Aesthetic appreciation: Convergence from experimental aesthetics and physiology

Chapter 4 Aesthetic appreciation: Convergence from experimental aesthetics and physiology 4.1 Introduction strange behaviors? Among the puzzling things that humans do which at first sight have no apparent biological use are many of those activities we subsume under the concept of culture. For example, humans spend hours in theaters and concert halls and they also visit art exhibitions, which often require traveling and other costs. In more abstract terms, regarding art, people spend time in quiet spaces and look at human-made two- or three-dimensional structures that remain unchanged and deliver a stable flow of rather invariable sensory stimulation. In the present chapter we discuss the study of processes involved in such behaviors, specifically in art appreciation, and how researchers can measure psychological processes and states of aesthetic appreciation by using behavioral and psychophysiological methods. The object of our studies, the consumption of art by mankind, is quite old. Available evidence suggests that art is a unique human feature (Davies 2012). Examples of prehistoric art range back at least 80 000 years. Thus, along with weapons and tools, artworks are among the oldest human artifacts. Although we know little about how the early artworks were perceived, and we have no evidence for the particular kinds of receptor perspectives involved (Dutton 2009), artistry is deeply embedded in our evolutionary past. For many people today, at least in many Western cultures, going to a museum is a normal behavior, for some even a frequent behavior. Nevertheless, the concept of a museum as we know it today is a relatively new invention (Shiner 2003). The Capitolian museum that opened in 1471 in Rome can be regarded as the first modern museum. Independently of whether cave paintings are seen as ancestors of the modern concept of museums, or whether museums are classified as a recent development in history, the existence of art raises the question of what kind of behavior the perception of art is, and what kind of experiences it yields. Art s appeal might be due to various distinct possibilities. Psychology of the arts has focused on the quality of experiences that explain why this behavior is demonstrated (Leder et al. 2004). In this vein, artworks in exhibitions represent an unquestioned aesthetic value to visitors, and promise exciting, enlightening, or uplifting experiences. This relates to the most-often assumed explanation of aesthetic behavior: that art deploys hedonic states of positive experiences of beauty, pleasure, or the sublime. 04-Huston-Chap04.indd 57

58 However, there are various examples of contemporary art exhibitions for which such simple, straightforward effects can hardly be true. For example, consider Teresa Margolles contribution to the Mexican Pavillion at the Venice Biennale in 2009, named What else can we talk about?. In Margolles installation family members of homicide victims of the Mexican drugs war cleaned the floors of a Venetian palace with water that was taken from the burying grounds of the victims. This could not be said to offer the viewer a first-hand pleasurable experience. Thus, at least when it comes to contemporary art, alternative explanations of the appeal of art should be discussed: art can promise interesting, arousing, sometimes challenging experiences, by presenting new, innovative, and even provoking artworks, which do not explicitly aim to display the beautiful, the sublime, or even beauty as commonly understood. However, these artworks might receive their appeal from situations that require some resolution that is, perceivers are trying to understand and make sense of what is displayed by the artwork and intended by the artist, similar to the solution of a cognitive puzzle or the resolution of visual ambiguity. Such processes might be selfrewarding (Leder et al. 2004). This also relates to explanations that stress social functions of art. Artists adopt a certain view of the world; they often express a specific attitude. Consequently, early psychological approaches emphasized that the perceiver has to be able to empathize with the artwork or the artist (see Currie 2011, for a comprehensive discussion). Thus, art also allows for practising playfully in learning to perceive, interpret, and exchange viewpoints and opinions (Dissanayake 2007). Related to this explanation, evolutionary approaches argue that art making and appreciation serves an important ritualistic function that enhances social cohesion (Chatterjee 2011, p. 56). Psychological analysis of art is usually directed towards an inner perspective, such as the psychological states and processes during perception of art. In this chapter we aim to address two issues. First, we discuss how psychology understands the behavior of perceiving and enjoying art. Second, we deconstruct the experience of art into different components and discuss how psychological, experimental aesthetics, and particularly its use of psychophysiological measures all provide insight into the complex processes during aesthetic receptions of visual art. 4.2 How do psychologists understand aesthetic experiences of art? Psychology aims to describe, explain, and, through scientific understanding, predict human behavior and experience. The focus of most psychological theories is a kind of inner view of a person (James 1890). In this respect, psychology of aesthetics aims to determine the processes that take place in a perceiver, in order to explain what he or she likes aesthetically. Fechner (1876) in his Vorschule der Ästhetik expressed this in the following way: In considering the aesthetic effect of anything upon us, we do not only think about the concepts under which it can be subsumed, and its categorization within such a system, although such questions need to be asked. However, the most interesting and important question has to be: Why is something liked or disliked?... The answer to this question can only be found in the specific laws 04-Huston-Chap04.indd 58

Aesthetic appreciation 59 relating to liking and disliking, and in the laws of obligation, in just the same way that the particular movements of physical objects are explained by the laws of motion, and by the reasons and purposes of those movements. Movements cannot be explained alone by naming concepts or categorizing into different types of movement, but require precise laws of motion, whilst the purposes of the movements are also taken into account. Conceptual explanations in aesthetics will therefore be nothing but an empty framework, until they are confirmed by proper explanatory laws. Translation by Leder and Pleines, see Leder (2014, p. 117) Fechner founded the field of empirical aesthetics with his Vorschule der Ästhetik (1876). In this approach he distinguished bottom-up, stimulus-driven, from top-down, cognitivedriven, approaches and anticipated the interplay of these approaches during aesthetic receptions. This is much in accordance with what psychology today has identified as happening within a person when he or she experiences a work of art (Leder et al. 2004). Fechner s empirical studies were clearly concerned with aesthetic reception not production. He presented the first laboratory studies regarding the principle of the Golden Section in rectangles, according to which a ratio of 1 to 1.618 should be particularly beautiful, and conducted a first field study in a museum by comparing the appreciation of the two Madonnas painted by Holbein exhibited in the Semper Gallerie in Dresden. Interestingly, although Fechner made aesthetics one of his two main research topics (in addition to psychophysics), the progress of empirical aesthetics throughout the twentieth century was surprisingly slow. Mainstream psychology had a stronghold in perceptual studies, but other topics of clinical, social, and basic processes in perception, like attention and thinking, became more dominant themes. There were exceptions, such as Karl Bühler s (1913) approach to Gestalt psychology, which very explicitly starts with a reference to art history. However, only after the influence of behaviorism subsided did experimental aesthetics experience a revival, owing mainly to Berlyne s biological arousal approach in the 1970s. Since the cognitive revolution, the information processing concept has guided the study and understanding of human behavior for half a century. As a result, cognition, thinking, and internal representations are common to many psychological theories. The information-processing model developed by Leder and colleagues (2004), and Chatterjee s (2003) neuropsychological model are two examples of this. In both models, the aesthetic experience is defined as a sequence of and interaction between information processing stages. Figure 4.1 shows Vartanian and Nadal s (2007) integration of both models. Chatterjee (2003) distinguished aesthetic decisions and emotional responses as outputs of his model. Both are the result of processing stages of early and intermediate vision, a representational domain, and are modulated through attention. Leder and colleagues (2004) model aims to explain similar outputs, aesthetic judgments and aesthetic emotions. It proposed five information-processing stages that explain how an aesthetic experience develops from perception of an artwork to its interpretation and evaluation. Most of this model is concerned with processes that are not directly visible; they represent inner-states, and constitute what informally would be called thinking. According to the information-processing assumption, cognitive processes form representations and change psychological and physiological states in the person. For example, the combination 04-Huston-Chap04.indd 59

60 Attention Aesthetic Decision judgment Output Emotional response Aesthetic emotion Cognitive state Affective state Evaluation Stimulus Input Stimulus Early vision Intermediate vision Perceptual analysis Implicit memory Representational domain Explicit classification Cognitive mastering Continuous affective evaluation Figure 4.1 Combination of models of aesthetic experience. (Adapted from Vartanian and Nadal 2007.) AQ1 of text and image in Magritte s La trahison des images, an image of a pipe, and the text ceci nest pas une pipe ( this is not a pipe ), creates a perceptual representation of image and words that are instantly seen as contradictory. However, by means of interpretation the perceiver might resolve the seeming contradiction by realizing that the pipe is not a pipe, but an image of a pipe, and that images can represent an object, but images aren t objects. This way, due to cognitive processes, a state of resolved ambiguity replaces a state of ambiguity. In the model developed by Leder and colleagues (2004) these cognitive processes are also seen as eliciting changes in emotional states, mainly as a consequence or byproduct of the various cognitive activities (see also Leder 2013). Thus, an art experience is a complex episode, consisting of an interaction between an object and a perceiver, and the interplay of cognitive as well as emotional processes (see Figure 4.3). The focus on the recipient s individual states might be due to the proximity and relatedness of the research programs with the field of perception. From the beginning, perception was a focus for the new discipline of empirical psychology. The early works of Helmholtz or Wundt developed research approaches to human perception with high experimental vigor. If the goal of the methods subsumed under the heading psychophysics was to understand scientifically the relation between the external world and its mental representation, then perception was the natural place to start. Throughout the twentieth century, expertise and knowledge about perception increased alongside technological developments in image presentation and manipulation. With these developments the complexity of research programs in psychology increased by the consideration of more complex visual arrays, such as artworks. As a consequence, today s psychology demonstrates an increased interest in the arts. Moreover, the scope has also broadened through the integration of various methods, comprising experimental, cognitive-behavioral studies as well as psychophysiology and brain research. The combination of an outer psychophysics how object features affect inner states, and the relation between inner states and physiological changes, termed inner psychophysics by Fechner pretty much determines the current program of empirical aesthetics. 04-Huston-Chap04.indd 60

Aesthetic appreciation 61 4.2.1 Psychological methods Empirical psychology faces the problem that most of the cognitive and emotional processes it posits are not directly observable, and they take place in very fast succession. These successive processes are usually not even directly accessible to consciousness, this they are difficult to report. However, based on theoretical considerations a systematic variation of independent variables and measurement of the outcome by deploying one or more dependent variables allows us to infer these processes from observations and reporting. In experimental aesthetics, most commonly only behavioral measurements are used as dependent variables but due to technical advances in recent decades, these measurements are being complemented by psychophysiological methods. Regarding behavioral measurements, aesthetic evaluations are very often measured by preference decisions. These can be given using forced-choice preference tasks or Likert scales measuring liking or preference. The scale how much do you like this is frequently used, but must be seen as a somewhat vague and very general measure (Leder et al. 2005) of aesthetic evaluations. The questions about how much an object is preferred, how strongly it evokes an aesthetic experience, or how beautiful it is, can also be posed. Interestingly, different object classes differ due to the dimensions that are associated with their aesthetic evaluation (Augustin et al. 2012). This should be considered in experiments. Psychophysiology provides an interesting improvement to and extension of the behavioral approach as it allows testing hypotheses associated with physiological correlates. Physiological responses are closely linked to inner states. Some of the psychophysiological methods (e.g. EEG, electroencephalography; MEG, magnetencephalophraphy) allow the tracking of changes in processes in real time as they have very high temporal resolution. Additionally, psychophysiological measures of the peripheral nervous system function, such as electrodermal activity (EDA), heart rate (HR), facial electromyography (facial EMG), and pupillometry complement the experimental repertoire for psychologists and can contribute to a deeper understanding of psychological processes (see Figure 4.2). Regarding EDA, one commonly used measure is skin conductance, which records how well the skin conduces current flow in between a pair of skin electrodes. As sweat secretion increases, due to sympathetic activation, so does skin conductance. Both slow (tonic) changes in skin conductance level (SCL) and fast event-related (phasic) skin conductance responses (SCR) can be interpreted as physiological indexes of arousal. SCR increases with stimulus intensity, regardless of whether it has positive or negative valence (e.g. Bradley et al. 1990). In contrast to EDA, facial EMG can be used to measure responses associated with emotional valence because facial muscle activity is crucial to the expression of emotions (Ekman et al. 1980). Facial EMG can detect subtle changes in facial muscle activity related to emotional processing, even if no overt changes in facial expression are visible. The musculus zygomaticus major which raises the corner of the mouth and causes smiling, and the musculus corrugator supercilii which draws the eyebrows downwards while frowning, are reliable indicators of affective processes. Whereas affectively positive stimuli elicit an 04-Huston-Chap04.indd 61

62 EEG Cognition & emotion O 2 Brain activity Pupil size Cognition & emotion Facial EMG Emotion Eye-movements Heart rate Arousal Electrodermal activity Body posture Respiration... Figure 4.2 Illustration of the various methods that can be used to find correlates of the inner states of a person. increase in zygomaticus muscle activity, affectively negative stimuli increase corrugator muscle activity (Bradley et al. 2001; Dimberg and Karlsson 1997). These changes are usually in accordance with self-reports of subjectively felt emotions (Dimberg 1988). One should note that facial EMG is also sensitive to changes in cognitive processing (Scherer and Ellgring 2007). For example, a higher cognitive load leads to stronger musculus corrugator supercilii activation (Lishner et al. 2008). Pupillometry, the measurement of pupil dilation and contraction, is another psychophysiological measure that can be applied to understand processes in aesthetics. Pupil size is controlled by the iris sphincter and the dilator muscles. The former is innervated by the parasympathetic nervous system the latter by the sympathetic nervous system. Although pupil size variation mainly owes to changes in luminance (the pupillary light reflex), it is also associated with changes in arousal (Bradley et al. 2008) and cognitive load (e.g. Laeng et al. 2012). Highly arousing pictures elicit an increase in pupil dilation regardless of stimulus valence (Bradley et al. 2008). High cognitive loads while processing visual and auditory stimuli increases pupil diameter and decreases light reflex (Kahneman and Beatty 1966; Steinhauer et al. 2000). Pupillometry is usually measured with eye-tracking systems. Such systems also allow inferences about spatial (where?) and temporal (when and how long?) eye movement characteristics of information intake during visual exploration. 04-Huston-Chap04.indd 62

Aesthetic appreciation 63 During the last decade, brain activity underlying aesthetic experiences was examined using methods such as electroencephalography (EEG), magnetencephalophraphy (MEG), and functional magnetic resonance imaging (fmri). Electrical potentials of neurons produce electric and magnetic fields that can be measured from outside the skull. EEG electrodes attached to the scalp record the electrical activity; the highly sensitive magnetometers used in MEG measure the magnetic field. EEG can be used to measure different frequency bands of neural oscillations associated with specific psychological states. Alternatively, more specific psychological processes can be analysed with the event-related potential (ERP) technique, which averages EEG responses for a specific time-locked event. In comparison to other brain imaging techniques, EEG and MEG have the advantage of a high temporal resolution. Functional MRI, on the other hand, has a high spatial resolution. The demand for oxygen due to the increase in the activity of specific brain regions raises the regional cerebral blood flow. Functional MRI detects these changes in blood oxygenation level and thus provides an indirect measure of brain activity. These techniques are not restricted to local measures and thereby they enable the identification of networks of brain regions involved in aesthetic processing. Taken together, fmri is suited to testing assumptions regarding the location and involvement of brain processes, while EEG and MEG shed light on temporal characteristics within a fraction of a second. Facial EMG, measures of EDA, and pupillometry are especially useful to test overt and subtle affective responses. Studies using these psychophysiological methods have corroborated results from behavioral studies and have revealed new insights regarding the components and processes involved in art experience. Higher-order thinking Emotion Artwork Perception Content Style Knowledge, Extra info Figure 4.3 Components of an aesthetic experience. 04-Huston-Chap04.indd 63

64 4.3 Components of aesthetic experience of art behavioral and physiological approaches Psychological theories of aesthetic experience disentangle the experience into distinctive processing stages, as noted earlier. Hereafter, we use examples concerned with some of the components in order to illustrate how their role in aesthetic episodes and art appreciation can be addressed using behavioral and psychophysiological methods. Figure 4.3 illustrates an aesthetic episode and the components that we consider in the next paragraphs. We discuss how contexts enable aesthetic processing, we study early perceptual processes including the important stages of style or content processing, and how knowledge and expertise, as well as emotion, shape aesthetic experiences from art. Regarding these components we present findings about the processes involved, by putting the emphasis on scientific studies that combined behavioral and physiological methods in their investigations. 4.3.1 The role of context As shown in Figures 4.1 and 4.3, the aesthetic experience is often associated with a specific context. Such contexts can be established by pre-classifying an object as art. This is accomplished by displaying the object in a museum or by being asked to evaluate it aesthetically in the laboratory. An aesthetic context encourages the perceiver to adopt what Cupchik and Laszlo (1992) called an aesthetic attitude, and it provides a necessary but not sufficient pre-condition for aesthetic experiences (Leder et al. 2004). Aesthetic experiences are then viewed as a special state of mind qualitatively different from those summoned by everyday experiences (Markovic 2012). Thus, context is a defining feature of aesthetic experiences. Many famous artists played with such contextual frames by placing everyday objects in an aesthetic context see, for example, Marcel Duchamp s Fountain (1917, which is a urinal), 1 Andy Warhol s Brillo Boxes (1964), 2 or Beuys blocks of fat (1968) 3 displayed in art museums. Indeed, seeing a urinal as an exhibit in a museum might lead to a different experience to seeing a urinal when visiting the rest room, but how do contextual cues change our aesthetic experiences? Contexts exert influence on our experience via several different routes. Critically, aesthetic experiences usually take place within rather safe environments (Frijda and Schram 1995) which pose no risk to the health and well-being to the individual. In such a context, everyday concerns can be suppressed (Cupchik and Winston 1996), allowing the individual to focus attention on the object, and to engage in an evaluative processing style of thought (Leder et al. 2004). Artists and art curators seem to be well aware of this; O Doherty (1976) characterized galleries as white spaces where the outside world must not come in. Everyday stimulus processing, on the contrary, is pragmatic and goal oriented. In order to secure well-being, perception usually ensures successful identification of and appropriate reaction to various objects in the visual field. In aesthetic processing this immediacy of generating unambiguous representations and goal-directed actions is far less pronounced. Given that aesthetic contexts reduce pragmatic and motivational considerations, the perceiver is allowed to enjoy a rather ambiguous state when seeing an 04-Huston-Chap04.indd 64

Aesthetic appreciation 65 artwork (Jakesch and Leder 2009). Indeed, artists produce works of art not just for the sake of mere representation of an object; they also exploit stylistic and formal properties in their artworks to provide the viewer with an engaging viewing experience. Pablo Picasso s cubist paintings, Rene Magritte s often semantically ambiguous paintings, or in the case of abstract art, paintings devoid of any representational content, are all examples of artworks exploiting such stylistic and formal properties. In addition, an aesthetic context might be marked by certain expectations, promising aesthetically pleasing and cognitively challenging experiences when looking at art (Leder et al. 2004). It is assumed that an aesthetic context encourages an evaluative processing style (Cupchik and Winston 1996; Leder et al. 2004). In a series of experiments, Jacobsen and colleagues directly tested this assumption. Participants evaluated abstract patterns in either an aesthetic task ( Is this stimulus beautiful? ), or in a pragmatic or at least non- aesthetic task ( Is this stimulus symmetrical? ), while physiological brain responses were recorded employing EEG (Jacobsen and Höfel 2003) or fmri (Jacobsen et al. 2006). An aesthetic context led to a strong right hemisphere lateralization of an ERP component called the late positive potential, LPP. The authors interpret this lateralization as reflecting an evaluative style of processing in aesthetic contexts, because the LPP is increased for stimuli of positive and negative valence (Cuthbert et al. 2000) and is associated with enhanced attentional processing for motivationally salient stimuli (Hajcak et al. 2012). Jacobsen and colleagues (2006) used fmri to show that an aesthetic context elicited stronger activity in frontal networks including the ventral prefrontal cortex (BA45/47) and the frontomedian cortex (BA10/9). These brain regions are linked to evaluative judgments of persons and actions. Activation of a similar region of the BA10 was also reported by Cupchik and colleagues (2009) in a study where artworks had to be contemplated in either an aesthetic ( focus on colors, tones, composition, and feelings elicited by the artwork ) or pragmatic ( focus on what is depicted and follow the visual narrative of the artwork ) context. Activation of BA10 is consistent with its role in top-down control of cognition, shifting processing towards more evaluative processing. Specifically, these areas are linked to the introspective evaluation of internal mental states (i.e. one s own thoughts and feelings), and might indicate self-referential processes in aesthetic evaluations (Cupchik et al. 2009). In another study, Kirk, Skov, Hulme, Christensen, and Zeki (2009) showed that topdown classification of stimuli by defining them as genuine artworks, and hence framing an aesthetic context, or as computer-generated which changed evaluative stimulus processing. Participants gave higher aesthetic ratings for the same material presented as art as compared with non-art. Analyses of the brain states revealed that this was accompanied by specific changes in brain activations in the medial orbitofrontal cortex, which indicates hedonic value. Additionally, the prefrontal cortex was more strongly activated. The authors argue that this pattern of activation is consistent with subjects expectations about the hedonic value of aesthetic experiences leading to an evaluative processing of the artwork. These studies also provided evidence for the prediction of enhanced stylistic and formal processing in aesthetic evaluations. For example, Cupchik and colleagues (2009) reported that painting style influenced emotional ratings only in an aesthetic context. This 04-Huston-Chap04.indd 65

66 was corroborated by a specific neuronal activation pattern. Perceptual visual areas associated with analyses of these structures like the left superior parietal lobule (BA 7) were more strongly activated in the aesthetic compared to pragmatic context. Additionally, Kirk, Skove, Hulme, and colleagues (2009) reported increased activations in the primary visual cortex during aesthetic evaluations. Enhanced activation in these areas is attributable to a greater cognitive load placed on the visual systems through systematic analyses of stylistic aspects and formal (i.e. visual narrative) aspects of the paintings. Moreover, the enthorhinal cortex and the temporal pole (see also Jacobsen et al. 2006) were activated more strongly. Such activations are linked with declarative memory processes (Kroll et al. 1997; Nakamura and Kubota 1995) and, according to the authors, they reflect recollection of contextually appropriate art-related and cultural information such as recalling the specific style of the painting. Noguchi and Murota (2013) studied the combined effects of two factors, a contextual manipulation (defining artworks, Renaissance sculptures of human bodies, as genuine or fake ) and, unknown by the participants, a stylistic manipulation (visually altering the proportions of the sculptures resulting in likenesses of deformed bodies) while recording EEG. As in the studies by Kirk, Skov, Hulme, and colleagues (2009), participants provided higher aesthetic ratings for artworks defined as genuine compared to those defined as non-genuine. Additionally, participants showed marginally higher aesthetic ratings for the original compared to the altered versions. Both manipulations also produced significant differences in the ERPs. The P200 component showed a more positive going waveform for genuine as compared to non-genuine artworks, and for original as compared to altered artworks. This suggests that both kinds of manipulations were integrated and processed in the brain very rapidly (after between 200 ms and 300 ms). The authors also conducted source analyses of the EEG data, which permits estimation of the locations in the brain where this activity was generated. The parietal and medial prefrontal cortices were the sources of activation when a genuine versus fake context or when the original versus altered stimuli were compared. Additionally, for the stylistic comparison the occipital cortex was more strongly activated. These findings are consistent with the studies of Cupchik and colleagues (2009), Jacobsen and his team (2006), and Kirk and co-workers (2009). Together, these studies confirm that context plays an important role in aesthetic experiences. It changes processing to a more evaluative style comprising activation changes in a widespread network of lower- and higher-order cognitive and emotional areas (Nadal et al. 2008; Chatterjee 2003; Vessel et al. 2012). It is a long-standing issue in the arts as to whether changes in emotional processing reflect genuine emotional reactions or whether these are specific to aesthetic experiences (Frijda 1988; Scherer 2005). In the next section we will discuss how emotional experience is affected by aesthetic processing. 4.3.2 Where physiology plays the decisive role: aesthetic emotions? Philosophers and psychologists have discussed whether emotions triggered by the arts are different from genuine, everyday emotions (Beardsley 1958; Frijda 1988; Scherer 2005). Scherer sees a crucial difference between aesthetic and everyday emotions, which 04-Huston-Chap04.indd 66

Aesthetic appreciation 67 he calls utilitarian. Utilitarian emotions serve adaptive and homeostatic purposes to secure the well-being of an individual in significant events in life. Besides appraisals of novelty and intrinsic pleasantness, they comprise motivational appraisals like approach or withdrawal, action appraisals like fight or flight, and appraisals of coping potential, all designed to fulfil current and future needs. Aesthetic emotions, on the other hand, lack this kind of necessary immediacy of adapting to current events. They are instead produced by intrinsic appraisals of the aesthetic qualities of the object. Thus, motivational and action appraisals and the need to cope with the current situation should be less pronounced, or even absent, in aesthetic emotions (Scherer 2005). Similarly, Frijda (1988) assumes that the intensity of emotional reactions is reduced when a person appraises an event as not realistic or not directly relevant to ones current or future needs. An aesthetic context, signaling as it does asafe experience detached from reality (Beardsley 1958), provides the frame for such appraisals. For example, one rarely runs away from a cinema during a frightening horror movie, although fear-related responses will be triggered, including bodily changes in facial expressions, increased heartbeat, and reported fear. Alternatively, some authors (Goldstein 2009; Mar and Oatley 2008) argue that because an aesthetic context lacks real-world consequences, it allows the individual to indulge in emotional reactions which are similar to or even stronger than those experienced in everyday life. Experiencing such emotions allows for risk-free practice (Dissanayake 2007; Tooby and Cosmides 2001) for later life events, if and when similar circumstances summon them. Only a few empirical studies tested whether and how aesthetic and everyday emotions differ (especially in the domain of visual art). Several studies employed a contextual manipulation, which is very similar to the distinction between aesthetic versus realistic processing, by telling participants that stimuli were either fictional (aesthetic processing has somehow a fictional character; Tan 2000) or realistic. Goldstein (2009) asked participants to watch films evoking sadness or anger under these two contexts and rated the intensity of their emotional experience. Additionally, participants rated the intensity of anger and sadness in personal events in real life. Overall, intensity ratings did not differ much. Participants reported equal amounts of anger and sadness when watching a film either in a realistic or fictional context. In relation to personal life events, participants reported more intense anger reactions but not more intense sadness reactions. This means that, at the very least, consciously reported intensity of emotional reactions is very similar under fictional, realistic, and real, life events. However, Mocaiber and colleagues (2010, 2011) reported that a fictional context attenuates components of the emotional reactions to negative stimuli as indicated by psychophysiological and behavioral parameters. A fictional, as compared to a realistic, context led to reduced amplitude of the LPP (Mocaiber et al. 2010), which is indicative of reduced emotional processing. Also, specific changes in heart rate a lower defensive bradycardia reaction were observed (Mocaiber et al. 2011) and reaction times slowed down. These results imply that emotional reactions to negative stimuli are attenuated within a fictional context, as a stimulus might not be judged to be relevant for oneself (Frijda 1988). 04-Huston-Chap04.indd 67

68 Critically, these studies used a very specific contextual manipulation by framing a fictional context (which is related to an aesthetic context, but is not the same; see discussion in section 4.3.1) and they did not use visual artworks. Thus, in the study by Gerger, Leder, and Kremer (2014), participants evaluated positively and negatively valenced visual contemporary artworks and pictures from the International Affective Picture System (IAPS; Lang et al. 1999) under an aesthetic ( These stimuli are artworks ) or a realistic context ( These stimuli are press photographs depicting real scenes ). To track changes of subjects experiences, positive reactions were measured by levels of liking, of joy, and recording of the musculus zygomaticus major activity, and negative reactions by measuring anger, disgust, shame, sorrow, sadness, fear ratings, and musculus corrugator supercilii activity. Comparable to Goldstein (2009) they did not find any changes due to context for negative reactions, suggesting a very similar emotional experience for negative aspects. However, they found some changes for the positive reactions. Specifically, in the aesthetic compared to realistic context, negatively valenced artworks received higher liking and joy ratings, which was accompanied by a stronger reaction of the musculus zygomaticus major. For the more realistic looking IAPS pictures only the joy ratings were influenced. Emotionally negative pictures were perceived less negatively (in accordance with Mocaiber s results), but also emotionally positive IAPS pictures were perceived less positively. These results are consistent with the following conclusions. An aesthetic context enables a more positive perception of negatively valenced stimuli, especially when they are artworks. At the same time the context does not erase negative reactions to emotionally negative stimuli. Thus, positive and negative aspects of the emotional experience are differentially affected by an aesthetic context. This is consistent with emotional models which posit that positive and negative aspects of emotional experiences are processed somewhat independently in different subsystems (Cacioppo et al. 1997; Norris et al. 2010). The results also imply that it is important to consider both positive and negative aspects of the emotional experience in order to answer the question whether and how an aesthetic context changes our emotions. We have seen that an aesthetic context changes our experiences on a cognitive as well as emotional level. In the next section we take a closer look at the early stage of information processing, from behavioral and physiological studies. 4.3.3 Early processes: changes in perceptual areas As was shown in Figure 4.1, at the beginning of each visual aesthetic experience is a stage of perceptual processing. All psychological theories of art emphasize these kinds of processes. Perception, however, comprises a fairly complex set of stages and processes and involves the transformation of sensory information into visual experiences which to some extent elicit aesthetic pleasure. Chatterjee (2003) summarized this as early and intermediate visual processing. Several studies have shown that aesthetic processing results in enhanced activities in the early and intermediate areas of visual processing. Fairhall and Ishai (2008) compared the perception of representational, ambiguous, and abstract artworks with scrambled control 04-Huston-Chap04.indd 68

Aesthetic appreciation 69 stimuli. The former showed stronger activation of the inferior occipital gyrus and gyrus fusiformis, where object form is processed, and the dorsal occipital cortex, which processes spatial properties of visual objects. Additionally, Cupchik and colleagues (2009) found indications for stronger visuo-spatial analyses in an aesthetic context compared to pragmatic context, and Kirk and his team (2009) reported stronger activations in early visual areas (primary visual cortex) under an aesthetic context (see section 4.3.1). Additionally, increased activations in these visual areas correlate with subjective aesthetic evaluations. Vartanian and Goel (2004), employing abstract and representational paintings, showed that activation in the occipital gyrus, including the fusiform gyrus, linearly increased as a function of subjective preference ratings. Similarly, Vessel and colleagues (2012) used abstract and representational paintings as well, and reported linear increases of activity in visual areas in the occipito-temporal cortex as a function of subjectively reported aesthetic value. However, activation patterns in these early and intermediate visual processing areas often show little overlap in different studies (e.g. Nadal et al. 2008). They depend critically on task demands, instructions, procedures, and stimulus materials used. For example, aesthetic processing of representational as compared to abstract artworks engages networks of object perception more strongly (Fairhall and Ishai 2008; Vartanian and Goel 2004). Overall, empirical studies provided some evidence that early and intermediate visual processing areas are consistently more strongly activated during aesthetic processing. This stronger activation might reflect a top-down enhancement due to attentional and/or emotional processes (Vuilleumier and Driver 2007). Further research is required in order to clarify the exact nature of these processes further research. The stage of perceptual analyses (Figure 4.1) considers the effects of a number of variables that influence a given image s aesthetic appeal. Among these variables are complexity, contrast, symmetry, order, and grouping. The combination of psychophysiological and behavioral methods can be fruitful in clarifying the contribution of such perceptual factors. For example, according to Berlyne (1971), visual complexity affects aesthetic responses through changes in arousal. However, attempts to test this prediction by using abstract patterns and psychophysiological measures of EDA failed to detect any effect of visual complexity on physiological arousal (Berlyne and Lawrence 1964). Krupinski and Locher (1988) manipulated complexity with different levels of symmetry. Their participants rated asymmetrical, single-, and double-axes symmetrical patterns according to their hedonic value while SCR was recorded. In contrast to Berlyne and colleagues (1963), they found a linear increase in SCR and hedonic ratings with an increase in complexity. These diverging results regarding the relationship between visual complexity and physiological arousal during the aesthetic evaluation of abstract patterns resemble results concerning the effects of visual complexity on art appreciation. Nadal and colleagues (2010) sought to resolve the ambiguity concerning the results of complexity in art appreciation ambiguous, because some studies found that simplicity was preferred. Nadal and colleagues (2010) combined various methods. They employed a large set of abstract and representational artworks as well as non-artistic images which 04-Huston-Chap04.indd 69

70 were selected to represent three levels of complexity, high, medium, or low. When participants rated these stimuli for perceived beauty, there was no clear pattern of influences of complexity on aesthetics. Interestingly, Nadal and colleagues (2010) classified their stimuli according to several dimensions that in the literature had been discussed as determinants of complexity, such as the number and variety of elements, an image s three- dimensionality, the general appeal of symmetry, etc. When they analysed the effects according to these dimensions, none of them had a general effect on beauty. The authors concluded that the perceived complexity might have more or less individual causes in different dimensions and that these variations might conceal clear effects of complexity on appreciation of artworks. However, the nature of the processes that determine how complexity affects the perception of artworks still remains an interesting topic for future studies. In Leder and colleagues (2004) approach, the specificity of art is not so much related to formal features such as complexity, but rather it is assigned to the distinction between context and style. 4.3.3.1 Representations: content and style? art specific? Kuchinke, Trapp, Jacobs, and Leder (2009) showed how physiology can inform the processes concerned with explicit content recognition. They tested the hypothesis that at the moment of insight when content of an artwork is recognized, changes in emotional responses can be observed. They looked at Cubist paintings, whose content takes the viewer time to recognize (Hekkert and vanwieringen 1996). Pablo Picasso and Georges Braque developed a style which depicts the content in simultaneously overlaid or combined perspective views. This style slows recognition because the constituting elements cannot easily be assigned to one view. In some cases, recognition of the depicted object exceeds several seconds. In everyday life, however, object recognition is usually very fast and takes only fractions of a second. Kuchinke and colleagues (2009) measured pupillary responses and showed that the moment of content recognition was indeed accompanied by an emotional response (Figure 4.1). Thus, at exactly the same time as an artwork s subject was identified, participants showed a pupillary response that could be interpreted as an elicited emotion of pleasurable insight. Style is considered an important feature in the processing of art (Leder et al. 2004). Regarding the processes underlying the recognition of style and content Augustin, Leder, Hutzler, and Carbon (2008) studied the time it takes for style and content to affect the perception of artworks by presenting images for very short times. Two artworks sometimes the same, sometimes different, in content, style, by the same or a different artist had to be evaluated according to their perceived similarity. With a presentation time of 10 ms, viewers could only distinguish matters of content. However, perceived similarity also was affected by painting style with 50 ms. Thus, temporally, style follows content. In a later study, Augustin, Defranceschi, Fuchs, Carbon, and Hutzler (2011) employed ERP to study the neural time-course underlying the perception of content or style. Participants viewed pictures that systematically varied in style and content and had to perform a go/no-go dual-choice task. Results again supported the hypotheses that in the processing of art, style follows content. Style-related information 04-Huston-Chap04.indd 70

Aesthetic appreciation 71 was available later than content-related information as indicated by certain ERP components (lateralized readiness potential LRP, and N200). In the future, physiological measures will serve to improve our understanding of the aesthetic experiences as they are experienced over time, and as we become able to measure short-term effects and dynamic changes continuously. 4.3.3.2 Art expertise and knowledge Experts are experts because they know more about a given subject than most individuals, and their knowledge is deeper. Knowledge about an artist s work and life, specific art styles and movements, and the process of artistic production aids stylistic and conceptual interpretation (Leder et al. 2004). The increased ability to interpret art, in turn, facilitates understanding and art appreciation. Psychological research on art expertise mainly used two approaches: (i) studying the effect of art-specific information on laypeople s art experience, or (ii) contrasting laypeople and art experts (people with formal training in the arts or art professionals). Both approaches revealed how knowledge and art expertise affects processing on a cognitive, perceptual, and affective level. Art-specific knowledge provided by additional information shapes processing on a higher-order cognitive level such as finding meaning, resolving ambiguity, and developing understanding. Due to their lack of pre-existing knowledge, laypeople are particularly suitable for projects which systematically manipulated art-specific knowledge. Several behavioral studies using this approach showed that additional information positively affects art appreciation and understanding (Leder et al. 2006; Millis 2001; Russell 2003; Swami 2013). However, this depends on relative expertise level and art style. Especially with regard to abstract art, additional information helps laypeople adopt style-focused processing, opens new perspectives for interpretation, and therefore increases their understanding of what they are viewing. Swami (2013) emphasized that increased understanding mediates the positive effect of additional information on art appreciation. However, the impact of additional information on art appreciation also relies on the pre-existing (relative) level of expertise. For example, Belke, Leder, and Augustin (2006) presented abstract artworks with or without stylistic information to participants differing in their relative level of art expertise. Participants with low art expertise showed an increased liking for abstract art when style-related information was given; participants with high expertise, in contrast, showed a decrease in liking. This negative effect of information on the appreciation of art was presumably due to its interference with pre-existing knowledge of the experts. Only a few studies examined the effect of additional information on art appreciation by employing psychophysiological measures. Lengger, Fischmeister, Leder, and Bauer (2007) studied the effect of stylistic information on the evaluation of abstract and representational art. Understanding increased when stylistic information was provided. However, on a neuronal level, artworks presented with stylistic information compared to artworks without stylistic information evoked a weaker activation mainly in left frontal and parietal networks. This could indicate that art-specific information, by increasing knowledge and understanding, helps to process artworks more efficiently on a neuronal level (Haier et al. 04-Huston-Chap04.indd 71

72 1988; Lengger et al. 2007). Taken together, these studies show that specific information and knowledge significantly affect understanding of art and its appreciation. Choosing another approach to investigate the effect of knowledge on evaluation of art, Pang, Nadal, Müller-Paul, Rosenberg, and Klein (2013) compared the neuronal response of laypeople and art experts while evaluating art and a set of control stimuli. Consistent with studies by Lengger and co-workers (2007), Pang and colleagues found that knowledge leads to more efficient processing on a neuronal level ERP amplitudes of the P3b and LPP, which are linked to knowledge retrieval, were reduced for art experts compared to laypeople. Interestingly, in the study by Pang and colleagues, these expertise effects were not limited to artworks; they also extended to control stimuli (filtered artworks and simple color-field stimuli). This suggests that art experts in general have a different style of visual processing. This is partly supported by a study by Long, Peng, Chen, Jin, and Yao (2011). They showed that early perceptual processes like color processing differs between art experts (artists) and laypeople on a neuronal level. First, MRI scans revealed structural differences between artists and laypeople indicated by a higher density of gray matter in brain areas involved in color processing (V4) for artists. Secondly, during a color-naming task, artists showed higher activity in V4 than laypeople. This contradicts the neural-efficiency hypothesis, in which we would expect a decreased activity for artists. These diverging results imply that the relation between expertise and efficient neural processing may be limited to specific tasks, brain regions, or subtypes of art expertise (but see Neubauer and Fink 2009). Nevertheless, a concurrent finding is that the level of art expertise affects visual processing on a neural level. On a behavioral level, experts and laypeople differ in how they visually explore artworks. Cupchik and Laszlo (1992) assume that art experts have learned to focus on stylistic and conceptual features whereas laypeople adopt an object-oriented style of processing typical for everyday perception. Several eye-tracking studies support this hypothesis. Laypeople directed their gaze more often to pictorial content, but art experts attended more to structural and compositional features (Nodine et al. 2009; Vogt and Magnussen 2007). Art expertise also influences affective components of aesthetic experiences. According to Leder and colleagues (2004), understanding and successful interpretation, both core factors of art expertise, are self-rewarding and affectively positive. In line with this, Kirk, Skov, Christensen, and Nygaard (2009) showed that architecture experts exhibited higher activity in brain areas associated with reward processing (medial orbitofrontal cortex and subcallosal cingulate gyrus) compared to non-experts when they evaluated pictures of buildings. If a high level of art expertise generally increases positive affective experiences of artworks, then art experts compared to laypeople should also feel more positive even when they look at negatively valenced art. Leder, Gerger, Brieber, and Schwarz (2014) addressed this issue by measuring emotional responses when art experts and laypeople looked at artworks and non-artworks (IAPS pictures) with affectively positive and negative content. In general, positive stimuli led to a higher zygomaticus activation and negative stimuli to a higher corrugator activation, and positive stimuli were liked more than negative ones, 04-Huston-Chap04.indd 72