The Effect of Color on Automaticity of Aesthetic ... - SAGE Journals

0 downloads 0 Views 175KB Size Report
The role of cognition in art appreciation has been studied by a number of .... Much of the early experimental work on color in the late 1800s and early 1900s.
Research Article

The Effect of Color on Automaticity of Aesthetic Judgments

Empirical Studies of the Arts 2016, Vol. 34(1) 8–34 ! The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0276237415621183 art.sagepub.com

John W. Mullennix1, Amy Varmecky1, Chi H. Chan2, Stephen Korenoski1, Zach Mickey1, and Lisa Polaski-Hoffman1

Abstract The cognitive processing mode used to process color while making aesthetic judgments was examined. In two experiments, participants rated artistic photographs for preference. Half the participants performed a concurrent memory preload task while rating the photographs to index whether automatic or controlled modes of processing was engaged. Overall, the results showed that preload had little effect on the pattern of ratings for color, black and white, and false color versions of the photographs. The results suggest that color processing during the judgment task was automatic and required few cognitive resources. Furthermore, the results indicate that the dual-mode processing framework can be a useful theoretical tool for examining the cognitive processes that underlie aesthetic experience. Keywords aesthetics, photograph processing, dual-mode model

The contemporary study of the psychological aspects of aesthetics can be traced back to Gustav Fechner’s 1871 book On Experimental Aesthetics (Seeley, 2011). Since the time of Fechner, empirical aesthetics (Konecˇni, 2012; Seeley, 2011),

1 2

University of Pittsburgh at Johnstown, PA, USA Chatham University, Pittsburgh, PA, USA

Corresponding Author: J. W. Mullennix, University of Pittsburgh at Johnstown, PA, USA. Email: [email protected]

Mullennix et al.

9

that is, using the methodologies from experimental psychology to study aesthetics, has proved useful. In recent years, the methodologies used in neuroscience have also begun to exert an influence on the study of aesthetics, with neuroaesthetics becoming an important area of research as well (Chatterjee, 2010; Chatterjee & Vartanian, 2014; Cinzia & Vittorio, 2009; Leder, 2013; Nadal & Skov, 2013). Currently, there is much discussion about the research topics studied in both areas, and where the research is heading toward next (Chatterjee, 2010; Leder, 2013; Leder & Nadal, 2014; Vartanian, 2014). In the present study, we approach our research question from an empirical aesthetics perspective. The basic issue we address is that of the cognitive processes used to process color information when people make judgments about visual art. As described below, the theoretical framework, we approach this question from, is that of cognitive dual-mode processing.

Cognition, Dual-Processing, and Aesthetics The role of cognition in art appreciation has been studied by a number of researchers. Silvia (2005, 2007) suggested that cognitive appraisal plays an important role in the aesthetic experience, with appraisal defined as a cognitive evaluation that leads to interest in the aesthetic work. Other researchers have emphasized the role of cognitive classification and categorization processes in the aesthetic experience (Whitfield, 1983, 2000; Whitfield & Slatter, 1979). Leder, Belke, Oeberst, and Augustin (2004) proposed an information processing model of aesthetic experience that describes the sequence of stages of perceptualcognitive processing that lead to an aesthetic experience. This model has served as a useful framework from which to further investigate the cognitive aspects of aesthetic experience (Leder & Nadal, 2014). In the current investigation, we utilized a somewhat different theoretical framework. This framework is rooted in the notion of dual-mode processing. Dualmode processing is a concept used to explain the human attentional system (Schneider & Shiffrin, 1977; Shiffrin & Schneider, 1977; Tresiman & Gelade, 1980; Wickens, 1991), reasoning, judgment, decision-making processes (Evans, 2007, 2008; Kahneman, 2011), and social cognition and social inference (Chaiken & Trope, 1999; Kruglanski & Orehek, 2007; Smith & Collins, 2009). As summarized by Evans (2008), there are two cognitive information processing modes available to perform mental tasks: System 1 and System 2. System 1 processes are unconscious, implicit, automatic, low effort, rapid, high capacity, holistic, perceptual and are typically the default process. System 2 processes are conscious, explicit, controlled, high effort, slow, low capacity, analytic, and inhibitive. A simplified version of this dichotomy pits implicit, automatic cognitive processes versus explicit, controlled cognitive processes. Automatic processes are fast, operate independently of a person’s control and do not require conscious effort or measureable mental resources. Controlled processes are slow,

10

Empirical Studies of the Arts 34(1)

characterized by conscious expenditure of mental effort and compete for mental resources (Wickens, 1991). Brain imaging data suggest that these two systems have a neurophysiological basis, with different brain networks underlying automatic and controlled processes, respectively (e.g., Sridharan, Levitin, & Menon, 2008).

Dual-Mode Processing and Aesthetics A dual-mode processing model of aesthetic preference was alluded to by Hekkert, Snelders, and van Wieringen (2003) to explain aesthetic preference for consumer products such as telephones and automobiles. Hekkert et al. suggested that there is an automatic or immediate mechanism that processes easyto-classify stimuli and a controlled and cognitively mediated mechanism that processes incongruous or novel stimuli. Either process can be invoked depending on the situation and the circumstances that are in place when participants make judgments about aesthetic preference. More recently, a comprehensive dual-processing model of aesthetic liking called the Pleasure-Interest (PIA) Model was proposed by Graf and Landwehr (2015). Their model proposes two hierarchical processes that produce pleasure and interest in the aesthetic object. The first process is stimulus-driven and characterized by automatic processing. This process occurs immediately when the aesthetic object is encountered and produces an affective reaction in the viewer. This reaction is then interpreted as pleasure or displeasure. Graf and Landwehr claim that when people are asked to make a quick, spontaneous judgment about an aesthetic object, the output of their judgment is derived from this initial process. The second process is perceiver-driven and characterized by controlled processing. This process occurs when a person focuses deliberate attention on the aesthetic object in order to engage the object at a conceptual level, along with attention devoted to detailed stimulus and meaning analysis. The output from this process produces interest, boredom, or confusion in the viewer. Graf and Landwehr suggest that activation of this controlled process can “overwrite” the automatic response produced by the first process. Overall, they suggest that evaluation of an aesthetic object may differ depending on the processing approach engaged. This, in turn, can explain inconsistent preference patterns related to the complexity of aesthetic stimuli and the cognitive approach adopted by a person forming an aesthetic judgment. The role of dual-mode processes in aesthetic judgments was examined specifically in an empirical study by Mullennix et al. (2013). Mullennix et al. assessed whether automatic or controlled processes were used while naı¨ ve viewers (those will little experience in photography) made aesthetic preference judgments about artistic photographs. While participants made their judgments, they also performed a concurrent memory load task. This task required participants to hold series of random digits in memory for later recall while they simultaneously

Mullennix et al.

11

performed the preference task. This concurrent task has been used frequently in studies where a cognitive load is imposed on participants (Baddeley, 1986; Baddeley & Hitch, 1974; Navon & Gopher, 1979). The assumption is that if the concurrent task does not affect the primary task, the processing performed for the primary task is automatic and resource free. However, if the concurrent task affects the primary task, processing is considered to be controlled and resource intensive. Mullennix et al. found that the concurrent task had little effect on the preference judgments, a result consistent with an automatic processing mode explanation for processes used by viewers to make aesthetic judgments.

Color Processing In general, color plays an important role in people’s lives, helping to determine people’s aesthetic preferences for visual art, housing, consumer products, and so forth. As succinctly put by Johann Wolfgang von Goethe, “All nature manifests itself by means of colors to the sense of sight” (cited from Lopes, 1999, p. 416). There is a long history of experimental study on color and aesthetics (Ball, 1965). Much of the early experimental work on color in the late 1800s and early 1900s examined color preferences (Ball, 1965). Although research on color preferences continues to the present day (Hurlbert & Ling, 2007; Palmer & Schloss, 2010; Taylor, Clifford, & Franklin, 2013), over time the experimental study of color expanded into other areas. For example, recent research using brain imaging technology has allowed us to explore the neurophysiological underpinnings of color processing in the human brain. This has resulted in identifying specific brain areas tasked with processing information pertaining to color (Beauchamp, Haxby, Jennings, & DeYoe, 1999; Brama˜o, Faı´ sca, Forkstam, Reis, & Petersson, 2010; McKreefry & Zeki, 1997; Tanaka, Weiskopf, & Williams, 2001; Zeki & Marini, 1998). Other examples of experimental work on color examine color processing as part of the perceptual-cognitive processing of visual stimuli. In this work, the role that color plays in identifying visual objects and retaining visual information in memory has been elucidated (Biederman & Ju, 1988; Brama˜o et al., 2010; Castelhano & Henderson, 2008; Gegenfurtner & Rieger, 2000; Oliva & Schyns, 2000; Ostergaard & Davidoff, 1985; Polzella, Hammar, & Hinkle, 2005; Spence, Wong, Rusan, & Rastegar, 2006; Suzuki & Takahashi, 1997; Tanaka et al., 2001; Tanka & Presnell, 1999; Wichmann, Sharpe, & Gegenfurtner, 2002). There is evidence that color information facilitates the cognitive processing of visual scenes. Castelhano and Henderson (2008) found that normal colors associated with real world visual scenes helped to activate scene gist, with activation happening fairly rapidly. Gegenfurtner and Rieger (2000) showed that color scenes are remembered better than black and white scenes, with color facilitating recognition at both early and later levels in the cognitive system. Spence et al.

12

Empirical Studies of the Arts 34(1)

(2006) found a similar advantage for color in recognition memory for visual scenes but suggested that the effects of color take place very early in the visual processing system. However, other research indicates that color has little effect on visual object identification and classification (Biederman & Ju, 1988; Cave, Bost, & Cobb, 1996; Ostergaard & Davidoff, 1985). Tanaka et al. (2001) attempted to explain the discrepancies between studies by suggesting that both early-level processing of object color and higher level processing of knowledge stored about object color occur. In support of this idea, at the neurophysiological level, there appear to be different stages of processing of color in the brain related to the distinction between early-level and higher level processing. As stated by Zeki and Marini (1998), . . . Memory, learning and judgment are important additional faculties used by the colour system when colours invest objects and are part of them. The latter is the more usual condition and recruits additional cortical areas, well beyond the automatic computational stage that computes colours without reference to the actual objects. (pp. 1681–1682)

This division of color processing into an early, automatic, computational stage and a later stage that involves cortical networks responsible for memory, learning, and judgment seems similar in flavor to the division of cognitive processes into automatic and controlled modes of processing as posited by dual-mode processing theory.

The Present Study Our study was designed to examine how color is processed and used to make aesthetic judgments via a paradigm that indexed dual-mode processing (i.e., whether automatic or controlled processes were engaged). The basic question was the following: Does the presence of color in an artwork require the engagement of controlled (resource intensive) cognitive processes? The casual observer may believe that judgments about artworks containing color are “automatic” and performed without thinking. However, without precise behavioral measurements to support this casual observation, it remains an open question. There is some suggestion in the literature that color processing is not entirely automatic, in terms of cognitive processes. Wichmann et al. (2002) commented on one experiment in their study, “Experiment 4 shows that the well-known saliency of color (Davidoff, 1991) can improve recognition memory, most probably owing to a nonspecific increase in attention” (p. 518). The idea here is that the presence of color increases attention to stimuli, resulting in stronger memory traces. This suggests that controlled processes may be involved (although it is possible that the attraction of attention to color could also be handled via automatic processes). Other research suggests that color processing

Mullennix et al.

13

is influenced by top-down information stored about color (Tanaka et al., 2001). This utilization of top-down information could also involve higher level controlled processes, although it is also conceivable that the integration of this information into color processing is automatic. Zeki and Marini (1998) showed that high-level cortical areas at V4 and beyond related to higher level cognitive processes such as learning and memory are engaged in color processing. When higher level brain areas are involved, the likelihood that higher level resource-demanding controlled processes are invoked becomes greater, although interpretive care must be taken when reverse inference is involved (Poldrack, 2006). A second rationale for the present study was to extend the findings of Mullennix et al. (2013) on automaticity of processing for aesthetic judgments. Their results were based on participants making preference judgments for black and white artistic photographs. The study outlined below provides another important empirical test of the automaticity of processing idea by introducing an additional stimulus dimension (color) into the situation. If color does not affect processing, this would suggest that preference judgments are made automatically despite additional (color) variation in the stimulus parameters. However, if the introduction of color results in controlled processing, this would substantially change the way we think about dual-mode processing during the aesthetic experience. Regardless of which finding is observed, the results will provide valuable additional information about the cognitive processes used when people view and judge visual art. Finally, since participants in the present study made preference judgments about photographic visual art, the study possesses a degree of ecological validity. Preference decisions about visual art are typically made by people when they peruse art in galleries, print media, and online media. If we have a better understanding of the mental processes used when people make decisions about what they like and what they do not like, this could be useful to visual artists and those who market and display visual art. Perhaps, the composition, execution, and display of artworks could be influenced by knowledge about how color is processed under certain conditions. Also, the present study uses short-time frames for people to view and make decisions. As discussed later, this may have important implications for viewers who do not spend substantial time perusing artworks in galleries or museums. In Experiment 1, we compared preference judgments for color versions and black and white versions of artistic photographs. Comparing color to black and white versions of photographs or paintings is a useful way to assess the processing of color information (Biederman & Ju, 1988; Brama˜o et al., 2010; Castelhano & Henderson, 2008; Gegenfurtner & Rieger, 2000; Oliva & Schyns, 2000; Ostergaard & Davidoff, 1985; Polzella et al., 2005; Spence et al., 2006; Suzuki & Takahashi, 1997; Tanka & Presnell, 1999; Tanaka et al., 2001; Wichmann et al., 2002). We adapted the preload paradigm used by Mullennix

14

Empirical Studies of the Arts 34(1)

et al. (2013) for use. Specifically, participants with little experience in photography performed two tasks: A preference-rating task and a semantic differentialrating task. For participants in the preload condition, they were told to perform a concurrent memory task consisting of memorizing random sets of letters at the same time they made preference judgments for the photographs. The comparison of no-load to preload conditions indexed automaticity of processing. Each participant viewed the same photographs twice, once in color and once in black and white. Color was blocked by session, with the two sessions separated by 2 weeks. We felt it necessary to examine color as a within-subjects variable in order to possess enough power to pick up any subtle differences in processing between color and black and white versions of the photographs. However, the negative aspect of this arrangement is possible “contamination” effects produced by rating the same photographs twice. This issue is addressed later. Although preference judgements were the primary data related to our predictions, semantic differential ratings were also collected for each of the photographs. These ratings (see Appendix 2) provided additional information about the attributes of the photographs that may affect preference and brought to bear information about qualitative aspects of the photographs that could vary across color and black and white versions. In terms of predictions, the critical pattern of results for preference judgements lies with the two-way interaction of preload condition (no load vs. preload) and color condition (color vs. black and white), or the three-way interaction of preload condition with color condition and photograph. If controlled processing is engaged to process color information, then the pattern of preference ratings for photographs should differ significantly across preload conditions for color versions but not black and white versions (as illustrated by a two-way interaction). Alternatively, the pattern of preference ratings for photographs could differ across preload and color (as illustrated by a three-way interaction). However, if automatic processing is engaged, then the pattern of preference ratings should be similar across preload conditions for both color and black and white versions of the photographs.

Experiment 1 Method Participants. A total of 29 undergraduate students (4 men and 25 women) at the University of Pittsburgh at Johnstown participated in the study. The average age of the participants was 18.7 years (SD ¼ 1.2). Participants were solicited from Introductory Psychology courses and received course credit for their participation. Each person took part in two experimental sessions; 14 participants were run in the zero-letter condition (no-load condition), and 15 participants were run in the eight-letter condition (preload condition).

Mullennix et al.

15

To assess the amount of experience and knowledge that participants had with photography, they were asked six yes or no questions (see Appendix 1). One participant answered “yes” to four questions, two participants answered “yes” to two questions, and nine participants answered “yes” to one question. The sparse affirmative responses to the photography questions suggested that only 1 of 29 participants possessed any significant experience with photography. As well, the age range of participants precluded any lengthy track record of expertise with photography. Materials. A total of 32 color artistic photographs printed on commercially available standard sized postcards were used. Photographs were chosen that varied as much as possible across the attributes of Pleasantness, Clarity, Liveliness, and Familiarity (see Mullennix et al., 2013). Mullennix et al. found that these four attributes played a large role in terms of how participants formed their impressions of the photographs. Black and white versions of the photographs were created via Photoshop, with color removed using a gradient mapping option that translated color to grayscale and preserved luminosity values. All 64 photographs were printed on heavy photographic paper at a 400  600 size for use. The color and black and white versions of the photographs used are shown in Appendix 3. Design and procedure. The study was conducted across two sessions. For the preference task, we used a 2  2 mixed design, with a between-subjects variable of preload condition (zero or eight letters) and a within-subjects variable of color condition (color or black and white). For both sessions, a preference-scaling task was followed by a computerrating task. Participants received color versions of the photographs in one session and black and white versions of the photographs in the other session. The order of versions was counterbalanced across participants. The 32 photographs were randomly ordered and given to participants in a stack. A 60 in visual preference scale was used, with scale endpoints of 0 and 10 (0 corresponding to least liked and 10 corresponding to most liked). Participants were told to examine each photograph and place it on the scale in a way that corresponded to their personal preference. They were told that the photographs did not have to be placed at equal distances from each other on the scale and that the photographs could be bunched together in clumps or groups if they preferred. In terms of the scale, they were told that a value of 0 meant that they could not think of any photograph they had ever seen that they would prefer less. A value of 10 meant that they could not think of any photograph they had ever seen that they would prefer more. When each participant finished arranging the photographs, the experimenter registered the scale values by asking the participant to tell them what scale value, with one decimal point, she or he intended for each photograph.

16

Empirical Studies of the Arts 34(1)

For half the participants, they performed the preference task as described earlier. For the other half, a concurrent preload task was imposed during the preference task. The preload task was a letter memory task. Participants in the eight-letter condition were told that they must remember random series of letters at the same time they performed the preference task. The experimenter stressed the importance of accurately remembering the letters and said that they should focus their attention on the letters and remember them to the best of their ability. A randomizing program was used to produce sequences of eight-letter sets that were randomly cycled through participants. Participants in the eight-letter condition were given a set of letters to remember when they began the preference task. After five minutes elapsed, they were asked to recall the letter set. They were then assigned a new set of letters to remember for another 5-min period. This continued until the participant was finished with the preference task. Participants varied in terms of how long they took to complete the preference task and as a result how many letter sets they received. The time it took to perform the preference task was also recorded. For the second part of the experiment, all participants rated each photograph on a computer using 27 nine-point semantic differential items from Mullennix et al. (2013). The rating scale adjective items were presented on a computer using the E-Prime 2.0 software package (Schneider, Eschman, & Zuccolotto, 2002). For each photograph, the 27 rating scale items were presented in a randomized sequence, one at a time. Participants selected a point on the scale that corresponded to their impression of the photograph and responded by pressing one of the number keys from 1 to 9. A response of “1” corresponded to one endpoint item and a response of “9” corresponded to the other endpoint item (see Appendix 3 for a list of items and an example scale). Thirteen of the scales were reverse ordered. At the end of this rating task, the six questions concerning photography experience were presented on computer and responded to by indicating “yes” or “no.”).

Results and Discussion Memory Task Data Participants in the zero-letter conditions did not perform the concurrent memory task. In the eight-letter condition, participants correctly recalled 87.5% of the letters. Given that according to some metrics, the average shortterm memory span is seven items (Miller, 1956), this result indicated that participants were attending to the memory task and were devoting a significant amount of mental resources toward that end. The time it took to complete the preference or concurrent task was also recorded for participants in the eightletter conditions. There was no significant difference in time between the color

Mullennix et al.

17

and black and white conditions, t < .4, M ¼ 271.2 sec for color and M ¼ 262.0 sec for black and white.

Preference-Rating Data Main effects. A four-way repeated measures ANOVA for the between-subjects variable of preload (zero or eight letters), the within-subjects variable of color condition (color or black and white), photograph (32 photographs), and order (whether the color or black and white condition came first) was conducted on the preference-scaling data.1 The main effect of photograph was significant, F(31, 775) ¼ 20.64, p < .001, p2 ¼ .45. As expected, photographs varied significantly in terms of preference, with preference values ranging from a low of 1.51 to a high of 8.55. The main effect of color condition was also significant, F(1, 25) ¼ 9.92, p < .02, p2 ¼ .28. Color versions were preferred more than black and white versions. The main effect of preload was not significant (F < .2) and neither was the effect of order (F < .2). Interaction of color condition  Photograph. The interaction of color condition with photograph was significant, F(31, 775) ¼ 2.83, p < .001, p2 ¼ .10. The interaction was probed via a series of paired-comparison t-tests. The results showed that ratings for color and black and white versions of seven photographs were significantly different from each other (two-tailed), with color versions rated higher than black and white versions for all (photographs shown in Appendix 3). The preference ratings for these seven photographs across color conditions are shown in Table 1. Interaction of preload  Photograph. The interaction of preload with photograph was also significant, F(31, 775) ¼ 1.68, p < .02, p2 ¼ .06. The interaction was

Table 1. Mean Preference Ratings for Photographs That Differed Across Color Conditions in Experiment 1. Photograph #9 #11 #12 #16 #21 #25 #30

Black and white

Color

1.57 2.49 8.17 2.21 2.46 4.76 3.67

2.09 3.52 8.92 3.71 3.65 5.94 5.99

Note. All comparisons significantly different at p < .05, two-tailed.

18

Empirical Studies of the Arts 34(1)

probed via a series of independent-samples t-tests. The results showed that three photographs significantly differed (two-tailed) across preload, with one photograph rated higher under the eight-letter condition and two photographs rated lower under the eight-letter condition. Since only 3 of 32 photographs varied, and the direction of the interaction varied as well, this interaction was viewed as having little impact. Interaction of order  Photograph. The interaction of photograph with order was significant, F(31, 775) ¼ 1.47, p < .05, p2 ¼ .06. Probing of the interaction via independent-samples t-tests revealed that only 5 of 32 photographs differed based on whether the color or black and white was presented first. Thus, the interaction was viewed as ancillary to the overall findings. Nonsignificant interactions. All other interactions were not significant. Of particular interest was the absence of a two-way interaction of preload and color condition (F < 1.3) and the absence of a three-way interaction of preload with color condition and photograph (F < 1.2). In terms of our original predictions, these results suggest that automatic processing occurred for both color and black and white versions of the photographs. This is in line with the idea that the introduction of color as an added stimulus dimension does not result in the engagement of controlled processes.

Semantic Differential-Rating Data The differences in preference ratings across color and black and white conditions were explored further via analyses of the semantic differential ratings. A fourway repeated-measures ANOVA for preload group, color, photograph, and rating scale was conducted on the rating data. Main effects. The analysis showed no significant difference in overall ratings across preload groups (F < .1; M ¼ 5.02 for the zero-letter condition and M ¼ 4.99 for the eight-letter condition). One concern in assessing the effects of preload across color conditions was that different groups of participants were in the preload and no-load conditions. The absence of an overall difference in semantic differential ratings between preload groups cannot be viewed as a direct test of the confound issue with preference ratings. However, the absence of an overall difference across preload groups for semantic ratings is consistent with the idea that participants in each group treated both rating tasks in the experiment with similar approaches. There was no significant main effect of color on the ratings, (F < 1.5). Significant main effects of photograph F(6, 162) ¼ 47.58, p < .001, p2 ¼ .64 and rating item F(26, 702) ¼ 12.83, p < .001, p2 ¼ .32 were observed. These two main effects indicated that some photographs were ranked higher or

Mullennix et al.

19

lower than others and that ratings varied across the 27 scales, which was expected. Interactions. There were a number of significant interactions. These included the interaction of color with photograph, F(6, 162) ¼ 6.17, p < .001, p2 ¼ .19, photograph with rating, F(156, 4212) ¼ 8.54, p < .001, p2 ¼ .24, color with rating, F(26, 702) ¼ 12.88, p < .001, p2 ¼ .32, and the three-way interaction of photograph, color, and rating, F(156, 4212) ¼ 6.22, p < .001, p2 ¼ .19. There was also a four-way interaction of preload, photograph, color, and rating, F(156, 4212) ¼ 1.38, p < .01, p2 ¼ .05. Interaction of color  Photograph  Rating item. This interaction was germane to examining how semantic differential ratings informed us about the underlying basis of the effect of color on preference ratings. The three-way interaction was probed via a series of paired samples t-tests for only the seven photographs whose preference ratings differed across color conditions. The results of these tests are shown in Table 2 and are discussed later.

Automaticity of Processing For the preference ratings, the absence of a main effect of preload condition and the absence of an interaction of preload with color condition (as well as the absence of a three-way interaction between preload, color, and photograph) was consistent with an automatic mode processing explanation. There is little evidence that controlled processing was engaged when color was introduced as an additional stimulus dimension. The results extend the findings of Mullennix et al. (2013) to photographs that contain color.

The Effect of Color Despite the fact that stimuli were processed via an automatic cognitive mode, color still affected preference. This is important, as it shows that color affected a behavioral measure of aesthetic preference, yet color was processed automatically in this situation. Preference ratings for seven photographs varied across color conditions, with color versions preferred for all (Table 1). The entire set of photographs is shown in Appendix 3. Overall, these seven photographs appear fairly abstract except for one natural scenery photograph (Photograph #12). This casual observation is borne out by the ratings on the abstract or concrete semantic differential rating scale, with ratings for the color versions of these photographs (except Photograph #12) leaning toward the abstract rating endpoint (see Table 3). When examining the semantic differential rating data, a number of trends emerge for the six abstract photographs. In general, the black and white versions

20

Pleasant/unpleasant Balanced/unbalanced Tasteful/yasteless Comfortable/uncomfortable Appealing/repulsive Expressive/expressionless Full of feeling/without feeling Soulful/soulless Significant/insignificant Full of Life/lifeless Interesting/ininteresting Familiar/unfamiliar Common/rare Expected/unexpected Comprehensible/ uncomprehensible Simple/complex

– – – 6.52 – – – – – 71.4 – – – – – 5.45 –

6.28 –

B/W

– – – 5.76 – – – – – 6.07 – – – – –

Color

#9

5.93 –

6.00 –

– – – – – – – – – 5.93 – –

Color

#11

4.34 –

– – – – – – – – – 4.55 – – – 4.65 –

B/W

3.90 2.03

3.31 –

– – 1.83 1.86 1.52 2.21 – – – 2.76 – –

Color

#12

1.55 1.48

– – 2.93 2.45 2.12 3.31 – – – 1.69 – – – 1.86 –

B/W

– –

– – – – – 5.03 – 5.55 – 6.59 4.79 – 6.45 6.52 6.83

Color

#16



– – – – – 6.79 – 6.62 – 5.55 6.65 – 4.93 4.93 5.38

B/W

Photograph

– –

6.62 6.69

– – 5.07 – – 5.45 6.48 5.55 – – 5.59 –

Color

#21

– –

5.21 5.14

– – 6.62 – – 6.79 5.55 6.69 – – 6.69 –

B/W

– 5.31

5.72 –

– – – – 3.07 – – – 5.48 3.96 – –

Color

#25

– 2.69

– – – – 4.28 – – – 3.76 3.76 – – – 3.76 –

B/W

#30

– 3.72

5.03 – 4.93 – 4.69 4.86 3.28 – 4.24 – 4.93 – 5.17 4.10 –

B/W

(continued)

– 5.28

3.90 – 3.86 – 4.10 3.14 4.41 – 5.69 – 3.52 – 7.55 6.41 –

Color

Table 2. Semantic Differential Ratings for the Seven Photographs Only That Differed Across Color and Black and White (B/W) Conditions.

21

– – – – – – – 4.14

– 6.79

– 5.93

B/W

– – – – – – – 5.65

Color

#9

– –

– 4.28 7.03 – 4.83 6.34 3.34 –

Color

#11

– –

– 6.41 5.52 – 6.10 5.07 5.62 –

B/W

– –

– 8.17 5.59 2.52 4.65 3.31 6.97 7.69

Color

#12

– –

– 2.62 2.31 3.90 1.86 1.48 2.24 1.76

B/W

6.24 5.07

– 3.17 – – – – – –

Color

#16

4.10 6.34

7.38 – – – – – –

B/W

Photograph

Note. Numbers are only shown for the comparisons that were significantly different at p < .05, two-tailed.

Unified theme/ clashing themes Clear meaning/vague meaning Obscure/obvious Moving/still Eventful/uneventful Lively/quiet Active/passive Abstract/concrete Leave nothing to the imagination/provoke imagination Realistic/unrealistic Beautiful/ugly

Table 2. Continued

6.41 –

– 3.31 – – – 7.31 2.49 –

Color

#21

4.59 –

– 6.62 – – – 5.52 4.41 –

B/W

6.76 –

– 3.83 – 3.65 3.90 – 2.65 7.17

Color

#25

4.55 –

– 5.97 – 4.55 5.86 – 4.45 3.21

B/W

6.41 –

– 4.07 6.14 4.07 4.10 – 2.31 6.55

Color

#30

5.21 –

– 6.31 7.34 5.62 5.62 – 4.83 3.72

B/W

22

Empirical Studies of the Arts 34(1) Table 3. Mean Semantic Differential Ratings for the AbstractConcrete Scale (1.00–9.00 scale) for Color Versions of Photographs Differing Across Color Conditions in Experiment 1. Photograph Photograph Photograph Photograph Photograph Photograph Photograph

#9 #11 #12 #16 #21 #25 #30

2.66 3.34 6.96 2.76 2.59 2.65 2.31

were rated as more unexpected, obvious, full of life, concrete, expressionless, quiet, and realistic than the color versions. To a lesser degree, black and white versions were rated as more tasteless, repulsive, uninteresting, simple, unified, uneventful, and active. Photograph #12 bucked a few of these trends, with color versions rated as more obvious, lively, and concrete, which probably spoke to people’s familiarity with natural scenes and what to expect in terms of color. Overall, the difference due to color as illustrated through these analyses seems best described as “vibrancy,” with color versions viewed as more vibrant and black and white versions viewed as more lifeless. It is also important to consider that the original photographs were in color, and color was “subtracted out.” With photographs deliberately shot in black and white for artistic effect, the results could be different. Although the ratings for only a portion of the photographs (7 of 32) differed across color conditions, this was not entirely unexpected. Depending on the content of the information in the photograph (whether there was a landscape, man-made objects, abstracts, people and animals, etc.), people may find color and black and white versions of certain photographs equally pleasing.

Experiment 2 The results of Experiment 1 suggest that although color plays a role in making judgments of preference for artistic photographs, the cognitive processing of color is accomplished via automatic processes. This finding argues against the view that the processing of color requires controlled processes and cognitive resources. In Experiment 2, we decided to conduct a further test of this idea. Specifically, in Experiment 2, a third “false color” condition was included. “False,” “abnormal,” or “incongruent” colors for objects and scenes have been used to examine perception and memory for visual stimuli (Castelhano & Henderson, 2008; Oliva & Schyns, 2000; Tanka & Presnell, 1999; Wichmann et al., 2002; Zeki & Marini, 1998). In their study on

Mullennix et al.

23

activation of visual scene gist, Castelhano and Henderson (2008) found that abnormally colored scenes produced interference in terms of activating gist. Furthermore, they found that the more abnormal the hues were for a scene, the greater the interference. Oliva and Scnyns (2000) found that abnormally colored scenes interfered with scene memory more than colored scenes or black and white scenes. Wichmann et al. (2002) showed that falsely colored images were recognized worse than normally colored images. Together, these studies suggest that the processing of false colors is producing perceptualcognitive interference. If interference arises, it is possible that controlled processes become engaged in response to this unusual situation where colors do not match up with what the viewer expects in a given situation. Thus, we viewed it worthwhile to repeat the preference task portion of Experiment 1 with a false color condition to examine whether automaticity of processing is disrupted by false colors. To accomplish this, the original color photographs used in Experiment 1 were digitally processed to produce colors different than the natural colors present in the original photographs in order to produce the falsely colored items. A second rationale for Experiment 2 was to address possible concerns about the within-subject manipulation of color in Experiment 1, where photographs were viewed twice. In Experiment 2, color was manipulated as a betweensubjects variable, with participants in each condition only viewing photographs once. This arrangement addressed any possible contamination effects that could have occurred in Experiment 1 by viewing the same photographs twice, such as a participant remembering their rating for the previous version of a photographs and simply applying that from memory. Only the preference task was performed in Experiment 2. Participants did not rate the photographs using the semantic differential scales as they did in Experiment 1. The predictions for the color and black and white conditions in Experiment 2 were similar to Experiment 1. In addition, if ratings across preload conditions differed for the false color condition, this would suggest that controlled processing is engaged as a result of interference arising from the processing of falsely colored photographs.

Method Participants A total of 170 undergraduate students from the same subject pool as Experiment 1 participated in this study. Age and gender information were not recorded, but based on our experience with the same subject pool over years of studies, it was reasonable to assume that age and gender distributions were similar to Experiment 1; 87 participants were run in the preload conditions and 83 participants were run in the no-load conditions.

24

Empirical Studies of the Arts 34(1)

Materials The same color and black and white versions of photographs used in Experiment 1 were used again. In addition, false color versions were created via the procedure described by Wichmann et al. (2002). This procedure involved changing the original color photographs using the Photoshop software package. Specifically, red, green, blue, and yellow pixels in the color photographs were exchanged at a 180 degree rotation to produce false color versions equal in luminance to the original color versions. Thus, a total of 96 photographs were used in the experiment, with 32 photographs per color condition.

Design and Procedure The study was conducted in one session. For the preference task, a 2  3 mixed design was used, with a between-subjects variable of preload (zero or eight letters) and a between-subjects variable of color (color, black and white, or false color). All procedures used for conducting the preference task were similar to those used for Experiment 1, except that each participant only made preference judgments for one set of 32 photographs. Also, the time it took to complete the preference task was not recorded, since there was no effect of duration of the preference task observed for Experiment 1.

Results and Discussion Participants in the eight-letter condition correctly recalled 84.6% of the letters, comparable to Experiment 1. A three-way repeated measures ANOVA for the between-subjects variable of color condition (color, black and white, and false color), the between-subjects variable of preload (preload or no-load), and photograph (32 photographs) was conducted on the preference scaling data.2 The main effect of color was not significant (F < .8). Mean ratings across color conditions were M ¼ 5.17 (SD ¼ 0.73) for color, M ¼ 5.06 (SD ¼ 0.85) for black and white, and M ¼ 4.99 (SD ¼ 0.86) for false color. The main effect of preload was not significant (F < .2), with mean ratings for the preload and no-load conditions M ¼ 5.10 (SD ¼ 0.81) and M ¼ 5.05 (SD ¼ 0.83), respectively. The main effect of photograph was significant, F(31, 5084) ¼ 61.66, p < .001, p2 ¼ .27. As expected, photographs varied significantly in terms of preference, with preference values ranging from a low of 2.17 to a high of 8.84. The interaction of color condition with photograph was significant, F(62, 5084) ¼ 4.40, p < .001, p2 ¼ .05, as well as the interaction of photograph with preload, F(31, 5084) ¼ 1.79, p < .01, p2 ¼ .01. The interaction of color condition with preload was not significant (F < .6), and the three-way interaction of preload with color condition and photograph was not significant (F < .6).

Mullennix et al.

25

Interaction of Preload  Photograph A series of independent samples t-tests were conducted to assess which photographs varied as a function of cognitive load. The analyses showed that ratings for 5 of 32 photographs varied significantly across load conditions. Three photographs were rated higher under the preload condition, and two photographs were rated higher under no-load condition. Since the ratings across preload did not show a consistent pattern, we concluded this interaction was of minimal importance.

Interaction of Color Condition  Photograph To assess how color condition affected preference, the interaction between color and photograph was probed via a series of independent samples t-tests. For each of the 32 photographs, three sets of tests were conducted: one set for the black and white or color comparison, one for the color or false color comparison, and one for the black and white or false color comparison. The comparisons for photographs that differed across color conditions are shown in Table 4. Preference ratings for 12 photographs differed across color or black and white conditions (with color preferred for 7 of 12), ratings for 13 photographs differed across color or false color conditions (with color preferred for 8 of 13), and ratings for 5 photographs differed across black and white or false color conditions (with black and white preferred for 3 of 5).

Automaticity of Processing The results from Experiment 2 mirror those of Experiment 1 in terms of automaticity. Similar to Experiment 1, there was no main effect of preload and no significant interaction between preload and color condition, and the three-way interaction of preload, color condition, and photograph was also not significant. There was little evidence that cognitive load affected the false color condition. These results support the findings of Experiment 1, in terms of color information being processed automatically and without the need of cognitive resources. In addition, it does not appear that any interference that may have arisen from falsely colored photographs caused a measureable decrement in processing resources, with the absence of such a decrement being a classic hallmark of automatic processing.

The Effect of Color Similar to Experiment 1, although an automatic cognitive processing mode was used during the preference judgment task, the presence of color (or false color) affected preference ratings for photographs. For the black and white and false

26

Empirical Studies of the Arts 34(1)

Table 4. Mean Preference Ratings for Photographs That Differed Across Color Conditions in Experiment 2. Photograph

C vs.

BW

C vs.

FC

BW vs.

FC

#1 #3 #4 #7 #8 #10 #11 #13 #16 #19 #21 #23 #25 #27 #28 #29

4.00 6.98 3.76 4.09 5.34 4.89 4.40 – 5.17 – 4.57 3.81 6.64 – – 4.92

4.94 4.70 2.56 5.65 4.33 6.04 3.36 – 3.21 – 2.60 4.76 5.26 – – 6.44

4.00 6.90 3.76 – 5.34 – – 6.56 5.17 6.79 4.57 3.81 6.64 6.84 4.55 4.92

5.18 3.98 2.92 – 4.12 – – 7.40 2.43 5.68 3.15 4.95 5.57 5.60 6.53 6.33

– – – 5.65 – – 3.36 – 3.21 – – – – 6.73 5.22 –

– – – 4.78 – – 4.61 – 2.43 – – – – 5.60 6.53 –

Note. C ¼ color, FC ¼ false color, BW ¼ black and white. Numbers are only shown for the comparisons that were significantly different at p < .05, two-tailed. “–” indicates that the comparison was not significantly different.

color condition comparison, five photographs differed, with preference split 3/2; 12 photographs differed significantly across the color or black and white conditions, and 13 photographs differed across the color or false color conditions. Although we did not explicitly classify the stimuli a priori into categories such as abstract, man-made, natural scenes, and so forth, and upon inspection, there appear to be some trends related to these categories. For the color or black and white comparison, color versions, more or less, appeared to be preferred for abstract photographs. On the other hand, black and white versions appeared to be preferred for man-made objects (e.g., sign, shoe, car). For the color or false color comparison, there was also a tendency for color versions of abstract photographs to be preferred, while false color versions appeared to be preferred more for man-made objects. It should be noted that color diagnosticity (i.e., the degree to which a color is highly predictive of an object or scene) can be a factor in color processing (Nagai & Yokosawa, 2003; Oliva & Schyns, 2000; Price & Humphreys, 1989; Tanka & Presnell, 1999; Wichmann et al., 2002). When considering the absence of preload effects in the false color condition, if the false colors did not differ substantially

Mullennix et al.

27

from normal colors in diagnosticity, one might argue that the false colors would not produce interference. Certainly, some falsely colored photographs were just as plausible as color photographs. For example, for Photograph #13, buildings colored in bluish tones are just as plausible as the reddish tones in the original photograph. For Photograph #23, an automobile dashboard and figurine are just as plausible in greenish or blue as in red. However, there were also many photographs where the falsely colored versions obviously could not exist in reality (such as photographs #5, #12, and #18), with no differences across color conditions in preference observed for them. When comparing color to false color, we observed that most of the color versions preferred over false color versions were abstract photographs. We also observed that most of the false color versions preferred over color versions were man-made objects. Although difficult to assess, there does not appear to be substantial evidence that diagnosticity affected the results in the false color condition. However, in future research, it may be worthwhile to manipulate diagnosticity to examine how this possibly interacts with preference.

General Discussion Overall, both experiments showed that a concurrent task involving a substantial cognitive load did not affect the pattern of preference ratings provided for artistic photographs across various color conditions. If controlled processing were required to process color or false color versions of the photographs, interactions should have been observed between preload and color condition, or preload with color condition and photograph, with the pattern of preferences shifting as a result. However, those interaction effects were not significant and neither was the main effect of preload. The results are consistent with an automatic processing explanation of color processing during the preference judgment task. The results support the findings that Mullennix et al. (2013) observed for black and white stimuli, and extend them to stimuli possessing the added dimension of color. Obviously, the task of making an aesthetic judgment is quite different than tasks assessing the effects of color on perception and memory for visual objects and scenes (Biederman & Ju, 1988; Castelhano & Henderson, 2008; Cave et al., 1996; Gegenfurtner & Rieger, 2000; Oliva & Schyns, 2000; Ostergaard & Davidoff, 1985; Spence et al., 2006; Tanka & Presnell, 1999; Wichmann et al., 2002). Unlike many of the studies that found a perceptual or memory “cost” when color information was introduced, we found that although color affects preference, the processing of color information does not introduce a cognitive cost. One possible explanation is that attention is automatically drawn to color during an aesthetic judgment task in a manner similar to the way attention is automatically drawn to reading color words in the classic Stroop (1935) paradigm. Just as one cannot ignore reading color words, perhaps one cannot ignore

28

Empirical Studies of the Arts 34(1)

the color information in an artistic photograph. For aesthetic judgments, if color attracts attention, then an automatic processing mode that results in the viewer having little or no control over the allocation of attention may be involved. The present findings also suggest that the dual-mode processing framework is useful in terms of exploring various aspects of aesthetic experience. In Leder et al.’s (2004) model, there is a division between early processes that involve perception and implicit memory integration and later processes that involve explicit classification, cognitive mastering, and evaluation. The early processes are “bottom-up” and appear to proceed regardless of the viewer’s expertise or knowledge about art. The later processes are “top-down” and rely on stored knowledge and expertise accessed from long-term memory. In some sense, the dual-mode processing framework roughly overlaps with the early and late stages in Leder et al.’s model, with automatic processes engaged early on and controlled processes later on. However, this division is not clear-cut. For example, one could envision top-down processes that operate either in automatic or controlled modes, in terms of cognitive resource requirements. We believe that the dual-mode framework complements the stages outlined by Leder et al., and that it may be useful to incorporate the automatic or controlled distinction into a future expanded model of aesthetic appreciation. As well, the present findings support the dual-processing model of Graf and Landwehr (2015), in terms of their idea about a hierarchical arrangement of automatic and controlled processes involved in aesthetic judgments. Their description of an early, immediate, automatic, stimulus-driven process dovetails with our characterization of the processes used to produce aesthetic judgments in this study. Indeed, our experimental conditions primed the engagement of an early, automatic process as Graf and Landwehr would describe it. The preference task required a simple judgment of liking, which minimized the attentional requirements of evaluating the stimuli. The judgments were also quick, illustrated by the fact (as noted below) that participants spent little time examining the photographs. And finally, since our participants possessed little expertise in the area of artistic photographs, it is likely that they did not engage higher level controlled processes that would produce evaluations based on interest, boredom, or confusion. Overall, the present empirical findings fit with Graf and Landwehr’s aesthetic liking model, although obviously the results are only germane to the automatic process component of their dual-process framework. In terms of effects of expertise and experience, if we were to replicate the present study with trained visual artists and photographers, or with people who possessed an extensive knowledge of art and art history, we may find that controlled processes are engaged by them to process color and to make preference decisions. There is also evidence that reorienting the viewer in terms of their approach to an artwork shifts their processing mode. Cupchik, Vartanian, Crawley, and Mikulis (2009) conducted a functional magnetic resonance imaging study where they provided viewers with instructions that stressed early-level

Mullennix et al.

29

versus later-level processes. They found that brain areas that were activated differed based on the instruction set, with Cupchik et al. concluding that aesthetic-based instructions prompted orienting of top-down attention to the artworks. To us, this is consistent with the idea that a controlled processing mode was engaged. In future research, if we were to reorient participants’ attention specifically to color, it is possible that this could shift processing to a controlled mode. Focusing on the preference task in the present study, we demonstrated that “on the fly,” when naı¨ ve viewers of artistic photographs judged them for preference, they do it via an automatic cognitive processing mode which handles color information automatically and without a requirement for measureable cognitive resources. In terms of the ecological validity mentioned earlier, we view this as important. We believe our experimental situation preserved many elements of the real-world situation present when viewers with little expertise decide whether they like an aesthetic object. However, we note that in our experimental situation, the time frame for the aesthetic episodes is on the short side. On the basis of time recorded for the preference task in Experiment 1, we calculated that each photograph was observed for an average of 8.33s. This is a shorter time frame than observed in studies examining the time people spend in front of artworks at galleries (median times ranging roughly from 11s to 38.8s, as summarized by Leder & Nadal, 2014). However, people do not view visual art just in galleries. In fact, that is probably the exception rather than the rule in contemporary society, at least for artistic photographs. Online presentation sites, social media sites, and print media probably account for a greater number of viewings of visual artworks from the casual observer than an art gallery or art museum. Can aesthetic experiences be generated by brief viewings? Leder and Nadal (2014) suggest not It could even be argued that what makes an experience aesthetic is its long extension in time, which allows for several cycles of feedback and feedforward influence among processes related to perception, cognition and emotion. This does not just mean that aesthetic episodes can last longer, but that the nature of an aesthetic episode is, precisely, an extended time devoted to perception-cognition-emotion interactions. (p. 449)

In terms of our thoughts on the issue, we would argue that if a viewer views an artwork and makes an aesthetic judgment, then this constitutes an aesthetic experience regardless of the amount of time spent viewing it. The model of Graf and Landwehr (2015) would also suggest that an aesthetic judgment, albeit just based on pleasure, is produced in a brief time frame. No one would argue that an artistic photograph can garner an immediate affective response, such as Dorthea Lange’s “Destitute Pea Pickers in California: Mother of Seven Children.” At any rate, a brief aesthetic experience is probably just a different

30

Empirical Studies of the Arts 34(1)

type of aesthetic experience. Is it a lower quality experience? Possibly. That debate is beyond the scope of the present study. In summary, the present results confirm that viewers of artistic photographs use automatic cognitive processes to make preference judgments about visual art. In addition, color information is also processed automatically. Future studies that utilize the dual-mode processing theoretical framework could prove useful in advancing our knowledge about the cognitive basis of the aesthetic experience. Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The authors received no financial support for the research, authorship, and/or publication of this article.

Note 1. Although photograph and order were not manipulated as independent variables in the design, they were included in the analyses to examine possible interactions with the key variables of interest.

References Baddeley, A. D. (1986). Working memory. Oxford, England: Oxford Science Publications. Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 8, pp. 47–89). London, England: Academic Press. Ball, V. K. (1965). The aesthetics of color: A review of 50 years of experimentation. The Journal of Aesthetics and Art Criticism, 23(4), 441–452. Beauchamp, M. S., Haxby, J. V., Jennings, J. E., & DeYoe, E. A. (1999). An fMRI version of the Farnsworth-Munsell 100-hue test reveals multiple color-selective areas in human ventral occipitotemporal cortex. Cerebral Cortex, 9, 257–263. Biederman, I., & Ju, G. (1988). Surface versus edge-based determinants of visual recognition. Cognitive Psychology, 20, 38–64. Brama˜o, I., Faı´ sca, L., Forkstam, C., Reis, A., & Petersson, K. M. (2010). Cortical brain regions associated with color processing: An fMRI study. The Open Neuroimaging Journal, 4, 164–173. Castelhano, M. S., & Henderson, J. M. (2008). The influence of color on the perception of scene gist. Journal of Experimental Psychology: Human Perception and Performance, 34, 660–675. Cave, C. B., Bost, P. R., & Cobb, R. E. (1996). Effects of color and pattern on implicit and explicit picture memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22(3), 639–653. Chaiken, S., & Trope, Y. T. (1999). Dual-process theories in social psychology. New York, NY: Guilford Press.

Mullennix et al.

31

Chatterjee, A. (2010). Neuroaesthetics: A coming of age story. Journal of Cognitive Neuroscience, 23(10), 53–62. Chatterjee, A., & Vartanian, O. (2014). Neuroaesthetics. Trends in Cognitive Sciences, 18(7), 370–375. Cinzia, D. D., & Vittorio, G. (2009). Neuroaesthetics: A review. Current Opinion in Neurobiology, 19, 682–687. Cupchik, G. C., Vartanian, O., Crawley, A., & Mikulis, D. J. (2009). Viewing artworks: Contributions of cognitive control and perceptual facilitation to aesthetic experience. Brain and Cognition, 70, 84–91. Davidoff, J. B. (1991). Cognition through color. Cambridge, MA: MIT Press. Evans, J. St. B. T. (2007). Hypothetical thinking: Dual processes in reasoning and judgment. Hove, Engalnd: Psychology Press. Evans, J. St. B. T. (2008). Dual-processing accounts of reasoning, judgment, and social cognition. Annual Review of Psychology, 59, 255–278. Gegenfurtner, K. R., & Rieger, J. (2000). Sensory and cognitive contributions of color to the recognition of natural scenes. Current Biology, 10, 805–808. Graf, L. K. M., & Landwehr, J. R. (2015). A dual-process perspective on fluency-based aesthetics: The pleasure-interest model of aesthetic liking. Personality and Social Psychology Review, 19(4), 395–410. Hekkert, P., Snelders, D., & van Wieringen, P. C. W. (2003). Most advanced, yet acceptable: Typicality and novelty as joint predictors of aesthetic preference in industrial design. British Journal of Psychology, 94, 111–124. Hurlbert, A. C., & Ling, Y. (2007). Biological components of sex differences in color preference. Current Biology, 17(16), R623–R625. Kahneman, D. (2011). Thinking, fast and slow. New York, NY: Farrar, Straus, and Giroux. Konecˇni, V. J. (2012). Empirical psycho-aesthetics and her sisters: Substantive and methodological issues. Proceedings of European Society for Aesthetics, 4, 271–315. Kruglanski, A. W., & Orehek, E. (2007). Partitioning the domain of social inference: Dual-mode and systems models and their alternatives. Annual Review of Psychology, 58, 291–316. Leder, H. (2013). Next steps in neuroaesthetics: Which processes and processing stages to study? Psychology of Aesthetics, Creativity, and the Arts, 7(1), 27–37. Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. British Journal of Psychology, 95, 489–508. Leder, H., & Nadal, M. (2014). Ten years of a model of aesthetic appreciation and aesthetic judgments: The aesthetic episode – Developments and challenges in empirical aesthetics. British Journal of Psychology, 105, 443–464. Lopes, D. M. M. (1999). Pictorial color: Aesthetics and cognitive science. Philosophical Psychology, 12(4), 415–428. McKreefry, D. J., & Zeki, S. (1997). The position and topography of the human colour centre as revealed by functional magnetic resonance imaging. Brain, 120, 2229–2242. Miller, G. E. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81–97.

32

Empirical Studies of the Arts 34(1)

Mullennix, J. W., Foytik, L. R., Chan, C. H., Dragun, B. R., Maloney, M., & Polaski, L. (2013). Automaticity and the processing of artistic photographs. Empirical Studies of the Arts, 31(2), 145–171. Nadal, M., & Skov, M. (2013). Introduction to the special issue: Toward an interdisciplinary neuroaesthetics. Psychology of Aesthetics, Creativity, and the Arts, 7(1), 1–12. Nagai, J., & Yokosawa, K. (2003). What regulates the surface color effect in object recognition: Color diagnosticity or category? Technical Report on Attention and Cognition, 28, 1–4. Navon, D., & Gopher, D. (1979). On the economy of the human-processing system. Psychological Review, 86, 214–255. Oliva, A., & Schyns, P. G. (2000). Diagnostic colors mediate scene recognition. Cognitive Psychology, 41, 176–210. Ostergaard, A. L., & Davidoff, J. B. (1985). Some effects of color on naming and recognition of objects. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11(3), 579–587. Palmer, S. E., & Schloss, K. B. (2010). An ecological valence theory of human color preference. Proceedings of the National Academy of Sciences in the United States of America, 107(19), 8877–8882. Poldrack, R. A. (2006). Can cognitive processes be inferred from neuroimaging data? Trends in Cognitive Science, 10, 59–63. Polzella, D. J., Hammar, S. H., & Hinkle, C. W. (2005). The effect of color on viewers’ ratings of paintings. Empirical Studies of the Arts, 23(2), 153–163. Price, C. J., & Humphreys, G. W. (1989). The effects of surface detail on object categorization and naming. Quarterly Journal of Experimental Psychology A, 41, 797–828. Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-Prime user’s guide. Pittsburgh, PA: Psychology Software Tools, Inc. Schneider, W., & Shiffrin, R. M. (1977). Controlled and automatic human information processing: I. Detection, search, and attention. Psychological Review, 84(1), 1–66. Seeley, W. P. (2011). What is the cognitive neuroscience of art. . . and why should we care? American Society of Aesthetics Newsletter, Summer 2011. Retrieved from http:// www.aesthetics-online.org/articles/index.php?articles_id¼53 Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory. Psychological Review, 84(2), 127–190. Silvia, P. (2005). Cognitive appraisals and interest in visual art: Exploring an appraisal theory of aesthetic emotions. Empirical Studies of the Arts, 23, 119–133. Silvia, P. (2007). Knowledge-based assessment of expertise in the arts: Exploring aesthetic fluency. Psychology of Aesthetics, Creativity, and the Arts, 1, 247–249. Smith, E. R., & Collins, E. C. (2009). Dual-process models: A social psychological perspective. In J. B. T. Evans & K. Frankish (Eds), Two minds: Dual processes and beyond (pp. 197–216). Oxford, England: Oxford University Press. Spence, I., Wong, P., Rusan, M., & Rastegar, N. (2006). How color enhances visual memory for natural scenes. Psychological Science, 17(1), 1–6. Sridharan, D., Levitin, D. J., & Menon, V. (2008). A critical role for the right frontoinsular cortex in switching between central-executive and default-mode networks. Proceedings of the National Academy of Sciences of the United States of America, 105(34), 12569–12574.

Mullennix et al.

33

Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643–662. Suzuki, K., & Takahashi, R. (1997). Effectiveness of color in picture recognition memory. Japanese Psychological Research, 39, 25–32. Tanka, J. W., & Presnell, L. M. (1999). Color diagnosticity in object recognition. Perception & Psychophysics, 61(6), 1140–1153. Tanaka, J., Weiskopf, D., & Williams, P. (2001). The role of color in high-level vision. Trends in Cognitive Sciences, 5(5), 211–215. Taylor, C., Clifford, A., & Franklin, A. (2013). Color preferences are not universal. Journal of Experimental Psychology: General, 142(4), 1015–1027. Tresiman, A. M., & Gelade, G. (1980). A feature-integration theory of attention. Cognitive Psychology, 12, 97–136. Vartanian, O. (2014). Toward a cumulative psychological science of aesthetics, creativity, and the arts. Psychology of Aesthetics, Creativity, and the Arts, 8(1), 15–17. Whitfield, T. W. A. (1983). Predicting preference for familiar, everyday objects: An experimental confrontation between two theories of aesthetic behaviour. Journal of Environmental Behavior, 3, 221–237. Whitfield, T. W. A. (2000). Beyond prototypicality: Toward a categorical-motivation model of aesthetics. Empirical Studies of the Arts, 18(1), 1–11. Whitfield, T. W. A., & Slatter, P. E. (1979). The effects of categorization and prototypicality on aesthetic choice in a furniture selection task. British Journal of Psychology, 70, 5–75. Wichmann, F. A., Sharpe, L. T., & Gegenfurtner, K. R. (2002). The contributions of color to recognition memory for natural scenes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(3), 509–520. Wickens, C. D. (1991). Processing resources and attention. In D. L. Damos (Ed.), Multiple-task performance (pp. 3–34). London, England: Taylor & Francis. Zeki, S., & Marini, L. (1998). Three cortical stages of colour processing in the brain. Brain, 121, 1669–1685.

Author Biographies John W. Mullennix is a Professor of Psychology at the University of Pittsburgh at Johnstown, USA. He has published research in the areas of speech perception, speech technology, memory, and empirical aesthetics. Amy Varmecky is currently an undergraduate student in Psychology at the University of Pittsburgh at Johnstown. Chi H. Chan is a doctoral student at Carlow University earning his Psy.D. in Counseling Psychology. He graduated with a Bachelors of Science degree in Psychology from the University of Pittsburgh at Johnstown, and then earned his Masters of Science degree in Counseling Psychology at Chatham University.

34

Empirical Studies of the Arts 34(1)

Stephen Korenoski graduated with a Bachelors of Science degree in Psychology from the University of Pittsburgh at Johnstown. Zach Mickey graduated with a Bachelors of Science degree in Psychology from the University of Pittsburgh at Johnstown. Lisa Polaski-Hoffman graduated with a Bachelors of Science degree in Psychology from the University of Pittsburgh at Johnstown.