, 35, 345–359. http://dx.doi.org/10.1002/ ejsp.253 Fromkin, H. L. (1970). Effects of experimentally aroused feelings of undistinctiveness upon valuation of scarce and novel experiences. Jour- nal of Personality and Social Psychology, 16, 521–529. http://dx.doi .org/10.1037/h0030059 Fromkin, H. L. (1972). Feelings of interpersonal undistinctiveness: An unpleasant affective state. Journal of Experimental Research in Person- ality, 6, 178 –182. Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 85, 453– 466. http://dx.doi.org/10.1037/0022-3514.85.3.453 Gocłowska, M. A., Crisp, R. J., & Labuschagne, K. (2013). Can counter- stereotypes boost flexible thinking? Group Processes & Intergroup Relations, 16, 217–231. http://dx.doi.org/10.1177/1368430212445076 Gombrich, E. H. (1985). A sense of order (2nd ed.). London, UK: Phaidon. Gombrich, E. H. (1995). The story of art (16th ed.). London, UK: Phaidon. Guyer, P. (2005). Values of beauty: Historical essays in aesthetics. Cam- bridge, UK: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511840876 Harari, Y. N. (2015). Sapiens: A brief history of humankind. New York, NY: Harper Collins Publishers. Hayes, A. F. (2012). PROCESS: A versatile computational tool for ob- served variable mediation, moderation, and conditional process model- ing [White paper]. Retrieved from http://www.afhayes.com/public/ process2012.pdf Heerdink, M. W., van Kleef, G. A., Homan, A. C., & Fischer, A. H. (2013). On the social influence of emotions in groups: Interpersonal effects of anger and happiness on conformity versus deviance. Journal of Person- ality and Social Psychology, 105, 262–284. http://dx.doi.org/10.1037/ a0033362 Heine, S. J. (2015). Cultural psychology (3rd ed.). New York, NY: Norton. Heinrichs, R. W., & Cupchik, G. C. (1985). Individual differences as predictors of preference in visual art. Journal of Personality, 53, 502– 515. http://dx.doi.org/10.1111/j.1467-6494.1985.tb00379.x Helweg-Larsen, M., & LoMonaco, B. L. (2008). Queuing among U2 fans: Reactions to social norm violations. Journal of Applied Social Psychology, 38, 2378 –2393. http://dx.doi.org/10.1111/j.1559-1816.2008.00396.x Herbig, P. A., & Palumbo, F. A. (1996). Innovation - Japanese style. Industrial Management & Data Systems, 96, 11–20. http://dx.doi.org/10 .1108/02635579610123299 Hollander, E. P. (1958). Conformity, status, and idiosyncrasy credit. Psy- chological Review, 65, 117–127. http://dx.doi.org/10.1037/h0042501 Hornsey, M. J., Jetten, J., McAuliffe, B. J., & Hogg, M. A. (2006). The impact of individualist and collectivist group norms on evaluations of dissenting group members. Journal of Experimental Social Psychology, 42, 57– 68. http://dx.doi.org/10.1016/j.jesp.2005.01.006 Howkins, J. (2001). The creative economy: How people make money from ideas. London, UK: Penguin. Huron, D. (2006). Sweet anticipation: Music and the psychology of expec- tation. Cambridge, MA: MIT Press. Inglis, I. (1996). Ideology, trajectory & stardom: Elvis Presley & The Beatles. International Review of the Aesthetics and Sociology of Music, 27, 53–78. http://dx.doi.org/10.2307/3108371 Jetten, J., & Hornsey, M. J. (2014). Deviance and dissent in groups. Annual Review of Psychology, 65, 461– 485. http://dx.doi.org/10.1146/annurev- psych-010213-115151 Jonas, E., McGregor, I., Klackl, J., Agroskin, D., Fritsche, I., Holbrook, C., . . . Quirin, M. (2014). Threat and defense: From anxiety to approach. Advances in Experimental Social Psychology, 49, 219 –286. http://dx.doi .org/10.1016/B978-0-12-800052-6.00004-4 Kam, C. C-S., & Bond, M. H. (2009). Emotional reactions of anger and shame to the norm violation characterizing episodes of interpersonal harm. British Journal of Social Psychology, 48, 203–219. http://dx.doi .org/10.1348/014466608X324367 Kant, I. (2007). Critique of judgement (J. C. Meredith, Trans.). Oxford, UK: Oxford University Press. (Original work published 1952) Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition. Psychological Review, 110, 265–284. http://dx.doi.org/ 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 Psychol- ogy, 77, 785– 800. http://dx.doi.org/10.1037/0022-3514.77.4.785 Koffka, K. (1935). Principles of Gestalt psychology. New York, NY: Harcourt-Brace. Lakens, D., & Evers, E. R. K. (2014). Sailing from the seas of chaos into the corridor of stability: Practical recommendations to increase the informational value of studies. Perspectives on Psychological Science, 9, 278 –292. http://dx.doi.org/10.1177/1745691614528520 Landau, M. J., Greenberg, J., Solomon, S., Pyszczynski, T., & Martens, A. (2006). Windows into nothingness: Terror management, meaningless- T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 298 STAMKOU, VAN KLEEF, AND HOMAN http://dx.doi.org/10.1017/S0140525X12000489 http://dx.doi.org/10.1017/S0140525X12000489 http://dx.doi.org/10.1037/0022-3514.86.1.19 http://dx.doi.org/10.1037/0022-3514.86.1.19 http://dx.doi.org/10.1037/0022-3514.58.6.1015 http://dx.doi.org/10.1002/9780470774922 http://dx.doi.org/10.1002/9780470774922 http://dx.doi.org/10.1523/JNEUROSCI.2248-10.2010 http://dx.doi.org/10.5465/AMR.2010.48463334 http://dx.doi.org/10.2190/UD7Y-GN8P-Q0EV-Q13J http://dx.doi.org/10.2190/UD7Y-GN8P-Q0EV-Q13J http://dx.doi.org/10.3758/BF03193146 http://dx.doi.org/10.3758/BF03193146 http://dx.doi.org/10.1207/s15327957pspr0204_5 http://dx.doi.org/10.1002/ejsp.253 http://dx.doi.org/10.1002/ejsp.253 http://dx.doi.org/10.1037/h0030059 http://dx.doi.org/10.1037/h0030059 http://dx.doi.org/10.1037/0022-3514.85.3.453 http://dx.doi.org/10.1177/1368430212445076 http://dx.doi.org/10.1017/CBO9780511840876 http://dx.doi.org/10.1017/CBO9780511840876 http://www.afhayes.com/public/process2012.pdf http://www.afhayes.com/public/process2012.pdf http://dx.doi.org/10.1037/a0033362 http://dx.doi.org/10.1037/a0033362 http://dx.doi.org/10.1111/j.1467-6494.1985.tb00379.x http://dx.doi.org/10.1111/j.1559-1816.2008.00396.x http://dx.doi.org/10.1108/02635579610123299 http://dx.doi.org/10.1108/02635579610123299 http://dx.doi.org/10.1037/h0042501 http://dx.doi.org/10.1016/j.jesp.2005.01.006 http://dx.doi.org/10.2307/3108371 http://dx.doi.org/10.1146/annurev-psych-010213-115151 http://dx.doi.org/10.1146/annurev-psych-010213-115151 http://dx.doi.org/10.1016/B978-0-12-800052-6.00004-4 http://dx.doi.org/10.1016/B978-0-12-800052-6.00004-4 http://dx.doi.org/10.1348/014466608X324367 http://dx.doi.org/10.1348/014466608X324367 http://dx.doi.org/10.1037/0033-295X.110.2.265 http://dx.doi.org/10.1037/0033-295X.110.2.265 http://dx.doi.org/10.1037/0022-3514.77.4.785 http://dx.doi.org/10.1177/1745691614528520 ness, and negative reactions to modern art. Journal of Personality and Social Psychology, 90, 879 – 892. http://dx.doi.org/10.1037/0022-3514 .90.6.879 Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. British Journal of Psy- chology, 95, 489 –508. http://dx.doi.org/10.1348/0007126042369811 Lena, J. C., & Pachucki, M. C. (2013). The sincerest form of flattery: Innovation, repetition and status in an art movement. Poetics, 41, 236 – 264. http://dx.doi.org/10.1016/j.poetic.2013.02.002 Levinson, J. (2007). Aesthetic contextualism. Postgraduate Journal of Aesthetics, 4, 1–12. Leyens, J. P., Yzerbyt, V. Y., & Rogier, A. (1997). Personality traits that distinguish you and me are better memorized. European Journal of Social Psychology, 27, 511–522. http://dx.doi.org/10.1002/(SICI)1099- 0992(199709/10)27:5�511::AID-EJSP827 3.0.CO;2-7 Lindell, A., & Mueller, J. (2011). Can science account for taste? Psycho- logical insights into art appreciation. Journal of Cognitive Psychology, 23, 453– 475. http://dx.doi.org/10.1080/20445911.2011.539556 Maddux, W. W., Adam, H., & Galinsky, A. D. (2010). When in Rome . . . learn why the Romans do what they do: How multicultural learning experiences facilitate creativity. Personality and Social Psychology Bul- letin, 36, 731–741. http://dx.doi.org/10.1177/0146167210367786 Maddux, W. W., & Galinsky, A. D. (2009). Cultural borders and mental barriers: The relationship between living abroad and creativity. Journal of Personality and Social Psychology, 96, 1047–1061. http://dx.doi.org/ 10.1037/a0014861 Magee, J. C. (2009). Seeing power in action: The roles of deliberation, implementation, and action in inferences of power. Journal of Experimental Social Psychology, 45, 1–14. http://dx.doi.org/10.1016/j.jesp.2008.06.010 Martindale, C. (1990). A clockwork muse: The predictability of artistic change. New York, NY: Basic Books. Martindale, C., & Moore, K. (1988). Priming, prototypicality, and preference. Journal of Experimental Psychology: Human Perception and Performance, 14, 661– 670. http://dx.doi.org/10.1037/0096-1523.14.4.661 Martindale, C., Moore, K., & West, A. (1988). Relationship of preference judgments to typicality, novelty, and mere exposure. Empirical Studies of the Arts, 6, 79 –96. http://dx.doi.org/10.2190/MCAJ-0GQT-DJTL- LNQD Mueller, J. S., Melwani, S., & Goncalo, J. A. (2012). The bias against creativity: Why people desire but reject creative ideas. Psychological Science, 23, 13–17. http://dx.doi.org/10.1177/0956797611421018 Nunn, N. (2008). The long-term effects of Africa’s slave trades. The Quarterly Journal of Economics, 123, 139 –176. http://dx.doi.org/10.1162/qjec.2008 .123.1.139 Ohbuchi, K. O., Tamura, T., Quigley, B. M., Tedeschi, J. T., Madi, N., Bond, M. H., & Mummendey, A. (2004). Anger, blame, and dimensions of perceived norm violations: Culture, gender, and relationships. Journal of Applied Social Psychology, 34, 1587–1603. http://dx.doi.org/10.1111/ j.1559-1816.2004.tb02788.x Okimoto, T. G., & Brescoll, V. L. (2010). The price of power: Power seeking and backlash against female politicians. Personality and Social Psychology Bulletin, 36, 923–936. http://dx.doi.org/10.1177/01461 67210371949 Palmer, S. E., Schloss, K. B., & Sammartino, J. (2013). Visual aesthetics and human preference. Annual Review of Psychology, 64, 77–107. http://dx.doi.org/10.1146/annurev-psych-120710-100504 Popa, M., Phillips, B. J., & Robertson, C. (2014). Positive outcomes of social norm transgressions. Journal of Consumer Behaviour, 13, 351– 363. http://dx.doi.org/10.1002/cb.1483 Price, R. H., & Bouffard, D. L. (1974). Behavioral appropriateness and situational constraint as dimensions of social behavior. Journal of Per- sonality and Social Psychology, 30, 579 –586. http://dx.doi.org/10.1037/ h0037037 Proulx, T., Heine, S. J., & Vohs, K. D. (2010). When is the unfamiliar the uncanny? Meaning affirmation after exposure to absurdist literature, humor, and art. Personality and Social Psychology Bulletin, 36, 817– 829. http://dx.doi.org/10.1177/0146167210369896 Rank, O. (1989). Art and artist: Creative urge and personality development (C. Atkinson, Trans.). New York, NY: Norton. (Original work published 1932) Raudenbush, S. W. (2009). Analyzing effect sizes: Random-effects mod- els. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (pp. 295–315). New York, NY: Russell Sage Foundation. Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and aesthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review, 8, 364 –382. http://dx.doi .org/10.1207/s15327957pspr0804_3 Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective judgments. Psychological Science, 9, 45– 48. http:// dx.doi.org/10.1111/1467-9280.00008 Rosch, E. (1975). Cognitive representations of semantic categories. Jour- nal of Experimental Psychology: General, 104, 192–233. http://dx.doi .org/10.1037/0096-3445.104.3.192 Rosenblum, N. (1989). A world history of photography. New York, NY: Abbeville Press. Sammartino, J., & Palmer, S. E. (2012). Aesthetic issues in spatial com- position: Representational fit and the role of semantic context. Percep- tion, 41, 1434 –1457. http://dx.doi.org/10.1068/p7233 Schonberg, H. C. (1997). The lives of the great composers. New York, NY: Norton. Schönfeld, S., & Reinstaller, A. (2007). The effects of gallery and artist reputation on prices in the primary market for art: A note. Journal of Cultural Economics, 31, 143–153. http://dx.doi.org/10.1007/s10824-007- 9031-1 Silvia, P. J. (2005). What is interesting? Exploring the appraisal structure of interest. Emotion, 5, 89 –102. http://dx.doi.org/10.1037/1528-3542.5 .1.89 Silvia, P. J. (2006). Exploring the psychology of interest (1st ed.). Oxford, UK: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/ 9780195158557.001.0001 Silvia, P. J. (2008). Interest - the curious emotion. Current Directions in Psychological Science, 17, 57– 60. http://dx.doi.org/10.1111/j.1467-8721 .2008.00548.x Sloboda, J. A. (1991). Music structure and emotional response: Some empirical findings. Psychology of Music, 19, 110 –120. http://dx.doi.org/ 10.1177/0305735691192002 Stamkou, E., & Van Kleef, G. A. (2014). Do we give power to the right people? When and how norm violators rise to the top. In J.-W. van Prooijen & P. A. M. van Lange (Eds.), Power, politics, and paranoia: Why people are suspicious about their leaders (pp. 33–52). Cambridge, UK: Cambridge University Press. http://dx.doi.org/10.1017/CBO97 81139565417.005 Stamkou, E., Van Kleef, G. A., Homan, A. C., & Galinsky, A. D. (2016). How norm violations shape social hierarchies: Those who stand on top block norm violators from rising up. Group Processes & Intergroup Relations, 19, 608 – 629. http://dx.doi.org/10.1177/1368430216641305 Stecker, R. (2003). Value in art. In J. Levinson (Ed.), The Oxford handbook of aesthetics (pp. 307–324). Oxford, UK: Oxford University Press. Stone, T. H., & Cooper, W. H. (2009). Emerging credits. The Leadership Quarterly, 20, 785–798. http://dx.doi.org/10.1016/j.leaqua.2009.06.007 Swafford, J. (2014). Beethoven: Anguish and triumph. London, UK: Faber & Faber. Tadmor, C. T., Galinsky, A. D., & Maddux, W. W. (2012). Getting the most out of living abroad: Biculturalism and integrative complexity as key drivers of creative and professional success. Journal of Personality T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 299ART OF INFLUENCE http://dx.doi.org/10.1037/0022-3514.90.6.879 http://dx.doi.org/10.1037/0022-3514.90.6.879 http://dx.doi.org/10.1348/0007126042369811 http://dx.doi.org/10.1016/j.poetic.2013.02.002 http://dx.doi.org/10.1002/%28SICI%291099-0992%28199709/10%2927:5%3C511::AID-EJSP827%3E3.0.CO;2-7 http://dx.doi.org/10.1002/%28SICI%291099-0992%28199709/10%2927:5%3C511::AID-EJSP827%3E3.0.CO;2-7 http://dx.doi.org/10.1080/20445911.2011.539556 http://dx.doi.org/10.1177/0146167210367786 http://dx.doi.org/10.1037/a0014861 http://dx.doi.org/10.1037/a0014861 http://dx.doi.org/10.1016/j.jesp.2008.06.010 http://dx.doi.org/10.1037/0096-1523.14.4.661 http://dx.doi.org/10.2190/MCAJ-0GQT-DJTL-LNQD http://dx.doi.org/10.2190/MCAJ-0GQT-DJTL-LNQD http://dx.doi.org/10.1177/0956797611421018 http://dx.doi.org/10.1162/qjec.2008.123.1.139 http://dx.doi.org/10.1162/qjec.2008.123.1.139 http://dx.doi.org/10.1111/j.1559-1816.2004.tb02788.x http://dx.doi.org/10.1111/j.1559-1816.2004.tb02788.x http://dx.doi.org/10.1177/0146167210371949 http://dx.doi.org/10.1177/0146167210371949 http://dx.doi.org/10.1146/annurev-psych-120710-100504 http://dx.doi.org/10.1002/cb.1483 http://dx.doi.org/10.1037/h0037037 http://dx.doi.org/10.1037/h0037037 http://dx.doi.org/10.1177/0146167210369896 http://dx.doi.org/10.1207/s15327957pspr0804_3 http://dx.doi.org/10.1207/s15327957pspr0804_3 http://dx.doi.org/10.1111/1467-9280.00008 http://dx.doi.org/10.1111/1467-9280.00008 http://dx.doi.org/10.1037/0096-3445.104.3.192 http://dx.doi.org/10.1037/0096-3445.104.3.192 http://dx.doi.org/10.1068/p7233 http://dx.doi.org/10.1007/s10824-007-9031-1 http://dx.doi.org/10.1007/s10824-007-9031-1 http://dx.doi.org/10.1037/1528-3542.5.1.89 http://dx.doi.org/10.1037/1528-3542.5.1.89 http://dx.doi.org/10.1093/acprof:oso/9780195158557.001.0001 http://dx.doi.org/10.1093/acprof:oso/9780195158557.001.0001 http://dx.doi.org/10.1111/j.1467-8721.2008.00548.x http://dx.doi.org/10.1111/j.1467-8721.2008.00548.x http://dx.doi.org/10.1177/0305735691192002 http://dx.doi.org/10.1177/0305735691192002 http://dx.doi.org/10.1017/CBO9781139565417.005 http://dx.doi.org/10.1017/CBO9781139565417.005 http://dx.doi.org/10.1177/1368430216641305 http://dx.doi.org/10.1016/j.leaqua.2009.06.007 and Social Psychology, 103, 520 –542. http://dx.doi.org/10.1037/a002 9360 Tennie, C., Call, J., & Tomasello, M. (2009). Ratcheting up the ratchet: On the evolution of cumulative culture. Philosophical Transactions of the Royal Society of London Series B, Biological Sciences, 364, 2405–2415. http://dx.doi.org/10.1098/rstb.2009.0052 Triandis, H. C. (1987). Collectivism vs. individualism: A reconceptualiza- tion of a basic concept in cross-cultural social psychology. In C. Bagley & G. K. Verma (Eds.), Personality, cognition and values: Cross-cultural perspectives of childhood and adolescence. London, UK: Macmillan. Trilling, L. (1971). Sincerity and authenticity. Cambridge, MA: Harvard University Press. Van de Cruys, S., & Wagemans, J. (2011). Putting reward in art: A tentative prediction error account of visual art. i-Perception, 2, 1035– 1062. http://dx.doi.org/10.1068/i0466aap Van Kleef, G. A., Homan, A. C., Finkenauer, C., Blaker, N. M., & Heerdink, M. W. (2012). Prosocial norm violations fuel power affor- dance. Journal of Experimental Social Psychology, 48, 937–942. http:// dx.doi.org/10.1016/j.jesp.2012.02.022 Van Kleef, G. A., Homan, A. C., Finkenauer, C., Gündemir, S., & Stam- kou, 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, 500 –507. http://dx.doi.org/10.1177/1948550611398416 Van Kleef, G. A., Wanders, F., Stamkou, E., & Homan, A. C. (2015). The social dynamics of breaking the rules: Antecedents and consequences of norm-violating behavior. Current Opinion in Psychology, 6, 25–31. http://dx.doi.org/10.1016/j.copsyc.2015.03.013 Van Tilburg, W. A. P., & Igou, E. R. (2014). From Van Gogh to Lady Gaga: Artist eccentricity increases perceived artistic skill and art appre- ciation. European Journal of Social Psychology, 44, 93–103. http://dx .doi.org/10.1002/ejsp.1999 Vignoles, V. L., Chryssochoou, X., & Breakwell, G. M. (2000). The distinctiveness principle: Identity, meaning and the bounds of cultural relativity. Personality and Social Psychology Review, 4, 337–354. http:// dx.doi.org/10.1207/S15327957PSPR0404_4 Voigtländer, N., & Voth, H.-J. (2012). Persecution perpetuated: The medieval origins of anti-Semitic violence in Nazi Germany. The Quarterly Journal of Economics, 127, 1339 –1392. http://dx.doi.org/10.1093/qje/qjs019 Whitfield, T., & Slatter, P. (1979). The effects of categorization and proto- typicality on aesthetic choice in a furniture selection task. British Journal of Psychology, 70, 65–75. http://dx.doi.org/10.1111/j.2044-8295.1979 .tb02144.x Wiersema, D. V., Van Der Schalk, J., & van Kleef, G. A. (2012). Who’s afraid of red, yellow, and blue? Need for cognitive closure predicts aesthetic preferences. Psychology of Aesthetics, Creativity, and the Arts, 6, 168 –174. http://dx.doi.org/10.1037/a0025878 Yard, S. (2007). Willem De Kooning: Works, writings and interviews. Barcelona, Spain: Ediciones Poligrafa. Yukl, G. (2010). Leadership in organizations (7th ed.). Englewood Cliffs, NJ: Prentice Hall. Zhou, H., & Fishbach, A. (2016). The pitfall of experimenting on the web: How unattended selective attrition leads to surprising (yet false) research conclusions. Journal of Personality and Social Psychology, 111, 493– 504. http://dx.doi.org/10.1037/pspa0000056 Zuckerman, E. W. (1999). The categorical imperative: Securities analysts and the illegitimacy discount. American Journal of Sociology, 104, 1398 –1438. http://dx.doi.org/10.1086/210178 T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 300 STAMKOU, VAN KLEEF, AND HOMAN http://dx.doi.org/10.1037/a0029360 http://dx.doi.org/10.1037/a0029360 http://dx.doi.org/10.1098/rstb.2009.0052 http://dx.doi.org/10.1068/i0466aap http://dx.doi.org/10.1016/j.jesp.2012.02.022 http://dx.doi.org/10.1016/j.jesp.2012.02.022 http://dx.doi.org/10.1177/1948550611398416 http://dx.doi.org/10.1016/j.copsyc.2015.03.013 http://dx.doi.org/10.1002/ejsp.1999 http://dx.doi.org/10.1002/ejsp.1999 http://dx.doi.org/10.1207/S15327957PSPR0404_4 http://dx.doi.org/10.1207/S15327957PSPR0404_4 http://dx.doi.org/10.1093/qje/qjs019 http://dx.doi.org/10.1111/j.2044-8295.1979.tb02144.x http://dx.doi.org/10.1111/j.2044-8295.1979.tb02144.x http://dx.doi.org/10.1037/a0025878 http://dx.doi.org/10.1037/pspa0000056 http://dx.doi.org/10.1086/210178 Appendix A Artworks Used in Study 2 Artworks used are listed per condition: No intrapersonal deviance—Only realistic style (top row), No intrapersonal deviance—Only nonrealistic style (middle row), and Intrapersonal deviance—Mixed style (bottom row). Still life, Willem De Kooning Portrait of Elaine, Willem De Kooning Untitled (still life), Willem De Kooning Drawings, Winter-Spring, Graham White Untitled (cow jumps over the moon), Willem De Kooning Untitled (charcoal drawing), Willem De Kooning Secretary, Willem De Kooning Figure, Willem De Kooning Copyright on works of visual artists affiliated to a CISAC organization has been arranged with Pictoright in Amsterdam. © c/o Pictoright Amsterdam 2018. © Graham White, 2018. (Appendices continue) T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 301ART OF INFLUENCE Appendix B Artworks Used in Study 3 Artworks used are listed per condition: No intrapersonal deviance/Realistic early style (top row), No intrapersonal deviance/Nonrealistic early style (upper middle row), Intrapersonal deviance/Realistic early style (lower middle row), and Intrapersonal deviance/Nonrealistic early style (bottom row). Bull (plate II), Bull (plate III), Bull (plate VII), Bull (plate IX) by Pablo Picasso Copyright on works of visual artists affiliated to a CISAC organization has been arranged with Pictoright in Amsterdam. © c/o Pictoright Amsterdam 2018. See the online article for the color version of this figure. (Appendices continue) T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 302 STAMKOU, VAN KLEEF, AND HOMAN Appendix C Artworks Used in Study 4 Artworks used are listed per condition: No interpersonal deviance/Contemporaries’ realistic style (top row), No interpersonal deviance/Contemporaries’ nonrealistic style (upper middle row), Interpersonal deviance/Contemporaries’ realistic style (lower middle row), Interpersonal deviance/Contemporaries’ nonrealistic style (bottom row). Note. The focal artwork appears in the middle position of each row. Braids (Helga Terstorf), Andrew Wyeth Portrait of the Artist’s Mother, Pablo Picasso Portrait of a Woman, Boris Kustodiev Portrait of Katharina Cornell, Salvador Dali Portrait of Jacqueline Roque With Her Hands Crossed, Pablo Picasso Ridens, Francis Picabia Copyright on works of visual artists affiliated to a CISAC organization has been arranged with Pictoright in Amsterdam. © c/o Pictoright Amsterdam 2018. See the online article for the color version of this figure. Received November 10, 2016 Revision received December 22, 2017 Accepted December 29, 2017 � T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 303ART OF INFLUENCE The Art of Influence: When and Why Deviant Artists Gain Impact How Deviance Shapes Artistic Impact Considering Artistic Deviance in Its Social Context Intrapersonal Deviance Interpersonal Deviance Considering Artistic Deviance in Its Historical Context Overview of Model and Hypotheses Overview of Studies Study 1 Method Results Discussion Study 2 Method Sample Design, materials, and procedure Results Perceived influence Price estimation Discussion Study 3 Method Sample Design, materials, and procedure Results Perceived intrapersonal deviance (manipulation check) Perceived influence Perceived will-power Moderated mediation Discussion Study 4 Method Sample Design, materials, and procedure Results Discussion Study 5 Method Sample Design, materials, and procedure Results Perceived interpersonal deviance (manipulation check) Perceived influence Purchase intention Visual attention Perceived will-power Moderated mediation Discussion Study 6 Method Sample Design, materials, and procedure Results Perceived interpersonal deviance (manipulation check) Perceived influence Perceived will-power Mediation Discussion Meta-Analytic Synthesis of Findings Intrapersonal Deviance Main Effect Interpersonal Deviance Main and Interaction Effects General Discussion Contributions and Implications Strengths, Limitations, and Future Directions Conclusion References Appendix AArtworks Used in Study 2 Appendix BArtworks Used in Study 3 Appendix CArtworks Used in Study 4