This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Brain and Cognition 70 (2009) 84 91 Contents lists available at ScienceDirect Brain and Cognition journal homepage: www.elsevier.com/locate/b&c Viewing artworks: Contributions of cognitive control and perceptual facilitation to aesthetic experience Gerald C. Cupchik a, *, Oshin Vartanian b, Adrian Crawley c, David J. Mikulis c a Department of Psychology, University of Toronto at Scarborough, 1265 Military Trail, Scarborough, Ont., Canada M1C 1A4 b Department of Psychology, York University, 4700 Keele Street, Toronto, Ont., Canada M3J 1P3 c Department of Medical Imaging, University of Toronto, Toronto Western Hospital, 399 Bathurst Street, Toronto, Ont., Canada M5T 2S8 article info abstract Article history: Accepted 11 January 2009 Available online 14 February 2009 Keywords: Aesthetics Art Attention Emotion Perception When we view visual images in everyday life, our perception is oriented toward object identification. In contrast, when viewing visual images as artworks, we also tend to experience subjective reactions to their stylistic and structural properties. This experiment sought to determine how cognitive control and perceptual facilitation contribute to aesthetic perception along with the experience of emotion. Using functional MRI, we show that aesthetic perception activated bilateral insula which we attribute to the experience of emotion. Moreover, while adopting the aesthetic orientation activated the left lateral prefrontal cortex, paintings that facilitated visuospatial exploration activated the left superior parietal lobule. The results suggest that aesthetic experience is a function of the interaction between top-down orienting of attention and bottom-up perceptual facilitation. Ó 2009 Elsevier Inc. All rights reserved. 1. Introduction Everyday perception is pragmatic, oriented toward the identification of objects in visual scenes. Our inclination for assuming this orientation is so strong and automatic that subjects report perceiving objects in pictorial compositions even when those compositions are devoid of recognizable objects (Ishai, Fairhall, & Pepperell, 2007; see also Treisman, 1964). However, when artists create paintings, they seek to provide the viewer with more than an opportunity to engage in object recognition per se. Rather, they attempt to evoke subjective reactions in the viewer to the stylistic and structural properties of their works. In support of this view, a recent literature review demonstrated that the neural systems underpinning aesthetic perception are not limited to those involved in object recognition (Nadal, Munar, Capó, Rosselló, & Cela-Conde, 2008). Our current challenge is to highlight processes that distinguish perception geared toward aesthetic experience from perception geared toward object identification in everyday perception. Aesthetic experience is currently viewed as a special, psychological process involving attention focused on the object and the suppression of everyday concerns (see Cupchik & Winston, 1996). Anthony, Earl of Shaftesbury (1711/1964), a British philosopher of the Enlightenment, contrasted a disinterested enjoyment of something for its own sake and interested enjoyment of anticipated benefits. His student, Frances Hutcheson (1725), proposed a * Corresponding author. E-mail address: cupchik@utsc.utoronto.ca (G.C. Cupchik). more fully developed psychological theory of the experience of beauty along with a sharper division between everyday and aesthetic perception. He argued that the experience of beauty involves a special taste faculty, a special criterion of beauty and the disinterested pleasure that results from it. Kant (1790/1914) also used the concept of disinterested satisfaction in which practical concerns played no role, but he rejected the notion of a separate faculty of taste that operated subsequent to ordinary cognitive faculties. For him, the same cognitive faculties used in the experience of ordinary objects would search for a satisfying structure in the object associated with beauty, thereby yielding pleasure. Schopenhauer (1818/1969) elaborated Kantian notions in such a way that a genuine theory of aesthetic attitude began to emerge. According to Schopenhauer, everyday consciousness is in the service of the will, but if the intellect can temporarily subdue the will, a state of disinterested detachment can be achieved in which aesthetic qualities can be suitably contemplated. Beardsley s (see 1958, 1966) theory represents the most fully developed modern notion of aesthetic experience as a distinct process. He follows in the tradition of disinterest, arguing that it involves affect that is slightly detached and a sense of freedom from concerns about past and future. But he did not posit a special mental function or faculty that produces disinterest. Rather, it emerges from existing cognitive functions and processes. Modern psychological aesthetics has incorporated the themes that developed through Shaftesbury, Hutcheson, Kant, Schopenhauer and Beardsley and formulated aesthetic experience as a special kind of experience in which pleasure is produced by the disinterested contemplation of objects. 0278-2626/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2009.01.003

G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 85 One implication of this formulation is that adopting an aesthetic viewing orientation may require an intentional shift to overcome the automatic cuing of semantic categories and to reinvest attention in sensory experience elicited by the stylistic properties of artworks (Cupchik, 1992; Cupchik & Winston, 1996). The Russian Formalists understood that everyday perceptual activity leads to automaticity and habituation. Shklovsky (1917/1988) expressed these ideas very clearly: The purpose of art is to impart the sensation of things as they are perceived and not as they are known... Art is a way of experiencing the artfulness of an object. The object is not important (p. 20). Because subjects untrained in the visual arts automatically apply the object-identification habit to viewing artworks (Cupchik & Gebotys, 1988; Cupchik, Winston, & Herz, 1992; Winston & Cupchik, 1992), this suggests that top-down control geared toward attenuating this habit may be necessary to engage in aesthetic perception. Much evidence links the lateral prefrontal cortex (PFC) to cognitive control (Burgess, Dumontheil, & Gilbert, 2007; Ridderinkhof, Ullsperger, Crone, & Niewenhuis, 2004; Sakai & Passingham, 2003; Simons, Schölvinck, Gilbert, Frith, & Burgess, 2006; van Veen & Carter, 2006). Prima facie, in subjects untrained in the visual arts, the lateral PFC would appear to be a critical structure for orienting the cognitive and perceptual apparatus toward aesthetic perception. A second implication of the formulation is that aesthetic perception requires the active integration of physical sensory properties that constitute style to construct unified image structures (Arnheim, 1971). Palmer (1999) has contrasted two processes of region segmentation whereby images are divided into mutually exclusive and therefore distinguishable areas. A boundary-based approach to region segmentation involves the application of edge-detection algorithms that focus on luminance edges to locate closed contours. In contrast, a Gestalt-based treatment of segmentation assumes a more global approach to grouping discrete elements based on common properties and mutual attraction. Precisely this distinction was made in the early 20th century by Wolfflin s (1950/1915) description of the psychological processes underlying the linear (e.g., hard-edge Neoclassical) versus painterly (e.g., soft-edge Baroque) dimension of artistic style. This contrast is between outline and surfaces and mere visual appearance (Wolfflin, 1950/1915, p. 14). In operational terms, this is equivalent to a contrast between edges (Hubel & Weisel, 1962) or luminance, line ends, tilt, curvature (Treisman, 1985) and elongated blobs (Julesz, 1981) or spatial frequencies (Graham, 1981). Multidimensional scaling studies of same-difference judgments involving pairs of paintings representing diverse styles have shown that the hard-edge and soft-edge distinction is fundamental and prior to representational-abstract discriminations (Berlyne & Ogilvie, 1974; Cupchik, 1974, 1976 1977). Whereas hard-edge paintings use contours to clearly define boundaries and isolate objects, soft-edge paintings have porous boundaries that engage viewers in an attempt to resolve forms. If aesthetic experience were a function of coherent image construction, then one would expect soft-edge paintings by virtue of containing ill-defined forms to facilitate aesthetic experience by stimulating active image construction. Given the role that the parietal lobes are known to play in the generation of abstract mental imagery (see Cavanna & Trimble, 2006; Goebel, Khorram-Sefat, Muckli, Hacker, & Singer, 1998), one would expect their engagement in aesthetic perception. This prediction is consistent with the results of a recent study in which activation in the precuneus was linked to the generation of mental imagery while subjects viewed paintings that lacked recognizable content, compared to a condition involving paintings that were rich with suggestive objects (Fairhall & Ishai, 2008). We contend that the two implications outlined above are not incompatible, but, rather, complementary (see Cupchik & Winston, 1996). The essential idea is that percepts develop from the global in the direction of the particular (Navon, 1977). This was proposed in the 1920s by researchers of the Aktualgenese (perceptual microgenesis) school who believed in the intrinsic structuredness of perception (see Flavell & Draguns, 1957) and reiterated by Neisser s (1967) notion that a pre-attentive stage of global processing is followed by a focal attentive stage during which the details of stimuli are identified. The physical/sensory information, derived from preattentive processing during the first glance, provide an expressive context within which more detailed features are perceived (Cupchik & Winston, 1996; see also Millis & Larson, 2008). This information includes first-order visual elements such as colour, tone, or texture, second-order properties related to collative properties (Berlyne, 1971) such as complexity and orderliness, and Gestalt properties including symmetry, good figure, harmony and contrast. After just a single glance at a painting (50 ms exposure), participants were able to discern the relative orderliness of two paintings and decide against a second viewing of High Arousal paintings with conflicting ordering principles (Cupchik & Berlyne, 1979). After longer exposure durations of 10+ s, participants disposed to absorption in art, music, literature and so on, favored multilayered images that evoked personal memories and thoughts about personal growth (Cupchik & Gignac, 2007). This complementarity is also embodied in Leder, Belke, Oeberst, and Augustin s (2004) recent information-processing model for the computation of aesthetic experience. Briefly, this model is comprised of five information-processing stages (perceptual analyses, implicit memory integration, explicit classification, cognitive mastering, evaluation) that are connected in sequence and through feedback loops. The final stage in this model (i.e., evaluation) involves appraising the meaning or interpretation placed on the artwork during cognitive mastering and generating two outputs: aesthetic judgment and aesthetic emotion. These evaluations constitute the endpoints of aesthetic experience. A critical feature of Leder et al. s model involves the context within which the object is viewed. Specifically, for the predictions of the model to hold, the person must engage the object as a work of art. While engagement is not sufficient for aesthetic evaluation, it is a necessary condition for aesthetic experience to occur. This is in line with our viewpoint that aesthetic experience necessitates an intentional orienting of perception toward distilling the properties of artworks. Moreover, the first stage in the model involves a stimulus-driven analysis of perceptual features, highlighting the role that bottom-up processes play in image construction. Thus, Leder et al. s (2004) model creates a framework within which topdown and bottom-up influences interact to generate aesthetic experiences. We relied on neural evidence to determine the relative contribution of intentional (top-down) and constructive (bottom-up) processes to aesthetic experience. Recently, Höfel and Jacobsen (2007) employed event-related brain potentials to tackle a similar question. In one condition they instructed subjects to simply view graphic patterns, whereas in another condition they instructed subjects to contemplate the beauty of those graphic patterns. The neural data revealed that while subjects engaged spontaneously in symmetry analysis in both conditions, evaluative categorization required contemplation and did not occur spontaneously. This demonstrates that when people contemplate the aesthetic properties (i.e., beauty) of visual images, a different set of neural processes is engaged than when they merely view the same visual images. In the present study, we instructed subjects to view paintings in the scanner for 10 s each under object-identification and aesthetic viewing orientations. We predicted that lateral PFC would be activated relatively more in the aesthetic viewing orientation, due to the increased need to exercise cognitive control to direct attention in that condition. In contrast, we predicted that structures known to mediate object recognition would be activated more under the

86 G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 object-identification viewing orientation, particularly the fusiform gyrus (see Fairhall & Ishai, 2008). We also predicted that soft-edge paintings would facilitate aesthetic perception, and that this process would be mediated by the parietal lobes. Finally, we were interested in examining the role of the experience of emotion in aesthetic perception. Given that artists seek to evoke subjective reactions to their works in the viewer, we predicted that aesthetic perception would be more likely to engage the brain s emotional circuitry, especially in structures related to the experience of emotion such as insula (Critchley, Wiens, Rotshtein, Ohman, & Dolan, 2004; Critchley et al., 2005; Damasio et al., 2000) and the orbitofrontal cortex (OFC) (Kringelbach & Rolls, 2004). 2. Method 2.1. fmri study 2.1.1. Subjects We recruited 16 right-handed subjects (8 males, 8 females) with normal vision and no formal training in visual arts. The institutional review board at the University of Toronto approved the experimental protocol, and all subjects gave written informed consent. 2.1.2. Materials Each stimulus was presented once in the course of the experiment. The experimental stimuli consisted of 32 representational paintings in social (8 group portrait and nudes) and non-social (8 still-life and landscape) motifs. To control for the activation of background knowledge, previous studies had employed abstract graphic geometric patterns as stimuli (Höfel & Jacobsen, 2007; Jacobsen, Schubotz, Höfel, & von Cramon, 2006). We chose to rely on representational paintings instead because they are better suited than abstract graphic geometric patterns for affective evocation as well as object recognition, which formed the primary foci of our study. The representational paintings were presented in four blocks, with eight paintings in each block (2 nude, group portrait, landscape, still-life). Two experimental blocks included paintings executed in a linear, hard-edge style emphasizing contour and composition, and two experimental blocks included paintings executed in a painterly, soft-edge, and expressive style (Berlyne & Ogilvie, 1974; Cupchik, 1974, 1976 1977). The higher-order ability to discriminate contour has been shown to underlie musical as well as spatial judgment (Cupchik, Phillips, & Hill, 2001), consistent with the central role played by the resolution of contour (i.e., outline) in discriminating fractal qualities in the style of gestural expressionist painters such as Jackson Pollock or members of the Quebec Automatiste school (Mureika, Cupchik, & Dyer, 2004; Mureika, Dyer, & Cupchik, 2005). The stimuli in the baseline blocks consisted of 16 non-representational paintings (8 fractal paintings by the American gestural expressionist artist Jackson Pollock, non-fractal paintings by the Israeli scholar Dr. Tsion Avital [Mureika et al., 2004; Mureika et al., 2005]). The non-representational paintings were also presented in four blocks, and each block consisted of four stimuli (2 Jackson Pollock, non-fractal paintings by Dr. Tsion Avital). While lacking representational content, non-representational paintings incorporate variations in colour, shape, and form. Therefore, comparing each experimental condition with the baseline condition served to isolate areas involved in viewing representational content under varying task instructions. 2.1.3. Procedure Prior to entering the scanner, subjects received thorough instructions on pragmatic and aesthetic viewing orientations. Specifically, for the pragmatic condition, they were instructed to apply an everyday informational criterion for viewing the paintings, and to approach the images in an objective and detached manner to obtain information about the content of the painting and visual narrative. In contrast, for the aesthetic condition, they were instructed to approach the paintings in a subjective and engaged manner, experiencing the mood of the work and the feelings it evokes, and to focus on its colours, tones, composition, and shapes. Note that neither condition required subjects to make any behavioural response in the scanner. Following instructions, subjects were given two practice trials under each condition to familiarize themselves with the process. The paintings used in the practice trials were not used in the experiment. Once inside the scanner, the experiment was conducted in eight blocks: Four blocks involving the representational paintings (i.e., experimental) and four blocks involving the non-representational paintings (i.e., baseline). Across pairs of subjects, the order of blocks was counterbalanced. The first block immediately following practice trials involved non-representational paintings. The four non-representational blocks were interspersed between the four representational blocks of paintings. The non-representational blocks served as baseline for the viewing of visual materials. The non-representational blocks were not accompanied by instructions to engage in any viewing orientation. Subjects were instead instructed not to engage in aesthetic or pragmatic perception while viewing the non-representational images. Images were presented using a rear projection screen inside the magnet. Viewing instructions were presented for 5 s. Each trial was initiated by a 1000 ms display of four + s at the centre of the viewing screen, and followed by a 200 ms blank screen. Each stimulus was presented for 10 s, followed by a 500 ms blank screen. Blank screens were inserted for backward masking. 2.1.4. fmri data acquisition and analysis All MRI data were acquired at a field strength of 1.5 T using a General Electric Signa scanner equipped with high-speed gradients. A high resolution 3D T1-weighted scan of each subject s brain anatomy was acquired with a 256 256 matrix, field-ofview = 20 cm, covering the brain with 60 axial slices at a slice thickness of 2.2 mm. Functional images were acquired using a spiral-out sequence (Glover, 1999) with a 64 64 matrix, field-ofview = 20 cm, covering the brain with 28 axial slices at a slice thickness of 4.5 mm. Each fmri scan consisted of 3 initial frames to allow the magnetization to reach equilibrium, followed by 286 frames of data acquired at a repetition time of 2.24 s (total scan time 10.5 min). Images were smoothed using a 7.6 mm FWHM Gaussian kernel. We collapsed across content levels (nude, group portrait, landscape, still-life) for all analyses. Therefore, the design included three factors: Orientation (pragmatic, aesthetic), painting style (representational, non-representational), and edge (hard, soft). A regressor was created to model the BOLD response to each representational block compared to all other conditions. Thus, a two-factor ANOVA analysis using all four regressors completely models the BOLD response to representational images (modulated by orientation and edge) relative to the baseline response to nonrepresentational images. The frames corresponding to viewing instructions were excluded from the analysis. Each regressor was shifted by 6 s to account for the hemodynamic delay. The fmri data were analyzed using AFNI (Cox, 1996). The twofactor ANOVA analysis was performed on each subject s data using the AFNI program 3dDeconvolve. Contrasts of interest included viewing orientation (pragmatic, aesthetic), painting style (representational, non-representational), and edge (hard, soft). After transforming each subject s results into Talairach space, the results of each contrast of interest were averaged over subjects and a final t-test statistic map was calculated (corresponding to a random-effects analysis with subject as a random factor). Clusters of voxels

G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 87 are reported as significant if the spatial extent of the activation exceeded a threshold of t > 2.1 and 4500 mm 3 volume, and the activation survived whole-brain family-wise correction for multiple comparisons (FWE). This corresponds to a map-wise one-sided p- value of.05 according to AFNI s AlphaSim Monte Carlo simulation. For our specific a priori hypotheses, we conducted the requisite region-of-interest (ROI) analyses. The ROI were selected based on earlier published reports, centred on peaks of maximum activation with a spherical radius of 10 mm. Specifically, we show the amplitude of the response during the relevant experimental conditions within lateral PFC ( 39, 57, 3 and 39, 54, 15; Simon et al., 2006), insula (30, 26, 12; Di Dio, Macaluso, & Rizzolatti, 2007), OFC ( 2, 36, 22; Kawabata & Zeki, 2004), superior parietal lobule ( 25, 54, 52; Goebel et al., 1998), and fusiform gyrus (38, 57, 19 and 35, 56, 21; Fairhall & Ishai, 2008). We did not use any threshold for the reported signal changes (i.e., we report raw signal changes compared to baseline). 2.1.5. Post-scan behavioural study Recall that while in the scanner, subjects were not instructed to make behavioural responses to the stimuli. The decision not to collect behavioural responses in the scanner was made to facilitate maximal immersion in the task. To ensure that subjects engaged in the task as instructed, we relied on thorough instructions and the completion of practice trials prior to data collection. Nevertheless, immediately following the completion of the scans, each participant completed a modified version of the task outside of the scanner. Specifically, each participant viewed the same representational stimuli under the aesthetic and pragmatic viewing orientations, presented in the same block structure as in the scanner. Stimulus presentation was self-paced. After viewing each stimulus subjects were asked Did the image evoke emotion in you? to which they responded using a 7-point scale ranging from 1 (not at all) to 7 (extremely). We also recorded reaction time (RT). We analyzed these data as part of our task analysis to determine whether emotion evocation is facilitated in relation to viewing soft-edge paintings in the aesthetic orientation, compared to viewing hard-edge paintings in the pragmatic orientation. 3. Results 3.1. Post-scan behavioural study First, we conducted an ANOVA with viewing orientation (pragmatic, aesthetic), edge (hard, soft), and image (nude, group portrait, landscape, still-life) as within-subjects variables, and RT to make emotion ratings (i.e., Did the image evoke emotion in you?) as the dependent variable. None of the main or interaction effects reached significance. Next, we repeated the above analysis but with the actual emotion ratings as the dependent variable. Confirming our prediction, the interaction between viewing orientation and edge was significant, F (1, 15) = 9.42, p <.01. Although neither simple main effect reached significance, in accordance with our prediction there was a trend for soft-edge images to receive higher emotion ratings in the context of the aesthetic than pragmatic viewing orientation. 3.2. fmri study 3.2.1. Viewing under aesthetic and pragmatic conditions We first conducted an overall contrast between all experimental (pragmatic and aesthetic) versus baseline trials. This analysis revealed activation in a distributed network including left insula ( 35, 17, 6; t = 4.33) (BA 13), left cingulate gyrus ( 1, 11, 42; t = 4.06) (BA 32/24), left lingual gyrus ( 37, 78, 2; t = 5.94) (BA 19), and right inferior temporal gyrus (31, 72, 7; t = 5.68) (BA 37/19) (Fig. 1). Next, we conducted a set of planned contrasts, geared toward testing specific a priori hypotheses within designated ROI. We discuss our structures of interest within those ROI, but for the sake of completeness also report activations elsewhere in the brain if they exceeded a threshold of t > 2.1 and 4500 mm 3 volume and survived whole-brain FWE correction. For each ROI we illustrate the amplitude of the response (i.e., percent signal change) during the relevant experimental conditions. 3.2.2. Viewing under aesthetic or pragmatic conditions We contrasted the aesthetic and pragmatic task conditions individually versus the baseline condition. For the aesthetic-baseline contrast we predicted activations in OFC and insula. Results revealed activation in right (39, 21, 6; t = 4.93) and left ( 37, 24, 9; t = 4.15) insula (BA 13) (Fig. 2). For the pragmatic-baseline contrast we predicted activation in the fusiform gyrus, and the results revealed activation in right fusiform gyrus (27, 52, 19; t = 3.80) (BA 37) (Fig. 3). In addition, significant activations were detected in the bilateral occipital gyri in the aesthetic (33, 76, 4; t = 6.01; 38, 80, 0; t = 3.48) (BA 18/19) and pragmatic (30, 77, 6; t = 4.10; 42, 75, 10; t = 6.64) (BA 19) conditions compared to baseline. 3.2.3. Engaging in aesthetic perception For the direct comparison of aesthetic and pragmatic conditions we predicted activation in lateral PFC, and the results revealed that activation in left lateral PFC ( 44, 37, 7; t = 3.50) (BA 10) was rel- Fig. 1. Categorical contrast between the experimental (aesthetic, pragmatic) and baseline conditions revealed significant activation in a distributed network including (a) left insula ( 35, 17, 6) (BA 13), left lingual gyrus ( 37, 78, 2) (BA 19), right inferior temporal gyrus (31, 72, 7) (BA 37/19), and (b) left cingulate gyrus ( 1, 11, 42) (BA 32/ 24). Parametric maps superimposed on to axial (a) and sagittal (b) MRI. Images are presented in radiological orientation. All AFNI coordinates converted into true Talairach coordinates. The colour bar indicates T-value.

88 G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 Fig. 3. The graph depicts the amplitude of the response in right fusiform gyrus (27, 52, 19) corresponding to the pragmatic and aesthetic conditions. The figure depicts the mean and standard errors of measurement (SEM). 4. Discussion Fig. 2. The graphs depict the amplitude of the response in (a) right (39, 21, 6) and (b) left ( 37, 24, 9) insula corresponding to the pragmatic and aesthetic conditions. The figures depict the mean and standard errors of measurement (SEM). atively higher in the aesthetic than pragmatic condition (Fig. 4). The reverse contrast did not reveal any significant area of activation. 3.2.4. Hard-edge versus soft-edge paintings For the direct contrast of soft versus hard-edge paintings we predicted activation in the superior parietal lobule, and the results revealed significant activation in left superior parietal lobule ( 20, 43, 47; t = 5.82) (BA 7). The reverse contrast did not reveal any significant area of activation. Importantly, to test whether softedge paintings facilitate aesthetic experience, we contrasted viewing soft-edge paintings under the aesthetic condition to viewing hard-edge paintings under the pragmatic condition. The results revealed that the left superior parietal lobule ( 34, 40, 57; t = 3.63) (BA 7) was activated relatively more when subjects viewed softedge paintings under the aesthetic condition than hard-edge paintings under the pragmatic condition (Fig. 5). The reverse contrast did not reveal any significant area of activation. The aesthetic and pragmatic conditions were accompanied by explicit instructions requiring interaction with paintings from a particular orientation. In contrast, the baseline condition was not accompanied by such instruction as subjects were asked to simply view the images. The contrast involving the pragmatic and aesthetic conditions versus the baseline condition revealed activation in a distributed network including left insula (BA 13), left cingulate gyrus (BA 32/24), left lingual gyrus (BA 19), and right inferior temporal gyrus (BA 37/19). We attribute activations in left insula and left cingulate gyrus in the experimental conditions to engagement in the task, as both structures have been shown to be activated across a wide set of paradigms whenever subjects engage in tasks relative to baseline conditions (Dosenbach et al., 2006). In addition, activations in right inferior temporal gyrus (BA 37/19) and left lingual gyrus (BA 19) can be attributed to mnemonic processing (Fletcher & Henson, 2001) while viewing representational paintings (Vartanian & Goel, 2004), respectively. To test whether the experience of emotion plays a larger role in the context of aesthetic perception, we contrasted the aesthetic and pragmatic task conditions individually versus the baseline condition. We expected to observe activations in OFC and/or insula in the aesthetic-baseline contrast. The results revealed activation in bilateral insula (BA 13) only (Fig. 2). We attribute this activation to the experience of emotion while viewing paintings, consistent with the role of the insula in the feeling of emotion (Critchley et al., 2004, 2005; Damasio et al., 2000). Interestingly, a recent fmri Fig. 4. (a) Categorical contrast between aesthetic and pragmatic conditions revealed significant activation in left lateral PFC ( 44, 37, 7) (BA 10). Parametric map superimposed on to axial MRI. Image is presented in radiological orientation. All AFNI coordinates converted into true Talairach coordinates. The colour bar indicates T-value. (b) The graph depicts the amplitude of the response in left lateral PFC corresponding to the pragmatic and aesthetic conditions. The figure depicts the mean and standard errors of measurement (SEM).

G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 89 Fig. 5. (a) Left superior parietal lobule ( 34, 40, 57) (BA 7) was activated relatively more when soft-edge paintings were viewed under the aesthetic task orientation. Parametric map superimposed on to coronal MRI. Image is presented in radiological orientation. All AFNI coordinates converted into true Talairach coordinates. The colour bar indicates T-value. (b) The graph depicts the amplitude of the response in the left superior parietal lobule corresponding to the four experimental conditions. The figure depicts the mean and standard errors of measurement (SEM). study by Di Dio et al. (2007) showed the right anterior insula to be activated when subjects viewed objectively beautiful sculptures, compared to a condition in which the sculptures were modified to be less proportional. The authors proposed that positive feelings elicited by viewing the canonical sculptures was mediated by a cortical network involving the anterior insula. We believe that the insula may play a similar role in our study, where feelings are evoked in response to viewing paintings, although given insula s involvement in the experience of negative and positive emotions we make no claim about the valence of the emotions experienced by our subjects (see Phan, Wager, Taylor, & Liberzon, 2002). In contrast, the pragmatic-baseline contrast revealed activation in right fusiform gyrus (BA 37) (Fig. 3). We attribute this activation to object recognition in representational paintings, including faces (Grill-Spector & Malach, 2004), and visual and spatial search for imposing situational models on pictorial scenes (Sieborger, Ferstl, & von Cramon, 2007). In addition, bilateral occipital gyri were activated more in the aesthetic (BA 18/19) and pragmatic (BA 19) conditions compared to baseline, which we attribute to greater load placed on the visual system when processing higher levels of visual detail in representational paintings. This study was conducted to address a specific question that we raised at the beginning of this paper: What are the processes that distinguish perception geared toward aesthetic experience from perception geared toward object identification? More specifically, does adopting the aesthetic viewing orientation require cognitive control, or is it the case that aesthetic perception is facilitated as a function of the perceptual features of the stimuli? The results suggest that aesthetic perception involves both processes. First, the direct comparison of aesthetic and pragmatic conditions resulted in activation in left lateral PFC (BA 10) (Fig. 4). We attribute this activation to top-down control in directing perception toward aesthetic orientation, consistent with the role of lateral PFC (BA 10) in top-down control of cognition (Ridderinkhof et al., 2004) and the maintenance of a main goal while performing concurrent sub-goals (Koechlin, Basso, Pietrini, Panzer, & Grafman, 1999). This interpretation is consistent with recent evidence linking lateral PFC to higher-order self-referential processing and the evaluation of internally generated information (Burgess et al., 2007; Christoff, Ream, Geddes, & Gabrieli, 2003; Northoff et al., 2006). According to this view, lateral PFC constitutes the interface between self-referential and cognitive demands of tasks, as would be expected under instructions to approach paintings from a subjective orientation. Interestingly, Jacobsen et al. (2006) reported activation in the frontomedian BA 9/10 when subjects judged the beauty of abstract graphic geometric patterns compared to a condition in which they judged the same patterns for symmetry. Although BA 10 has been linked to the introspective evaluation of internal mental states, evidence from research on prospective memory suggests that there may be a finer grained dissociation between the functions of the medial and lateral aspects of BA 10. Prospective memory tasks involve maintaining intention in memory for delayed execution. Whereas medial BA 10 may be involved in suppressing attention from external stimuli, lateral BA 10 may be involved in maintaining attention on internally generated cognitions (Burgess, Scott, & Frith, 2003; see also Christoff et al., 2003). Because the area activated in our study was lateral rather than medial, the activation may have been due to the maintenance and execution of internally generated goals to approach the artworks from an aesthetic orientation. Our results suggest that given the inherently self-referential nature of aesthetic perception, maintaining attention on internally generated cognitions may be an important component of this experience. We also addressed the extent to which aesthetic perception is preferentially engaged as a function of edge. This speaks to the bottom-up modulation of perceptual orientation as a function of stimulus features. The results revealed that left superior parietal lobule (BA 7) was activated relatively more when subjects viewed softedge paintings in the aesthetic orientation than hard-edge paintings in the pragmatic orientation (Fig. 5). Given that soft-edge paintings facilitate flexible visuospatial exploration of the stimuli (Berlyne & Ogilvie, 1974; Cupchik, 1974, 1976 1977), we attribute the engagement of the left superior parietal lobule (BA 7) to viewers attempts to resolve the indeterminate forms in soft-edge paintings to construct coherent images in the aesthetic condition, consistent with the role that the parietal lobes are known to play in spatial cognition and visual imagery (Fairhall & Ishai, 2008; Marshall & Fink, 2001). 5. Conclusion Our results highlight the role of bilateral insula in aesthetic perception and the role of right fusiform gyrus in pragmatic perception. We attribute the involvement of bilateral insula in aesthetic perception to subjective experience of emotion. In turn, we attribute the involvement of the fusiform gyrus in pragmatic perception to the identification of meaningful objects and the imposition of situational models on pictorial scenes. Furthermore, our results extend the functions of left lateral PFC in top-down control of internally oriented and self-referential goals to the domain of aesthetic perception, while highlighting the involvement of left superior parietal lobule in the process of active image construction. Supporting recent theoretical formulations (Cupchik & Winston,

90 G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 1996; Leder et al., 2004), our results suggest that aesthetic experience is a function of the interaction between top-down orienting of attention and bottom-up perceptual input. Acknowledgments We thank Morris Moscovitch for helpful comments on an earlier draft of this manuscript, and Tonya Hendriks for help with statistical analysis. Ken Seergobin wrote the program that orchestrated stimulus presentation in the magnet as well as data collection in the separate stimulus viewing part of the study. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bandc.2009.01.003. References Arnheim, R. (1954/1971). Art and visual perception. Berkeley, CA: University of California Press. Beardsley, M. C. (1958). Aesthetics: Problems in the philosophy of criticism. New York: Harcourt. Beardsley, M. C. (1966). Aesthetics from classical Greece to the present. New York: MacMillan. Berlyne, D. E. (1971). Aesthetics and psychobiology. New York: Appleton-Century- Crofts. Berlyne, D. E., & Ogilvie, J. C. (1974). Dimensions of perception of paintings. In D. E. Berlyne (Ed.), Studies in the new experimental aesthetics: Steps toward an objective psychology of aesthetic appreciation (pp. 181 226). Washington: Hemisphere. Burgess, P. W., Dumontheil, I., & Gilbert, S. J. (2007). The gateway hypothesis of rostral prefrontal cortex (area 10) function. Trends in Cognitive Sciences, 11, 290 298. Burgess, P. W., Scott, F. K., & Frith, A. D. (2003). The role of the rostral frontal cortex (area 10) in prospective memory: A lateral versus medial dissociation. Neuropsychologia, 41, 906 918. Cavanna, A. E., & Trimble, M. R. (2006). The precuneus: A review of its functional anatomy and behavioural correlates. Brain, 129, 564 583. Christoff, K., Ream, J. M., Geddes, L. P., & Gabrieli, J. D. (2003). Evaluating selfgenerated information: Anterior prefrontal contributions to human cognition. Behavioral Neuroscience, 117, 1161 1168. Cox, R. W. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research, 29, 162 173. Critchley, H. D., Rotshtein, P., Nagai, Y., O Doherty, J., Mathias, C. J., & Dolan, R. J. (2005). Activity in the human brain predicting differential heart rate responses to emotional facial expressions. NeuroImage, 24, 751 762. Critchley, H. D., Wiens, S., Rotshtein, P., Ohman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7, 189 195. Cupchik, G. C. (1974). An experimental investigation of perceptual and stylistic dimensions of paintings suggested by art history. In D. E. Berlyne (Ed.), Studies in the new experimental aesthetics: Steps toward an objective psychology of aesthetic appreciation (pp. 235 257). Washington: Hemisphere. Cupchik, G. C. (1976 1977). Perspective théorique et empirique sur la peinture impressionniste. Bulletin de Psychologie, 30, 720 729. Cupchik, G. C. (1992). From perception to production: A multilevel analysis of the aesthetic process. In G. C. Cupchik & J. Laszlo (Eds.), Emerging visions of the aesthetic process: Psychology, semiology, philosophy (pp. 83 99). New York: Cambridge University Press. Cupchik, G. C., & Berlyne, D. E. (1979). The perception of collative properties in visual stimuli. Scandinavian Journal of Psychology, 20, 93 104. Cupchik, G. C., & Gebotys, R. J. (1988). The search for meaning in art: Interpretive styles and judgments of quality. Visual Arts Research, 14, 38 50. Cupchik, G. C., & Gignac, A. (2007). Finding meaning and expressing emotion in response to artworks. Visual Arts Research, 33(1), 56 71. Cupchik, G. C., Phillips, K., & Hill, D. S. (2001). Shared processes in spatial rotation and musical permutation. Brain and Cognition, 46, 373 382. Cupchik, G. C., & Winston, A. S. (1996). Confluence and divergence in empirical aesthetics, philosophy, and mainstream psychology. In E. C. Carterette & M. P. Friedman (Eds.), Handbook of Perception & Cognition, Cognitive Ecology (pp. 62 85). San Diego, CA: Academic Press. Cupchik, G. C., Winston, A. S., & Herz, R. (1992). Judgments of similarity and difference between paintings. Visual Arts Research, 18, 37 50. Damasio, A. R., Grabowski, T. J., Bechara, A., Damasio, H., Ponto, L. L., Parvizi, J., et al. (2000). Subcortical and cortical brain activity during the feeling of selfgenerated emotions. Nature Neuroscience, 3, 1049 1056. Di Dio, C., Macaluso, E., & Rizzolatti, G. (2007). Brain response to classical and renaissance sculptures. PloS Biology, 11, e1201. Dosenbach, N. U., Visscher, K. M., Palmer, E. D., Miezin, F. M., Wenger, K. K., Kang, H. C., et al. (2006). A core system for the implementation of task sets. Neuron, 50, 799 812. Earl of Shaftesbury, A. (1711/1964). Characteristics of men, manners, opinions, and times (Vol. 1). Indianapolis, IN: Bobbs-Merrill. Fairhall, S. L., & Ishai, A. (2008). Neural correlates of object indeterminacy in art compositions. Consciousness and Cognition, 17, 923 932. Flavell, J., & Draguns, J. (1957). A microgenetic approach to perception and thought. Psychological Bulletin, 54, 197 217. Fletcher, P. C., & Henson, R. N. A. (2001). Frontal lobes and human memory: Insights from functional neuroimaging. Brain, 124, 849 881. Glover, G. H. (1999). Simple analytic spiral K-space algorithm. Magnetic Resonance in Medicine, 42, 412 415. Goebel, R., Khorram-Sefat, D., Muckli, L., Hacker, H., & Singer, W. (1998). The constructive nature of vision: Direct evidence from functional magnetic resonance imaging studies of apparent motion and motion imagery. European Journal of Neuroscience, 10, 1563 1573. Graham, N. (1981). Psychophysics of spatial-frequency channels. In M. Kubovy & J. R. Pomerantz (Eds.), Perceptual organization (pp. 1 25). Hillsdale, NJ: Erlbaum. Grill-Spector, K., & Malach, R. (2004). The human visual cortex. Annual Reviews of Neuroscience, 27, 649 677. Höfel, L., & Jacobsen, T. (2007). Electrophysiological indices of processing aesthetics: Spontaneous or intentional processes? International Journal of Psychophysiology, 65, 20 31. Hubel, D. H., & Weisel, T. N. (1962). Receptive fields, binocular interaction and functional architecture in the cat s visual cortex. Journal of Physiology, 160, 106 154. Hutcheson, F. (1725). An inquiry into the original of our ideas of beauty and virtue. London: D. Midwinter and Others. Ishai, A., Fairhall, S. L., & Pepperell, R. (2007). Perception, memory and aesthetics of indeterminate art. Brain Research Bulletin, 73, 319 324. Jacobsen, T., Schubotz, R. I., Höfel, L., & von Cramon, D. Y. (2006). Brain correlates of aesthetic judgment of beauty. NeuroImage, 29, 276 285. Julesz, B. (1981). Figure and ground perception in briefly presented isodipole textures. In M. Kubovy & J. R. Pomerantz (Eds.), Perceptual organization (pp. 27 54). Hillsdale, NJ: Erlbaum. Kant, I. (1914). Critique of judgment (J. H. Bernard, Trans.). London: MacMillan (original work published, 1790). Kawabata, H., & Zeki, S. (2004). Neural correlates of beauty. Journal of Neurophysiology, 91, 1699 1705. Koechlin, E., Basso, G., Pietrini, P., Panzer, S., & Grafman, J. (1999). The role of the anterior prefrontal cortex in human cognition. Nature, 339, 148 151. Kringelbach, M. L., & Rolls, E. T. (2004). The functional neuroanatomy of the human orbitofrontal cortex: Evidence from neuroimaging and neuropsychology. Progress in Neurobiology, 72, 341 372. Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgements. British Journal of Psychology, 95, 489 508. Marshall, J. C., & Fink, G. R. (2001). Spatial cognition: Where we were and where we are. NeuroImage, 14, S4 S7. Millis, K., & Larson, M. (2008). Applying the construction integration framework to aesthetic responses to representational artworks. Discourse Processes, 45, 263 287. Mureika, J. R., Cupchik, G. C., & Dyer, C. C. (2004). Multifractal fingerprints in the visual arts. Leonardo, 37, 53 56. Mureika, J. R., Dyer, C. C., & Cupchik, G. C. (2005). Multifractal structure in nonrepresentational art. Physical Review E, 72, 046101. Nadal, M., Munar, E., Capó, M. A., Rosselló, J., & Cela-Conde, C. J. (2008). Towards a framework for the study of the neural correlates of aesthetic preference. Spatial Vision, 21, 379 396. Navon, D. (1977). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9, 353 383. Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century-Crofts. Northoff, G., Heinzel, A., de Greck, M., Bermpohl, F., Dobrowolny, H., & Panksepp, J. (2006). Self-referential processing in our brain A meta-analysis of imaging studies on the self. NeuroImage, 31, 440 457. Palmer, S. E. (1999). Vision science. Photons to phenomenology. Cambridge, MA: MIT Press. Phan, K. L., Wager, T., Taylor, S. F., & Liberzon, I. (2002). Functional neuroanatomy of emotion: A meta-analysis of emotional activation studies in PET and fmri. NeuroImage, 16, 331 348. Ridderinkhof, K. R., Ullsperger, M., Crone, E. A., & Niewenhuis, S. (2004). The role of the medial frontal cortex in cognitive control. Science, 306, 443 447. Sakai, K., & Passingham, R. E. (2003). Prefrontal interactions reflect future task operations. Nature Neuroscience, 6, 75 81. Schopenhauer, A. (1969). The world as will and representation (Vol. 1) (E. J. Payne, Trans.). New York: Dover (original published, 1818). Shklovsky, V. (1988). Art as technique. In D. Lodge (Ed.), Modern criticism and theory (pp. 16 30). New York: Longman (original published, 1917). Sieborger, F. T., Ferstl, E. C., & von Cramon, D. Y. (2007). Making sense of nonsense: An fmri study of task induced inference processes during discourse comprehension. Brain Research, 1166, 77 91. Simons, J. S., Schölvinck, M. L., Gilbert, S. J., Frith, C. D., & Burgess, P. W. (2006). Differential components of prospective memory? Evidence from fmri. Neuropsychologia, 44, 1388 1397.

G.C. Cupchik et al. / Brain and Cognition 70 (2009) 84 91 91 Treisman, A. (1964). Monitoring and storage of irrelevant messages in selective attention. Journal of Verbal Learning and Verbal Behavior, 3, 449 459. Triesman, A. (1985). Preattentive processing in vision. Computer Vision, Graphics, and Image Processing, 31, 156 177. van Veen, V., & Carter, C. S. (2006). Conflict and cognitive control in the brain. Current Directions in Psychological Science, 15, 237 240. Vartanian, O., & Goel, V. (2004). Neuroanatomical correlates of aesthetic preference for paintings. NeuroReport, 15, 893 897. Winston, A. S., & Cupchik, G. C. (1992). The evaluation of high art and popular art by naive and experienced viewers. Visual Arts Research, 18, 1 14. Wolfflin, H. (1950). Principles of art history. New York: Dover (original published, 1915).